Skip to main content

Full text of "USPTO Patents Application 09848616"

See other formats


MOLECULAR ANTIGEN ARRAY 



CROSS-REFERENCE TO RELATED APPLICATIONS 

[0001] This application claims priority benefit of U.S. provisional application no. 

60/202,341, filed May 5, 2000, which is incorporated by reference in its entirety 

BACKGROUND OF THE INVENTION 

Field of the Invention 

[0002] The present invention is related to the fields of molecular biology, 

virology, immunology and medicine. The invention provides a composition 
comprising an ordered and repetitive antigen or antigenic determinant array. The 
invention also provides a process for producing an antigen or antigenic 
determinant in an ordered and repetitive array The ordered and repetitive antigen 
or antigenic determinant is useful in the production of vaccines for the treatment 
of infectious diseases, the treatment of allergies and as a pharmaccine to prevent 
or cure cancer and to generate defined self- specific antibodies and specific immune 
responses of the Th2 type. 

Background Art 

[0003] Vaccine development for the prevention of infectious disease has had the 

greatest impact on human health of any medical invention. It is estimated that 
three million deaths per year are prevented worldwide by vaccination (Hillemann, 
Nature Medicine 4:501 (1 998)). The most common vaccination strategy, the use 
of attenuated (i.e., less virulent) pathogens or closely related organisms, was first 
demonstrated by Edward Jenner in 1796, who vaccinated against smallpox by the 
administration of a less dangerous cowpox virus. Although a number of live 
attenuated viruses (e.g., measles, mumps, rubella, varicella, adenovirus, polio, 



-2- 



influenza) and bacteria (e.g., bacille Calmette-Guerin (BCG) against tuberculosis) 
are successfully administered for vaccination, there is a risk for the development 
of serious complications related to a reversion to virulence and infection by the 
'vaccine' organism, in particular in immunocompromised individuals. 
[0004] The specific design of attenuated viruses is now enabled by recombinant 

DNA technology (i.e., genetic engineering) through the generation of deletion or 
mutation variants. For example, the administration of an engineered Simian 
Immunodeficiency Virus (SIV) with a deletion within the nef gene was shown to 
protect macaques from subsequent infection with a pathogenic SIV strain (Daniel 
et al., Science 255:1938-1941 (1992)). However, the progression of acquired 
immunodeficiency syndrome (AIDS)-like symptoms in animals administered 
attenuated SIV raises safety concerns (Baba et al, Science 267: 1820- 1825 
(1995)). 

[0005] As an alternative approach, attenuated viruses or bacteria may be used as 

carriers for the antigen-encoding genes of a pathogen that is considered too unsafe 
to be administered in an attenuated form (e.g., Human Immunodeficiency Virus 
(HIV)). Upon delivery of the antigen-encoding gene to the host, the antigen is 
synthesized in situ. Vaccinia and related avipox viruses have been used as such 
carriers for various genes in preclinical and clinical studies for a variety of diseases 
(e.g., Shen etal, Science 252:440 (1991)). One disadvantage ofthis vaccination 
strategy is that it does not mimic the virion surface, because the recombinant 
protein is expressed on the surface of the host cell. Additionally, complications 
may develop in immunocompromised individuals, as evidenced by life-threatening 
disseminated vaccinia infections (Redfield, N. Eng. J. Med. 316:673 (1998)). 

[0006] A fourth vaccination approach involves the use of isolated components of 

a pathogen, either purified from the pathogen grown in vitro (e.g., influenza 
hemagglutinin or neuraminidase) or after heterologous expression of a single viral 
protein (e.g., Hepatitis B surface antigen). For example, recombinant, mutated 
toxins (detoxified) are used for vaccination against diphtheria, tetanus, cholera 
and pertussis toxins (Levine et al., New generation vaccines, 2nd edn., Marcel 



Dekker, Inc., New York 1 997), and recombinant proteins of HIV (gp 1 20 and full- 
length gp 1 60) were evaluated as a means to induce neutralizing antibodies against 
HIV with disappointing results (Connor et al, J. Virol. 72:1552 (1998)). 
Recently, promising results were obtained with soluble oligomeric gp 1 60, that can 
induce CTL response and elicit antibodies with neutralizing activity against HIV- 1 
isolates (Van Cortt et al. , J. Virol. 71:4319 (1997)). Inaddition, peptide vaccines 
may be used in which known B- or T-cell epitopes of an antigen are coupled to 
a carrier molecule designed to increase the immunogenicity of the epitope by 
stimulating T-cell help. However, one significant problem with this approach is 
that it provides a limited immune response to the protein as a whole. Moreover, 
vaccines have to be individually designed for different MHC haplotypes. The 
most serious concern for this type of vaccine is that protective antiviral antibodies 
recognize complex, three-dimensional structures that cannot be mimicked by 
peptides. 

[0007] A more novel vaccination strategy is the use of DNA vaccines (Donnelly 

et al, Ann. Rev. Immunol. 75:617 (1997)), which may generate MHC Class I- 
restricted CTL responses (without the use of a live vector). This may provide 
broader protection against different strains of a virus by targeting epitopes from 
conserved internal proteins pertinent to many strains of the same virus. Since the 
antigen is produced with mammalian post-translational modification, conformation 
and oligomerization, it is more likely to be similar or identical to the wild-type 
protein produced by viral infection than recombinant or chemically modified 
proteins. However, this distinction may turn out to be a disadvantage for the 
application of bacterial antigens, since non-native post-translational modification 
may result in reduced immunogenicity. In addition, viral surface proteins are not 
highly organized in the absence of matrix proteins. 

[0008] In addition to applications for the prevention of infectious disease, vaccine 

technology is now being utilized to address immune problems associated with 
allergies. In allergic individuals, antibodies of the IgE isotype are produced in an 
inappropriate humoral immune response towards particular antigens (allergens). 



The treatment of allergies by allergy immunotherapy requires weekly 
administration of successively increasing doses of the particular allergen over a 
period of up to 3-5 years. Presumably, 'blocking' IgG antibodies are generated 
that intercept allergens in nasal or respiratory secretions or in membranes before 
they react with IgE antibodies on mast cells. However, no constant relationship 
exists between IgG titers and symptom relief. Presently, this is an extremely time- 
and cost-consuming process, to be considered only for patients with severe 
symptoms over an extended period each year. 
[0009] It is well established that the administration of purified proteins alone is 

usually not sufficient to elicit a strong immune response; isolated antigen generally 
must be given together with helper substances called adjuvants. Within these 
adjuvants, the administered antigen is protected against rapid degradation, and the 
adjuvant provides an extended release of a low level of antigen. 
[0010] Unlike isolated proteins, viruses induce prompt and efficient immune 

responses in the absence of any adjuvants both with and without T-cell help 
(Bachmann & Zinkernagel, Ann. Rev. Immunol. 15:235-270 (1997)). Although 
viruses often consist of few proteins, they are able to trigger much stronger 
immune responses than their isolated components. For B cell responses, it is 
known that one crucial factor for the immunogenicity of viruses is the 
repetitiveness and order of surface epitopes. Many viruses exhibit a quasi- 
crystalline surface that displays a regular array of epitopes which efficiently 
crosslinks epitope-specific immunoglobulins on B cells (Bachmann & Zinkernagel, 
Immunol. Today 77:553-558 (1996)). This crosslinking of surface 
immunoglobulins onB cells is a strong activation signal that directly induces cell- 
cycle progression and the production of IgM antibodies. Further, such triggered 
B cells are able to activate T helper cells, which in turn induce a switch from IgM 
to IgG antibody production in B cells and the generation of long-lived B cell 
memory - the goal of any vaccination (Bachmann & Zinkernagel, Ann. Rev. 
Immunol. 75:23 5-270 ( 1 997)). Viral structure is even linked to the generation of 
anti-antibodies in autoimmune disease and as a part of the natural response to 



pathogens {see Fehr, T., et al., J. Exp. Med. 755:1785-1792 (1997)). Thus, 
antigens on viral particles that are organized in an ordered and repetitive array are 
highly immunogenic since they can directly activate B cells. 

[001 1] In addition to strong B cell responses, viral particles are also able to induce 

the generation of a cytotoxic T cell response, another crucial arm of the immune 
system. These cytotoxic T cells are particularly important for the elimination of 
non-cytopathic viruses such as HIV or Hepatitis B virus and for the eradication 
of tumors. Cytotoxic T cells do not recognize native antigens but rather recognize 
their degradation products in association with MHC class I molecules (Townsend 
& Bodmer, Ann. Rev. Immunol. 7:601-624 (1989)). Macrophages and dendritic 
cells are able to take up and process exogenous viral particles (but not their 
soluble, isolated components) and present the generated degradation product to 
cytotoxic T cells, leading to their activation and proliferation (Kovacsovics- 
Bankowski et al, Proc. Natl. Acad. Sci. USA 90:4942-4946 (1993); Bachmann 
etal, Eur. J. Immunol. 25:2595-2600 (1996)). 

[0012] Viral particles as antigens exhibit two advantages over their isolated 

components: (1) Due to their highly repetitive surface structure, they are able to 
directly activate B cells, leading to high antibody titers and long-lasting B cell 
memory; and (2) Viral particles but not soluble proteins are able to induce a 
cytotoxic T cell response, even if the viruses are non-infectious and adjuvants are 
absent. 

[0013] Several new vaccine strategies exploit the inherent immunogenicity of 

viruses. Some of these approaches focus on the particulate nature of the virus 
particle; for example see Harding, C. V. and Song, R., (J. Immunology 753:4925 
(1994)), which discloses a vaccine consisting of latex beads and antigen, 
Kovacsovics-Bankowski, M., et al. (Proc. Natl. Acad. Sci. USA 90:4942-4946 
( 1 993)), which discloses a vaccine consisting of iron oxide beads and antigen; U. S. 
Patent No 5,334,394 to Kossovsky, N., et al, which discloses core particles 
coated with antigen, U.S. Patent No. 5,871,747, which discloses synthetic 
polymer particles carrying on the surface one or more proteins covalently bonded 



thereto, and a core particle with a non-covalently bound coating, which at least 
partially covers the surface of said core particle, and at least one biologically 
active agent in contact with said coated core particle (see, e.g., WO 94/15585). 
[0014] However, a disadvantage of these viral mimicry systems is that they are 

not able to recreate the ordered presentation of antigen found on the viral surface. 
Antigens coupled to a surface in a random orientation are found to induce CTL 
response and no or only weak B-cell response. For an efficient vaccine, both arms 
of the immune system have to be strongly activated, as described above and in 
Bachmann & Zinkernagel, Ann. Rev. Immunol. 75:235 (1997). 

[0015] In another example, recombinant viruses are being utilized for antigen 

delivery. Filamentous phage virus containing an antigen fused to a capsid protein 
has been found to be highly immunogenic {see Perham R.N., et al., FEMS 
Microbiol. Rev. 77:25-31 (1995); Willis et al., Gene 725:85-88 (1993); 
Minenkova et al, Gene 725:85-88 (1993)). However, this system is limited to 
very small peptides (5 or 6 amino acid residues) when the fusion protein is 
expressed at a high level (Iannolo et al, J. Mol. Biol. 245:835-844 (1995)) or 
limited to the low level expression of larger proteins (de la Cruz et al, J. Biol. 
Chem. 263:43 1 8-4322 (1 988)) For small peptides, so far only the CTL response 
is observed and no or only weak B-cell response. 

[0016] In yet another system, recombinant alphaviruses are proposed as a means 

of antigen delivery (see U.S. Patent Nos. 5,766,602; 5,792,462; 5,739,026; 
5;789,245 and 5,814,482). Problems with the recombinant virus systems 
described so far include a low density expression of the heterologous protein on 
the viral surface and/or the difficulty of successfully and repeatedly creating a new 
and different recombinant viruses for different applications. 

[0017] In a further development, virus-like particles (VLPs) are being exploited 

in the area of vaccine production because of both their structural properties and 
their non-infectious nature. VLPs are supermolecular structures built in a 
symmetric manner from many protein molecules of one or more types. They lack 



the viral genome and, therefore, are noninfectious. VLPs can often be produced 
in large quantities by heterologous expression and can be easily be purified. 

[0018] Examples of VLPs include the capsid proteins of Hepatitis B virus (Ulrich, 

et al., Virus Res. 50:141-182 (1998)), measles virus (Warnes, et al, Gene 
760:173-178 (1995)), Sindbis virus, rotavirus (U.S. Patent Nos. 5,071,651 and 
5,374,426), foot-and-mouth-disease virus (Twomey, et al, Vaccine 
73:1603-1610, (1995)), Norwalk virus (Jiang, X., et al. , Science 250:1580-1583 
(1990); Matsui, S.M., etal.,J. Clin. Invest. 87: 1456-1461 (1991)), the retroviral 
GAG protein (PCT Patent Appl. No. WO 96/30523), the retrotransposon Ty 
protein pi, the surface protein of Hepatitis B virus (WO 92/1 1291) and human 
papilloma virus (WO 98/15631). In some instances, recombinant DNA 
technology may be utilized to fuse a heterologous protein to a VLP protein 
(Kratz, P. A., etal, Proc. Natl Acad. Sci. USA 96: 19151920 (1999)). 

[0019] Thus, there is a need in the art for the development of new and improved 

vaccines that promote a strong CTL and B-cell immune response as efficiently as 
natural pathogens. 



BRIEF SUMMARY OF THE INVENTION 



[0020] The invention provides a versatile new technology that allows production 

of particles or pili coated with any desired antigen. The technology allows the 
creation of highly efficient vaccines against infectious diseases and for the creation 
of vaccines for the treatment of allergies and cancers. The invention also provides 
compositions suited for the induction of Th type 2 T-helper cells (Th2 cells) 
Thus, efficient vaccines for the treatment of chronic diseases induced or 
accelerated by a Thl type immune response, such as arthritis, colitis, diabetes and 
multiple sclerosis can be produced with the technology provided by this invention. 

[0021] In a first embodiment, the invention provides a novel composition 

comprising (A) a non-natural molecular scaffold and (B) an antigen or antigenic 
determinant 



[0022] The non-natural molecular scaffold comprises, or alternatively consists of, 

(i) a core particle selected from the group consisting of (1) a core particle of non- 
natural origin and (2) a core particle of natural origin; and (ii) an organizer 
comprising at least one first attachment site, wherein said organizer is connected 
to said core particle by at least one covalent bond. 

[0023] In certain specific embodiments, the core particle naturally contains an 

organizer. One example of an embodiment of the invention where the organizer 
is naturally occurring is the bacterial pilus or pilin protein. The antigenic 
determinant may be linked by a cysteine to a naturally occurring lysine residue of 
the bacterial pili or pilin protein. 

[0024] The antigen or antigenic determinant has at least one second attachment 

site which is selected from the group consisting of (i) an attachment site not 
naturally occurring with said antigen or antigenic determinant; and (ii) an 
attachment site naturally occurring with said antigen or antigenic determinant. 

[0025] The invention provides for an ordered and repetitive antigen array through 

an association of the second attachment site to the first attachment site by way of 
at least one non-peptide bond. Thus, the antigen or antigenic determinant and the 
non-natural molecular scaffold are brought together through this association of the 
first and the second attachment site to form an ordered and repetitive antigen 
array. 

[0026] In another embodiment, the core particle of the aforementioned 

composition comprises a virus, a virus-like particle, a bacterial pilus, a structure 
formed from bacterial pilin, a bacteriophage, a viral capsid particle or a 
recombinant form thereof. Alternatively, the core particle may be a synthetic 
polymer or a metal. 

[0027] In yet another embodiment, the core particle comprises, or alternatively 

consists of, one or more different Hepatitis core (capsid) proteins (HBcAgs). In 
a related embodiment, one or more cysteine residues of these HBcAgs are either 
deleted or substituted with another amino acid residue (e.g., a serine residue). In 
a specific embodiment, the cysteine residues of the HBcAg used to prepare 



compositions of the invention which correspond to amino acid residues 48 and 
107 in SEQ ID NO: 134 are either deleted or substituted with another amino acid 
residue (e.g., a serine residue). 

[0028] Further, the HBcAg variants used to prepare compositions of the invention 

will generally be variants which retain the ability to associate with other HBcAgs 
to form dimeric or multimeric structures that present ordered and repetitive 
antigen or antigenic determinant arrays. 

[0029] In another embodiment, the non-natural molecular scaffold comprises, or 

alternatively consists of, pili or pilus-like structures that have been either produced 
from pilin proteins or harvested from bacteria. When pili or pilus-like structures 
are used to prepare compositions of the invention, they may be formed from 
products of pilin genes which are naturally resident in the bacterial cells but have 
been modified by genetically engineered (e.g. , by homologous recombination) or 
pilin genes which have been introduced into these cells. 

[0030] In a related embodiment, the core particle comprises, or alternatively 

consists of, pili or pilus-like structures that have been either prepared from pilin 
proteins or harvested from bacteria. These core particles may be formed from 
products of pilin genes naturally resident in the bacterial cells Further, antigens 
or antigenic determinants may be linked to these core particles naturally containing 
an organizer. In such a case, the core particles will generally be linked to a second 
attachment site of the antigen or antigenic determinant. In most embodiments of 
the invention, the pili or pilus-like structures will be able to form an ordered and 
repetitive antigen array with the antigen or antigenic determinant linked to the 
core particle at a specific or preferred location (e.g., a specific amino acid 
residue). 

[0031] In a particular embodiment, the organizer may comprise at least one first 

attachment site. The first and the second attachment sites are particularly 
important elements of compositions of the invention. In various embodiments of 
the invention, the first and/or the second attachment site may be an antigen and an 
antibody or antibody fragment thereto; biotin and avidin; strepavidin and biotin; 



-10- 



a receptor and its ligand; a ligand-binding protein and its ligand; interacting leucine 
zipper polypeptides; an amino group and a chemical group reactive thereto; a 
carboxyl group and a chemical group reactive thereto; a sulfhydryl group and a 
chemical group reactive thereto; or a combination thereof. 

[0032] In one embodiment, the invention provides the coupling of almost any 

antigen of choice to the surface of a virus, bacterial pilus, structure formed from 
bacterial pilin, bacteriophage, virus-like particle or viral capsid particle. By 
bringing an antigen into a quasi-crystalline 'virus-like' structure, the invention 
exploits the strong antiviral immune reaction of a host for the production of a 
highly efficient immune response, i.e. , a vaccination, against the displayed antigen. 

[0033] In another embodiment, the core particle may be selected from the group 

consisting of: recombinant proteins of Rotavirus, recombinant proteins of 
Norwalk virus, recombinant proteins of Alphavirus, recombinant proteins of Foot 
and Mouth Disease virus, recombinant proteins of Retrovirus, recombinant 
proteins of Hepatitis B virus, recombinant proteins of Tobacco mosaic virus, 
recombinant proteins of Flock House Virus, and recombinant proteins of human 
Papilomavirus. 

[0034] In yet another embodiment, the antigen may be selected from the group 

consisting of: (1) a protein suited to induce an immune response against cancer 
cells; (2) a protein suited to induce an immune response against infectious 
diseases; (3) a protein suited to induce an immune response against allergens, and 
(4) a protein suited to induce an immune response in pets or farm animals. 

[0035] In one embodiment, the invention relates to the induction of specific Th 

type 2 T-helper cells (Th2 cells) using antigens attached to Pili. The induction of 
Th2 responses may be beneficial for the treatment of a number of diseases. For 
example, many chronic diseases in humans an animals, such as arthritis, colitis, 
diabetes and multiple sclerosis are dominated by Thl response, where T cells 
secrete IFN r and other pro-inflammatory cytokines precipitating disease. 

[0036] In a particularly embodiment of the invention, the first attachment site 

and/or the second attachment site comprise an interacting leucine zipper 



polypeptide. In a related embodiment, the first attachment site and/or the second 
attachment site are selected from the group comprising. (1) the JUN leucine 
zipper protein domain; and (2) the FOS leucine zipper protein domain, 

[0037] In another embodiment, the first attachment site and/or the second 

attachment site are selected from the group comprising: (1) a genetically 
engineered lysine residue and (2) a genetically engineered cysteine residue, two 
residues that may be chemically linked together. 

[0038] The invention also includes embodiments where the organizer particle has 

only a single first attachment site and the antigen or antigenic determinant has only 
a single second attachment site. Thus, when an ordered and repetitive antigen 
array is prepared using such embodiments, each organizer will be bound to a 
single antigen or antigenic determinant. 

[0039] In one aspect, the invention provides compositions comprising, or 

alternatively consisting of, (a) a non-natural molecular scaffold comprising (i) a 
core particle selected from the group consisting of a core particle of non-natural 
origin and a core particle of natural origin, and (ii) an organizer comprising at least 
one first attachment site, wherein the core particle comprises, or alternatively 
consists of, a bacterial pilus, a pilus-like structure, or a modified HBcAg, or 
fragment thereof, and wherein the organizer is connected to the core particle by 
at least one covalent bond, and (b) an antigen or antigenic determinant with at 
least one second attachment site, the second attachment site being selected from 
the group consisting of (i) an attachment site not naturally occurring with the 
antigen or antigenic determinant and (ii) an attachment site naturally occurring 
with the antigen or antigenic determinant, wherein the second attachment site is 
capable of association through at least one non-peptide bond to the first 
attachment site, and wherein the antigen or antigenic determinant and the scaffold 
interact through the association to form an ordered and repetitive antigen array. 

[0040] Other embodiments of the invention include processes for the production 

of compositions of the invention and a methods of medical treatment using vaccine 
compositions described herein. 



-12- 



[0041] It is to be understood that both the foregoing general description and the 

following detailed description are exemplary and explanatory only and are 
intended to provide further explanation of the invention as claimed. 

BRIEF DESCRIPTION OF THE DRAWINGS 

[0042] Figure 1 shows a Western blot demonstrating the production of viral 

particles containing the E2-JUN fusion protein using the pCYTts::E2/MV 
expression vector. 

[0043] Figure 2 shows a Western blot demonstrating the production of viral 

particles containing the E2-JUN fusion protein expressed from pTE5'2J::E2J6W 
expression vector 

[0044] Figure 3 shows a Western dot blot demonstrating bacterial and eukaryotic 

expression of the FOS-hgh antigen. 
[0045] Figure 4 shows the expression of HBcAg-JUN in E. coli cells. 

[0046] Figure 5 shows a Western blot demonstrating that FfficAg-JUN is soluble 

in E. coli lysates 

[0047] Figure 6 shows an SDS-PAGE analysis of enrichment of HBcAg-JUN 

capsid particles on a sucrose density gradient. 

[0048] Figure 7 shows a non-reducing SDS-PAGE analysis of the coupling of 

hGH-FOS and HBcAg-JUN particles. 

[0049] Figure 8 depicts an analysis by SDS-PAGE of the coupling reaction of the 

FLAG peptide to HBcAG-Lys treated with iodacetamide and activated with 
Sulfo-MB S The excess of cr o ss-linker and of peptide over HBcAg-Lys monomer 
is indicated below the figure. 

[0050] Figure 9 depicts an analysis of coupling of the FLAG peptide to type-1 

bacterial pili by SDS-PAGE. Lane 1 shows the unreacted pili subunitFimA. Lane 
3 shows the purified reaction mixture of the pili with the FLAG peptide. The 
upper band corresponds to the coupled product, while the lower band corresponds 
to the unreached subunit. 



[0051] Figure 10 depicts an analysis by SDS-PAGE of the derivatization of 

HBcAg-Lys with SPDP. 

[0052] Figure 11 depicts an analysis by SDS-PAGE of the derivatization of 

HBcAg-Lys with Sulfo-MBS. 

[0053] Figure 12 depicts an analysis by SDS-PAGE of the coupling of HBcAg- 

Lys-2cyc-Mut to the FLAG peptide. The arrow shows the bands corresponding 
to the coupling of one and two FLAG peptides, respectively, to one subunit of 
FfBcAgLys-2cyc-Mut. Lane M corresponds to the marker, lane 1 to the 
unreached HBcAg-Lys-2cyc-Mut, lane 2 to HBcAg-Lys-2cyc-Mut activated with 
Sulfo-MBS, and lane 3 activated HBcAg-Lys-2cyc-Mut after reaction with the 
FLAG peptide containing an N-terminal cysteine. 

[0054] Figure 13 depicts an analysis by SDS-PAGE of the coupling of pili to the 

p33 peptide. 

[0055] Figure 14A shows an analysis of coupling of DP 178c peptide by SDS- 

PAGE analysis and Coomassie staining. Lane 1 corresponds to the supernatant of 
the coupling reaction after centrifugation, while lane 2 corresponds to the pellet. 
Figure 14B show an ELISA data and subtype analysis of mice, sera immunized 
with Pili-DP178c. The OD (450 nm) of the ELISA signal obtained at a fifty-fold 
dilution of the sera is shown in the diagram. For each subtype determination, mice 
sera were titrated from a fifty-fold dilution in two-fold dilution steps. The ELISA 
titer of the IgGl subtype (OD50 dilution) was 1 :400, while the titer of the IgG2b 
subtype was 1 100. The other subtypes all had titers inferior to 1 :50. The IgG 
isotype pattern is characteristic of a Th2 response, with a high IgGl titer and a 
low IgG2a titer. 

[0056] Figure 1 5 A shows an analysis of Coupling of GRA2 to Pili by SDS-PAGE 

analysis and Coomassie staining. Figure 1 5B relates to immunization of mice with 
Pili-GRA2 and IgG subtype determination. Depicted is an analysis of total IgG 
titer and IgG subtype titers by ELISA. The ELISA titer is given by the dilution 
of sera at which OD50 is obtained. The result of the immunization of two 



-14- 



individual mice is shown. A high IgGl titer and a low IgG2a titer is characteristic 
of a Th2 response. 

[0057] Figure 16A shows an analysis of coupling of B2 and D2 peptides to Pili 

by SDS-PAGE analysis and Coomassie staining. Figure 16B relates to 
immunization of mice with Pili-B2 and IgG subtype determination. The OD (450 
nm) of the ELIS A signal obtained at a fifty-fold dilution of the sera is shown in the 
diagram. For each subtype determination, mice sera were titrated from a fifty-fold 
dilution in two-fold dilution steps. The titer of the IgGl subtype (dilution at 
which the signal corresponds to OD 50) wasl:250, while the other subtypes all 
had titers inferior to 1:50. The titer of the IgGl subtype is much higher than the 
titer of the IgG2a subtype, a pattern typical for a Th2 response. 

[0058] Figure 17 relates to the measurement of antibodies specific for TNFa 

protein in the serum of mice immunized with the muTNFa peptide coupled to 
type-1 Pili. As a control, preimmune sera of two mice were assayed for binding 
to TNFa protein. Sera were added at three different dilutions (1 :50, 1 : 100 and 
1 :200), and bound IgG was detected with a horseradish peroxidase-conjugated 
anti-murine IgG antibody Results from four individual mice are shown on day 2 1 
and day 43 OD (450 nm): optical density at 450 nm. 

[0059] Figure 1 8A shows an analysis of coupling of 5'-TNF II and 3-TNF II by 

SDS-PAGE and Coomassie staining. Lane M is the marker lane. Untreated Pili 
were loaded on lane 1, Pili-5'-TNF II before dialysis on lane 2, Pili-3'-TNF II 
before dialysis on lane 3, Pili-5-TNF II after dialysis on lane 4, pili-3-TNF II after 
dialysis on lane 5. The arrow indicates the size at which the coupled product 
migrates. 

[0060] Figure 1 8B shows an ELISA analysis of sera of mice immunized with Pili- 

5'-TNF II and Pili-3'-TNF II: Anti-TNFa ELISA. IgG antibodies specific for 
native TNFa protein were measured in a specific ELISA. 2 //g/ml native TNFa 
protein was coated on ELISA plates. Sera were added at different dilutions and 
bound IgG was detected with a horseradish peroxidase-conjugated anti-murine 
IgG antibody. Results from four individual mice are shown on day 21 and day 43 



-15- 



OD (450 nm): optical density at 450 nm. The data show that mice immunized 
with the TNF peptides coupled to pili mount an antibody response against native 
TNFa protein, thus breaking self-tolerance. 

[0061] Figure 1 8C shows an ELISA analysis of sera of mice immunized with Pili- 

5-TNF II and Pili-3-TNF II' Anti-TNFa peptide ELISA. IgG antibodies specific 
for the 5 'TNF II and 3 'TNF II peptides were measured in a specific ELISA: 10 
^g/ml Ribonuclease A coupled to 5'TNF II or 3 'TNF II peptide was coated on 
ELISA plates. Sera were added at different dilutions and bound IgG was detected 
with a horseradish peroxidaseconjugated anti-murine IgG antibody. Results from 
four individual mice are shown on day 2 1 . 

[0062] Figure 18D shows that IgG subtype analysis of anti-TNF peptide 

antibodies in mice vaccinated with the corresponding TNF-peptides coupled to 
Pili. Results from four individual mice (no. 1-4) are shown on day 50. ELISA 
titer: dilution step at which half- maximal optical density was reached (-log 2 of 
40-fold prediluted sera). The high IgGl titer obtained as compared to the very 
low IgG2a titer is typical of a Th2 response. 

[0063] Figure 19A shows an analysis of coupling of M2 peptide to Pili by SDS- 

P AGE analysis and Coomassie staining The bands corresponding to non-coupled 
Pili and to the coupling product, Pili~M2, are indicated by arrows. Figure 19B 
shows an ELISA analysis and IgG subtype determination of mice vaccinated with 
Pili-M2. Sera were diluted eighty-fold, and titrated down in two-fold dilution 
steps. For the IgGl subtype, a titer of 1 :2560 was obtained, while for the IgG2a 
and IgG2b subtypes, titers below 1:100 were obtained. The titer for the IgG3 
subtype was below 1 :80. Titers were calculated as the serum dilution resulting in 
half-maximal optical density (OD 50 ). A strong IgGl titer in conjunction with a low 
IgG2a titer is characteristic for a Th2 type response. Average results from two 
mice are shown as optical densities obtained with a 1 .80 dilution of the serum. 

[0064] Figure 20 shows an ELISA analysis and IgG subtype determination of sera 

from mice immunized with HBcAg-Lys-2cys-Mut coupled to the Flag peptide. 
Ribonuclease A coupled to Flag peptide was coated at 10 Mg/ml, and serum was 



-16- 



added at a 1 :40 dilution. In contrast to experiments where mice were immunized 
with antigens coupled to Pili, there is no predominance of the IgGl subtype over 
the other IgG subtypes. 

DETAILED DESCRIPTION OF THE INVENTION 

1 . Definitions 

[0065] The following definitions are provided to clarify the subject matter which 

the inventors consider to be the present invention. 

[0066] Alphavirus: As used herein, the term "alphavirus" refers to any of the RNA 

viruses included within the genus Alphavirus Descriptions of the members of this 
genus are contained in Strauss and Strauss, Microbiol. Rev., 55:491-562 (1994). 
Examples of alphaviruses include Aura virus, Bebaru virus, Cabassou virus, 
Chikungunya virus, Easter equine encephalomyelitis virus, Fort morgan virus, 
Getah virus, Kyzylagach virus, Mayoaro virus, Middleburg virus, Mucambo virus, 
Ndumu virus, Pixuna virus, Tonate virus, Triniti virus, Una virus, Western equine 
encephalomyelitis virus, Whataroa virus, Sindbis virus (SIN), Semliki forest virus 
(SFV), Venezuelan equine encephalomyelitis virus (VEE), and Ross River virus. 

[0067] Antigen: As used herein, the term "antigen" is a molecule capable of being 

bound by an antibody. An antigen is additionally capable of inducing a humoral 
immune response and/or cellular immune response leading to the production of B- 
and/or T-lymphocytes. An antigen may have one or more epitopes (B- and T- 
epitopes). The specific reaction referred to above is meant to indicate that the 
antigen will react, in a highly selective manner, with its corresponding antibody 
and not with the multitude of other antibodies which may be evoked by other 
antigens. 

[0068] Antigenic determinant: As used herein, the term"antigenic determinant" 

is meant to refer to that portion of an antigen that is specifically recognized by 
either B- or T-lymphocytes. B-lymphocytes respond to foreign antigenic 



-17- 



determinants via antibody production, whereas T-lymphocytes are the mediator 
of cellular immunity. Thus, antigenic determinants or epitopes are those parts of 
an antigen that are recognized by antibodies, or in the context of an MHC, by T- 
cell receptors. 

[0069] Association. As used herein, the term "association" as it applies to the first 

and second attachment sites, is used to refer to at least one non-peptide bond. 
The nature of the association may be covalent, ionic, hydrophobic, polar or any 
combination thereof 

[0070] Attachment Site, First: As used herein, the phrase "first attachment site" 

refers to an element of the "organizer", itself bound to the core particle in a non- 
random fashion, to which the second attachment site located on the antigen or 
antigenic determinant may associate The first attachment site may be a protein, 
a polypeptide, an amino acid, a peptide, a sugar, a polynucleotide, a natural or 
synthetic polymer, a secondary metabolite or compound (biotin, fluorescein, 
retinol, digoxigenin, metal ions, phenylmethylsulfonylfluoride), or a combination 
thereof, or a chemically reactive group thereof. Multiple first attachment sites are 
present on the surface of the non-natural molecular scaffold in a repetitive 
configuration. 

[0071] Attachment Site, Second. As used herein, the phrase "second attachment 

site" refers to an element associated with the antigen or antigenic determinant to 
which the first attachment site of the "organizer" located on the surface of the 
non-natural molecular scaffold may associate. The second attachment site of the 
antigen or antigenic determinant may be a protein, a polypeptide, a peptide, a 
sugar, a polynucleotide, a natural or synthetic polymer, a secondary metabolite or 
compound (biotin, fluorescein, retinol, digoxigenin, metal ions, 
phenylmethylsulfonylfluoride), or a combination thereof, or a chemically reactive 
group thereof. At least one second attachment site is present on the antigen or 
antigenic determinant. 

[0072] Core particle: As used herein, the term "core particle" refers to a rigid 

structure with an inherent repetitive organization that provides a foundation for 



attachment of an "organizer". A core particle as used herein may be the product 
of a synthetic process or the product of a biological process. 

[0073] In certain embodiments of the invention, the antigens or antigenic 

determinants are directly linked to the core particle. 

[0074] Cis-acting: As used herein, the phrase Vis-acting" sequence refers to 

nucleic acid sequences to which a replicase binds to catalyze the RNA-dependent 
replication of RNA molecules. These replication events result in the replication 
of the full-length and partial RNA molecules and, thus, the alpahvirus subgenomic 
promoter is also a "c/s-acting" sequence. Qs-acting sequences may be located at 
or near the 5' end, 3' end, or both ends of a nucleic acid molecule, as well as 
internally. 

[0075] Fusion- As used herein, the term "fusion" refers to the combination of 

amino acid sequences of different origin in one polypeptide chain by in-frame 
combination of their coding nucleotide sequences. The term "fusion" explicitly 
encompasses internal fusions, i. e. , insertion of sequences of different origin within 
a polypeptide chain, in addition to fusion to one of its termini. 

[0076] Heterologous sequence: As used herein, the term "heterologous sequence" 

refers to a second nucleotide sequence present in a vector of the invention. The 
term "heterologous sequence" also refers to any amino acid or RNA sequence 
encoded by a heterologous DNA sequence contained in a vector of the invention. 
Heterologous nucleotide sequences can encode proteins or RNA molecules 
normally expressed in the cell type in which they are present or molecules not 
normally expressed therein {e.g., Sindbis structural proteins). 

[0077] Isolated. As used herein, when the term "isolated" is used in reference to 

a molecule, the term means that the molecule has been removed from its native 
environment. For example, a polynucleotide or a polypeptide naturally present in 
a living animal is not "isolated," but the same polynucleotide or polypeptide 
separated from the coexisting materials of its natural state is "isolated." Further, 
recombinant DNA molecules contained in a vector are considered isolated for the 
purposes of the present invention Isolated RNA molecules include in vivo or in 



-19- 



vitro RNA replication products of DNA and RNA molecules. Isolated nucleic 
acid molecules further include synthetically produced molecules. Additionally, 
vector molecules contained in recombinant host cells are also isolated. Thus, not 
all "isolated" molecules need be "purified." 
[0078] Immunotherapeutic: As used herein, the term "immunotherapeutic" is a 

composition for the treatment of diseases or disorders. More specifically, the term 
is used to refer to a method of treatment for allergies or a method of treatment for 
cancer. 

[0079] Individual: As used herein, the term "individual" refers to multicellular 

organisms and includes both plants and animals. Preferred multicellular organisms 
are animals, more preferred are vertebrates, even more preferred are mammals, 
and most preferred are humans. 

[0080] Low or undetectable: As used herein, the phrase "low or undetectable," 

when used in reference to gene expression level, refers to a level of expression 
which is either significantly lower than that seen when the gene is maximally 
induced {e.g. , at least five fold lower) or is not readily detectable by the methods 
used in the following examples section. 

[0081] Lectin. As used herein, proteins obtained particularly from the seeds of 

leguminous plants, but also from many other plant and animal sources, that have 
binding sites for specific mono- or oligosaccharides. Examples include 
concanavalin A and wheat-germ agglutinin, which are widely used as analytical 
and preparative agents in the study of glycoprotein. 

[0082] Natural origin: As used herein, the term "natural origin" means that the 

whole or parts thereof are not synthetic and exist or are produced in nature. 

[0083] Non-natural: As used herein, the term generally means not from nature, 

more specifically, the term means from the hand of man. 

[0084] Non-natural origin: As used herein, the term "non-natural origin" generally 

means synthetic or not from nature; more specifically, the term means from the 
hand of man. 



-20- 



[0085] Non-natural molecular scaffold: As used herein, the phrase "non-natural 

molecular scaffold" refers to any product made by the hand of man that may serve 
to provide a rigid and repetitive array of first attachment sites. Ideally but not 
necessarily, these first attachment sites are in a geometric order. The non-natural 
molecular scaffold may be organic or non-organic and may be synthesized 
chemically or through a biological process, in part or in whole. The non-natural 
molecular scaffold is comprised of: (a) a core particle, either of natural or non- 
natural origin; and (b) an organizer, which itself comprises at least one first 
attachment site and is connected to a core particle by at least one covalent bond. 
In a particular embodiment, the non-natural molecular scaffold may be a virus, 
virus-like particle, a bacterial pilus, a virus capsid particle, a phage, a recombinant 
form thereof, or synthetic particle. 

[0086] Ordered and repetitive antigen or antigenic determinant array: As used 

herein, the term "ordered and repetitive antigen or antigenic determinant array" 
generally refers to a repeating pattern of antigen or antigenic determinant, 
characterized by a uniform spacial arrangement of the antigens or antigenic 
determinants with respect to the non-natural molecular scaffold. In one 
embodiment of the invention, the repeating pattern may be a geometric pattern. 
Examples of suitable ordered and repetitive antigen or antigenic determinant 
arrays are those which possess strictly repetitive paracrystalline orders of antigens 
or antigenic determinants with spacings of 5 to 15 nanometers. 

[0087] Organizer: As used herein, the term "organizer" is used to refer to an 

element bound to a core particle in a non-random fashion that provides a 
nucleation site for creating an ordered and repetitive antigen array. An organizer 
is any element comprising at least one first attachment site that is bound to a core 
particle by at least one covalent bond. An organizer may be a protein, a 
polypeptide, a peptide, an amino acid (/. e. , a residue of a protein, a polypeptide 
or peptide), a sugar, a polynucleotide, a natural or synthetic polymer, a secondary 
metabolite or compound (biotin, fluorescein, retinol, digoxigenin, metal ions, 



-21- 



phenylmethylsulfonylfluoride), or a combination thereof, or a chemically reactive 
group thereof 

[0088] Permissive temperature: As used herein, the phrase "permissive 

temperature" refers to temperatures at which an enzyme has relatively high levels 
of catalytic activity. 

[0089] Pili: As used herein, the term "pili" (singular being "pilus") refers to 

extracellular structures of bacterial cells composed of protein monomers (e.g., 
pilin monomers) which are organized into ordered and repetitive patterns. 
Further, pili are structures which are involved in processes such as the attachment 
of bacterial cells to host cell surface receptors, inter-cellular genetic exchanges, 
and cell-cell recognition. Examples of pili include Type-1 pili, P-pili, F1C pili, 
S-pili, and 987P-pili. Additional examples of pili are set out below. 

[0090] Pilus-like structure: As used herein, the phrase "pilus-like structure" refers 

to structures having characteristics similar to that of pili and composed of protein 
monomers. One example of a "pilus-like structure" is a structure formed by a 
bacterial cell which expresses modified pilin proteins that do not form ordered and 
repetitive arrays that are essentially identical to those of natural piliA 

[0091] Purified: As used herein, when the term "purified" is used in reference to 

a molecule, it means that the concentration of the molecule being purified has been 
increased relative to molecules associated with it in its natural environment. 
Naturally associated molecules include proteins, nucleic acids, lipids and sugars 
but generally do not include water, buffers, and reagents added to maintain the 
integrity or facilitate the purification of the molecule being purified. For example, 
even if mRNA is diluted with an aqueous solvent during oligo dT column 
chromatography, mRNA molecules are purified by this chromatography if 
naturally associated nucleic acids and other biological molecules do not bind to the 
column and are separated from the subject mRNA molecules. 

[0092] Receptor: As used herein, the term "receptor" refers to proteins or 

glycoproteins or fragments thereof capable of interacting with another molecule, 
called the ligand. The ligand may belong to any class of biochemical or chemical 



-22- 



compounds. The receptor need not necessarily be a membrane-bound protein. 
Soluble protein, like e.g., maltose binding protein or retinol binding protein are 
receptors as well. 

[0093] Residue: As used herein, the term "residue" is meant to mean a specific 

amino acid in a polypeptide backbone or side chain. 

[0094] Temperature-sensitive: As used herein, the phrase "temperature-sensitive" 

refers to an enzyme which readily catalyzes a reaction at one temperature but 
catalyzes the same reaction slowly or not at all at another temperature. An 
example of a temperature- sensitive enzyme is the replicase protein encoded by the 
pCYTts vector, which has readily detectable replicase activity at temperatures 
below 34 °C and has low or undetectable activity at 37 °C. 

[0095] Transcription: As used herein, the term "transcription" refers to the 

production of RNA molecules from DNA templates catalyzed by RNA 
polymerase. 

[0096] Recombinant host cell: As used herein, the term "recombinant host cell" 

refers to a host cell into which one ore more nucleic acid molecules of the 
invention have been introduced. 

[0097] Recombinant virus: As used herein, the phrase "recombinant virus" refers 

to a virus that is genetically modified by the hand of man. The phrase covers any 
virus known in the art. More specifically, the phrase refers to a an alphavirus 
genetically modified by the hand of man, and most specifically, the phrase refers 
to a Sinbis virus genetically modified by the hand of man. 

[0098] Restrictive temperature: As used herein, the phrase "restrictive 

temperature" refers to temperatures at which an enzyme has low or undetectable 
levels of catalytic activity. Both "hot" and "cold" sensitive mutants are known 
and, thus, a restrictive temperature may be higher or lower than a permissive 
temperature. 

[0099] RNA-dependent RNA replication event: As used herein, the phrase 

"RNA-dependent RNA replication event" refers to processes which result in the 
formation of an RNA molecule using an RNA molecule as a templateA 



-23- 



[0100] RNA-Dependent RNA polymerase: As used herein, the phrase "RNA- 

Dependent RNA polymerase" refers to a polymerase which catalyzes the 
production of an RNA molecule from another RNA molecule. This term is used 
herein synonymously with the term "replicase." 

[0101] Untranslated RNA: As used herein, the phrase "untranslated RNA" refers 

to an RNA sequence or molecule which does not encode an open reading frame 
or encodes an open reading frame, or portion thereof, but in a format in which an 
amino acid sequence will not be produced (e.g., no initiation codon is present). 
Examples of such molecules are tRNA molecules, rRNA molecules, and 
ribozymes. 

[0102] Vector: As used herein, the term "vector" refers to an agent (e.g., a 

plasmid or virus) used to transmit genetic material to a host cell. A vector may 
be composed of either DNA or RNA. 

[0103] one, a, or an: When the terms "one," "a," or "an" are used in this 

disclosure, they mean "at least one" or "one or more," unless otherwise indicated. 

2. Compositions of Ordered and Repetitive Antigen or Antigenic 
Determinant Arrays and Methods to Make the Same 

[0104] The disclosed invention provides compositions comprising an ordered and 

repetitive antigen or antigenic determinant. Furthermore, the invention 
conveniently enables the practitioner to construct ordered and repetitive antigen 
or antigenic determinant arrays for various treatment purposes, which includes the 
prevention of infectious diseases, the treatment of allergies and the treatment of 
cancers. The invention also enables the practitioner to construct compositions 
comprising Pili inducing Th2 immune responses, useful in the treatment of chronic 
diseases. 

[0105] Compositions ofthe invention essentially comprise, or alternatively consist 

of, two elements: (1) a non-natural molecular scaffold; and (2) an antigen or 
antigenic determinant with at least one second attachment site capable of 
association through at least one non-peptide bond to said first attachment site. 



-24- 



[0106] The non-natural molecular scaffold comprises, or alternatively consists of: 

(a) a core particle selected from the group consisting of (1) a core particle of non- 
natural origin and (2) a core particle of natural origin, and (b) an organizer 
comprising at least one first attachment site, wherein said organizer is connected 
to said core particle by at least one covalent bond. 

[0107] Compositions of the invention also comprise, or alternatively consist of, 

core particles to which antigens or antigenic determinants are directly linked. 

[0108] The antigen or antigenic determinant has at least one second attachment 

site which is selected from the group consisting of (a) an attachment site not 
naturally occurring with said antigen or antigenic determinant, and (b) an 
attachment site naturally occurring with said antigen or antigenic determinant. 

[0109] The invention provides for an ordered and repetitive antigen array through 

an association of the second attachment site to the first attachment site by way of 
at least one non-peptide bond. Thus, the antigen or antigenic determinant and the 
non-natural molecular scaffold are brought together through this association of the 
first and the second attachment site to form an ordered and repetitive antigen 
array 

[0110] The practioner may specifically design the antigen or antigenic determinant 

and the second attachment site such that the arrangement of all the antigens or 
antigenic determinants bound to the non-natural molecular scaffold or, in certain 
embodiments, the core particle will be uniform. For example, one may place a 
single second attachment site on the antigen or antigenic determinant at the 
carboxyl or amino terminus, thereby ensuring through design that all antigen or 
antigenic determinant molecules that are attached to the non-natural molecular 
scaffold are positioned in a uniform way. Thus, the invention provides a 
convenient means of placing any antigen or antigenic determinant onto a 
non-natural molecular scaffold in a defined order and in a manner which forms a 
repetitive pattern. 

[0111] As will be clear to those skilled in the art, certain embodiments of the 

invention involve the use of recombinant nucleic acid technologies such as cloning, 



-25- 



polymerase chain reaction, the purification of DNA and RNA, the expression of 
recombinant proteins in prokaryotic and eukaryotic cells, etc. Such 
methodologies are well known to those skilled in the art and may be conveniently 
found in published laboratory methods manuals (e.g., Sambrook, J. et ah, eds., 
Molecular Cloning, A Laboratory Manual, 2nd. edition, Cold Spring 
Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989); Ausubel, F et ah, 
eds., Current Protocols in MolecularBiology, JohnH. Wiley & Sons, Inc. 
(1997)). Fundamental laboratory techniques for working with tissue culture cell 
lines (Celis, J., ed., CELL BIOLOGY, Academic Press, 2 nd edition, (1998)) and 
antibody-based technologies (Harlow, E. andLane,D., "Antibodies: ALaboratory 
Manual," Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. (1988); 
Deutscher, MP, "Guide to Protein Purification, "Me th. Enzymoh 128, Academic 
Press San Diego (1990); Scopes, R.K., "Protein Purification Principles and 
Practice," 3 rd ed., Springer- Verlag, New York (1994)) are also adequately 
described in the literature, all of which are incorporated herein by reference. 

A. Construction of a Non-Natural Molecular Scaffold 

[0112] One element in compositions of the invention is a non-natural molecular 

scaffold comprising, or alternatively consisting of, a core particle and an organizer 
As used herein, the phrase "non-natural molecular scaffold" refers to any product 
made by the hand of man that may serve to provide a rigid and repetitive array of 
first attachment sites. More specifically, the non-natural molecular scaffold 
comprises, or alternatively consists of, (a) a core particle selected from the group 
consisting of (1) a core particle of non-natural origin and (2) a core particle of 
natural origin; and (b) an organizer comprising at least one first attachment site, 
wherein said organizer is connected to said core particle by at least one covalent 
bond. 

[0113] As will be readily apparent to those skilled in the art, the core particle of 

the non-natural molecular scaffold of the invention is not limited to any specific 



-26- 



form. The core particle may be organic or non-organic and may be synthesized 
chemically or through a biological process. 

In one embodiment, a non-natural core particle may be a synthetic 
polymer, a lipid micelle or a metal Such core particles are known in the art, 
providing a basis from which to build the novel non-natural molecular scaffold of 
the invention. By way of example, synthetic polymer or metal core particles are 
described in U.S. Patent No. 5,770,380, which discloses the use of a calixarene 
organic scaffold to which is attached a plurality of peptide loops in the creation of 
an 'antibody mimic', and U.S. Patent No. 5,334,394 describes nano crystalline 
particles used as a viral decoy that are composed of a wide variety of inorganic 
materials, including metals or ceramics. Suitable metals include chromium, 
rubidium, iron, zinc, selenium, nickel, gold, silver, platinum Suitable ceramic 
materials in this embodiment include silicon dioxide, titanium dioxide, aluminum 
oxide, ruthenium oxide and tin oxide. The core particles of this embodiment may 
be made from organic materials including carbon (diamond). Suitable polymers 
include polystyrene, nylon and nitrocellulose. For this type of nanocrystalline 
particle, particles made from tin oxide, titanium dioxide or carbon (diamond) are 
may also be used A lipid micelle may be prepared by any means known in the art. 
For example micelles may be prepared according to the procedure of Baiselle and 
Millar (Biophys. Chem. 4:355-361 (1975)) or Corti et al. (Chem. Phys. Lipids 
35:197-214 (1981)) or Lopez et al. (FEBS Lett. ¥2*5:314-318 (1998)) or 
Topchieva and Karezin (J. Colloid Interface Sci. 273:29-35 (1999)) or Morein et 
al, {Nature 308:457-460 (1984)), which are all incorporated herein by reference. 

The core particle may also be produced through a biological process, 
which may be natural or non-natural. By way of example, this type of 
embodiment may includes a core particle comprising, or alternatively consisting 
of, a virus, virus-like particle, a bacterial pilus, a phage, a viral capsid particle or 
a recombinant form thereof. In a more specific embodiment, the core particle may 
comprise, or alternatively consist of, recombinant proteins of Rotavirus, 
recombinant proteins of Norwalk virus, recombinant proteins of Alphavirus, 



-27- 



recombinant proteins which form bacterial pili or pilus-like structures, 
recombinant proteins of Foot and Mouth Disease virus, recombinant proteins of 
Retrovirus, recombinant proteins of Hepatitis B virus (e.g., a HBcAg), 
recombinant proteins of Tobacco mosaic virus, recombinant proteins of Flock 
House Virus, and recombinant proteins of human Papilomavirus. 

[0116] Whether natural or non-natural, the core particle of the invention will 

generally have an organizer that is attached to the natural or non-natural core 
particle by at least one covalent bond. The organizer is an element bound to a 
core particle in a non- random fashion that provides a nucleation site for creating 
an ordered and repetitive antigen array. Ideally, but not necessarily, the organizer 
is associated with the core particle in a geometric order. Minimally, the organizer 
comprises a first attachment site. 

[0117] In some embodiments of the invention, the ordered and repetitive array is 

formed by association between (1) either core particles or non-natural molecular 
scaffolds and (2) an antigen or antigenic determinant. For example, bacterial pili 
or pilus-like structures are formed from proteins which are organized into ordered 
and repetitive structures. Thus, in many instances, it will be possible to form 
ordered arrays of antigens or antigenic determinants by linking these constituents 
to bacterial pili or pili-like structures. 

[0118] As previously stated, the organizer may be any element comprising at least 

one first attachment site that is bound to a core particle by at least one covalent 
bond. An organizer may be a protein, a polypeptide, a peptide, an amino acid 
(i.e., a residue of a protein, a polypeptide or peptide), a sugar, a polynucleotide, 
a natural or synthetic polymer, a secondary metabolite or compound (biotin, 
fluorescein, retinol, digoxigenin, metal ions, phenylmethylsulfonylfluoride), or a 
combination thereof, or a chemically reactive group thereof. In a more specific 
embodiment, the organizer may comprise a first attachment site comprising an 
antigen, an antibody or antibody fragment, biotin, avidin, strepavidin, a receptor, 
a receptor ligand, a ligand, a ligand-binding protein, an interacting leucine zipper 
polypeptide, an amino group, a chemical group reactive to an amino group; a 



-28- 



carboxyl group, chemical group reactive to a carboxyl group, a sulfhydryl group, 
a chemical group reactive to a sulfhydryl group, or a combination thereof. 

[0119] In one embodiment, the core particle of the non-natural molecular scaffold 

comprises a virus, a bacterial pilus, a structure formed from bacterial pilin, a 
bacteriophage, a virus-like particle, a viral capsid particle or a recombinant form 
thereof Any virus known in the art having an ordered and repetitive coat and/or 
core protein structure may be selected as a non-natural molecular scaffold of the 
invention; examples of suitable viruses include: sindbis and other alphaviruses; 
vesicular somatitis virus; rhabdo-, (e.g. vesicular stomatitis virus), picorna-, toga-, 
orthomyxo-, polyoma-, parvovirus, rotavirus, Norwalk virus, foot and mouth 
disease virus, a retrovirus, Hepatitis B virus, Tobacco mosaic virus, flock house 
virus, human papilomavirus (for example, see Table 1 in Bachman, M.F. and 
Zinkernagel, R.M., Immunol. Today 77 553-558 (1996)). 

[0120] In one embodiment, the invention utilizes genetic engineering of a virus to 

create a fusion between an ordered and repetitive viral envelope protein and an 
organizer comprising a heterologous protein, peptide, antigenic determinant or a 
reactive amino acid residue of choice. Other genetic manipulations known to 
those in the art may be included in the construction of the non-natural molecular 
scaffold; for example, it may be desirable to restrict the replication ability of the 
recombinant virus through genetic mutation. The viral protein selected for fusion 
to the organizer (i.e., first attachment site) protein should have an organized and 
repetitive structure. Such an organized and repetitive structure include 
paracrystalline organizations with a spacing of 5- 1 5 nm on the surface of the virus. 
The creation of this type of fusion protein will result in multiple, ordered and 
repetitive organizers on the surface of the virus. Thus, the ordered and repetitive 
organization of the first attachment sites resulting therefrom will reflect the normal 
organization of the native viral protein 

[0121] As will be discussed in more detail herein, in another embodiment of the 

invention, the non-natural molecular scaffold is a recombinant alphavirus, and 
more specifically, a recombinant Sinbis virus. Alphaviruses are positive stranded 



-29- 



RNA viruses that replicate their genomic RNA entirely in the cytoplasm of the 
infected cell and without a DNA intermediate (Strauss, J. and Strauss, E., 
Microbiol. Rev. 55:491-562 (1994)). Several members of the alphavirus family, 
Sindbis (Xiong, C. etal, Science 243- 1 188-1 191 (1989); Schlesinger, S., Trends 
Biotechnol. 77:18-22 (1993)), Semliki Forest Virus (SFV) (Liljestrom, P. & 
Garoff, H., Bio/Technology 9: 1356-1361 (1991)) and others (Davis, N.L. etal, 
Virology 777:189-204 (1989)), have received considerable attention for use as 
virus-based expression vectors for a variety of different proteins (Lundstrom, K., 
Curr. Opin. Biotechnol. 5:578-582 (1997); Liljestrom, P., Curr. Opin. 
Biotechnol. 5:495-500 (1994)) and as candidates for vaccine development. 
Recently, a number of patents have issued directed to the use of alphaviruses for 
the expression of heterologous proteins and the development of vaccines {see U. S . 
Patent Nos 5,766,602; 5,792,462; 5,739,026; 5,789,245 and 5,814,482). The 
construction of the alphaviral scaffold of the invention may be done by means 
generally known in the art of recombinant DNA technology, as described by the 
aforementioned articles, which are incorporated herein by reference. 

A variety of different recombinant host cells can be utilized to produce a 
viral-based core particle for antigen or antigenic determinant attachment. For 
example, Alphaviruses are known to have a wide host range; Sindbis virus infects 
cultured mammalian, reptilian, and amphibian cells, as well as some insect cells 
(Clark, H., J. Natl. Cancer Inst. 57:645 (1973); Leake, C, J. Gen. Virol. 35:335 
(1977); Stollar, V. in THE TOGA VIRUSES, R.W Schlesinger, Ed., Academic Press, 
(1980), pp. 583-621). Thus, numerous recombinant host cells can be used in the 
practice of the invention. BHK, COS, Vero, HeLa and CHO cells are particularly 
suitable for the production of heterologous proteins because they have the 
potential to glycosylate heterologous proteins in a manner similar to human cells 
(Watson, E. etal, Glycobiology 4:221, (1994)) and can be selected (Zang, M. et 
al, Bio/Technology 13 3 89 (1995)) or genetically engineered (Renner W. et al, 
Biotech. Bioeng. 4:476 (1995); LeeK etal. Biotech. Bioeng. 50336 (1996)) to 
grow in serum-free medium, as well as in suspension. 



-30- 



[0123] Introduction of the polynucleotide vectors into host cells can be effected 

by methods described in standard laboratory manuals (see, e.g. , Sambrook, J. et 
al. , eds , Molecular Cloning, A Laboratory Manual, 2nd. edition, Cold 
Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989), Chapter 9; 
Ausubel, F. etal, eds., Current Protocols in Molecular Biology, JohnH. 
Wiley & Sons, Inc. (1997), Chapter 16), including methods such as 
electroporation, DEAE-dextran mediated transfection, transfection, 
microinjection, cationic lipid-mediated transfection, transduction, scrape loading, 
ballistic introduction, and infection. Methods for the introduction of exogenous 
DNA sequences into host cells are discussed in Feigner, P. etal.,U S. Patent No. 
5,580,859. 

[0124] Packaged RNA sequences can also be used to infect host cells. These 

packaged RNA sequences can be introduced to host cells by adding them to the 
culture medium. For example, the preparation of non-infective alpahviral particles 
is described in a number of sources, including "Sindbis Expression System", 
Version C (Invitrogen Catalog No. K750-1) 

[0125] When mammalian cells are used as recombinant host cells for the 

production of viral-based core particles, these cells will generally be grown in 
tissue culture. Methods for growing cells in culture are well known in the art (see, 
e.g., Celis, J., ed., Cell Biology, Academic Press, 2 nd edition, (1998); Sambrook, 
J. et al, eds., Molecular Cloning, A Laboratory Manual, 2nd. edition, 
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989); Ausubel, 
F et al, eds., CURRENT PROTOCOLS IN MOLECULAR BIOLOGY, John H. Wiley & 
Sons, Inc. (1997); Freshney, R., CULTURE OF ANIMAL CELLS, Alan R. Liss, Inc. 
(1983)) 

[0126] As will be understood by those in the art, the first attachment site may be 

or be a part of any suitable protein, polypeptide, sugar, polynucleotide, peptide 
(amino acid), natural or synthetic polymer, a secondary metabolite or combination 
thereof that may serve to specifically attach the antigen or antigenic determinant 
of choice to the non-natural molecular scaffold. In one embodiment, the 



-31- 



attachment site is a protein or peptide that may be selected from those known in 
the art. For example, the first attachment site may selected from the following 
group: a ligand, a receptor, a lectin, avidin, streptavidin, biotin, an epitope such 
as an HA or T7 tag, Myc, Max, immunoglobulin domains and any other amino 
acid sequence known in the art that would be useful as a first attachment site. 

It should be further understood by those in the art that with another 
embodiment of the invention, the first attachment site may be created secondarily 
to the organizer (/'. e. , protein or polypeptide) utilized in constructing the in-frame 
fusion to the capsid protein. For example, a protein may be utilized for fusion to 
the envelope protein with an amino acid sequence known to be glycosylated in a 
specific fashion, and the sugar moiety added as a result may then serve at the first 
attachment site of the viral scaffold by way of binding to a lectin serving as the 
secondary attachment site of an antigen. Alternatively, the organizer sequence 
may be biotinylated in vivo and the biotin moiety may serve as the first attachment 
site of the invention, or the organizer sequence may be subjected to chemical 
modification of distinct amino acid residues in vitro, the modification serving as 
the first attachment site. 

One specific embodiment of the invention utilizes the Sinbis virus. The 
Sinbis virus RNA genome is packaged into a capsid protein that is surrounded by 
a lipid bilayer containing three proteins called El, E2, and E3 . These so-called 
envelope proteins are glycoproteins, and the glycosylated portions are located on 
the outside of the lipid bilayer, where complexes of these proteins form the 
"spikes" that can be seen in electron micrographs to project outward from the 
surface of the virus. In another embodiment of the invention, the first attachment 
site is selected to be the JUN or FOS leucine zipper protein domain that is fused 
in frame to the E2 envelope protein. However, it will be clear to all individuals 
in the art that other envelope proteins may be utilized in the fusion protein 
construct for locating the first attachment site in the non-natural molecular 
scaffold of the invention. 



-32- 



[0129] In a specific embodiment of the invention, the first attachment site is 

selected to be the JUN-FOS leucine zipper protein domain that is fused in frame 
to the Hepatitis B capsid (core) protein (HBcAg). However, it will be clear to all 
individuals in the art that other viral capsid proteins may be utilized in the fusion 
protein construct for locating the first attachment site in the non-natural molecular 
scaffold of the invention. 

[0130] In another specific embodiment of the invention, the first attachment site 

is selected to be a lysine or cysteine residue that is fused in frame to the HBcAg. 
However, it will be clear to all individuals in the art that other viral capsid or 
virus-like particles may be utilized in the fusion protein construct for locating the 
first attachment in the non-natural molecular scaffold of the invention. 

[0131] Example 1 is provided to demonstrate the construction of an in-frame 

fusion protein between the Sinbis virus E2 envelope protein and the JUN leucine 
zipper protein domain using the pTE5'2 J vector of Hahn et al. [Proc. Natl. Acad. 
Sci. USA 59:2679-2683 (1992)). The JUN amino acid sequence utilized for the 
first attachment site is the following: CGGRIARLEEKVKTLKAQNSE 
LAS T ANMLREQ V AQLKQKVMNHVGC (SEQ ID NO: 59). In this instance, 
the anticipated second attachment site on the antigen would be the FOS leucine 
zipper protein domain and the amino acid sequence would be the following: 
CGGLTDTLQAETDQVEDEKSALQTEIANLLKEKEKLEFILAAHGGC(SEQ 
ID NO: 60) 

[0132] These sequences are derived from the transcription factors JUN and FOS, 

each flanked with a short sequence containing a cysteine residue on both sides. 
These sequences are known to interact with each other. The original hypothetical 
structure proposed for the JUN-FOS dimer assumed that the hydrophobic side 
chains of one monomer interdigitate with the respective side chains of the other 
monomer in a zipper-like manner (Landschulz et al, Science 240:1759-1764 

(1988) ). However, this hypothesis proved to be wrong, and these proteins are 
known to form an a-helical coiled coil (O'Shea et al, Science 243 53 8-542 

(1989) ; O'Shea et al, Cell 65:699-708 (1992); Cohen & Parry, Trends Biochem. 



-33- 



Sci. 1 7:245-248 (1986)). Thus, the term "leucine zipper" is frequently used to 
refer to these protein domains for more historical than structural reasons. 
Throughout this patent, the term "leucine zipper" is used to refer to the sequences 
depicted above or sequences essentially similar to the ones depicted above. The 
terms JUN and FOS are used for the respective leucine zipper domains rather than 
the entire JUN and FOS proteins. 

In one embodiment, the invention provides for the production of a Sinbis 
virus E2-JUN scaffold using the pCYTts expression system (WO 99/50432). The 
pCYTts expression system provides novel expression vectors which permit tight 
regulation of gene expression in eucaryotic cells. The DNA vectors of this system 
are transcribed to form RNA molecules which are then replicated by a 
temperature- sensitive replicase to form additional RNA molecules. The RNA 
molecules produced by replication contain a nucleotide sequence which may be 
translated to produce a protein of interest or which encode one or more 
untranslated RNA molecules. Thus the expression system enables the production 
of recombinant Sinbis virus particles. 

Example 2 provides details on the production of the E2-JUN Sinbis non- 
natural molecular scaffold of the invention. Additionally provided in Example 3 
is another method for the production of recombinant E2-J r LWSinbis virus scaffold 
using the pTE5 '2 JE2 JUN vector produced in Example 1. Thus the invention 
provides two means, the pCYTts expression system (Example 2) and the pTE5" 2J 
vector system (Example 3) by which recombinant Sinbis virus B2-JUN non-natural 
molecular scaffold may be produced. An analysis of viral particles produced in 
each system is provided in Figure 1 and Figure 2. 

As previously stated, the invention includes viral-based core particles 
which comprise, or alternatively consist of, a virus, virus-like particle, a phage, a 
viral capsid particle or a recombinant form thereof. Skilled artisans have the 
knowledge to produce such core particles and attach organizers thereto. By way 
of providing other examples, the invention provides herein for the production of 
Hepatitis B virus-like particles and measles viral capsid particles as core particles 



-34- 



(Examples 17 to 22). In such an embodiment, the JUN leucine zipper protein 
domain or FOS leucine zipper protein domain may be used as an organizer, and 
hence as a first attachment site, for the non-natural molecular scaffold of the 
invention. 

[0136] Examples 23-29 provide details of the production of Hepatitis B core 

particles carrying an in-frame fused peptide with a reactive lysine residue and 
antigens carrying a genetically fused cysteine residue, as first and second 
attachment site, respectively. 

[0137] In other embodiments, the core particles used in compositions of the 

invention are composed of a Hepatitis B capsid (core) protein (HBcAg), or 
fragment thereof, which has been modified to either eliminate or reduce the 
number of free cysteine residues. Zhou et al. (J. Virol. 66:5393-5398 (1992)) 
demonstrated that HBcAgs which have been modified to remove the naturally 
resident cysteine residues retain the ability to associate and form multimeric 
structures. Thus, core particles suitable for use in compositions of the invention 
include those comprising modified HBcAgs, or fragments thereof, in which one 
or more of the naturally resident cysteine residues have been either deleted or 
substituted with another amino acid residue (e.g., a serine residue). 

[0138] The HBcAg is a protein generated by the processing of a Hepatitis B core 

antigen precursor protein. A number of isotypes of the HBcAg have been 
identified. For example, the HBcAg protein having the amino acid sequence 
shown in SEQ ID NO: 132 is 183 amino acids in length and is generated by the 
processing of a 212 amino acid Hepatitis B core antigen precursor protein. This 
processing results in the removal of 29 amino acids from the N-terminus of the 
Hepatitis B core antigen precursor protein. Similarly, the HBcAg protein having 
the amino acid sequence shown in SEQ ID NO 134 is 185 amino acids in length 
and is generated by the processing of a 214 amino acid Hepatitis B core antigen 
precursor protein The amino acid sequence shown in SEQ ID NO: 134, as 
compared to the amino acid sequence shown in SEQ ID NO: 132, contains a two 
amino acid insert at positions 152 and 153 in SEQ ID NO: 134. 



-35- 



[0139] In most instances, vaccine compositions of the invention will be prepared 

using the processed form of a HBcAg (i.e., a HBcAg from which the N-terminal 
leader sequence (e.g., the first 29 amino acid residues shown in SEQ ID NO: 134) 
of the Hepatitis B core antigen precursor protein have been removed). 

[0140] Further, when HBcAgs are produced under conditions where processing 

will not occur, the HBcAgs will generally be expressed in "processed" form. For 
example, bacterial systems, such as E. coli, generally do not remove the leader 
sequences of proteins which are normally expressed in eukaryotic cells. Thus, 
when an E. coli expression system is used to produce HBcAgs of the invention, 
these proteins will generally be expressed such that the N-terminal leader sequence 
of the Hepatitis B core antigen precursor protein is not present. 

[0141] In one embodiment of the invention, a modified HBcAg comprising the 

amino acid sequence shown in SEQ ID NO: 1 34, or subportion thereof, is used to 
prepare non-natural molecular scaffolds. In particular, modified HBcAgs suitable 
for use in the practice of the invention include proteins in which one or more of 
the cysteine residues at positions corresponding to positions 48, 61, 107 and 185 
of a protein having the amino acid sequence shown in SEQ ID NO: 134 have been 
either deleted or substituted with other amino acid residues (e.g. , a serine residue) 
As one skilled in the art would recognize, cysteine residues at similar locations in 
HBcAg variants having amino acids sequences which differ from that shown in 
SEQ ID NO: 134 could also be deleted or substituted with other amino acid 
residues. The modified HBcAg variants can then be used to prepare vaccine 
compositions of the invention. 

[0142] The present invention also includes HBcAg variants which have been 

modified to delete or substitute one or more additional cysteine residues which are 
not found in polypeptides having the amino acid sequence shown in SEQ ID 
NO: 134. Examples of such HBcAg variants have the amino acid sequences 
shown in SEQ ID NOs:90 and 132. These variant contain cysteines residues at 
a position corresponding to amino acid residue 147 in SEQ ID NO: 134. Thus, the 
vaccine compositions of the invention include compositions comprising HBcAgs 



-36- 



in which cysteine residues not present in the amino acid sequence shown in SEQ 
ID NO: 134 have been deleted. 

[0143] Under certain circumstances (e.g. , when aheterobifunctional cross-linking 

reagent is used to attach antigens or antigenic determinants to the non-natural 
molecular scaffold), the presence of free cysteine residues in the HBcAg is 
believed to lead to covalent coupling of toxic components to core particles, as 
well as the cross-linking of monomers to form undefined species. 

[0144] Further, in many instances, these toxic components may not be detectable 

with assays performed on compositions of the invention. This is so because 
covalent coupling of toxic components to the non-natural molecular scaffold 
would result in the formation of a population of diverse species in which toxic 
components are linked to different cysteine residues, or in some cases no cysteine 
residues, of the HBcAgs. In other words, each free cysteine residue of each 
HBcAg will not be covalently linked to toxic components. Further, in many 
instances, none of the cysteine residues of particular HBcAgs will be linked to 
toxic components. Thus, the presence of these toxic components may be difficult 
to detect because they would be present in a mixed population of molecules. The 
administration to an individual of HBcAg species containing toxic components, 
however, could lead to a potentially serious adverse reaction. 

[0145] It is well known in the art that free cysteine residues can be involved in a 

number of chemical side reactions. These side reactions include disulfide 
exchanges, reaction with chemical substances or metabolites that are, for example, 
injected or formed in a combination therapy with other substances, or direct 
oxidation and reaction with nucleotides upon exposure to UV light. Toxic 
adducts could thus be generated, especially considering the fact that HBcAgs have 
a strong tendency to bind nucleic acids Detection of such toxic products in 
antigen-capsid conjugates would be difficult using capsids prepared using HBcAgs 
containing free cysteines and heterobifunctional cross-linkers, since a distribution 
of products with a broad range of molecular weight would be generated. The 
toxic adducts would thus be distributed between a multiplicity of species, which 



-37- 



individually may each be present at low concentration, but reach toxic levels when 
together. 

[0146] In view of the above, one advantage to the use of HBcAgs in vaccine 

compositions which have been modified to remove naturally resident cysteine 
residues is that sites to which toxic species can bind when antigens or antigenic 
determinants are attached to the non-natural molecular scaffold would be reduced 
in number or eliminated altogether. Further, a high concentration of cross-linker 
can be used to produce highly decorated particles without the drawback of 
generating a plurality of undefined cross-linked species of HBcAg monomers (i. e. , 
a diverse mixture of cross-linked monomeric HbcAgs). 

[0147] A number of naturally occurring HBcAg variants suitable for use in the 

practice of the present invention have been identified. Yuan et al, (J. Virol. 
73:10122-10128 (1999)), for example, describe variants in which the isoleucine 
residue at position corresponding to position 97 in SEQ ID NO: 134 is replaced 
with either a leucine residue or a phenylalanine residue. The amino acid sequences 
of a number of HBcAg variants, as well as several Hepatitis B core antigen 
precursor variants, are disclosed in GenBank reports AAF121240 (SEQ ID 
NO:89), AF121239 (SEQ ID NO:90), X85297 (SEQ ID NO:91), X02496 (SEQ 
ID NO:92), X85305 (SEQ ID NO:93), X85303 (SEQ ID NO:94), AF1 51735 
(SEQ ID NO:95), X85259 (SEQ ID NO:96), X85286 (SEQ IDNO:97), X85260 
(SEQ ID NO:98), X85317 (SEQ ID NO:99), X85298 (SEQ ID NO: 100), 
AF043593 (SEQ ID NO: 101), M20706 (SEQ ID NO: 102), X85295 (SEQ ID 
NO: 103), X80925 (SEQ ID NO: 104), X85284 (SEQ ID NO: 105), X85275 (SEQ 
ID NO: 106), X72702 (SEQ ID NO: 107), X85291 (SEQ ID NO: 108), X65258 
(SEQ ID NO: 109), X85302 (SEQ ID NO: 110), M32138 (SEQ ID NO: 111), 
X85293 (SEQ ID NO' 112), X85315 (SEQ ID NO: 113), U95551 (SEQ ID 
NO 1 14), X85256 (SEQ ID NO: 1 15), X853 16 (SEQ ID NO: 1 16), X85296 (SEQ 
ID NO : 1 1 7), AB033 559 (SEQ ID NO: 1 1 8), X59795 (SEQ ID NO: 1 19), X85299 
(SEQ ID NO: 120), X85307 (SEQ ID NO: 121), X65257 (SEQ ID NO: 122), 
X85311 (SEQ ID NO: 123), X85301 (SEQ ID NO: 124), X85314 (SEQ ID 



-38- 



NO: 125), X85287 (SEQ ID NO: 126), X85272 (SEQ ID NO: 127), X853 19 (SEQ 
ID NO: 128), ABO 10289 (SEQ ID NO: 129), X85285 (SEQ ID NO: 130), 
AB010289 (SEQ ID NO: 131), AF121242 (SEQ ID NO: 132), M90520 (SEQ ID 
NO:135),P03153(SEQIDNO:136), AF110999(SEQIDNO:137),andM95589 
(SEQ ID NO: 138), the disclosures of each of which are incorporated herein by 
reference. These HBcAg variants differ in amino acid sequence at a number of 
positions, including amino acid residues which corresponds to the amino acid 
residues located at positions 12, 13, 21, 22, 24, 29, 32, 33, 35, 38, 40, 42, 44, 45, 
49, 51, 57, 58, 59, 64, 66, 67, 69, 74, 77, 80, 81, 87, 92, 93, 97, 98, 100, 103, 
105, 106, 109, 113, 116, 121, 126, 130, 133, 135, 141, 147, 149, 157, 176, 178, 
182 and 183 in SEQ ID NO: 134. The invention is also directed to amino acid 
sequences that are at least 65, 0, 75, 80, 85, 90 or 95% identical to the above 
Hepatitis B viral capsid protein sequences. HBcAgs suitable for use in the present 
invention may be derived from any organism so long as they are able to associate 
to form an ordered and repetitive antigen array 

[0148] Asnotedabove, generally processed HBcAgs (i.e., those which lack leader 

sequences) will be used in the vaccine compositions of the invention. Thus, when 
HBcAgs having amino acid sequence shown in SEQ ID NOs: 136, 137, or 138 are 
used to prepare vaccine compositions of the invention, generally 30, 35-43, or 
35-43 amino acid residues at the N-terminus, respectively, of each of these 
proteins will be omitted. 

[0149] The present invention includes vaccine compositions, as well as methods 

for using these compositions, which employ the above described variant HBcAgs 
for the preparation of non-natural molecular scaffolds. 

[0150] Further included withing the scope of the invention are additional HBcAg 

variants which are capable of associating to form dimeric or multimeric structures. 
Thus, the invention further includes vaccine compositions comprising HBcAg 
polypeptides comprising, or alternatively consisting of, amino acid sequences 
which are at least 80%, 85%, 90%, 95%, 97%, or 99% identical to any of the 
amino acid sequences shown in SEQ ID NOs :8 9- 132 and 134-138, and forms of 



-39- 



these proteins which have been processed, where appropriate, to remove the 
N-terminal leader sequence. 
[0151] Whether the amino acid sequence of a polypeptide has an amino acid 

sequence that is at least 80%, 85%, 90%, 95%, 97%, or 99% identical to one of 
the amino acid sequences shown in SEQ ID NOs:89-132 and 134-138, or a 
subportion thereof, can be determined conventionally using known computer 
programs such the Bestfit program. When using Bestfit or any other sequence 
alignment program to determine whether a particular sequence is, for instance, 
95% identical to a reference amino acid sequence according to the present 
invention, the parameters are set such that the percentage of identity is calculated 
over the full length of the reference amino acid sequence and that gaps in 
homology of up to 5% of the total number of amino acid residues in the reference 
sequence are allowed. 

[0152] The HBcAg variants and precursors having the amino acid sequences set 

out in SEQ ID NOs:89-132 and 134-136 are relatively similar to each other. 
Thus, reference to an amino acid residue of a HBcAg variant located at a position 
which corresponds to a particular position in SEQ ID NO : 1 34, refers to the amino 
acid residue which is present at that position in the amino acid sequence shown in 
SEQ ID NO: 134. The homology between these HBcAg variants is for the most 
part high enough among Hepatitis B viruses that infect mammals so that one 
skilled in the art would have little difficulty reviewing both the amino acid 
sequence shown in SEQ ID NO: 134 and that of a particular HBcAg variant and 
identifying "corresponding" amino acid residues. For example, the HBcAg amino 
acid sequence shown in SEQ ID NO: 135, which shows the amino acid sequence 
of a HBcAg derived from a virus which infect woodchucks, has enough homology 
to the HBcAg having the amino acid sequence shown in SEQ ID NO: 134 that it 
is readily apparent that a three amino acid residue insert is present in SEQ ID 
NO: 135 between amino acid residues 155 and 156 of SEQ ID NO: 134 

[0153] The HBcAgs of Hepatitis B viruses which infect snow geese and ducks 

differ enough from the amino acid sequences of HBcAgs of Hepatitis B viruses 



-40- 



which infect mammals that alignment of these forms of this protein with the amino 
acid sequence shown in SEQ ID NO: 134 is difficult. However, the invention 
includes vaccine compositions which comprise HBcAg variants of Hepatitis B 
viruses which infect birds, as wells as vaccine compositions which comprise 
fragments of these HBcAg variants. HBcAg fragments suitable for use in 
preparing vaccine compositions of the invention include compositions which 
contain polypeptide fragments comprising, or alternatively consisting of, amino 
acid residues selected from the group consisting of 36-240, 36-269, 44-240, 
44-269, 36-305, and 44-305 of SEQ ID NO: 137 or 36-240, 36-269, 44-240, 
44-269, 36-305, and 44-305 of SEQ ID NO: 138 As one skilled in the art would 
recognize, one, two, three or more of the cysteine residues naturally present in 
these polypeptides (e.g., the cysteine residues at position 153 is SEQ ID NO:137 
or positions 34, 43, and 196 in SEQ ID NO: 138) could be either substituted with 
another amino acid residue or deleted prior to their inclusion in vaccine 
compositions of the invention. 

[0154] In one embodiment, the cysteine residues at positions 48 and 107 of a 

protein having the amino acid sequence shown in SEQ ID NO: 134 are deleted or 
substituted with another amino acid residue but the cysteine at position 61 is left 
in place. Further, the modified polypeptide is then used to prepare vaccine 
compositions of the invention. 

[0155] As set out below in Example 3 1 , the cysteine residues at positions 48 and 

107, which are accessible to solvent, may be removed, for example, by 
site-directed mutagenesis. Further, the inventors have found that the Cys-48-Ser, 
Cys-107-Ser HBcAg double mutant constructed as described in Example 3 1 can 
be expressed in E. coli. 

[0156] As discussed above, the elimination of free cysteine residues reduces the 

number of sites where toxic components can bind to the HBcAg, and also 
eliminates sites where cross-linking of lysine and cysteine residues of the same or 
of neighboring HBcAg molecules can occur. The cysteine at position 61, which 
is involved in dimer formation and forms a disulfide bridge with the cysteine at 



-41- 



position 61 of another HBcAg, will normally be left intact for stabilization of 
HBcAg dimers and multimers of the invention. 

[0157] As shown in Example 32, cross-linking experiments performed with 

(1 ) HBcAgs containing free cysteine residues and (2) HBcAgs whose free cysteine 
residues have been made unreactive with iodacetamide, indicate that free cysteine 
residues of the HBcAg are responsible for cross-linking between HBcAgs through 
reactions between heterobifunctional cross-linker derivatized lysine side chains, 
and free cysteine residues Example 32 also indicates that cross-linking of HBcAg 
subunits leads to the formation of high molecular weight species of undefined size 
which cannot be resolved by SDS-polyacrylamide gel electrophoresis. 

[0158] When an antigen or antigenic determinant is linked to the non-natural 

molecular scaffold through a lysine residue, it may be advantageous to either 
substitute or delete one or both of the naturally resident lysine residues located at 
positions corresponding to positions 7 and 96 in SEQ ID NO: 134, as well as other 
lysine residues present in HBcAg variants. The elimination of these lysine residues 
results in the removal of binding sites for antigens or antigenic determinants which 
could disrupt the ordered array and should improve the quality and uniformity of 
the final vaccine composition. 

[0159] In many instances, when both of the naturally resident lysine residues at 

positions corresponding to positions 7 and 96 in SEQ ID NO: 134 are eliminated, 
another lysine will be introduced into the HBcAg as an attachment site for an 
antigen or antigenic determinant. Methods for inserting such a lysine residue are 
set out, for example, in Example 23 below. It will often be advantageous to 
introduce a lysine residue into the HBcAg when, for example, both of the naturally 
resident lysine residues at positions corresponding to positions 7 and 96 in SEQ 
ID NO .134 are altered and one seeks to attach the antigen or antigenic 
determinant to the non-natural molecular scaffold using a heterobifunctional 
cross-linking agent 

[0160] The C-terminus of the HBcAg has been shown to direct nuclear 

localization of this protein (Eckhardt et al., J. Virol. 65:575-582 (1991).) 



-42- 



Further, this region of the protein is also believed to confer upon the HBcAg the 
ability to bind nucleic acids. 
[0161] In some embodiments, vaccine compositions of the invention will contain 

HBcAgs which have nucleic acid binding activity (e.g., which contain a naturally 
resident HBcAg nucleic acid binding domain). HBcAgs containing one or more 
nucleic acid binding domains are useful for preparing vaccine compositions which 
exhibit enhanced T-cell stimulatory activity. Thus, the vaccine compositions of 
the invention include compositions which contain HBcAgs having nucleic acid 
binding activity. Further included are vaccine compositions, as well as the use of 
such compositions in vaccination protocols, where HBcAgs are bound to nucleic 
acids. These HBcAgs may bind to the nucleic acids prior to administration to an 
individual or may bind the nucleic acids after administration. 
[0162] In other embodiments, vaccine compositions of the invention will contain 

HBcAgs from which the C-terminal region (e.g., amino acid residues 145-185 or 
150-185 of SEQ ID NO: 134) has been removed and do not bind nucleic acids. 
Thus, additional modified HBcAgs suitable for use in the practice of the present 
invention include C-terminal truncation mutants. Suitable truncation mutants 
include HBcAgs where 1, 5, 10, 15, 20, 25, 30, 34, 35, 36, 37, 38, 39 40, 41, 42 
or 48 amino acids have been removed from the C-terminus. 
[0163] HBcAgs suitable for use in the practice of the present invention also 

include N-terminal truncation mutants. Suitable truncation mutants include 
modified HBcAgs where 1, 2, 5, 7, 9, 10, 12, 14, 15, and 17 amino acids have 
been removed from the N-terminus. 
[0164] Further HBcAgs suitable for use in the practice of the present invention 

include N- and C-terminal truncation mutants. Suitable truncation mutants include 
HBcAgs where 1, 2, 5, 7, 9, 10, 12, 14, 15, and 17 amino acids have been 
removed from the N-terminus and 1, 5, 10, 15, 20, 25, 30, 34, 35, 36, 37, 38, 39 
40, 41, 42 or 48 amino acids have been removed from the C-terminus. 
[0165] The invention further includes vaccine compositions comprising HBcAg 

polypeptides comprising, or alternatively consisting of, amino acid sequences 



-43- 



which are at least 80%, 85%, 90%, 95%, 97%, or 99% identical to the above 
described truncation mutants. 
[0166] As discussed above, in certain embodiments of the invention, a lysine 

residue is introduced as a first attachment site into a polypeptide which forms the 
non-natural molecular scaffold. In preferred embodiments, vaccine compositions 
of the invention are prepared using a HBcAg comprising, or alternatively 
consisting of, amino acids 1-144 or amino acids 1-149 of SEQ ID NO: 1 34 which 
is modified so that the amino acids corresponding to positions 79 and 80 are 
replaced with a peptide having the amino acid sequence of Gly-Gly-Lys-Gly-Gly 
(SEQ ID NO: 158) and the cysteine residues at positions 48 and 107 are either 
deleted or substituted with another amino acid residue, while the cysteine at 
position 61 is left in place. The invention further includes vaccine compositions 
comprising corresponding fragments of polypeptides having amino acid sequences 
shown in any of SEQ ID NOs:89-132 and 135-136 which also have the above 
noted amino acid alterations. 
[0167] The invention further includes vaccine compositions comprising fragments 

of a HBcAg comprising, or alternatively consisting of, an amino acid sequence 
other than that shown in SEQ ID NO: 134 from which a cysteine residue not 
present at corresponding location in SEQ ID NO: 134 has been deleted. One 
example of such a fragment would be a polypeptide comprising, or alternatively 
consisting of, amino acids amino acids 1-149 of SEQ ID NO:132 where the 
cysteine residue at position 147 has been either substituted with another amino 
acid residue or deleted. 
[0168] The invention further includes vaccine compositions comprising HBcAg 

polypeptides comprising, or alternatively consisting of, amino acid sequences 
which are at least 80%, 85%, 90%, 95%, 97%, or 99% identical to amino acids 
1-144 or 1-149 of SEQ ID NO. 134 and corresponding subportions of a 
polypeptide comprising an amino acid sequence shown in any of SEQ ID 
NOs:89-132 or 134-136, as well as to amino acids 1-147 or 1-152 of SEQ ID 
NO:158. 



-44- 



[0169] The invention also includes vaccine compositions comprising HBcAg 

polypeptides comprising, or alternatively consisting of, amino acid sequences 
which are at least 80%, 85%, 90%, 95%, 97%, or 99% identical to amino acids 
36-240, 36-269, 44-240, 44-269, 36-305, and 44-305 of SEQ ID NO: 137 or 
36-240, 36-269, 44-240, 44-269, 36-305, and 44-305 of SEQ ID NO: 138. 

[0170] Vaccine compositions of the invention may comprise mixtures of different 

HBcAgs Thus, these vaccine compositions may be composed of HBcAgs which 
differ in amino acid sequence. For example, vaccine compositions could be 
prepared comprising a "wild-type" HBcAg and a modified HBcAg in which one 
or more amino acid residues have been altered (e.g., deleted, inserted or 
substituted). In most applications, however, only one type of a HBcAg, or at least 
HBcAgs having essentially equivalent first attachment sites, will be used because 
vaccines prepared using such HBcAgs will present highly ordered and repetitive 
arrays of antigens or antigenic determinants. Further, preferred vaccine 
compositions of the invention are those which present highly ordered and 
repetitive antigen arrays. 

[0171] The invention further includes vaccine compositions where the non-natural 

molecular scaffold is prepared using a HBcAg fused to another protein. As 
discussed above, one example of such a fusion protein is a HBcAg/FOS fusion. 
Other examples of HBcAg fusion proteins suitable for use in vaccine compositions 
of the invention include fusion proteins where an amino acid sequence has been 
added which aids in the formation and/or stabilization of HBcAg dimers and 
multimers This additional amino acid sequence may be fused to either the N- or 
C-terminus of the HBcAg. One example, of such a fusion protein is a fusion of 
a HBcAg with the GCN4 helix region of Saccharomyces cerevisiae (GenBank 
Accession No. P03069 (SEQ ID NO: 154)). 
[0172] The helix domain of the GCN4 protein forms homodimers via 

non-covalent interactions which can be used to prepare and stabilize HBcAg 
dimers and multimers. 



-45- 



[0173] In one embodiment, the invention provides vaccine compositions prepared 

using HBcAg fusions proteins comprising a HBcAg, or fragment thereof, with a 
GCN4 polypeptide having the sequence of amino acid residues 227 to 276 in SEQ 
ID NO 154 fused to the C-terminus. This GCN4 polypeptide may also be fused 
to the N-terminus of the HbcAg. 

[0174] HBcAg/src homology 3 (SH3) domain fusion proteins could also be used 

to prepare vaccine compositions of the invention. SH3 domains are relatively 
small domains found in a number of proteins which confer the ability to interact 
with specific proline-rich sequences in protein binding partners {see McPherson, 
Cell Signal 77:229-238 (1999). HBcAg/SH3 fusion proteins could be used in 
several ways First, the SH3 domain could form a first attachment site which 
interacts with a second attachment site of the antigen or antigenic determinant. 
Similarly, a proline rich amino acid sequence could be added to the HBcAg and 
used as a first attachment site for an SH3 domain second attachment site of an 
antigen or antigenic determinant. Second, the SH3 domain could associate with 
proline rich regions introduced into HBcAgs. Thus, SH3 domains and proline rich 
SH3 interaction sites could be inserted into either the same or different HBcAgs 
and used to form and stabilized dimers and multimers of the invention. 

[0175] In other embodiments, a bacterial pilin, a subportion of a bacterial pilin, or 

a fusion protein which contains either a bacterial pilin or subportion thereof is used 
to prepare vaccine compositions of the invention. Examples of pilin proteins 
include pilins produced by Escherichia coli, Haemophilus influenzae, Neisseria 
meningitidis, Neisseria gonorrhoeae, Caulobacter crescentus, Pseudomonas 
stutzeri, and Pseudomonas aeruginosa. The amino acid sequences of pilin 
proteins suitable for use with the present invention include those set out in 
GenBank reports AJ000636 (SEQ ID NO: 139), AJ 132364 (SEQ ID NO: 140), 
AF229646 (SEQ ID NO:141), AF051814 (SEQ ID NO: 142), AF051815 (SEQ 
ID NO: 143), and X0098 1 (SEQ ID NO: 155), the entire disclosures of which are 
incorporated herein by reference. 



-46- 



[0176] Bacterial pilin proteins are generally processed to remove N-terminal 

leader sequences prior to export of the proteins into the bacterial periplasm. 
Further, as one skilled in the art would recognize, bacterial pilin proteins used to 
prepare vaccine compositions of the invention will generally not have the naturally 
present leader sequence. 

[0177] One specific example of a pilin protein suitable for use in the present 

invention is the P-pilin ofE. coli (GenBank report AF237482 (SEQ ID NO: 144)). 
An example of a Type-1 E. coli pilin suitable for use with the invention is a pilin 
having the amino acid sequence set out in GenBank report P04128 (SEQ ID 
NO: 146), which is encoded by nucleic acid having the nucleotide sequence set out 
in GenBank report M27603 (SEQ ID NO: 145). The entire disclosures of these 
GenBank reports are incorporated herein by reference. Again, the mature form 
of the above referenced protein would generally be used to prepare vaccine 
compositions of the invention. Another example of a pilin protein is SEQ ID NO: 
184 , which is identical to SEQ ID NO: 146, except that in SEQ ID NO: 146, 
amino acid 20 is threonine, but in SEQ ID NO: 184, amino acid 20 is alanine. 

[0178] Bacterial pilins or pilin subportions suitable for use in the practice of the 

present invention will generally be able to associate to form non-natural molecular 
scaffolds. 

[0179] Methods for preparing pili and pilus-like structures in vitro are known in 

the art. Bullitt et al, Proc. Natl. Acad. Sci. USA 93:12890-12895 (1996), for 
example, describe the in vitro reconstitution of E. coli P-pili subunits. Further, 
Eshdat et al, J. Bacteriol. / 45:308-314 (1981) describe methods suitable for 
dissociating Type-1 pili of E. coli and the reconstitution of pili. In brief, these 
methods are as follows: pili are dissociated by incubation at 37 °C in saturated 
guanidine hydrochloride. Pilin proteins are then purified by chromatography, after 
which pilin dimers are formed by dialysis against 5 mM tris(hydroxymethyl) 
aminomethane hydrochloride (pH 8.0). Eshdat et al. also found that pilin dimers 
reassemble to form pili upon dialysis against the 5 mM tris(hydroxymethyl) 
aminomethane (pH 8.0) containing 5 mM MgCl 2 . 



-47- 



[0180] Further, using, for example, conventional genetic engineering and protein 

modification methods, pilin proteins may be modified to contain a first attachment 
site to which an antigen or antigenic determinant is linked through a second 
attachment site. Alternatively, antigens or antigenic determinants can be directly 
linked through a second attachment site to amino acid residues which are naturally 
resident in these proteins. These modified pilin proteins may then be used in 
vaccine compositions of the invention. 

[0181] Bacterial pilin proteins used to prepare vaccine compositions of the 

invention may be modified in a manner similar to that described herein for HBcAg. 
For example, cysteine and lysine residues may be either deleted or substituted with 
other amino acid residues and first attachment sites may be added to these 
proteins. Further, pilin proteins may either be expressed in modified form or may 
be chemically modified after expression. Similarly, intact pili may be harvested 
from bacteria and then modified chemically. 

[0182] In another embodiment, pili or pilus-like structures are harvested from 

bacteria (e.g., E. coli) and used to form vaccine compositions of the invention. 
One example of pili suitable for preparing vaccine compositions is the Type- 1 pilus 
of E. coli, which is formed from pilin monomers having the amino acid sequence 
set out in SEQ ID NO: 146. 

[0183] A number of methods for harvesting bacterial pili are known in the art. 

Bullitt and Makowski (Biophys. J. 74 623-632 (1998)), for example, describe a 
pilus purification method for harvesting P-pili from E. coli. According to this 
method, pili are sheared from hyperpiliated E. coli containing a P-pilus plasmid 
and purified by cycles of solubilization and MgCl 2 (1 .0 M) precipitation. A similar 
purification method is set out below in Example 33. 

[0184] Once harvested, pili or pilus-like structures may be modified in a variety 

of ways. For example, a first attachment site can be added to the pili to which 
antigens or antigen determinants may be attached through a second attachment 
site. In other words, bacterial pili or pilus-like structures can be harvested and 
modified to form non-natural molecular scaffolds. 



-48- 



[0185] Pili or pilus-like structures may also be modified by the attachment of 

antigens or antigenic determinants in the absence of a non-natural organizer. For 
example, antigens or antigenic determinants could be linked to naturally occurring 
cysteine resides or lysine residues. In such instances, the high order and 
repetitiveness of a naturally occurring amino acid residue would guide the 
coupling of the antigens or antigenic determinants to the pili or pilus-like 
structures. For example, the pili or pilus-like structures could be linked to the 
second attachment sites of the antigens or antigenic determinants using a 
heterobifunctional cross-linking agent. 

[0186] When structures which are naturally synthesized by organisms (e.g., pili) 

are used to prepare vaccine compositions of the invention, it will often be 
advantageous to genetically engineer these organisms so that they produce 
structures having desirable characteristics. For example, when Type-1 pili of E. 
coli are used, the E. coli from which these pili are harvested may be modified so 
as to produce structures with specific characteristics. Examples of possible 
modifications of pilin proteins include the insertion of one or more lysine residues, 
the deletion or substitution of one or more of the naturally resident lysine residues, 
and the deletion or substitution of one or more naturally resident cysteine residues 
(e.g., the cysteine residues at positions 44 and 84 in SEQ ID NO: 146). 

[0187] Further, additional modifications can be made to pilin genes which result 

in the expression products containing a first attachment site other than a lysine 
residue (e.g., a FOS or JUN domain). Of course, suitable first attachment sites 
will generally be limited to those which do not prevent pilin proteins from forming 
pili or pilus-like structures suitable for use in vaccine compositions of the 
invention. 

[0188] Pilin genes which naturally reside in bacterial cells can be modified in vivo 

(e.g. , by homologous recombination) or pilin genes with particular characteristics 
can be inserted into these cells. For examples, pilin genes could be introduced into 
bacterial cells as a component of either a replicable cloning vector or a vector 



-49- 



which inserts into the bacterial chromosome. The inserted pilin genes may also 
be linked to expression regulatory control sequences (e.g., a lac operator). 

[0189] In most instances, the pili or pilus-like structures used in vaccine 

compositions of the invention will be composed of single type of a pilin subunit. 
Pili or pilus-like structures composed of identical subunits will generally be used 
because they are expected to form structures which present highly ordered and 
repetitive antigen arrays. 

[0190] However, the compositions of the invention also include vaccines 

comprising pili or pilus-like structures formed from heterogenous pilin subunits. 
The pilin subunits which form these pili or pilus-like structures can be expressed 
from genes naturally resident in the bacterial cell or may be introduced into the 
cells. When a naturally resident pilin gene and an introduced gene are both 
expressed in a cell which forms pili or pilus-like structures, the result will generally 
be structures formed from a mixture of these pilin proteins. Further, when two or 
more pilin genes are expressed in a bacterial cell, the relative expression of each 
pilin gene will typically be the factor which determines the ratio of the different 
pilin subunits in the pili or pilus-like structures. 

[0191] When pili or pilus-like structures having a particular composition of mixed 

pilin subunits is desired, the expression of at least one of the pilin genes can be 
regulated by a heterologous, inducible promoter. Such promoters, as well as 
other genetic elements, can be used to regulate the relative amounts of different 
pilin subunits produced in the bacterial cell and, hence, the composition of the pili 
or pilus-like structures. 

[0192] In additional, while in most instances the antigen or antigenic determinant 

will be linked to bacterial pili or pilus-like structures by a bond which is not a 
peptide bond, bacterial cells which produce pili or pilus-like structures used in the 
compositions of the invention can be genetically engineered to generate pilin 
proteins which are fused to an antigen or antigenic determinant. Such fusion 
proteins which form pili or pilus-like structures are suitable for use in vaccine 
compositions of the invention. 



-50- 



[0193] The inventors surprisingly found that bacterial Pili induced an antibody 

response dominated by the IgGl isotype in mince. This type of antibodies is 
indicative for a Th2 response. Moreover, antigens coupled to Pili also induced a 
IgGl response indicating that coupling of antigens to Pili was sufficient for 
induction of antigen-specific Th2 responses. 

B. Construction of an Antigen or Antigenic Determinant with a 
Second Attachment Site 

[0194] The second element in the compositions of the invention is an antigen or 

antigenic determinant possessing at least one second attachment site capable of 
association through at least one non-peptide bond to the first attachment site of 
the non-natural molecular scaffold. The invention provides for compositions that 
vary according to the antigen or antigenic determinant selected in consideration 
of the desired therapeutic effect. Other compositions are provided by varying the 
molecule selected for the second attachment site. 

[0195] However, when bacterial pili, or pilus-like structures, pilin proteins are 

used to prepare vaccine compositions of the invention, antigens or antigenic 
determinants may be attached to pilin proteins by the expression of pilin/antigen 
fusion proteins. Antigen and antigenic determinants may also be attached to 
bacterial pili, or pilus-like structures, pilin proteins through non-peptide bonds. 

[0196] Antigens of the invention maybe selected from the group consisting of the 

following: (a) proteins suited to induce an immune response against cancer cells; 
(b) proteins suited to induce an immune response against infectious diseases; (c) 
proteins suited to induce an immune response against allergens ,(d) proteins suited 
to induce an immune response in farm animals, and (e) fragments (e.g. , a domain) 
of any of the proteins set out in (a)-(d). 

[0197] In one specific embodiment of the invention, the antigen or antigenic 

determinant is one that is useful for the prevention of infectious disease. Such 
treatment will be useful to treat a wide variety of infectious diseases affecting a 
wide range of hosts, e.g., human, cow, sheep, pig, dog, cat, other mammalian 



-51- 



species and non-mammalian species as well. Treatable infectious diseases are well 
known to those skilled in the art, examples include infections of viral etiology such 
as HIV, influenza, Herpes, viral hepatitis, Epstein Bar, polio, viral encephalitis, 
measles, chicken pox, etc.; or infections of bacterial etiology such as pneumonia, 
tuberculosis, syphilis, etc.; or infections of parasitic etiology such as malaria, 
trypanosomiasis, leishmaniasis, trichomoniasis, amoebiasis, etc. Thus, antigens or 
antigenic determinants selected for the compositions of the invention will be well 
known to those in the medical art; examples of antigens or antigenic determinants 
include the following: the HIV antigens gp 1 40 and gp 1 60; the influenaza antigens 
hemagglutinin and neuraminidase, Hepatitis B surface antigen, circumsporozoite 
protein of malaria. 

[0198] In another specific embodiment, compositions of the invention are an 

immunotherapeutic that may be used for the treatment of allergies or cancer. 

[0199] The selection of antigens or antigenic determinants for compositions and 

methods of treatment for allergies would be known to those skilled in the medical 
art treating such disorders; representative examples of this type of antigen or 
antigenic determinant include the following: bee venom phospholipase A 2 , Bet v 
I (birch pollen allergen), 5 Dol m V (white-faced hornet venom allergen), Der p 
I (House dust mite allergen). 

[0200] The selection of antigens or antigenic determinants for compositions and 

methods of treatment for cancer would be known to those skilled in the medical 
art treating such disorders; representative examples of this type of antigen or 
antigenic determinant include the following. Her2 (breast cancer), GD2 
(neuroblastoma), EGF-R (malignant glioblastoma), CEA (medullary thyroid 
cancer), CD52 (leukemia). 

[0201] In a particular embodiment of the invention, the antigen or antigenic 

determinant is selected from the group consisting of: (a) a recombinant protein of 
HIV, (b) a recombinant protein of Influenza virus, (c) a recombinant protein of 
Hepatitis B virus, (d) a recombinant protein of Toxoplasma, (e) a recombinant 
protein of Plasmodium falciparum, (f) a recombinant protein of Plasmodium 



-52- 



vivax, (g) a recombinant protein of Plasmodium ovale, (h) a recombinant protein 
of Plasmodium malariae, (i) a recombinant protein of breast cancer cells, (j) a 
recombinant protein of kidney cancer cells, (k) a recombinant protein of prostate 
cancer cells, (1) a recombinant protein of skin cancer cells, (m) a recombinant 
protein of brain cancer cells, (n) a recombinant protein of leukemia cells, (o) a 
recombinant profiling, (p) a recombinant protein of bee sting allergy, (q) a 
recombinant proteins of nut allergy, (r) a recombinant proteins of food allergies, 
(s) recombinant proteins of asthma, (t) a recombinant protein of Chlamydia, and 
(u) a fragment of any of the proteins set out in (a)-(t). 
[0202] Once the antigen or antigenic determinant of the composition is chosen, 

at least one second attachment site may be added to the molecule in preparing to 
construct the organized and repetitive array associated with the non-natural 
molecular scaffold of the invention. Knowledge of what will constitute an 
appropriate second attachment site will be known to those skilled in the art. 
Representative examples of second attachment sites include, but are not limited 
to, the following: an antigen, an antibody or antibody fragment, biotin, avidin, 
strepavidin, a receptor, a receptor ligand, a ligand, a ligand-binding protein, an 
interacting leucine zipper polypeptide, an amino group, a chemical group reactive 
to an amino group; a carboxyl group, chemical group reactive to a carboxyl 
group, a sulfhydryl group, a chemical group reactive to a sulfhydryl group, or a 
combination thereof. 

[0203] The association between the first and second attachment sites will be 

determined by the characteristics of the respective molecules selected but will 
comprise at least one non-peptide bond. Depending upon the combination of first 
and second attachment sites, the nature of the association may be covalent, ionic, 
hydrophobic, polar, or a combination thereof 

[0204] In one embodiment of the invention, the second attachment site may be the 

FOS leucine zipper protein domain or the JUN leucine zipper protein domain. 

[0205] In a more specific embodiment of the invention, the second attachment site 

selected is the FOS leucine zipper protein domain, which associates specifically 



-53- 



with the JUN leucine zipper protein domain of the non-natural molecular scaffold 
of the invention The association of the JUN and FOS leucine zipper protein 
domains provides a basis for the formation of an organized and repetitive antigen 
or antigenic determinant array on the surface of the scaffold. The FOS leucine 
zipper protein domain may be fused in frame to the antigen or antigenic 
determinant of choice at either the amino terminus, carboxyl terminus or internally 
located in the protein if desired. 

[0206] Several FOS fusion constructs are provided for exemplary purposes 

Human growth hormone (Example 4), bee venom phospholipase A 2 (PLA) 
(Example 9), ovalbumin (Example 10) and HIV gpl40 (Example 12). 

[0207] In order to simplify the generation of FOS fusion constructs, several 

vectors are disclosed that provide options for antigen or antigenic determinant 
design and construction (see Example 6). The vectors pAVl-4 were designed for 
the expression of FOS fusion in E. coli; the vectors pAV5 and pAV6 were 
designed for the expression of FOS fusion proteins in eukaryotic cells. Properties 
of these vectors are briefly described. 

[0208] 1 . pAVl : This vector was designed for the secretion of fusion proteins 

with FOS at the C-terminus into the E. coli periplasmic space. The gene of 
interest (g.o.i.) may be ligated into the StuI/NotI sites of the vector. 

[0209] 2. pAV2 : This vector was designed for the secretion of fusion proteins 

with FOS at the N-terminus into the E. coli periplasmic space. The gene of 
interest (g.o.i.) ligated into the Notl/EcoRV (or Notl/Hindlll) sites of the vector. 

[0210] 3. pAV3 : This vector was designed for the cytoplasmic production of 

fusion proteins with FOS at the C-terminus inE. coli The gene of interest (g.o.i.) 
may be ligated into the EcoRV/NotI sites of the vector. 

[0211] 4. pAV4 : This vector is designed for the cytoplasmic production of 

fusion proteins with FOS at the N-terminus in E. coli. The gene of interest (g. o i ) 
may be ligated into the Notl/EcoRV (or Notl/Hindlll) sites of the vector. The 
N-terminal methionine residue is proteolytically removed upon protein synthesis 
(Hirel et al, Proc. Natl. Acad. Sci. USA 5(5:8247-8251 (1989)). 



-54- 



[0212] 5. pAV5 . This vector was designed for the eukaryotic production of 

fusion proteins with FOS at the C-terminus. The gene of interest (g.o.i.) may be 
inserted between the sequences coding for the hGH signal sequence and the FOS 
domain by ligation into the Eco47III/NotI sites of the vector. Alternatively, a 
gene containing its own signal sequence may be fused to the FOS coding region 
by ligation into the StuI/NotI sites. 

[0213] 6. pAV6 : This vector was designed for the eukaryotic production of 

fusion proteins with FOS at the N-terminus. The gene of interest (g.o.i.) may be 
ligated into the Notl/StuI (or Notl/Hindlll) sites of the vector. 

[0214] As will be understood by those skilled in the art, the construction of a 

FO^-antigen or -antigenic determinant fusion protein may include the addition of 
certain genetic elements to facilitate production of the recombinant protein. 
Example 4 provides guidance for the addition of certain E. coli regulatory 
elements for translation, and Example 7 provides guidance for the addition of a 
eukaryotic signal sequence. Other genetic elements may be selected, depending 
on the specific needs of the practioner. 

[0215] The invention is also seen to include the production of theFOS-antigen or 

FO^-antigenic determinant fusion protein either in bacterial (Example 5) or 
eukaryotic cells (Example 8). The choice of which cell type in which to express 
the fusion protein is within the knowledge of the skilled artisan, depending on 
factors such as whether post-translational modifications are an important 
consideration in the design of the composition. 

[0216] As noted previously, the invention discloses various methods for the 

construction of a FOS-antigen or -FQS-antigenic determinant fusion protein 
through the use of the pAV vectors. In addition to enabling prokaryotic and 
eukaryotic expression, these vectors allow the practitioner to choose between Isl- 
and C-terminal addition to the antigen of the FOS leucine zipper protein domain. 
Specific examples are provided wherein N- and C-terminal FOS fusions are made 
to PL A (Example 9) and ovalbumin (Example 10). Example 1 1 demonstrates the 
purification of the PLA and ovalbumin FOS fusion proteins. 



-55- 



[0217] In a more specific embodiment, the invention is drawn to an antigen or 

antigenic determinant encoded by the HIV genome. More specifically, the HIV 
antigen is gpl40. As provided for in Examples 11-15, HIV gpl40 may be created 
with a FOS leucine zipper protein domain and the fusion protein synthesized and 
purified for attachment to the non-natural molecular scaffold of the invention. As 
one skilled in the art would know, other HIV antigens or antigenic determinants 
may be used in the creation of a composition of the invention. 

[0218] In a more specific embodiment of the invention, the second attachment site 

selected is a cysteine residue, which associates specifically with a lysine residue of 
the non-natural molecular scaffold of the invention. The chemical linkage of the 
lysine residue (Lys) and cysteine residue (Cys) provides a basis for the formation 
of an organized and repetitive antigen or antigenic determinant array on the 
surface of the scaffold The cysteine residue may be engineered in frame to the 
antigen or antigenic determinant of choice at either the amino terminus, carboxyl 
terminus or internally located in the protein if desired. By way of example, PLA 
and HIV gp 1 40 are provided with a cysteine residue for linkage to a lysine residue 
first attachment site. 

C. Preparation of the AlphaVaccine Particles 

[0219] The invention provides novel compositions and methods for the 

construction of ordered and repetitive antigen arrays. As one of skill in the art 
would know, the conditions for the assembly of the ordered and repetitive antigen 
array depend to a large extent on the specific choice of the first attachment site of 
the non-natural molecular scaffold and the specific choice of the second 
attachment site of the antigen or antigenic determinant. Thus, practitioner choice 
in the design of the composition (i.e. , selection of the first and second attachment 
sites, antigen and non-natural molecular scaffold) will determine the specific 
conditions for the assembly of the AlphaVaccine particle (the ordered and 
repetitive antigen array and non-natural molecular scaffold combined). 



-56- 



Information relating to assembly of the AlphaVaccine particle is well within the 
working knowledge of the practitioner, and numerous references exist to aid the 
practitioner (e.g., Sambrook, J et al, eds , MOLECULAR CLONING, A 
Laboratory Manual, 2nd. edition, Cold Spring Harbor Laboratory Press, Cold 
Spring Harbor, N.Y. (1989); Ausubel, F. et al, eds., CURRENT PROTOCOLS IN 
Molecular Biology, John H. Wiley & Sons, Inc. (1997); Celis, J., ed., CELL 
BIOLGY, Academic Press, 2 nd edition, (1998); Harlow, E. and Lane, D , 
"Antibodies: A Laboratory Manual," Cold Spring Harbor Laboratory, Cold Spring 
Harbor, N.Y. (1988), all of which are incorporated herein by reference. 

[0220] In a specific embodiment ofthe invention, the JUN and FOSleucine zipper 

protein domains are utilized for the first and second attachment sites of the 
invention, respectively. In the preparation of AlphaVaccine particles, antigen must 
be produced and purified under conditions to promote assembly of the ordered 
and repetitive antigen array onto the non-natural molecular scaffold. In the 
particular JUNIFOS leucine zipper protein domain embodiment, the FOS-mtigen 
or FOS-antigenic determinant should be treated with a reducing agent (e.g., 
Dithiothreitol (DTT)) to reduce or eliminate the incidence of disulfide bond 
formation (Example 15). 

[0221] For the preparation of the non-natural molecular scaffold (i.e., 

recombinant Sinbis virus) of the JUNIFOS leucine zipper protein domain 
embodiment, recombinant E2-JUN viral particles should be concentrated, 
neutralized and treated with reducing agent (see Example 16). 

[0222] Assembly of the ordered and repetitive antigen array in the JUNIFOS 

embodiment is done in the presence of a redox shuffle. E2-JUN viral particles are 
combined with a 240 fold molar excess of FCW-antigen or FOS-antigenic 
determinant for 10 hours at 4°C. Subsequently, the AlphaVaccine particle is 
concentrated and purified by chromatography (Example 16). 

[0223] In another embodiment of the invention, the coupling of the non-natural 

molecular scaffold to the antigen or antigenic determinant may be accomplished 
by chemical cross-linking. In a specific embodiment, the chemical agent is a 



-57- 



heterobifunctional cross-linking agent such as e-maleimidocaproic acid N- 
hydroxysuccinimide ester (Tanimori etal. , J. Pharm. Dyn. 4 812(1981); Fujiwara 
etal, J. Immunol. Meth. 45:195 (1981)), which contains (1) a succinimide group 
reactive with amino groups and (2) a maleimide group reactive with SH groups. 
A heterologous protein or polypeptide of the first attachment site may be 
engineered to contain one or more lysine residues that will serve as a reactive 
moiety for the succinimide portion of the heterobifunctional cross-linking agent. 
Once chemically coupled to the lysine residues of the heterologous protein, the 
maleimide group of the heterobifunctional cross-linking agent will be available to 
react with the SH group of a cysteine residue on the antigen or antigenic 
determinant. Antigen or antigenic determinant preparation in this instance may 
require the engineering of a cysteine residue into the protein or polypeptide chosen 
as the second attachment site so that it may be reacted to* the free maleimide 
function on the cross-linking agent bound to the non-natural molecular scaffold 
first attachment sites. Thus, in such an instance, the heterobifunctional 
cross-linking agent binds to a first attachment site of the non-natural molecular 
scaffold and connects the scaffold to a second binding site of the antigen or 
antigenic determinant. 

3 . Compositions, Vaccines, and the Administration Thereof, and Methods of 
Treatment 

[0224] In one embodiment, the invention provides vaccines for the prevention of 

infectious diseases in a wide range of species, particularly mammalian species such 
as human, monkey, cow, dog, cat, horse, pig, etc. Vaccines may be designed to 
treat infections of viral etiology such as HIV, influenza, Herpes, viral hepatitis, 
Epstein Bar, polio, viral encephalitis, measles, chicken pox, etc.; or infections of 
bacterial etiology such as pneumonia, tuberculosis, syphilis, etc.; or infections of 
parasitic etiology such as malaria, trypanosomiasis, leishmaniasis, trichomoniasis, 
amoebiasis, etc. 



-58- 



[0225] In another embodiment, the invention provides vaccines for the prevention 

of cancer in a wide range of species, particularly mammalian species such as 
human, monkey, cow, dog, cat, horse, pig, etc. Vaccines may be designed to treat 
all types of cancer: lymphomas, carcinomas, sarcomas, melanomas, etc. 

[0226] In another embodiment of the invention, compositions of the invention 

may be used in the design of vaccines for the treatment of allergies. Antibodies 
of the IgE isotype are important components in allergic reactions. Mast cells bind 
IgE antibodies on their surface and release histamines and other mediators of 
allergic response upon binding of specific antigen to the IgE molecules bound on 
the mast cell surface. Inhibiting production of IgE antibodies, therefore, is a 
promising target to protect against allergies. This should be possible by attaining 
a desired T helper cell response. T helper cell responses can be divided into type 
1 (T H 1) and type 2 (T H 2) T helper cell responses (Romagnani, Immunol. Today 
75:263-266 (1997)). T H 1 cells secrete interferon-gamma and other cytokines 
which trigger B cells to produce protective IgG antibodies. In contrast, a critical 
cytokine produced by T H 2 cells is IL-4, which drive B cells to produce IgE. In 
many experimental systems, the development of T H 1 and T H 2 responses is 
mutually exclusive sinceT H l cells suppress the induction of T H 2 cells and vice 
versa. Thus, antigens that trigger a strong T H 1 response simultaneously suppress 
the development of T H 2 responses and hence the production of IgE antibodies. 
Interestingly, virtually all viruses induce a T H 1 response in the host and fail to 
trigger the production of IgE antibodies (Coutelier etal.,J. Exp. Med. 165:64-69 
(1987)). This isotype pattern is not restricted to live viruses but has also been 
observed for inactivated or recombinant viral particles (Lo-Man et al, Eur. J. 
Immunol. 25:1401-1407 (1998)) Thus, by using the processes of the invention 
{e.g., AlphaVaccine Technology), viral particles can be decorated with various 
allergens and used for immunization Due to the resulting "viral structure" of the 
allergen, a T H 1 response will be elicited, "protective" IgG antibodies will be 
produced, and the production of IgE antibodies which cause allergic reactions will 
be prevented. Since the allergen is presented by viral particles which are 



-59- 



recognized by a different set of helper T cells than the allergen itself, it is likely 
that the allergen-specific IgG antibodies will be induced even in allergic individuals 
harboring pre-existing T H 2 cells specific for the allergen. The presence of high 
concentrations of IgG antibodies may prevent binding of allergens to mast cell 
bound IgE, thereby inhibiting the release of histamine. Thus, presence of IgG 
antibodies may protect from IgE mediated allergic reactions. Typical substances 
causing allergies include: grass, ragweed, birch or mountain cedar pollens, house 
dust, mites, animal danders, mold, insect venom or drugs (e.g., penicillin). Thus, 
immunization of individuals with allergen-decorated viral particles should be 
beneficial not only before but also after the onset of allergies Food allergies are 
also very common, and immunization of subjects with particles decorated with 
food allergens should be useful for the treatment of these allergies. 

[0227] In another embodiment, the invention relates to the induction of specific 

Th type 2 (Th2) cells. The inventors surprisingly found that bacterial Pili induce 
an antibody response dominated by the IgGl isotype in mice, indicative of a Th2 
response. Antigens coupled to Pili also induced a IgGl response indicating that 
coupling of antigens to Pili was sufficient for induction of antigen-specific Th2 
response. Many chronic diseases in humans an animals, such as arthritis, colitis, 
diabetes and multiple sclerosis are dominated by Thl response, where T cells 
secrete IFNy and other pro-inflammatory cytokines precipitating disease. By 
contrast, Th2 cells secrete 11-4, II- 1 3 and also II- 1 0 The latter cytokine is usually 
associated with immunosuppression and there is good evidence that specific Th2 
cells can suppress chronic diseases, such as arthritis, colitis, diabetes and multiple 
sclerosis in vivo. Thus, induction of antigen-specific Th2 cells is desirable for the 
treatment of such chronic diseases. 

[0228] It is known that induction of therapeutic self-specific antibodies may allow 

treating a variety of diseases It is, e.g., known that anti-TNF antibodies can 
ameliorate symptoms in arthritis or colitis and antibodies specific for the A(3- 
peptide may remove plaques from the brain of Alzheimers patients It will usually 
be beneficial for the patient if such antibodies can be induced in the absence of a 



-60- 



pro-inflammatory Thl response. Thus, self antigens coupled to Pili that induce 
a strong antibody response but no Thl response may be optimal for such 
immunotherapy. 

[0229] In a preferred embodiment, the antigen is the amyloid beta peptide (Ap\. 42 ) 

(D AEFRHD S GYE VHHQKL VFFAEDVGSNKGAIIGLMVGGVVIA (SEQ ID 
NO: 174), or a fragment thereof. The amyloid beta protein is SEQ ID NO: 172. 
The amyloid beta precursor protein is SEQ ID NO: 173 

[0230] The amyloid B peptide (Ap M2 ) has a central role in the neuropathology of 

Alzheimers disease. Region specific, extracellular accumulation of Ap peptide is 
accompanied by microgliosis, cytoskeletal changes, dystrophic neuritis and 
synaptic loss. These pathological alterations are thought to be linked to the 
cognitive decline that defines the disease. 

[0231] In a mouse model of Alzheimer disease, transgenic animals engineered to 

produce Ap M2 (PDAPP-mice), develop plaques and neuron damage in their 
brains. Recent work has shown immunization of young PDAPP-mice, using Ap\. 
42 , resulted in inhibition of plaque formation and associated dystrophic neuritis 
(Schenk, D. et al., Nature 400:113-11 (1999)). 

[0232] Furthermore immunization of older PDAPP mice that had already 

developed AD-like neuropathologies, reduced the extent and progression of the 
neuropathologies. The immunization protocol for these studies was as follows; 
peptide was dissolved in aqueous buffer and mixed 1 : 1 with complete Freunds 
adjuvant (for primary dose) to give a peptide concentration of lOOug/dose. 
Subsequent boosts used incomplete Freunds adjuvant. Mice received 11 
immunizations over an 1 1 month period. Antibodies titres greater than 1:10 000 
were achieved and maintained. Hence, immunization may be an effective 
prophylactic and therapeutic action against Alzheimer disease. 

[0233] In another study, peripherally administered antibodies raised against Ap w2 , 

were able to cross the blood-brain barrier, bind Ap peptide, and induce clearance 
of pre-existing amyloid (Bard, F. etal, Nature Medicine 6: 916-19 (2000)). This 
study utilized either polyclonal antibodies raised against Ap^, or monoclonal 



-61- 



antibodies raised against synthetic fragments derived from different regions of AfJ. 
Thus induction of antibodies can be considered as a potential therapeutic 
treatment for Alzheimer disease. 

[0234] In another more specific embodiment, the invention is drawn to vaccine 

compositions comprising at least one antigen or antigenic determinant encoded by 
an Influenza viral nucleic acid, and the use of such vaccine compositions to elicit 
immune responses. In an even more specific embodiment, the Influenza antigen 
or antigenic determinant may be an M2 protein (e.g. , an M2 protein having the 
amino acids shown in SEQ ID NO. 171, GenBank Accession No. P06821, or in 
SEQ ID NO: 170, PIR Accession No. MFIV62, or fragment thereof (e.g., amino 
acids from about 2 to about 24 in SEQ ID NO' 171, the amino acid sequence in 
SEQ ID NO 170. Further, influenza antigens or antigenic determinants may be 
coupled to pili or pilus-like structures. Portions of an M2 protein (e.g., an M2 
protein having the amino acid sequence in SEQ ID NO: 170), as well as other 
proteins against which an immunological response is sought, suitable for use with 
the invention may comprise, or alternatively consist of, peptides of any number of 
amino acids in length but will generally be at least 6 amino acids in length (e.g., 
peptides 6, 7, 8, 9, 10, 12, 15, 18, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 
80, 85, 90, 95, or 97 amino acids in length) 

[0235] In an even more specific embodiment, the Influenza antigen or antigenic 

determinant may be an M2 protein (e.g., an M2 protein having the amino acids 
shown in SEQ ID NO: 1 70, GenBank Accession No. P06821 , or in SEQ ID NO: 
212, PIR Accession No. MFIV62, or fragment thereof (e.g., amino acids from 
about 2 to about 24 in SEQ ID NO: 171, the amino acid sequence in SEQ ID NO: 
170). 

[0236] As would be understood by one of ordinary skill in the art, when 

compositions of the invention are administered to an individual, they may be in a 
composition which contains salts, buffers, adjuvants, or other substances which 
are desirable for improving the efficacy of the composition. Examples of materials 
suitable for use in preparing pharmaceutical compositions are provided in 



-62- 



numerous sources including Remington's Pharmaceutical Sciences (Osol, A, 
ed., Mack Publishing Co., (1980)). 

[0237] Compositions of the invention are said to be "pharmacologically accept- 

able" if their administration can be tolerated by a recipient individual. Further, the 
compositions of the invention will be administered in a "therapeutically effective 
amount" (i.e., an amount that produces a desired physiological effect) 

[0238] The compositions of the present invention may be administered by various 

methods known in the art, but will normally be administered by injection, infusion, 
inhalation, oral administration, or other suitable physical methods. The 
compositions may alternatively be administered intramuscularly, intravenously, or 
subcutaneously Components of compositions for administration include sterile 
aqueous (e.g., physiological saline) or non-aqueous solutions and suspensions. 
Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, 
vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. 
Carriers or occlusive dressings can be used to increase skin permeability and 
enhance antigen absorption. 

[0239] The present invention also provides a composition comprising a bacterial 

pilin polypeptide to which an antigen or antigenic determinant has been attached 
by a covalent bond. 

[0240] The present invention also provides a composition comprising a fragment 

of a bacteriophage coat protein to which an antigen or antigenic determinant has 
been attached by a covalent bond. 

[0241] The present invention also provides a composition comprising (a) non- 

natural molecular scaffold comprising (i) a core particle selected from the group 
consisting of (1) a bacterial pilus or pilin protein; and (2) a recombinant form of 
a bacterial pilus or pilin protein; and (ii) an organizer comprising at least one first 
attachment site, wherein the organizer is connected to the core particle by at least 
one covalent bond; and (b) an antigen or antigenic determinant with at least one 
second attachment site, the second attachment site being selected from the group 
consisting of (i) an attachment site not naturally occurring with the antigen or 



-63- 



antigenic determinant; and (ii) an attachment site naturally occurring with the 
antigen or antigenic determinant, wherein the second attachment site is capable of 
association through at least one non-peptide bond to the first attachment site; and 
wherein the antigen or antigenic determinant and the scaffold interact through the 
association to form an ordered and repetitive antigen array. 

The present invention also provides a composition comprising (a) a non- 
natural molecular scaffold comprising (i) a core particle selected from the group 
consisting of: (1) a bacterial pilus; and (2) a recombinant form of a bacterial pilus; 
and (ii) an organizer comprising at least one first attachment site, wherein the 
organizer is connected to the core particle by at least one covalent bond; and (b) 
an antigen or antigenic determinant with at least one second attachment site, the 
second attachment site being selected from the group consisting of (i) an 
attachment site not naturally occurring with the antigen or antigenic determinant; 
and (ii) an attachment site naturally occurring with the antigen or antigenic 
determinant, wherein the second attachment site is capable of association through 
at least one non-peptide bond to the first attachment site; and wherein the antigen 
or antigenic determinant and the scaffold interact through the association to form 
an ordered and repetitive antigen array. 

The present invention also provides a composition comprising (a) a non- 
natural molecular scaffold comprising (i) a virus-like particle that is a dimer or a 
multimer of a polypeptide comprising amino acids 1-147 of SEQ ID NO: 158 as 
core particle; and (ii) an organizer comprising at least one first attachment site, 
wherein the organizer is connected to the core particle by at least one covalent 
bond; and (b) an antigen or antigenic determinant with at least one second 
attachment site, the second attachment site being selected from the group 
consisting of (i) an attachment site not naturally occurring with the antigen or 
antigenic determinant; and (ii) an attachment site naturally occurring with the 
antigen or antigenic determinant, wherein the second attachment site is capable of 
association through at least one non-peptide bond to the first attachment site; and 



-64- 



wherein the antigen or antigenic determinant and the scaffold interact through the 
association to form an ordered and repetitive antigen array. 
[0244] The present invention also provides a pharmaceutical composition 

comprising any of compositions of the present invention, and a pharmaceutical^ 
acceptable carrier. 

[0245] The present invention also provides a vaccine composition comprising any 

of compositions of the present invention. The vaccine composition may further 
comprise at least one adjuvant. The present invention also provides a method of 
immunizing, comprising administering to a subject a vaccine composition of the 
present invention. 

[0246] The present invention also provides a composition comprising (a) a non- 

natural molecular scaffold comprising (i) Hepatitis B virus capsid protein 
comprising an amino acid sequence selected from the group consisting of (1) the 
amino acid sequence of SEQ ID NO: 89, (2) the amino acid sequence of SEQ ID 
NO: 90 (3) the amino acid sequence of SEQ ID NO 93, (4) the amino acid 
sequence of SEQ ID NO:98, (5) the amino acid sequence of SEQ ID NO:99, (6) 
the amino acid sequence of SEQ ID NO: 102, (7)the amino acid sequence of SEQ 
ID NO: 104, (8) the amino acid sequence of SEQ ID NO: 105, (9) the amino acid 
sequence of SEQ ID NO: 106, (10) the amino acid sequence of SEQ ID NO: 1 19, 
(1 1) the amino acid sequence of SEQ ID NO: 120, (12) the amino acid sequence 
of SEQ ID NO: 123, (13) the amino acid sequence of SEQ ID NO: 125, (14) the 
amino acid sequence of SEQ ID NO: 131, (15) the amino acid sequence of SEQ 
ID NO:132, (16) the amino acid sequence of SEQ ID NO:134, (17) the amino 
acid sequence of SEQ ID NO:157, and (18) the amino acid sequence of SEQ ID 
NO: 158; and (ii) an organizer comprising at least one first attachment site, 
wherein the organizer is connected to the core particle by at least one covalent 
bond; and (b) an antigen or antigenic determinant with at least one second 
attachment site, the second attachment site being selected from the group 
consisting of (i) an attachment site not naturally occurring with the antigen or 
antigenic determinant; and (ii) an attachment site naturally occurring with the 



-65- 



antigen or antigenic determinant, wherein the second attachment site is capable of 
association through at least one non-peptide bond to the first attachment site; and 
wherein the antigen or antigenic determinant and the scaffold interact through the 
association to form an ordered and repetitive antigen array. Preferably, the 
organizer is a polypeptide or residue thereof, wherein the second attachment site 
is a polypeptide or residue thereof, and wherein the first attachment site is a lysine 
residue and the second attachment site is a cysteine residue. Preferably, one or 
more cysteine residues of the Hepatitis B virus capsid protein have been either 
deleted or substituted with another amino acid residue. Preferably, the cysteine 
residues corresponding to amino acids 48 and 107 in SEQ ID NO: 134 have been 
either deleted or substituted with another amino acid residue. 

The present invention also provides a composition comprising: (1 ) a non- 
natural molecular scaffold comprising (i) a core particle selected from the group 
consisting of (1) a bacterial pilus, and (2) a recombinant form of a bacterial pilus 
or pilin protein; and (ii) an organizer comprising at least one first attachment site, 
wherein the organizer is connected to the core particle by at least one covalent 
bond; and (2) an antigen or antigenic determinant with at least one second 
attachment site, the second attachment site being selected from the group 
consisting of (i) an attachment site not naturally occurring with the antigen or 
antigenic determinant, and (ii) an attachment site naturally occurring with the 
antigen or antigenic determinant, wherein the second attachment site is capable of 
association through at least one non-peptide bond to the first attachment site, 
wherein the antigen or antigenic determinant and the scaffold interact through the 
association to form an ordered and repetitive antigen array, and wherein the 
antigen or antigenic determinant is selected from the group consisting of an 
influenza M2 peptide, the GRA2 polypeptide, the DP 178c peptide, the tumor 
necrosis factor polypeptide, a tumor necrosis factor peptide, the B2 peptide, the 
D2 peptide, and the A(3 peptide. 



-66- 



[0248] In the compositions and vaccines of the present invention, for a covalent 

bond between a first and second attachment site, the covalent bond is preferably 
not a peptide bond. 

[0249] If a bacterial pilus is present in a composition or vaccine of the present 

invention, the pilus is preferably a Type-1 pilus of Eschericia coli. More 
preferably, pilin subunits of the Type-1 pilus comprises the amino acid sequence 
shown in SEQ ID NO: 146. Preferably, the bacterial pilus and the antigen or 
antigen determinant are attached via either a naturally or non-naturally occurring 
attachment. Preferably, the first attachment site will be a lysine residue, while hte 
second attachment site will be a cysteine residue present or engineered on the 
antigen If the attachment comprises interacting leucine zipper polypeptides, the 
polypeptides are preferably JUN and/or FOS leucine zipper polypeptides. 
[0250] In the compositions and vaccines of the present invention that comprise 

an organizer having a first attachment site, attached to the second attachment site 
on the antigen, the organizer is preferably a polypeptide or a residue thereof, and 
the second attachment site is preferably a polypeptide or a residue thereof More 
preferably, the first and/or the second attachment sites comprise an antigen and an 
antibody or antibody fragment thereto, biotin and avidin, strepavidin and biotin, 
a receptor and its ligand, a ligand-binding protein and its ligand, interacting leucine 
zipper polypeptides, an amino group and a chemical group reactive thereto, a 
carboxyl group and a chemical group reactive thereto, a sulfhydryl group and a 
chemical group reactive thereto, or a combination thereof. More preferably, the 
first attachment site is an amino group, and the second attachment site is a 
sulfhydryl group 

[0251] In the compositions and vaccines of the present invention, the antigen is 

preferably selected from the group consisting of a protein suited to induce an 
immune response against cancer cells, a protein suited to induce an immune 
response against infectious diseases, a protein suited to induce an immune 
response against allergens, and a protein suited to induce an immune response in 
farm animals. Preferably, the antigen induces an immune response against one or 



-67- 



more allergens. More preferably, the antigen is a recombinant protein of HIV, a 
recombinant protein of Influenza virus, a recombinant protein of Hepatitis C virus, 
a recombinant protein of Toxoplasma, a recombinant protein of Plasmodium 
falciparum, a recombinant protein of Plasmodium vivax, a recombinant protein of 
Plasmodium ovale, a recombinant protein of Plasmodium malariae, a recombinant 
protein of breast cancer cells, a recombinant protein of kidney cancer cells, a 
recombinant protein of prostate cancer cells, a recombinant protein of skin cancer 
cells, a recombinant protein of brain cancer cells, a recombinant protein of 
leukemia cells, a recombinant profiling, a recombinant protein of bee sting allergy, 
a recombinant protein of nut allergy, a recombinant protein of food allergies, or 
a recombinant protein of asthma, or a recombinant protein of Chlamydia. 

[0252] In the method of immunizing provided by the present invention, the 

immunization produces an immune response in the subject. Preferably, the 
immunization produces a humoral immune response, a cellular immune response, 
a humoral and a cellular immune response, or a protective immune response. 

[0253] In the compositions and vaccines of the present invention, the antigen or 

antigenic determinant is attached to the non-natural molecular scaffold through the 
first attachment site, to form an antigen array or antigenic determinant array. 
Preferably, the array is ordered and/or repetitive. 

[0254] In the compositions and vaccines of the present invention, the first and/ or 

the second attachment sites are preferably attached via either a non-naturally 
occurring attachment, or by an attachment comprising interacting leucine zipper 
polypeptides. More preferably, the interacting leucine zipper polypeptides are 
JUN and/or FOS leucine zipper polypeptides. 

[0255] The present invention also provides a method for making the compositions 

and vaccines of the present invention, comprising combining the antigen or 
antigenic determinant with the non-natural molecular scaffold through the first 
attachment site and organizer present on the non-natural molecular scaffold. 



[0256] In addition to vaccine technologies, other embodiments of the invention 

are drawn to methods of medical treatment for cancer, allergies, and chronic 
diseases. 

[0257] Following is a protocol for analyzing pili by SDS-PAGE Analysis. Add 

trichloroacetic acid to a final concentration of 10% to the pili solution containing 
approx. 50 ug of pili Vortex and incubate for 10 minutes on ice. Centrifuge at 
maximal speed for 5 minutes in a microcentrifuge. Discard the supernatant and 
resuspend the pellet in 50 ul of a 8.5 M guanidiniumhydrochloride, pH 3 solution. 
Heat the sample for 15 minutes at 70°C Precipitate the protein by adding 1.5 ml 
of Ethanol precooled at -20°C, and centrifuge 5 minutes at RT at maximal speed. 
Resuspend the pellet in 15 ul of a 10 mM Tris, pH 8 buffer. Add SDS-PAGE 
sample buffer, vortex shortly and heat the sample 10 minutes at 100°C Load the 
sample on a 12% gel. 



EXAMPLES 

[0258] Enzymes and reagents used in the experiments that follow included' T4 

DNA ligase obtained from New England Biolabs; Taq DNA Polymerase, QIAprep 
Spin Plasmid Kit, QIAGEN Plasmid Midi Kit, QiaExII Gel Extraction Kit, 
QIAquick PCR Purification Kit obtained from QIAGEN; QuickPrep Micro 
mRNA Purification Kit obtained from Pharmacia; Superscript One-step RT PCR 
Kit, fetal calf serum (FCS), bacto-tryptone and yeast extract obtained from Gibco 
BRL; Oligonucleotides obtained from Microsynth (Switzerland); restriction 
endonucleases obtained from Boehringer Mannheim, New England Biolabs or 
MBI Fermentas; Pwo polymerase and dNTPs obtained from Boehringer 
Mannheim. HP-1 medium was obtained from Cell culture technologies 
(Glattbrugg, Switzerland). All standard chemicals were obtained from 
Fluka-Sigma-Aldrich, and all cell culture materials were obtained from TPP. 

[0259] DNA manipulations were carried out using standard techniques. DNA 

was prepared according to manufacturer instruction either from a 2 ml bacterial 



culture using the QIAprep Spin Plasmid Kit or from a 50 ml culture using the 
QIAGEN Plasmid Midi Kit. For restriction enzyme digestion, DNA was 
incubated at least 2 hours with the appropriate restriction enzyme at a 
concentration of 5-10 units (U) enzyme per mg DNA under manufacturer 
recommended conditions (buffer and temperature). Digests with more than one 
enzyme were performed simultaneously if reaction conditions were appropriate for 
all enzymes, otherwise consecutively. DNA fragments isolated for further 
manipulations were separated by electrophoresis in a 0.7 to 1.5% agarose gel, 
excised from the gel and purified with the QiaExII Gel Extraction Kit according 
to the instructions provided by the manufacturer. For ligation of DNA fragments, 
100 to 200 pg of purified vector DNA were incubated overnight with a threefold 
molar excess of the insert fragment at 1 6 °C in the presence of 1 U T4 DNA ligase 
in the buffer provided by the manufacturer (total volume. 10-20 ul). An aliquot 
(0.1 to 0.5 ul) of the ligation reaction was used for transformation of E. coli 
XLl-Blue (Stratagene) Transformation was done by electroporation using a 
Gene Pulser (BioRAD) and 0. 1 cm Gene Pulser Cuvettes (BioRAD) at 200 Q, 25 
uF, 1.7 kV. After electroporation, the cells were incubated with shaking for 1 h 
in 1 ml S.O.B. medium (Miller, 1972) before plating on selective S.O.B. agar. 

EXAMPLE 1 : 

Insertion of the JUN amphiphatic helix domain within E2 
[0260] In the vector pTE5^2J (Hahn et al, Proc. Natl. Acad. Sci. USA 

59:2679-2683, (1992)), MM and a BstEII restriction enzyme sites were 
introduced between codons 71 (Gin) and 74 (Thr) of the structural protein E2 
coding sequence, resulting in vector pTES^JBM. Introduction of these 
restriction enzymes sites was done by PCR using the following oligonucleotides: 



Oligo 1: 

E2insBstEII/BssHII: 

5'-ggggACGCGTGCAGCAggtaaccaccgTTAAAGAAGGCACC-3' (SEQ ID 
NO:l) 



-70- 



Oligo 2. 
E2insMluIStuI. 

5 '-cggtggttaccTGCTGCACGCGTTGCTTAAGCGACATGTAGCGG-3 ' (SEQ 
ID NO:2) 

Oligo 3 

E2insStuI: 5'-CCATGAGGCCTACGATACCC-3' (SEQ ID NO:3) 
01igo4: 

E2insBssHII: 5 -GGCACTCACGGCGCGCTTTACAGGC-3 ' (SEQ IDNO:4) 

[0261] For the PCR reaction, 100 pmol of each oligo was used with 5 ng of the 

template DNA in a 100 pi reaction mixture containing 4 units of Taq or Pwo 
polymerase, 0. 1 mM dNTPs and 1 . 5 mM MgS0 4 . All DNA concentrations were 
determined photometrically using the GeneQuant apparatus (Pharmacia) 
Polymerase was added directly before starting the PCR reaction (starting point 
was 95 °C). Temperature cycling was done in the following manner and order: 
95 °C for 2 minutes; 5 cycles of 95 °C (45 seconds), 53 °C (60 seconds), 72°C 
(80 seconds); and 25 cycles of 95 °C (45 seconds), 57°C (60 seconds), 72°C 
(80 seconds). 

[0262] The two PCR fragments were analyzed and purified by agarose 

gelelectrophoresis. Assembly PCR of the two PCR fragments using oligo 3 and 
4 for amplification was carried out to obtain the final construct. 

[0263] For the assembly PCR reaction, 100 pmol of each oligo was used with 

2 ng of the purified PCR fragments in a 100 ul reaction mixture containing 4 units 
of Taq or Pwo polymerase, 0.1 mM dNTPs and 1.5 mM MgS0 4 . All DNA 
concentrations were determined photometrically using the GeneQuant apparatus 
(Pharmacia). Polymerase was added directly before starting the PCR reaction 
(starting point was 95 ° C). Temperature cycling was done in the following manner 
and order- 95 °C for 2 minutes; 5 cycles of 95 °C (45 seconds), 57 °C 
(60 seconds), 72 °C (90 seconds); and 25 cycles of 95 °C (45 seconds), 59 °C 
(60 seconds), 72°C (90 seconds). 

[0264] The final PCR product was purified using Qia spin PCR columns (Qiagen) 

and digested in an appropriate buffer using 10 units each of BssHII and StuI 



-71- 



restriction endonucleases for 12 hours at 37° C. The DNA fragments were 
gel-purified and ligated intoBssHII/StuI digested and gel-purified pTE5 '2J vector 
(Hahn et al, Proc. Natl. Acad. Sci. USA 59:2679-2683). The correct insertion 
of the PCR product was first analyzed by BstEII and Mlul restriction analysis and 
then by DNA sequencing of the PCR fragment. 
[0265] The DNA sequence coding for the JUN amphiphatic helix domain was 

PCR-amplified from vector pJuFo (Crameriand Suter, Gene 757:69 (1993)) using 
the following oligonucleotides: 

Oligo 5- 
JMVBstEII: 

5'-CCTTCTTTAAcggtggttaccTGCTGGCAACCAACGTGGTTCATGAC-3' 
(SEQ ID NO 5) 

Oligo 6: 

MluL/W: 5 -AAGCATGCTGCacgcgtgTGCGGTGGTCGGATCGCCCGGC-3 ' 
(SEQ ID NO:6) 

[0266] For the PCR reaction, 100 pmol of each oligo was used with 5 ng of the 

template DNA in a 100 nl reaction mixture containing 4 units of Taq or Pwo 
polymerase, O.lmM dNTPs and 1 .5 mM MgS0 4 All DNA concentrations were 
determined photometrically using the GeneQuant apparatus (Pharmacia). 
Polymerase was added directly before starting the PCR reaction (starting point 
was 95 °C). Temperature cycling was done in the following order and manner: 
95 °C for 2 minutes; 5 cycles of 95 °C (45 seconds), 60°C (30 seconds), 72°C 
(25 seconds); and 25 cycles of 95°C (45 seconds), 68°C (30 seconds), 72°C 
(20 seconds). 

[0267] The final PCR product was gel-purified and ligated into EcoRV digested 

and gel-purified pBluescript II(KS"). From the resulting vector, the JUN sequence 
was isolated by cleavage withMM/5s*EII purified with QiaExII and ligated into 
vector pTE5 l 2JBM (previously cut with the same restriction enzymes) to obtain 
the vector pTE5^2J:E2JMV. 



-72- 



EXAMPLE 2; 

Production of viral particles containing E2-JUN using the pCYTts system 
The structural proteins were PCR amplified using pTE5'2J:E2JUN as 
template and the oligonucleotides XbalStruct 

(ctatcaTCTAGAATGAATAGAGGATTCTTTAAC (SEQ ID NO: 12)) and 
StructBspl201 (tcgaatGGGCCCTCATCTTCGTGTGCTAGTCAG (SEQ ID 
NO : 8 7)) . For the PCR 1 00 pmol of each loligo was used and 5 ng of the template 
DNA was used in the 100 ul reaction mixture, containing 4 units of Tac or Pwo 
polymerase, 0. 1 mM dNTPs and 1 .5 mM MgS0 4 . All DNA concentrations were 
determined photometrically using the GeneQuant apparatus (Pharmacia). The 
polymerase was added directly before starting the PCR reaction (starting point 
was 95 °C). The temperature cycles were as follows: 95 °C for 3 minutes, 
followed by 5 cycles of 92° C (30 seconds), 54 °C (35 seconds), 72 °C (270 
seconds) and followed by 25 cycles of 92°C (30 seconds), 63 °C (35 seconds), 
72° C (270 seconds. The PCR product was gel purified and digested with the 
restriction enzymes Xbal/Bspl201 and ligated into vector pCYTts previously 
cleaved with the same enzymes (WO 99/50432) 

Twenty ug of pCYTtsE2:JWV were incubated with 30 U of Seal in an 
appropriate buffer for at least 4 hours at 37°C. The reaction was stopped by 
phenol/chloroform extraction, followed by an isopropanol precipitation of the 
linerized DNA. The restriction reaction was checked by agarose gel 
eletrophoresis. For the transfection, 5.4 ug of linearized pCYTtsE2:JMV r was 
mixed with 0.6 ng of linearized pSV2Neo in 30 ul H 2 0 and 30 ul of 1 M CaCl 2 
solution were added. After addition of 60 ul phosphate buffer (50 mM HEPES, 
280 mM NaCl, 1.5 mM Na 2 HP0 4 , pH 7.05), the solution was vortexed for 5 
seconds, followed by an incubation at room temperature for 25 seconds. The 
solution was immediately added to 2 ml HP-1 medium containing 2% FCS (2% 
FCS medium). The medium of an 80% confluent BHK21 ceil culture in a 6-well 
plate was then replaced with the DNA containing medium. After an incubation 
for 5 hours at 37 °C in a C0 2 incubator, the DNA containing medium was 



-73- 



removed and replaced by 2 ml of 1 5% glycerol in 2% FCS medium. The glycerol 
containing medium was removed after a 30 second incubation phase, and the cells 
were washed by rinsing with 5 ml of HP-1 medium containing 10% FCS. Finally 
2 ml of fresh HP-1 medium containing 10% FCS was added. 
[0270] Stably transfected cells were selected and grown in selection medium 

(HP-1 medium, supplemented with G418) at 37 °C in a C0 2 incubator. When the 
mixed population was grown to confluency, the culture was split to two dishes, 
followed by a 12 hours growth period at 37°C. One dish of the cells was shifted 
to 30° C to induce the expression of the viral particles; the other dish was kept at 
37°C. 

[0271] The expression of viral particles was determined by Western blotting 

(Figure 1). Culture medium (0.5 ml) was methanol/chloroform precipitated, and 
the pellet was resuspended in SDS-PAGE sample buffer. Samples were heated 
for 5 minutes at 95 °C before being applied to 15% acrylamide gel. After 
SDS-PAGE, proteins were transferred to Protan nitrocellulose membranes 
(Schleicher & Schuell, Germany) as described by Bass and Yang, in Creighton, 
T.E., ed., Protein Function: A Practical Approach, 2nd Edn., IRL Press, Oxford 
(1997), pp. 29-55. The membrane was blocked with 1% bovine albumin (Sigma) 
in TBS (lOxTBS per liter: 87.7 g NaCl, 66. lg Trizma hydrochloride (Sigma) and 
9.7 g Trizma base (Sigma), pH 7.4) for 1 hour at room temperature, followed by 
an incubation with an anti-El /E2antibody (polyclonal serum) for 1 hour. The blot 
was washed 3 times for 10 minutes with TBS-T (TBS with 0.05% Tween20), and 
incubated for 1 hour with an alkaline-phosphatase-anti-rabbit IgG conjugate (0. 1 
ug/ml, Amersham Life Science, England). After washing 2 times for 10 minutes 
with TBS-T and 2 times for 10 minutes with TBS, the development reaction was 
carried out using alkaline phosphatase detection reagents (10 ml AP buffer (100 
mM Tris/HCl, 100 mM NaCl, pH 9.5) with 50 ^1 NBT solution (7.7% Nitro Blue 
Tetrazolium (Sigma) in 70% dimethylformamide) and 37 \i\ of X-Phosphate 
solution (5% of 5-bromo-4-chloro-3-indolyl phosphate in dimethylformamide). 



-74- 



[0272] The production of viral particles is shown in Figure 1 The Western Blot 

pattern showed that E2-JUN (lane 1) migrated to a higher molecular weight in 
SDS-PAGE compared to wild type E2 (lane 2) and the BHK21 host cell line did 
not show any background. 

EXAMPLE 3: 
Production of viral particles containing E2-JUN 
using the pTE5'2JE2.JUN vector 
[0273] RNase-free vector (1.0 ug) was linerarized by Pvul digestion. 

Subsequently, in vitro transcription was carried out using an SP6 in vitro 
transcription kit (Invitro scrip CAP by InvitroGen, Invitrogen BV, NV Leek, 
Netherlands). The resulting 5 '-capped mRNA was analyzed on a reducing 
agarose-gel. 

[0274] In vitro transcribed mRNA (5 ug) was electroporated into BHK 21 cells 

(ATCC: CCL10) according to Invitrogen's manual (Sindbis Expression system, 
Invitrogen BV, Netherlands). After 10 hours incubation at 37°C, the FCS 
containing medium was exchanged by HP- 1 medium without FCS, followed by an 
additional incubation at 37°C for 10 hours. The supernatant was harvested and 
analyzed by Western blot analysis for production of viral particles exactly as 
described in Example 2. 

[0275] The obtained result was identical to the one obtained with pCYTtsE2: JUN 

as shown in Figure 2. 

EXAMPLE 4: 

Fusion of human growth hormone (hGH) to the FOS leucine 
zipper domain (OmpA signal sequence) 
[0276] The hGH gene without the human leader sequence was amplified from the 

original plasmid (ATCC 3 1389) by PCR. Oligo 7 with an internal Xbal site was 
designed for annealing at the 5 ' end of the hGH gene, and oligo 9 with an internal 
EcoRI site primed at the 3' end of the hGH gene. For the PCR reaction, 100 



-75- 



pmol of each oligo and 5 ng of the template DNA was used in the 75 ul reaction 
mixture (4 units of Taq or Pwo polymerase, 0.1 raM dNTPs and 1 .5 mMMgSOJ. 

[0277] PCR cycling was performed in the following manner: 30 cycles with an 

annealing temperature of 60 °C and an elongation time of 1 minute at 72 °C. 

[0278] The gel purified and isolated PCR product was used as a template for a 

second PCR reaction to introduce the ompA signal sequence and the 
Shine-Dalgarno sequence For the PCR reaction, 100 pmol of oligo 8 and 9 and 
1 ng of the template PCR fragment was used in the 75 ul reaction mixture (4 units 
of Taq or Pwo polymerase, 0. 1 mM dNTPs and 1.5 mM MgSOJ. The annealing 
temperature for the first five cycles was 55 °C with an elongation time of 60 
seconds at 72 ° C; another 25 cycles were performed with an annealing temperature 
of 65°C and an elongation time of 60 seconds at 72°C. 

[0279] Oligol: gggtctagattcccaaccattcccttatccaggctttttgac aacgctatgctccgcgccc 

atcgtctgcaccagctggcctttgacacc(SEQ ID NO: 7) ;ohgo 8: gggtctagaaggaggtaaaaaa 
cgatgaaaaagacagctatcgcgattgcagtggcactggctggtttcgctaccgtagcgcaggccttcccaac 
cattcccttatcc (SEQ ID NO: 8); ohgo 9: cccgaattcctagaagccacagctgccctcc (SEQ ID 
NO:9). 

[0280] The resulting recombinant hGH gene was subcloned into pBluescript via 

Xbal/EcoRI The correct sequence of both strands was confirmed by DNA 
sequencing. 

[0281] The DNA sequence coding for the FOS amphiphatic helix domain was 

PCR-amplified from vector pJuFo (Crameri & Suter Gene 137:69 (1993)) using 
the oligonucleotides: 
orrvp-FOS: 

5'- ccTGCGGTGGTCTGACCGACACCC-3 ' (SEQ ID NO: 10) 
FOS-hgh: 

5'- ccgcggaagagccaccGCAACCACCGTGTGCCGCCAGGATG-3' (SEQ ID 
NO: 11) 



-76- 



[0282] For the PCR reaction, 100 pmol of each oligo and 5 ng of the template 

DNA was used in the 75 ul reaction mixture (4 units of Taq or Pwo polymerase, 
0.1 mM dNTPs and 1.5 mM MgSOJ. The temperature cycles were as follows 

[0283] 95°C for 2 minutes, followed by 5 cycles of 95°C (45 seconds), 60°C 

(30 seconds), 72° C (25 seconds) and followed by 25 cycles of 95 °C (45 seconds), 
68 °C (30 seconds), 72 °C (20 seconds). 

[0284] The PCR product was purified, isolated and cloned into the StuI digested 

pBluescript-ompA-hGH. The hybrid gene was then cloned into the pKK223-3 
Plasmid (Pharmacia). 

EXAMPLE 5: 
Bacterial expression of FOS-hGH 
[0285] The ompA-FOS-hGH in pkk223-3 was expressed under the control of the 

inducible IPTG-dependend promoter using JM101 as E. coli host strain 
Expression was performed in shaker flask. Cells were induced with 1 mM IPTG 
(final concentration) at an OD600 of 0.5. Expression was continued for 10 hours 
at 3 7 0 C. Cells were harvested by centrifugation at 3 600 at 1 0 ° C for 1 5 min. The 
cell pellet was frozen (-20 °C or liq. N 2 ) and stored for 16 hours. The pellet was 
then thawed at 4°C and resuspended in 10 ml 10 mM Tris-HCl, pH 7.4 containing 
600 mM sucrose. After stirring for 15 min at 4°C, peripiasmic proteins were 
released by an osmotic shock procedure. Chilled (4 ° C) deionized H 2 0 was added, 
and the suspension was stirred for 30 min at 4 °C The sludge was diluted, 
resuspended, and lysozyme was added to degrade the cell wall of the bacteria. 
The cells and the peripiasmic fraction spheroplasts were separated by 
centrifugation for 20 min at 11000 x g at 4 °C. The ^OS-hGH-containing 
supernatant was analyzed by reducing and non-reducing SDS-Page and Dot Blot. 
Dot Blot was carried out as described in Example 8, using an anti-hGH antibody 
(Sigma) as the first antibody and an alkaline phosphatase (AP)-anti-mouse 
antibody conjugate as the second antibody. 



-77- 



[0286] Full length, correctly processed FOS-hGU could be detected under 

reducing and non-reducing conditions. Part of FOS-hGH was bound to other, 
non-identified proteins due to the free cysteines present in the FOS amphiphatic 
helix. However, more than 50% of expressed FOS-hGH occurred in its native 
monomeric conformation ( Figure 3). 

[0287] Purified FOS-hGH will be used to perform first doping experiments with 

JUN containing viral particles. 

EXAMPLE 6. 

Construction of the pAV vector series for expression of FOS fusion proteins 
[0288] A versatile vector system was constructed that allowed either cytplasmic 

production or secretion of N- or C-terminal FOS fusion proteins in E. coli or 
production of N- or C-terminal FOS fusion proteins in eukaryotic cells. The 
vectors pAVl - pAV4 which was designed for production of FOS fusion proteins 
in E. coli, encompasses the DNA cassettes listed below, which contain the 
following genetic elements arranged in different orders, (a) a strong ribosome 
binding site and 5' -untranslated region derived from the E. coli ompA gene 
(aggaggtaaaaaacg) (SEQ ID NO: 13); (b) a sequence encoding the signal peptide 
of E. coli outer membrane protein OmpA (MKKT AIAJAVALAGF ATVAQA) 
(SEQ ID NO: 14); (c) a sequence coding for the FOS dimerization domain flanked 
on both sides by two glycine residues and a cystein residue 
(CGGLTDTLQAETDQVEDEKSALQTEIANLLKEKEKLEFILAAHGGC) 
(SEQ ID NO: 15), and (d) a region encoding a short peptidic linker (AAASGG 
(SEQ ID NO: 16) or GGSAAA (SEQ ID NO: 17)) connecting the protein of 
interest to the FOS dimerization domain. Relevant coding regions are given in 
upper case letters. The arrangement of restriction cleavage sites allows easy 
construction of FOS fusion genes with or without a signal sequence. The 
cassettes are cloned into the EcoRI/Hindlll restriction sites of expression vector 
pKK223-3 (Pharmacia) for expression of the fusion genes under control of the 
strong tac promotor. 



-78- 



pAVl 

[0289] This vector was designed for the secretion of fusion proteins with FOS at 

the C-terminus into the E. coli periplasmic space. The gene of interest (g.o.i.) 
may be ligated into the StuI/NotI sites of the vector. 



ECORI 

qaa ttc agg agg taa aaa acg ATG AAA AAG ACA GCT ATC GCG ATT GCA 
GTG GCA CTG GCT 

MKKTAIAIA 

V A L A 

61/21 StuI NotI 

GGT TTC GCT ACC GTA GCG CAG_GCC_tgg gtg ggg GCG GCC GCT TCT GGT 
GGT TGC GGT GGT 

GFATVAQA (goi) A A A S G 



121/41 151/51 

CTG ACC GAC ACC CTG CAG GCG GAA ACC GAC CAG GTG GAA GAC GAA AAA 
TCC GCG CTG CAA 

LTDTLQAETDQVEDEK 
S A L Q 

181/61 211/71 

ACC GAA ATC GCG AAC CTG CTG AAA GAA AAA GAA AAG CTG GAG TTC ATC 
CTG GCG GCA CAC 



241/81 Hindi I I 

GGT GGT TGC t aa get t (SEQ ID NO: 1 8) 

G G c * a (SEQ ID NOs- 14 and 19) 

pAV2 

[0290] This vector was designed for the secretion of fusion proteins with FOS at 

the N-terminus into the E. coli periplasmic space. The gene of interest (g.o.i.) 
ligated into the Notl/EcoRV (or Notl/Hindlll) sites of the vector. 



EcoRI 



31/11 



-79- 



craa ttc agg agg taa aaa acg ATG AAA AAG ACA GCT ATC GCG ATT GCA 
GTG GCA CTG GCT 



GGT TTC GCT ACC GTA GCG C AG GCC T GC GGT GGT CTG ACC GAC ACC CTG 
CAG GCG GAA ACC 



GAC CAG GTG GAA GAC GAA AAA TCC GCG CTG CAA ACC GAA ATC GCG AAC 
CTG CTG AAA GAA 



181/61 211/71 
Not I 

AAA GAA AAG CTG GAG TTC ATC CTG GCG GCA CAC GGT GGT TGC GGT GGT 
TCT GCG GCC GC T 



K 



H G G C G G 



241/81 EcoRV Hindi I I 

ggg tgt ggg gat ate aag ctt (SEQ ID NO:20) 

(goi) (SEQIDNO.21) 
pAV3 

[0291] This vector was designed for the cytoplasmic production of fusion proteins 

with FOS at the C-terminus in E. coli. The gene of interest (g.o i.) may be ligated 
into the EcoRV/NotI sites of the vector. 



-80- 



EcoRI EcoRV 

gaa ttc agg agg taa aaa gat ate 
GGT TGC GGT GGT 

G C G G 
61/21 

CTG ACC GAC ACC CTG CAG GCG GAA 
TCC GCG CTG CAA 

LTDTLQAE 
S A L Q 

121/41 

ACC GAA ATC GCG AAC CTG CTG AAA 
CTG GCG GCA CAC 

TE IANLLK 
L A A H 



Not I 

ggg tgt ggg GCG GCC GC T TCT GGT 
(goi) A A A S G 

91/31 

ACC GAC CAG GTG GAA GAC GAA AAA 
TDQVEDEK 

151/51 

GAA AAA GAA AAG CTG GAG TTC ATC 
EKEKLEFI 



181/61 Hindlll 

GGT GGT TGC t aa get t (SEQ ID NO:22) 

g G c * (SEQIDNO:23) 



pAV4 

[0292] This vector is designed for the cytoplasmic production of fusion proteins 

with FOS at the N-terminus in E. coli. The gene of interest (g.o.i ) may be ligated 
into the Notl/EcoRV (or Notl/Hindlll) sites of the vector. The N-terminal 
methionine residue is proteolytically removed upon protein synthesis (Hirel et al, 
Proc. Natl. Acad. Sci. USA 56:8247-8251 (1989)). 



EcoRI 




gaa ttc 


agg 


CTG CAG 


GCG 


E F 


R 


L Q 


A 


61/21 




ACC GAC 


CAG 


AAC CTG 


CTG 


T D 


Q 


N L 


L 



121/41 151/51 
Not I 

GAA AAA GAA AAG CTG GAG TTC ATC CTG GCG GCA CAC GGT GGT TGC GGT 
GGT TCT GCG GCC 

EKEKLEFILAAHGGCG 
G S A A 

181/61 EcoRV Hindlll 

GC T ggg tgt ggg crat ate aag ctt (SEQ ID NO: 24) 

A ( goi) (SEQ ID NOs: 88 and 25) 

The vectors pAV5 and pAV6, which are designed for eukaryotic 
production of FOS fusion proteins, encompasses the following genetic elements 
arranged in different orders: (a) a region coding for the leader peptide of human 
growth hormone (MATGSRT SLLL AFGLLCLPWLQEGS A) (SEQ ID NO:26); 

(b) a sequence coding for the FOS dimerization domain flanked on both sides by 
two glycine residues and a cysteine residue 

(CGGLTDTLQAETDQVEDEKSALQTEIANLLKEKEKLEFILAAHGGC) 
(SEQ ID NO 15); and 

(c) a region encoding a short peptidic linker (AAASGG (SEQ ID NO 16) or 
GGSAAA (SEQ ID NO: 17)) connecting the protein of interest to the FOS 
dimerization domain. Relevant coding regions are given in upper case letters. 
The arrangement of restriction cleavage sites allows easy construction of FOS 
fusion genes. The cassettes are cloned into the EcoRI/Hindlll restriction sites of 
the expression vector pMPSVEH (Artelt et al, Gene 65:213-219 (1988)). 



pAV5 

This vector is designed for the eukaryotic production of fusion proteins 
with FOS at the C-terminus. The gene of interest (g o.i.) may be inserted between 
the sequences coding for the hGH signal sequence and theFOS' domain by ligation 
into the Eco47III/NofI sites of the vector. Alternatively, a gene containing its 
own signal sequence may be fused to the FOS coding region by ligation into the 
StuI/NotI sites. 



-82- 



EcoRI StuI 31/11 

qaa ttc aqcr cct ATG GCT ACA GGC TCC CGG ACG TCC CTG CTC CTG GCT 
TTT GGC CTG CTC 



61/21 EC047III NotI 

TGC CTG CCC TGG CTT CAA GAG GGC AGC GCT ggg tgt ggg GCG GCC GC T 
TCT GGT GGT TGC 

CLPWLQEGSA (goi) AAA 



GGT GGT CTG ACC GAC ACC CTG CAG GCG GAA ACC GAC CAG GTG GAA GAC 
GAA AAA TCC GCG 



CTG CAA ACC GAA ATC GCG AAC CTG CTG AAA GAA AAA GAA AAG CTG GAG 
TTC ATC CTG GCG 



241/81 Hindlll 

GCA CAC GGT GGT TGC t aa get t (SEQ ID NO: 27) 

A h g g c * (SEQIDNO:28) 
pAV6 

[0295] This vector is designed for the eukaryotic production of fusion proteins 

with FOS at the N-terminus. The gene of interest (g o.i.) may be ligated into the 
Notl/StuI (or Notl/Hindlll) sites of the vector. 



EcoRI 31/11 

Qaa ttc ATG GCT ACA GGC TCC CGG ACG TCC CTG CTC CTG GCT TTT GGC 
CTG CTC TGC CTG 

MATGSRTSLLLAFG 
L L C L 

61/21 EC047III 91/31 



CCC TGG CTT CAA GAG GGC AGC GCT TGC GGT GGT CTG ACC GAC ACC CTG 
CAG GCG GAA ACC 



-83- 



PWLQEGSACGGLTDTL 
Q A E T 

121/41 151/51 

GAC CAG GTG GAA GAC GAA AAA TCC GCG CTG CAA ACC GAA ATC GCG AAC 
CTG CTG AAA GAA 

DQVEDEKSALQTEIAN 
L L K E 

181/61 211/71 
Not I 

AAA GAA AAG CTG GAG TTC ATC CTG GCG GCA CAC GGT GGT TGC GGT GGT 
TCT GCG GCC GC T 

KEKLEFILAAHGGCGG 
S A A A 

241/81 StuI Hindi I I 

ggg tgt ggg agg cct aag ctt (SEQIDNO:29) 
(goi) (SEQIDNO.30) 

Construction of expression vectors pAVl - pAV6 

[0296] The following oligonucleotides have been synthesized for construction of 

expression vectors pAVl - pAV6: 
FOS-FORl: 

CCTGGGTGGGGGCGGCCGCTTCTGGTGGTTGCGGTGGTCTGACC(SEQ 

IDNO:31); 

FOS-FOR2: 

GGTGGGAATTCAGGAGGTAAAAAGATATCGGGTGTGGGGCGGCC 

(SEQ ID NO 32); 

FOS-FOR3 

GGTGGGAATTCAGGAGGTAAAAAACGATGGCTTGCGGTGGTCTGACC 

(SEQIDNO:33); 

FOS-FOR4: 

GCTTGCGGTGGTCTGACC (SEQ ID NO:34); 
FOS-BEVl: 

CCACCAAGCTTAGCAACCACCGTGTGC (SEQ ID N0.35); 
FOS-REV2. 



-84- 



CCACCAAGCTTGATATCCCCACACCCAGCGGCCGCAGAACCACCGC 

AACCACCG (SEQ ID NO: 3 6); 

FOS-REV3: 

CCACCAAGCTTAGGCCTCCCACACCCAGCGGC (SEQ ID N0.37); 
OmpA-FORl. 

GGTGGGAATTCAGGAGGTAAAAAACGATG (SEQ ID N0.38); 
hGH-FORl . 

GGTGGGAATTCAGGCCTATGGCTACAGGCTCC (SEQ ID NO:39); and 
hGH-FOR2: 

GGTGGGAATTCATGGCTACAGGCTCCC (SEQ ID NO.40). 

[0297] For the construction of vector pAV2, the regions coding for the OmpA 

signal sequence and the FOS domain were amplified from the ompA-FOS-hGH 
fusion gene in vector pKK223-3 (see Example 5) using the primer pair 
OmpA-FORl/ FOS-KEV2. The PGR product was digested with EcoRI/Hindlll 
and ligated into the same sites of vector pKK223-3 (Pharmacia) 

[0298] For the construction of vector pAVl, the FOS coding region was 

amplified from the ompA-FOS-hGH fusion gene in vector pKK223-3 (see 
Example 5) usingthe primer pair FOS-F OR1 / FOS-KEV1. The PCR product was 
digested with Hindlll and ligated into Stul/Hindlll digested vector pAV2 

[0299] For the construction of vector pAV3, the region coding for the FOS 

domain was amplified from vector pAVl using the primer pair 
FOS-FOK2/FOS-RE V 1 . The PCR product was digested with EcoRI/Hindlll and 
ligated into the same sites of the vector pKK223-3 (Pharmacia). 

[0300] For the construction of vector pAV4, the region coding for the FOS 

domain was amplified from the omp A-FOS-hGH fusion gene in vector pKK223 -3 
(see Example 5) using the primer pair FOS-FOR3/FOS-KEV2. The PCR product 
was digested with EcoRI/Hindlll and ligated into the same sites of the vector 
pKK223-3 (Pharmacia). 



-85- 



[0301] For the construction of vector pAV5, the region coding for the hGH signal 

sequence is amplified from the hGH-FOS-hGH fusion gene in vector pSINrep5 
(see Example 7) using the primer pair hGH-FOR 1 /hGHRE V 1 . The PCR product 
is digested with EcoRI/NotI and ligated into the same sites of the vector pAVl. 
The resulting cassette encoding the hGH signal sequence and the FOS domain is 
then isolated by EcoRI/Hindlll digestion and cloned into vector pMPSVEH 
(Artelt et al., Gene 55:213-219 (1988)) digested with the same enzymes. 

[0302] For the construction of vector pAV6, the FOS coding region is amplified 

from vector pAV2 using the primer pair FOS-FOR4/FOSKEV3 . The PCR 
product is digested with Hindlll and cloned into Eco47III/HindIII cleaved vector 
pAV5. The entire cassette encoding the hGH signal sequence and the FOS 
domain is then reamplified from the resulting vector using the primer pair 
hGH-FOR2/FOSREV3, cleaved with EcoRI/Hindlll and ligated into vector 
pMPSVEH (Artelt et al, Gene 65:213-219 (1988)) cleaved with the same 
enzymes. 

EXAMPLE 7: 

Construction of FOS-hGH with human (hGH) signal sequence 
[0303] For eukaryotic expression of the FOS-hGH fusion protein, the 

OmpA-FOS-hGH fusion gene was isolated from pBluescript: :OmpA-F(3.S-hGH 
(see Example 4) by digestion with Xbal/B sp 1 201 and cloned into vector pSINrep 5 
(Invitrogen) cleaved with the same enzymes. The hGH signal sequence was 
synthesized by PCR (reaction mix: 50 pmol of each primer, dATP, dGTP, dTTP, 
dCTP (200 uM each), 2 5 U Taq DNA polymerase (Qiagen), 50 ul total volume 
in the buffer supplied by the manufacturer; amplification: 92 °C for 30 seconds, 
55 °C for 30 seconds, 72 °C for 30 seconds, 30 cycles) using the overlapping 
oligonucleotides Sig-hGH-FOR 

(GGGTCTAGAATGGCTACAGGCTCCCGGACGTCCCTGCTCCTGGCTT 
TTGGCCTGCTCTG) (SEQ ID NO:41) and Sig-hGH-REV 



| 



(CGCAGGCCTCGGCACTGCCCTCTTGAAGCCAGGGCAGGCAGAGCA 
GGCCAAAAGCCAG) (SEQ ID NO. 42). The PCR product was purified using 
the QiaExII Kit, digested with Stul/Xbal and ligated into vector 
pSrNrep5.:OmpA-F0,S , -hGH cleaved with the same enzymes. 

EXAMPLE 8: 
Eukaryotic expression of FOS-hGH 
[0304] RNase-free vector (1.0 ug) (pSINrep5: :OmpA-FOS-hGH) and 1 .0 ug of 

DHEB (Bredenbeek et al, J. Virol. 67:6439-6446 (1993)) were linerarized by 
Seal restriction digest. Subsequently, in vitro transcription was carried out using 
an SP6 in vitro transcription kit (InvitroscripCAP by InvitroGen, Invitrogen BV, 
NV Leek, Netherlands). The resulting 5 '-capped mRNA was analyzed on 
reducing agarose-gel. 

[0305] In vitro, transcribed mRNA 5 ug was electroporated into BHK 21 cells 

(ATCC: CCL10) according to Invitrogen' s manual (Sindbis Expression system, 
Invitrogen BV, Netherlands). After 10 hours incubation at 37°C the FCS 
containing medium was exchanged by HP- 1 medium without FCS, followed by an 
additional incubation at 37°C for 10 hours. The supernatant was harvested and 
analyzed by dot-blot analysis for production of FOS-hgh. 

[0306] Culture media (2.5 pi) was spotted on a nitrocellulose membrane and dried 

for 10 minutes at room temperature. The membrane was blocked with 1 % bovine 
albumin (Sigma) in TBS (lOxTBS per liter: 87.7 g NaCl, 66. lg Trizma 
hydrochloride (Sigma) and 9.7 g Trizma base (Sigma), pH 7.4) for 1 hour at room 
temperature, followed by an incubation with 2 ug rabbit anti-human hGH antibody 
(Sigma) in 10 ml TBS-T (TBS with 0.05% Tween20) for 1 hour. The blot was 
washed 3 times for 10 minutes with TBS-T and incubated for 1 hour with alkaline 
phosphatase conjugated anti-rabbit IgG (Jackson ImmunoResearch Laboratories, 
Inc.) diluted 1 :5000 in TBS-T. After washing 2 times for 10 minutes with TBS-T 
and 2 times for 10 minutes with TBS, the blot was developed by AP staining as 
described in Example 2. Results are shown in Figure 3. 



-87- 



EXAMPLE 9. 
Construction of FOS-PLA (N- and C-terminal) 
[0307] The following gene is constructed by chemical gene synthesis coding for 

a catalytically inactive variant (Forster et al, J. Allergy Clin. Immunol. 95. 
1229-1235 (1995)) of bee venom phospholipase A 2 (PLA) 

1/1 31/11 

ATC ATC TAC CCA GGT ACT CTG TGG TGT GGT CAC GGC AAC AAA TCT TCT 
GGT CCG AAC GAA 



CTC GGC CGC TTT AAA CAC ACC GAC GCA TGC TGT CGC ACC CAG GAC ATG 
TGT CCG GAC GTC 

LGRFKHTDACCRTQDM 



ATG TCT GCT GGT GAA TCT AAA CAC GGG TTA ACT AAC ACC GCT TCT CAC 
ACG CGT CTC AGC 

MSAGE S KHGLTNTASH 



TGC GAC TGC GAC GAC AAA TTC TAC GAC TGC CTT AAG AAC TCC GCC GAT 
ACC ATC TCT TCT 

CDCDDKFYDCLKNSAD 



TAC TTC GTT GGT AAA ATG TAT TTC AAC CTG ATC GAT ACC AAA TGT TAC 
AAA CTG GAA CAC 

YFVGKMYFNLIDTKCY 



CCG GTA ACC GGC TGC GGC GAA CGT ACC GAA GGT CGC TGC CTG CAC TAC 
ACC GTT GAC AAA 



TCT AAA CCG AAA GTT TAC CAG TGG TTC GAC CTG CGC AAA TAC (SEQ 

ID NO:43) 

SKPKVYQWFDLRKY (SEQ 

ID NO:44) 

For fusion of PL A to the N-terminus of the FOS dimerization domain, the 
region is amplified using the oligonucleotides PLA-FOR1 
(CCATCATCTACCCAGGTAC) (SEQ ID NO:45) and PLA-REV1 
(CCCACACCCAGCGGCCGCGTATTTGCGCAGGTCG) (SEQ ID N0 46). 
The PCR product is cleaved with NotI and ligated into vector pAVl previously 
cleaved with the restriction enzymes Stul/Notl. For fusion of PL A to the 
C-terminus of the FOS dimerization domain, the region is amplified using the 
oligonucleotides PLA-FOR2 

(CGGTGGTTCTGCGGCCGCTATCATCTACCCAGGTAC) (SEQ ID NO:47) 
and PLA-REV2 (TTAGTATTTGCGCAGGTCG) (SEQ ID NO:48). The PCR 
product is cleaved with NotI and ligated into vector p AV2 previously cleaved with 
the restriction enzymes Notl/EcoRV 

EXAMPLE 10: 

Construction of FOS-Ovalbumin fusion gene (N- and C-terminal) 
] For cloning of the ovalbumin coding sequence, mRNA from chicken 

oviduct tissue is prepared using the QuickPrep™ Micro mRNA Purification Kit 
(Pharmacia) according to manufacturer instructions Using the Superscript™ 
One-step RT PCR Kit (Gibco BRL), a cDNA encoding the mature part of 
ovalbumin (corresponding to nucleotides 68-1222 of the mRNA (McReynolds et 
al, Nature 275:723-728 (1978)) is synthesized using the primers Ova-FORl 
(CCGGCTCCATCGGTGCAG) (SEQ ID NO:49) and Ova-REVl 
(ACC ACCAGAAGCGGCCGC AGGGGAAAC ACATCTGCC) (SEQ ID NO : 50). 
The PCR product is digested with NotI and cloned into Stul/Notl digested vector 
pAVl for expression of the fusion protein with the FOS dimerization domain at 
the C terminus. For production of a fusion protein with the FOS dimerization 



domain at the N terminus, the Ovalbumin coding region is amplified from the 
constructed vector (p AVI:: Ova) using the primers Ova-FOR2 
(CGGTGGTTCTGCGGCCGCTGGCTCCATCGGTGCAG) (SEQ ID NO:51) 
and Ova-REV2 (TTAAGGGGAAACACATCTGCC) (SEQ ID NO: 52) The 
PCR product is digested with NotI and cloned into the Notl/EcoRV digested 
vector pAV2. Cloned fragments are verified by DNA sequence analysis 

EXAMPLE 1 1 
Production and purification of FOS-PLA and 
FOS ovalbumin fusion proteins 
For cytoplasmic production of FOS fusion proteins, an appropriate £. coli 
strain was transformed with the vectors p AV3 : :PL A, p AV4 : : PL A, p AV3 : : Ova or 
pAV4::Ova. The culture was incubated in rich medium in the presence of 
ampicillin at 37°C with shaking. At an optical density (550nm) of 1, 1 mM IPTG 
was added and incubation was continued for another 5 hours. The cells were 
harvested by centrifugation, resuspended in an appropriate buffer {e.g., tris-HCl, 
pH 7.2, 150 mM NaCl) containing DNase, RNase and lysozyme, and disrupted 
by passage through a french pressure cell. After centrifugation (Sorvall RC-5C, 
SS34 rotor, 15000 rpm, 10 min, 4°C), the pellet was resuspended in 25 ml 
inclusion body wash buffer (20 mM tris-HCl, 23% sucrose, 0.5% Triton X-100, 
1 mMEDTA, pH8) at 4°C and recentrifuged as described above. This procedure 
was repeated until the supernatant after centrifugation was essentially clear. 
Inclusion bodies were resuspended in 20 ml solubilization buffer (5.5 M 
guanidinium hydrochloride, 25 mM tris-HCl, pH 7.5) at room temperature and 
insoluble material was removed by centrifugation and subsequent passage of the 
supernatant through a sterile filter (0.45 um). The protein solution was kept at 

4 ° C for at least 1 0 hours in the presence of 1 0 mM EDT A and 1 00 mM DTT and 
then dialyzed three times against 1 0 volumes of 5 . 5 M guanidinium hydrochloride, 
25 mM tris-HCl, 10 mM EDT A, pH 6. The solution was dialyzed twice against 

5 liters of 2 M urea, 4 mM EDTA, 0. 1 M NH 4 C1, 20 mM sodium borate (pH 8.3) 



-90- 



in the presence of an appropriate redox shuffle (oxidized glutathione/reduced 
glutathione; cystine/cysteine). The refolded protein was then applied to an ion 
exchange chromatography. The protein was stored in an appropriate buffer with 
a pH above 7 in the presence of 2-10 mM DTT to keep the cysteine residues 
flanking the FOS domain in a reduced form. Prior to coupling of the protein with 
the alphavirus particles, DTT was removed by passage of the protein solution 
through a Sephadex G-25 gel filtration column. 

EXAMPLE 12: 
Constructions of gpl40-FOS 
[0311] jThe gpHO gene (Swiss-Prot:P03375) without the internal protease 

cleavage site was amplified by PCR from the original plasmid pAbT4674 (ATCC 
40829) containing the full length gp 1 60 gene using the following oligonucleotides: 
HIV-1. 

5'-ACTAGTCTAGAatgagagtgaaggagaaatatc-3' (SEQ ID NO: 53); 
fflV-end: 

5'-TAGCATGCTAGCACCGAAtttatctaattccaataattcttg-3' (SEQ ID NO:54); 
HIV-Cleav: 

5 '-gtagcacccaccaaggcaaagCTGAAAGCTACCCAGCTCGAGAAACTGgca-3 ' 

(SEQ ID NO 55); and 

HIV-Cleav2: 

5 ' -caaagctcctattcccactgcC AGTTTCTCGAGCTGGGT AGCTTTC AG-3 ' 
(SEQ ID NO 56). 

[0312] For PCR I, 100 pmol of oligo HIV-1 and HIV-Cleav2 and 5 ng of the 

template DNA were used in the 75 ul reaction mixture (4 units of Taq or Pwo 
polymerase, 0. 1 mM dNTPs and 1 .5 mM MgS0 4 ). PCR cycling was done in the 
following manner. 30 cycles with an annealing temperature of 60 °C and an 
elongation time of 2 minutes at 72 °C. 



-91- 



[0313] For PCR II, 100 pmol of oligo HIV-end and HIV-Cleav and 5 ng of the 

template DNA were used in the 75 \i\ reaction mixture, (4 units of Taq or Pwo 
polymerase, 0. 1 mM dNTPs and 1 .5 mM MgS0 4 ). PCR cycling was done in the 
following manner: 30 cycles with an annealing temperature of 60 °C and an 
elongation time of 50 seconds at 72 °C. 

[0314] Both PCR fragments were purified, isolated and used in an assembly PCR 

reaction. For the assembly PCR reaction, 100 pmol of oligo HIV-1 and HIV-end 
and 2 ng of each PCR fragment (PCRI and PCR II) were used in the 75 (4 units 
of Taq or Pwo polymerase, 0. 1 mM dNTPs and 1 .5 mM MgS0 4 ). PCR cycling 
was done in the following manner: 30 cycles with an annealing temperature of 
60 ° C and an elongation time of 2 . 5 minutes at 72 ° C . The assembly PCR product 
was digested Xbal and Nhel. The FOS amphiphatic helix was fused in frame to 
the C-terminal end of gp-140. 

[0315] The DNA sequence coding for the FOS amphiphatic helix domain was 

PCR-amplified from vector pJuFo (Crameri & Suter Gene 137.69 (1993)) using 
the oligonucleotides: 
FOS-HIV: 

5'-ttcggtgctagcggtggcTGCGGTGGTCTGACCGAC-3' (SEQ ID NO: 57); and 
FOS-Apa: 

5'-gatgctgggcccttaaccGCAACCACCGTGTGCCGCC-3' (SEQ IDNO:58). 

[0316] For the PCR reaction, 100 pmol of each oligo and 5 ng of the template 

DNA was used in the 75 ul reaction mixture (4 units of Taq or Pwo polymerase, 
0. 1 mM dNTPs and 1 . 5 mM MgSOJ. Temperature cycling was done as follows: 
95 °C for 2 minutes, followed by 5 cycles of 95 °C (45 seconds), 60°C (30 
seconds), 72 °C (25 seconds) and followed by 25 cycles of 95 °C (45 seconds), 
68 0 C (3 0 seconds) , 72 0 C (20 seconds) . The obtained PCR fragment was digested 
with Nhel and Bspl20L. 

[0317] The final expression vector for GPIA0-FOS was obtained in a 3 fragment 

ligation of both PCR fragments into pSinRep5. The resultant vector 
pSinRep5-GP 1 40-FOS was evaluated by restriction analysis and DNA sequencing. 



-92- 



[0318] GP 1 40-FOS was also cloned into pCYTts via Xbal and Bspl20L to obtain 

a stable, inducible GP14Q-FOS expressing cell line. 

EXAMPLE 13: 
Expression of GP140FOS using P SinRep5-GP140FOS 
[0319] RNase-free vector (1.0 ug)(pSinRep5-GP140-FOS) and 1.0 ugofDHEB 

(Bredenbeek et al, J. Virol. (57:6439-6446 (1993)) were linearized by restriction 
digestion. Subsequently, in vitro transcription was carried out using an SP6 in 
vitro transcription kit (InvitroscripCAP by InvitroGen, Invitrogen BV, NV Leek, 
Netherlands). The resulting 5 '-capped mRNA was analyzed on a reducing 
agarose-gel. 

[0320] In vitro transcribed mRNA (5 ug) was electroporated into BHK 21 cells 

(ATCC: CCL10) according to Invitrogen's manual (Sindbis Expression System, 
Invitrogen BV, Netherlands). After 10 hours incubation at 37° C, the FCS 
containing medium was exchanged by HP- 1 medium without FCS, followed by an 
additional incubation at 37 °C for 10 hours. The supernatant was harvested and 
analyzed by Western blot analysis for production of soluble GP140-FCW exactly 
as described in Example 2. 

EXAMPLE 14. 
Expression of GP140FOS using pCYTts-GP140FOS 
[0321] pCYT-GPMO-FCW 20 ug was linearized by restriction digestion. The 

reaction was stopped by phenol/chloroform extraction, followed by anisopropanol 
precipitation of the linearized DNA. The restriction digestion was evaluated by 
agarose gel eletrophoresis. For the transfection, 5.4 ug of linearized 
pCYTtsGP MO-FaS 1 was mixed with 0.6 ug of linearized pSV2Neo in 30 ul FLO 
and 30 pi of 1 M CaCl 2 solution was added. After addition of 60 ul phosphate 
buffer(50mMHEPES,280mMNaCl, 1 5 mMNa 2 HP0 4 ,pH 7.05), the solution 
was vortexed for 5 seconds, followed by an incubation at room temperature for 
25 seconds The solution was immediately added to 2 ml HP-1 medium 



-93- 



containing 2% FCS (2% FCS medium). The medium of an 80% confluent 
BHK21 cell culture (6-well plate) was then replaced by the DNA containing 
medium. After an incubation for 5 hours at 37° C in a C0 2 incubator, the DNA 
containing medium was removed and replaced by 2 ml of 1 5% glycerol in 2% FCS 
medium. The glycerol containing medium was removed after a 30 second 
incubation phase, and the cells were washed by rinsing with 5 ml of HP-1 medium 
containing 10% FCS. Finally 2 ml of fresh HP-1 medium containing 10% FCS 
was added. 

[0322] Stably transfected cells were selected and grown in selection medium 

(FIP-1 medium supplemented with G41 8) at 37 °C in a C0 2 incubator. When the 
mixed population was grown to confluency, the culture was split to two dishes, 
followed by a 12 h growth period at 37°C. One dish of the cells was shifted to 
30 °C to induce the expression of soluble GP140-FOS. The other dish was kept 
at37°C. 

[0323] The expression of soluble GP140-FOS was determined by Western blot 

analysis. Culture media (0.5 ml) was methanol/chloroform precipitated, and the 
pellet was resuspended in SDS-PAGE sample buffer. Samples were heated for 5 
minutes at 95 °C before being applied to a 1 5% acrylamide gel. After SDS-PAGE, 
proteins were transferred to Protan nitrocellulose membranes (Schleicher & 
Schuell, Germany) as described by Bass and Yang, in Creighton, T.E., ed , 
Protein Function: A Practical Approach, 2nd Edn., IRL Press, Oxford (1997), 
pp. 29-55. The membrane was blocked with 1 % bovine albumin (Sigma) in TBS 
(lOxTBS per liter: 87 7 g NaCl, 66. lg Trizma hydrochloride (Sigma) and 9.7 g 
Trizma base (Sigma), pH 7 4) for 1 hour at room temperature, followed by an 
incubation with an anti-GP140 or GP-160 antibody for 1 hour. The blot was 
washed 3 times for 10 minutes with TBS-T (TBS with 0.05% Tween20), and 
incubated for 1 hour with an alkaline-phosphatase-anti- 
mouse/rabbit/monkey/human IgG conjugate. After washing 2 times for 10 
minutes with TBS-T and 2 times for TO minutes with TBS, the development 
reaction was carried out using alkaline phosphatase detection reagents (10 ml AP 



-94- 



buffer (1 00 mM Tris/HCl, 1 00 mM NaCl, pH 9. 5) with 50 |ul NBT solution (7.7% 
Nitro Blue Tetrazolium (Sigma) in 70% dimethylformamide) and 37 ul of 
X-Phosphate solution (5% of 5-bromo-4-chloro-3-indolyl phosphate in 
dimethylformamide) . 

EXAMPLE 15: 
Production and purification of GP140FOS 

[0324] An anti-gp 1 20 antibody was covalently coupled to a NHS/EDC activated 

dextran and packed into a chromatography column. The supernatant, containing 
GP 140FOS is loaded onto the column and after sufficient washing, GP1 40FOS 
was eluted using 0. 1 M HC1. The eluate was directly neutralized during collection 
using 1 M Tris pH 7.2 in the collection tubes. 

[0325] Disulfide bond formation might occur during purification, therefore the 

collected sample is treated with 10 mM DTT in 10 mM Tris pH 7.5 for 2 hours 
at25°C. 

[0326] DTT is remove by subsequent dialysis against 10 mM Mes; 80 mM NaCl 

pH 6. 0 Finally GP 1 40FOS is mixed with alphavirus particles containing the JUN 
leucine zipper in E2 as described in Example 16. 

EXAMPLE 16: 
Preparation of the Alpha Vaccine Particles 
[0327] Viral particles {see Examples 2 and 3) were concentrated using Millipore 

Ultrafree Centrifugal Filter Devices with a molecular weight cut-off of 100 kD 
according to the protocol supplied by the manufacturer Alternatively, viral 
particles were concentrated by sucrose gradient centrifugation as described in the 
instruction manual of the Sindbis Expression System (Invitrogen, San Diego, 
California). The pH of the virus suspension was adjusted to 7.5 and viral particles 
were incubated in the presence of 2-10 mMDTT for several hours. Viral particles 
were purified from contaminating protein on a Sephacryl S-300 column 
(Pharmacia) (viral particles elute with the void volume) in an appropriate buffer. 



-95- 



[0328] Purified virus particles were incubated with at least 240 fold molar excess 

of FOS-antigen fusion protein in an appropriate buffer (pH 7.5-8.5) in the 
presence of a redox shuffle (oxidized glutathione/reduced glutathione; 
cystine/cysteine) for at least 10 hours at 4°C. After concentration of the particles 
using a Millipore Ultrafree Centrifugal Filter Device with a molecular weight 
cut-off of 1 00 kD, the mixture was passed through a Sephacryl S-300 gel filtration 
column (Pharmacia). Viral particles were eluted with the void volume. 

EXAMPLE 17 

Fusion of JUN amphipathic helix to the amino terminus of HBcAg( 1-144) 
[0329] The JUN helix was fused to the amino terminus of the EDBcAg amino acid 

sequence 1 to 144 (JUN-HBcAg construct). For construction of the JUN-F£BcAg 
DNA sequence, the sequences encoding the JUN helix and HBcAg(l -144) were 
amplified separately by PCR. The JUN sequence was amplified from the pJuFo 
plasmid using primers EcoRI-JUN(s) and JUN-SacII(as). The EcoRI-JUN(s) 
primer introduced an EcoRI site followed by a start ATG codon. The JUN- 
SacII(as) primer introduced a linker encoding the amino acid sequence GAAGS. 
The HBcAg (1-144) sequence was amplified from the pEco63 plasmid (obtained 
from ATCC No. 31518) using primers JUN-HBcAg(s) and 
HBcAg(l-144)Hind(as). JUN-HBcAg(s) contained a sequence corresponding to 
the 3' end of the sequence encoding the JUN helix followed by a sequence 
encoding the GAAGS linker and the 5 ' end of the FfficAg sequence. HBcAg(l- 
144)Hind(as) introduces a stop codon and a Hindlll site after codon 144 of the 
HBcAg gene. For the PCR reactions, 100 pmol of each oligo and 50 ng of the 
template DNAs were used in the 50 ^1 reaction mixtures with 2 units of Pwo 
polymerase, 0. 1 mM dNTPs and 2 mM MgS0 4 . For both reactions, temperature 
cycling was carried out as follows 94 °C for 2 minutes; and 30 cycles of 94° C (1 
minute), 50°C (1 minute), 72°C (2 minutes). 



-96- 



[0330] Primer sequences. 

EcoRI-JUN(s): 

(5 '-CCGGAATTCATGTGCGGTGGTCGGATCGCCCGG-3 ') (SEQ ID 
NO:61); 

JUN-SacII(as)- 

(5'-GTCGCTACCCGCGGCTCCGCAACCAACGTGGTTCATGAC-3 ') (SEQ 
IDNO:62); 

JUN-HBcAg(s): 

(5 '-GTTGGTTGCGGAGCCGCGGGTAGCGACATTGACCCTTATAAAGAATTTGG-3 ') 
(SEQ ID NO:63); 

HBcAg(l-144)Hind(as): 

(5 -CGCGTCCCAAGCTTCTACGGAAGCGTTGATAGGATAGG-3 ') (SEQ 
ID NO:64). 

[0331] Fusion of the two PCR fragments was performed by PCR using primers 

EcoRI-JUN(s) and HBc Ag(l-144)Hind(as) .100 pmol of each oligo was used with 
lOOng of the purified PCR fragments in a 50 ul reaction mixture containing 2 units 
of Pwo polymerase, 0. 1 mM dNTPs and 2 mM MgS0 4 . PCR cycling conditions 
were: 94°C for 2 minutes; and 35 cycles of 94°C (1 minute), 50°C (1 minute), 
72° C (2 minutes). The final PCR product was analyzed by agarose gel 
electrophoresis, purified and digested for 16 hours in an appropriate buffer with 
EcoRI and Hindlll restriction enzymes. The digested DNA fragment was ligated 
into EcoRI/Hindlll-digested pKK vector to generate pKK-JUN-HBcAg 
expression vector. Insertion of the PCR product was analyzed by EcoRI/Hindlll 
restriction analysis and by DNA sequencing of the insert. 



-97- 



EXAMPLE 18 

Fusion of JUN amphipathic helix to the carboxy terminus of HBcAg( 1-144) 
The JUN helix was fused to the carboxy terminus of the HBcAg amino 
acid sequence 1 to 1 44 (HBcAg-JUN construct) . For construction of the UBcAg- 
JUN DNA sequence, the sequences encoding the JUN helix and HBcAg(l-144) 
were amplified separately by PCR. The JUN sequence was amplified from the 
pJuFo plasmid with primers SacII-JUN(s) and JUN-Hindlll(as). SacII-JUN(s) 
introduced a linker encoding amino acids LAAG. This sequence also contains a 
SacII site. JUN-Hindlll(as) introduced a stop codon (TAA) followed by a 
Hindlll site. The HBcAg( 1 - 1 44) DNA sequence was amplified from the pEco63 
plasmid using primers EcoRI-HBcAg(s) and HBcAg(l-144)-JUN(as). EcoRI- 
HBcAg(s) introduced an EcoRI site prior to the Start ATG of the HBcAg coding 
sequence. HBcAg(l-144)-JUN(as) introduces a sequence encoding the peptide 
linker (LAAG), which also contains a SacII site. For the PCR reactions, 1 00 pmol 
of each oligo and 50 ng of the template DNAs were used in the 50 ul reaction 
mixtures with 2 units of Pwo polymerase, 0.1 mM dNTPs and 2 mM MgS0 4 . 
Temperature cycling was carried out as follows: 94 °C for 2 minutes; and 30 
cycles of 94°C (1 minute), 50°C (1 minute), 72°C (2 minutes). 
Primer sequences 

SacII- JUN(s): 

(5 '-CTAGCCGCGGGTTGCGGTGGTCGGATCGCCCGG-3 ') (SEQ ID 
NO:65); 

JTJN-Hindlll(as). 

(5'-CGCGTCCCAAGCTTTTAGCAACCAACGTGGTTCATGAC -3') (SEQ 
ID NO:66); 



EcoRI-HBcAg(s). 

(5 '-CCGGAATTCATGGACATTGACCCTTATAAAG-3 ') (SEQ ID NO:67), 
and 

HBcAg-JUN(as): 

(5 -CCGACCACCGCAACCCGCGGCTAGCGGAAGCGTTGATAGGATAGG-3 ') 
(SEQ ID NO. 68). 

[0334] Fusion of the two PCR fragments was performed by PCR using primers 

EcoRI-HBcAg(s) and JUN-Hindlll(as). For the PCR fusion, 100 pmol of each 
oligo was used with lOOng of the purified PCR fragments in a 50 ul reaction 
mixture containing 2 units of Pwo polymerase, 0. 1 mM dNTPs and 2 mM MgS0 4 . 
PCR cycling conditions were: 94°C for 2 minutes; and 35 cycles of 94°C (1 
minute), 50 °C (1 minute), 72 °C (2 minutes). The final PCR product was 
analyzed by agarose gel electrophoresis, and digested for 16 hours in an 
appropriate buffer with EcoRI and Hindlll restriction enzymes. The DNA 
fragment was gel purified and ligated into EcoRI/Hindlll-digested pKK vector to 
generate pKK-UBcAg- JUN expression vector. Insertion of the PCR product was 
analyzed by EcoRI/Hindlll restriction analysis and by DNA sequencing of the 
insert. 

EXAMPLE 19 

Insertion of JUN amphipathic helix into the c/el epitope of HBcAg(l-144) 
[0335] The c/el epitope (residues 72 to 88) of HBcAg is known to be located in 

the tip region on the surface of the Hepatitis B virus capsid. A part of this region 
(residues 76 to 82) of the protein was genetically replaced by the JUN helix to 
provide an attachment site for antigens (HBcAg-JUNIns construct). The FfJBcAg- 
JUNIns DNA sequence was generated by PCRs: The JUN helix sequence and 
two sequences encoding HBcAg fragments (amino acid residues 1 to 75 and 83 
to 1 44) were amplified separately by PCR. The JUN sequence was amplified from 



-99- 



the pJuFo plasmid with primers BamHI-JUN(s) and JUN-SacII(as). BamHI- 
JUN(s) introduced a linker sequence encoding the peptide sequence GSGGG that 
also contains a BamHI site. JUN-SacII(as) introduced a sequence encoding the 
peptide linker GAAGS followed by a sequence complementary to the 3 ' end of 
the JUN coding sequence. The HBcAg(l-75) DNA sequence was amplified from 
the pEco63 plasmid using primers EcoRIHBcAg(s) and HBcAg75-JUN(as). 
EcoRIHBcAg(s) introduced an EcoRI site followed by a sequence corresponding 
to the 5' end of the HBcAg sequence HBcAg75-JUN(as) introduced a linker 
encoding the peptide GSGGG after amino acid 75 of HBcAg followed by a 
sequence complementary to the 5 ' end of the sequence encoding the JUN helix. 
The HBcAg (83-144) fragment was amplified using primers JUN-HBcAg83(s) 
and HBcAg(l-144)Hind(as). JUN-HBcAg83(s) contained a sequence 
corresponding to the 3 ' end of the JUN-encoding sequence followed by a linker 
encoding the peptide, GAAGS and a sequence corresponding to the 5' end of the 
sequence encoding HBcAg (83-144). HBcAg(l-144)Hind(as) introduced a stop 
codon and a Hindlll site after codon 144 of the HBcAg gene. For the PCR 
reactions, 100 pmol of each oligo and 50 ng of the template DNAs were used in 
the 50 ul reaction mixtures (2 units of Pwo polymerase, 0 1 mM dNTPs and 2 
mM MgS0 4 ). Temperature cycling was performed as follows: 94°C for 2 
minutes; and 35 cycles of 94°C (1 minute), 50°C (1 minute), 72°C (2 minutes). 

[0336] Primer sequences: 

BamHI- JUN(s): 

(5 '-CTAATGGATCCGGTGGGGGCTGCGGTGGTCGGATCGCCCGGCTCGAG-3 ') 
(SEQ ID NO: 69); 

JUN-SacII(as): 

(5 '-GTCGCTACCCGCGGCTCCGCAACCAACGTGGTTCATGAC-3 ') (SEQ 
ID NO:70); 



-100- 



EcoRIHBcAg(s): 

(5'- CCGGAATTCATGGACATTGACCCTTATAAAG-3') (SEQ ID N0.71); 
HBcAg75-JUN (as): 

(5'-CCGACCACCGCAGCCCCCACCGGATCCATTAGTACCCACCCAGGTAGC-3') 
(SEQ ID NO:72); 

JUN-HBcAg83(s): 

(5 -GTTGGTTGCGGAGCCGCGGGTAGCGACCTAGTAGTCAGTTATGTC-3 ') 
(SEQ ID NO :73); and 

HBcAg(l-144)Hind(as): 

(5 '-CGCGTCCCAAGCTTCTACGGAAGCGTTGATAGGATAGG-3 ') (SEQ 
IDNO:74). 

[0337] Fusion of the three PCR fragments was performed as follows. First, the 

fragment encoding HBcAg 1-75 was fused with the sequence encoding JUN by 
PCR using primers EcoRIHBcAg(s) and JUN-SacII(as). Second, the product 
obtained was fused with the HBcAg(83-144) fragment by PCR using primers 
EcoRI HBcAg(s) and HBcAg Hindlll(as). For PCR fusions, 100 pmol of each 
oligo was used with 100 ng of the purified PCR fragments in a 50 ul reaction 
mixture containing 2 units of Pwo polymerase, 0. 1 mM dNTPs and 2 mM MgS0 4 . 
The same PCR cycles were used as for generation of the individual fragments. 
The final PCR product was digested for 16 hours in an appropriate buffer with 
EcoRI and Hindlll restriction enzymes. The DNA fragment was ligated into 
EcoRI/Hindlll-digested pKK vector, yielding the pKK-HBcAg-JUNIns vector. 
Insertion of the PCR product was analyzed by EcoRI/Hindlll restriction analysis 
and by DNA sequencing of the insert. 



-101- 



EXAMPLE 20 

Fusion of the JUN amphipathic helix to the carboxy terminus of the 
measles virus nucleocapsid (N) protein 
[0338] The JUN helix was fused to the carboxy terminus of the truncated measles 

virus N protein fragment comprising amino acid residues 1 to 473 (N473-JUN 
construct). For construction of the DNA sequence encoding N473-JUN the 
sequence encoding the JUN helix and the sequence encoding N473-JUN were 
amplified separately by PCR. The JUN sequence was amplified from the pJuFo 
plasmid with primers SacII-JUN(s) and JUN-Hindlll(as). SacII-JUN(s) 
introduced a sequence encoding peptide linker LAAG. This sequence also 
contained a SacII site. The JUN-Hindlll(as) anti-sense primer introduced a stop 
codon (TAA) followed by a Hindlll site. The N (1-473) sequence was amplified 
from the pSC-N plasmid containing the complete measles virus N protein coding 
sequence (obtained from M. Billeter, Zurich) using primers EcoRI-Nmea(s) and 
Nmea-JUN(as). EcoRI-N(mea)(s) introduced an EcoRI site prior to the Start 
ATG of the N coding sequence. N(mea)-JUN(as) was complementary to the 3 ' 
end of the N(l-473) coding sequence followed by a sequence complementary to 
the coding sequence for the peptide linker (LAAG) For the PCR reactions, 1 00 
pmol of each oligo and 50 ng of the template DNAs were used in the 50 ul 
reaction mixtures with 2 units of Pwo polymerase, 0.1 mM dNTPs and 2 mM 
MgS0 4 . Temperature cycling was performed as follows: 94°C for 2 minutes; and 
35 cycles of 94°C (1 minute), 55°C (1 minute), 72 C C (2 minutes). 

[0339] Primer sequences: 

SacII- JUN(s): 

(5 -CTAGCCGCGGGTTGCGGTGGTCGGATCGCCCGG-3 ') (SEQ ID 
NO:75), 



-102- 



JUN-Hindlll(as): 

(5'-CGCGTCCCAAGCTTTTAGCAACCAACGTGGTTCATGAC -3') (SEQ 
IDNO:76); 

EcoRI-Nmea(s): 

(S'-CCGGAATTCATGGCCACACTTTTAAGGAGC-S ') (SEQ IDNO:77); and 
Nmea-JUN(as): 

(5'-CGCGTCCCAAGCTTTTAGCAACCAACGTGGTTCATGAC-3 ') (SEQ ID 
NO:78). 

Fusion of the two PCR fragments was performed in a further PCR using 
primers EcoRI-Nmea(s) and Nmea-JUN(as). For the PCR fusion, 100 pmol of 
each oligo was used with 1 00 ng of the purified PCR fragments in a 50 ul reaction 
mixture containing 2 units of Pwo polymerase, 0 1 mM dNTPs and 2 mM MgS0 4 . 
Temperature cycling was performed as follows: 94 °C for 2 minutes; and 35 cycles 
of 94 °C (1 minute), 50° C (1 minute), 72 °C (2 minutes). The PCR product was 
digested for 16 hours in an appropriate buffer with EcoRI and Hindlll restriction 
enzymes. The DNA fragment was gel purified and ligated into EcoRI/Hindlll- 
digested pKK vector, yielding the pKK-N473-JUN plasmid Insertion of the PCR 
product was analyzed by EcoRI/Hindlll restriction analysis and by DNA 
sequencing of the insert. 

Example 21 

Expression and partial purification of HBcAg-JUN 
[0340] E. coli strain XL-1 blue was transformed with pKK-HBcAg-JUN. 1 ml 

of an overnight culture of bacteria was used to innoculate 100 ml of LB medium 
containing 1 00 ug/ml ampicillin. This culture was grown for 4 hours at 37 ° C until 
an OD at 600 nm of approximately 0.8 was reached. Induction of the synthesis 
of HBcAg-JUN was performed by addition of IPTG to a final concentration of 1 



-103- 



mM. After induction, bacteria were further shaken at 37 °C for 16 hours. 
Bacteria were harvested by centrifugation at 5000 x g for 15 minutes. The pellet 
was frozen at -20 °C. The pellet was thawed and resuspended in bacteria lysis 
buffer (10 mM Na 2 HP0 4 , pH 7.0, 30 mM NaCl, 0.25% Tween-20, 10 mM 
EDTA, 10 mM DTT) supplemented with 200 ug/ml lysozyme and 10 ul of 
Benzonase (Merck). Cells were incubated for 30 minutes at room temperature 
and disrupted using a French pressure cell. Triton X-l 00 was added to the lysate 
to a final concentration of 0.2%, and the lysate was incubated for 30 minutes on 
ice and shaken occasionally. Figure 4 shows HBcAg-JUN protein expression in 
E. coli upon induction with IPTG. E. coli cells harboring pKK-HBcAg-JUN 
expression plasmid or a control plasmid were used for induction of HBcAg-JUN 
expression with IPTG. Prior to the addition of IPTG, a sample was removed from 
the bacteria culture carrying the pKK -HBcAg-JUN plasmid (lane 3) and from a 
culture carrying the control plasmid (lane 1). Sixteen hours after addition of 
IPTG, samples were again removed from the culture containing pKK-HBcAg- 
JUN (lane 4) and from the control culture (lane 2). Protein expression was 
monitored by SDS-PAGE followed by Coomassie staining. 

The lysate was then centrifuged for 30 minutes at 12,000 x g in order to 
remove insoluble cell debris The supernatant and the pellet were analyzed by 
Western blotting using a monoclonal antibody against HBcAg (YVS1841, 
purchased from Accurate Chemical and Scientific Corp , Westbury, NY, USA), 
indicating that a significant amount of HBcAg-JUN protein was soluble (Fig. 5). 
Briefly, lysates from E. coli cells expressing HBcAg-JUN and from control cells 
were centrifuged at 14,000 x g for 30 minutes Supernatant (= soluble fraction) 
and pellet (= insoluble fraction) were separated and diluted with SDS sample 
buffer to equal volumes Samples were analyzed by SDS-PAGE followed by 
Western blotting with anti-HBcAg monoclonal antibody YVS 1841. Lane 1. 
soluble fraction, control cells; lane 2: insoluble fraction, control cells; lane 3: 
soluble fraction, cells expressing HBcAg-JUN; lane 4: insoluble fraction, cells 
expressing HbcAg-JUN. 



-104- 



[0342] The cleared cell lysate was used for step-gradient centrifugation using a 

sucrose step gradient consisting of a 4 ml 65% sucrose solution overlaid with 3 
ml 15% sucrose solution followed by 4 ml of bacterial lysate. The sample was 
centrifuged for 3 hrswith 100,000 x gat 4° C After centrifugation, 1 ml fractions 
from the top of the gradient were collected and analyzed by SDS-PAGE followed 
by Coomassie staining. (Fig. 6). Lane 1: total E. coli lysate prior to 
centrifugation. Lane 1 and 2: fractions 1 and 2 from the top of the gradient. 
Lane 4 to 7: fractions 5 to 8 (15% sucrose). The HBcAg-JUN protein was 
detected by Coomassie staining. 

[0343] The HBcAg-JUN protein was enriched at the interface between 1 5 and 

65% sucrose indicating that it had formed a capsid particle. Most of the bacterial 
proteins remained in the sucrose-free upper layer of the gradient, therefore step- 
gradient centrifugation of the HBcAg-JUN particles led both to enrichment and 
to a partial purification of the particles. 

EXAMPLE 22 
Covalent Coupling of hGH-FOS to HBcAg-JUN 
[0344] In order to demonstrate binding of a protein to HBcAg-JUN particles, we 

chose human growth hormone (hGH) fused with its carboxy terminus to the FOS 
helix as a model protein (hGH-FOS). HBcAg-JUN particles were mixed with 
partially purified hGH-FOS and incubated for 4 hours at 4°C to allow binding of 
the proteins. The mixture was then dialyzed overnight against a 3000-fold volume 
of dialysis buffer (150 raM NaCl, 10 mM Tris-HCl solution, pH 8.0) in order to 
remove DTT present in both the HBcAg-JUN solution and the hGH-FOS solution 
and thereby allow covalent coupling of the proteins through the establishment of 
disulfide bonds. As controls, the HBcAg-JUN and the hGH-FOS solutions were 
also dialyzed against dialysis buffer. Samples from all three dialyzed protein 
solutions were analyzed by SDS-PAGE under non-reducing conditions. Coupling 
of hGH-FOS to HBcAg-JUN was detected in an anti-hGH immunoblot (Fig. 7). 
hGH-FOS bound to HBcAg-JUN should migrate with an apparent molecular mass 



-105- 



of approximately 53 kDa, while unbound hGH-FOS migrates with an apparent 
molecular mass of 31 kDa. The dialysate was analyzed by SDS-PAGE in the 
absence of reducing agent (lane 3) and in the presence of reducing agent (lane 2) 
and detected by Coomassie staining As a control, hGH-FOS that had not been 
mixed with capsid particles was also loaded on the gel in the presence of reducing 
agent (lane 1). 

[0345] A shift of hGH-FOS to a molecular mass of approximately 53 kDa was 

observed in the presence of HBcAg-JUN capsid protein, suggesting that efficient 
binding of hGH-FOS to HBcAg-JUN had taken place. 

EXAMPLE 23 
Insertion of a peptide containing a Lysine residue into the 
c/el epitope of HBcAg(l-149) 

[0346] The c/el epitope (residues 72 to 88) of HBcAg is located in the tip region 

on the surface of the Hepatitis B virus capsid (HBcAg). A part of this region 
(Proline 79 and Alanine 80) was genetically replaced by the peptide Gly-Gly-Lys- 
Gly-Gly (HBcAg-Lys construct). The introduced Lysine residue contains a 
reactive amino group in its side chain that can be used for intermolecular chemical 
crosslinking of HBcAg particles with any antigen containing a free cysteine group. 

[0347] HBcAg-Lys DNA, having the amino acid sequence shown in SEQ ID 

NO : 1 5 8, was generated by PCRs: The two fragments encoding HBcAg fragments 
(amino acid residues 1 to 78 and 81 to 149) were amplified separately by PCR. 
The primers used for these PCRs also introduced a DNA sequence encoding the 
Gly-Gly-Lys-Gly-Gly peptide. The HBcAg (1 to 78) fragment was amplified from 
pEco63 using primers EcoRIHBcAg(s) and Lys-HBcAg(as). The HBcAg (81 to 
149) fragment was amplified from pEco63 using primers Lys-HBcAg(s) and 
HBcAg(l - 149)Hind(as). Primers Lys-HBcAg(as) and Lys-HBcAg(s) introduced 
complementary DNA sequences at the ends of the two PCR products allowing 
fusion of the two PCR products in a subsequent assembly PCR The assembled 



-106- 



fragments were amplified by PCR using primers EcoRIHBcAg(s) and HbcAg(l- 
149)Hind(as) 

[0348] For the PCRs, 100 pmol of each oligo and 50 ng of the template DNAs 

were used in the 50 ul reaction mixtures with 2 units of Pwo polymerase, 0. 1 mM 
dNTPs and 2 mM MgS04. For both reactions , temperature cycling was carried 
out as follows: 94°C for 2 minutes; 30 cycles of 94°C (1 minute), 50°C (1 
minute), 72 °C (2 minutes). 

[0349] Primer sequences' 

EcoRIHBcAg(s): 

(5'-CCGGAATTCATGGACATTGACCCTTATAAAG-3') (SEQ ID NO 79); 
Lys-HBcAg(as): 

(5'-CCTAGAGCCACCTTTGCCACCATCTTCTAAATTAGTACCCACCCAG 
GTAGC-3') (SEQ ID NO: 80); 

Lys-HBcAg(s) 

(5 ' -GAAGATGGTGGC AAAGGTGGCTCT AGGGACCT AGT AGTC AGTT AT 
GTC-3')(SEQ ID NO:81); 

HBcAg(l-149)Hind(as): 

(5'-CGCGTCCCAAGCTTCTAAACAACAGTAGTCTCCGGAAG-3')(SEQID 
NO: 82). 

[0350] For fusion of the two PCR fragments by PCR 100 pmol of primers 

EcoRIHBcAg(s) and HBcAg(l-149)Hind(as) were used with 100 ng of the two 
purified PCR fragments in a 50 ul reaction mixture containing 2 units of Pwo 
polymerase, 0. 1 mM dNTPs and 2 mM MgS0 4 . PCR cycling conditions were: 
94°C for 2 minutes; 30 cycles of 94°C (1 minute), 50°C (1 minute), 72°C (2 



-107- 



minutes). The assembled PCR product was analyzed by agarose gel 
electrophoresis, purified and digested for 19 hours in an appropriate buffer with 
EcoRI and Hindlll restriction enzymes. The digested DNA fragment was ligated 
into EcoRI/Hindlll-digested pKK vector to generate pKK-HBc Ag-Lys expression 
vector. Insertion of the PCR product into the vector was analyzed by 
EcoRI/Hindlll restriction analysis and DNA sequencing of the insert. 
[0351] The amino acid sequence of the HBcAg-Lys polypeptide is 

MDIDPYKEFGATVELLSFLPSDFFPSVRDLLDTASALYREAIESPEHCSP 
HHTALRQAILCWGELMTLATWVGTNLEDGGKGGSRDLVVSYVNTNM 
GLKIRQLLWFHI S CLTFGRET VLE YL VSFGVWIRTPP AYRPPN APIL S TL 
PETTVV (SEQ ID NO: 185). This sequence differs from SEQ ID NO. 134 at 
amino acid 74 (N in SEQ ID NO:13 14, T in SEQ ID NO:185) and at amino acid 
87(Nin SEQ ID NO: 134, Sin SEQ ID NO: 185). 

EXAMPLE 24 
Expression and partial purification of HBcAg-Lys 
[0352] E. coli strain XL-1 blue was transformed with pKK-HBcAg-Lys 1 ml of 

an overnight culture of bacteria was used to innoculate 100 ml of LB medium 
containing 100 ug/mlampicillin. This culture was grown for 4 hours at 37°C until 
an OD at 600 nm of approximately 0.8 was reached. Induction of the synthesis 
of HBcAg-Lys was performed by addition of IPTG to a final concentration of 1 
mM. After induction, bacteria were further shaken at 37 °C for 16 hours 
Bacteria were harvested by centrifugation at 5000 x g for 15 minutes. The pellet 
was frozen at -20 °C. The pellet was thawed and resuspended in bacteria lysis 
buffer (10 mM Na 2 HP0 4 , pH 7.0, 30 mM NaCl, 0.25% Tween-20, 10 mM 
EDTA, 10 mM DTT) supplemented with 200 ug/ml lysozyme and 10 ul of 
Benzonase (Merck). Cells were incubated for 30 minutes at room temperature 
and disrupted using a French pressure cell. Triton X-l 00 was added to the lysate 
to a final concentration of 0.2%, and the lysate was incubated for 30 minutes on 
ice and shaken occasionally. E. coli cells harboring pKK-HBcAg-Lys expression 



-108- 



plasmid or a control plasmid were used for induction of HBcAg-Lys expression 
with IPTG. Prior to the addition of IPTG, a sample was removed from the 
bacteria culture carrying the pKK-HBcAg-Lys plasmid and from a culture carrying 
the control plasmid. Sixteen hours after addition of IPTG, samples were again 
removed from the culture containing pKK-HBcAg-Lys and from the control 
culture. Protein expression was monitored by SDS-PAGE followed by Coomassie 
staining. 

[0353] The lysate was then centrifuged for 30 minutes at 12,000 x g in order to 

remove insoluble cell debris. The supernatant and the pellet were analyzed by 
Western blotting using a monoclonal antibody against HBcAg (YVS1841, 
purchased from Accurate Chemical and Scientific Corp., Westbury, NY, USA), 
indicating that a significant amount of HBcAg-Lys protein was soluble. Briefly, 
lysates from E. coli cells expressing HBcAg-Lys and from control cells were 
centrifuged at 14,000 x g for 30 minutes. Supernatant (= soluble fraction) and 
pellet (= insoluble fraction) were separated and diluted with SDS sample buffer 
to equal volumes. Samples were analyzed by SDS-PAGE followed by Western 
blotting with anti-HBcAg monoclonal antibody YVS 1841. 

[0354] The cleared cell lysate was used for step-gradient centrifugation using a 

sucrose step gradient consisting of a 4 ml 65% sucrose solution overlaid with 3 
ml 15% sucrose solution followed by 4 ml of bacterial lysate. The sample was 
centrifuged for 3 hrs with 100,000 x g at 4°C After centrifugation, 1 ml fractions 
from the top of the gradient were collected and analyzed by SDS-PAGE followed 
by Coomassie staining. The HBcAg-Lys protein was detected by Coomassie 
staining. 

[0355] The HBcAg-Lys protein was enriched at the interface between 15 and 

65% sucrose indicating that it had formed a capsid particle Most of the bacterial 
proteins remained in the sucrose-free upper layer of the gradient, therefore step- 
gradient centrifugation of the HBcAg-Lys particles led both to enrichment and to 
a partial purification of the particles. 



-109- 



EXAMPLE 25 
Chemical coupling of FLAG peptide to HBcAg-Lys 
using the heterobifiinctional cross-linker SPDP 

[0356] Synthetic FLAG peptide with a Cysteine residue at its amino terminus 

(amino acid sequence CGGDYKDDDDK (SEQ ID NO: 147)) was coupled 
chemically to purified HBcAg-Lys particles in order to elicit an immune response 
against the FLAG peptide. 600 u.1 of a 95% pure solution of HBcAg-Lys particles 
(2 mg/ml) were incubated for 30 minutes at room temperature with the 
heterobifiinctional cross-linker N-Succinimidyl 3-(2-pyridyldithio) propionate 
(SPDP) (0.5 mM). After completion of the reaction, the mixture was dialyzed 
overnight against 1 liter of 50 mM Phosphate buffer (pH 7.2) with 150 mM NaCl 
to remove free SPDP. Then 500 ul of derivatized HBcAg-Lys capsid (2 mg/ml) 
were mixed with 0 1 mM FLAG peptide (containing an amino -terminal cysteine) 
in the presence of 1 0 mM EDTA to prevent metal-catalyzed sulfhydryl oxidation. 
The reaction was monitored through the increase of the optical density of the 
solution at 343 nm due to the release of pyridine-2-thione from SPDP upon 
reaction with the free cysteine of the peptide. The reaction of derivatized Lys 
residues with the peptide was complete after approximately 30 minutes. 

[0357] The FLAG decorated particles were injected into mice. 



EXAMPLE 26 
Construction of pMPSV-gpl40cys 

[0358] The gpl40 gene was amplified by PCR from pCytTSgpl40FOS using 

oligos gpl40CysEcoRI and Sallgp 140. For the PCRs, 100 pmol of each oligo and 
50 ng of the template DNAs were used in the 50 jul reaction mixtures with 2 units 
of Pwo polymerase, 0.1 mM dNTPs and 2 mM MgS04. For both reactions , 
temperature cycling was carried out as follows: 94°C for 2 minutes; 30 cycles of 
94°C (0.5 minutes), 55°C (0.5 minutes), 72°C (2 minutes). 

[0359] The PCR product was purified using QiaEXII kit, digested with 

Sall/EcoRI and ligated into vector pMPSVHE cleaved with the same enzymes 



-110- 



[0360] Oligo sequences: 

Gpl40CysEcoRI: 

5 '-GCCGAATTCCTAGCAGCTAGCACCGAATTTATCTAA-3 ' (SEQ ID 
NO:83); 

Sallgpl40. 

5 '- GGTTAAGTCGAC ATGAGAGTGAAGGAGAAATAT-3 ' (SEQ ID NO: 84). 

EXAMPLE 27 
Expression of pMPSVgpl40Cys 
[0361] pMPSVgpl40Cys (20 ug) was linearized by restriction digestion. The 

reaction was stopped by phenol/chloroform extraction, followed by an isopropanol 
precipitation of the linearized DNA. The restriction digestion was evaluated by 
agarose gel eletrophoresis For the transfection, 5.4 fj.g of linearized 
pMPSVgpl40-Cys was mixed with 0.6 ug of linearized pSV2Neo in 30 H 2 0 
and 30 ul of 1 M CaCl 2 solution was added. After addition of 60 ul phosphate 
buffer (50 mM HEPES, 280 mM NaCl, 1 . 5 mM Na 2 HP0 4 , pH 7. 05), the solution 
was vortexed for 5 seconds, followed by an incubation at room temperature for 
25 seconds. The solution was immediately added to 2 ml HP-1 medium 
containing 2% FCS (2% FCS medium). The medium of an 80% confluent 
BHK21 cell culture (6-well plate) was then replaced by the DNA containing 
medium. After an incubation for 5 hours at 37°C in a C0 2 incubator, the DNA 
containing medium was removed and replaced by 2 ml of 1 5% glycerol in 2% FCS 
medium. The glycerol containing medium was removed after a 30 second 
incubation phase, and the cells were washed by rinsing with 5 ml of HP- 1 medium 
containing 10% FCS. Finally 2 ml of fresh HP-1 medium containing 10% FCS 
was added. 



-111- 



[0362] Stably transfected cells were selected and grown in selection medium 

(HP- 1 medium supplemented with G4 1 8) at 3 7 ° C in a C0 2 incubator. When the 
mixed population was grown to confluency, the culture was split to two dishes, 
followed by a 12 h growth period at 37°C. One dish of the cells was shifted to 
30 °C to induce the expression of soluble GP140-FOS. The other dish was kept 
at37°C 

[0363] The expression of soluble GP140-Cys was determined by Western blot 

analysis. Culture media (0.5 ml) was methanol/chloroform precipitated, and the 
pellet was resuspended in SDS-PAGE sample buffer Samples were heated for 5 
minutes at 95 0 C before being applied to a 1 5% acrylamide gel. After SDS-PAGE, 
proteins were transferred to Protan nitrocellulose membranes (Schleicher & 
Schuell, Germany) as described by Bass and Yang, in Creighton, T.E., ed., 
Protein Function: A Practical Approach, 2nd Edn., IRL Press, Oxford (1997), 
pp. 29-55. The membrane was blocked with 1 % bovine albumin (Sigma) in TBS 
(lOxTBS per liter: 87 7 g NaCl, 66. lg Trizma hydrochloride (Sigma) and 9 7 g 
Trizma base (Sigma), pH 7.4) for 1 hour at room temperature, followed by an 
incubation with an anti-GP140 or GP-160 antibody for 1 hour. The blot was 
washed 3 times for 10 minutes with TBS-T (TBS with 0.05% Tween20), and 
incubated for 1 hour with an alkaline-phosphatase-anti- 
mouse/rabbit/monkey/human IgG conjugate. After washing 2 times for 10 
minutes with TBS-T and 2 times for 10 minutes with TBS, the development 
reaction was carried out using alkaline phosphatase detection reagents (10 ml AP 
buffer (1 00 mM Tris/HCl, 1 00 mM NaCl, pH 9. 5) with 50 |xl NBT solution (7.7% 
Nitro Blue Tetrazolium (Sigma) in 70% dimethylformamide) and 37 ul of 
X-Phosphate solution (5% of 5-bromo-4-chloro-3-indolyl phosphate in 
dimethylformamide) . 



-112- 



EXAMPLE 28 
Purification of gpl40Cys 

[0364] An anti-gp 1 20 antibody was covalently coupled to a NHS/EDC activated 

dextran and packed into a chromatography column. The supernatant, containing 
GP \40Cys is loaded onto the column and after sufficient washing, GP 1 AOCys was 
eluted using 0. 1 M HC1. The eluate was directly neutralized during collection 
using 1 M Tris pH 7.2 in the collection tubes. 

[0365] Disulfide bond formation might occur during purification, therefore the 

collected sample is treated with 10 mM DTT in 10 mM Tris pH 7.5 for 2 hours 
at25°C 

[0366] DTT is remove by subsequent dialysis against 10 mM Mes; 80 mM NaCl 

pH 6.0. Finally GP140Cys is mixed with alphavirus particles containing the JUN 
residue in E2 as described in Example 16. 

EXAMPLE 29 
Construction of PLA2-Cys 

[0367] The PLA2 gene was amplified by PCR from pAV3PLAfos using oligos 

EcoRTPLA and PLA-Cys-hind. For the PCRs, 100 pmol of each oligo and 50 ng 
of the template DNAs were used in the 50 ul reaction mixtures with 2 units of 
Pwo polymerase, 0.1 mM dNTPs and 2 mM MgS04. For both reactions , 
temperature cycling was carried out as follows: 94 °C for 2 minutes; 30 cycles of 
94°C (0.5 minutes), 55 °C (0.5 minutes), 72°C (2 minutes) 

[0368] The PCR product was purified using QiaEXII kit, digested with 

EcoRI/HinDIII and ligated into vector pAV3 cleaved with the same enzymes. 

[0369] Oligos 
EcoRIPLA: 

5 '-TAACCGAATTC AGGAGGTAAAAAGATATGG-3 ' (SEQ ID NO: 85) 



-113- 



PLACys-hind: 

5 '-GAAGTAAAGCTTTTAACCACCGCAACCACCAGAAG-3 ' (SEQ ID 
NO 86). 

EXAMPLE 30 
Expression and Purification of PLA-Cys 
For cytoplasmic production of Cys tagged proteins, E. coli XL-l-Blue 
strain was transformed with the vectors pAV3 : :PLA and pPLA-Cys. The culture 
was incubated in rich medium in the presence of ampicillin at 37 °C with shaking. 
At an optical density (550nm) of, 1 mM IPTG was added and incubation was 
continued for another 5 hours The cells were harvested by centrifugation, 
resuspended in an appropriate buffer {e.g., Tris-HCl, pH 7.2, 150 mM NaCl) 
containing DNase, RNase and lysozyme, and disrupted by passage through a 
french pressure cell. After centrifugation (Sorvall RC-5C, SS34 rotor, 
15000 rpm, 10 min, 4°C), the pellet was resuspended in 25 ml inclusion body 
wash buffer (20 mM tris-HCl, 23% sucrose, 0.5% Triton X-100, 1 mM EDTA, 
pH8) at 4°C and recentrifuged as described above. This procedure was repeated 
until the supernatant after centrifugation was essentially clear. Inclusion bodies 
were resuspended in 20 ml solubilization buffer (5.5 M guanidinium 
hydrochloride, 25 mM tris-HCl, pH 7.5) at room temperature and insoluble 
material was removed by centrifugation and subsequent passage of the supernatant 
through a sterile filter (0 45 um). The protein solution was kept at 4°C for at 
least 10 hours in the presence of 10 mM EDTA and 100 mM DTT and then 
dialyzed three times against 10 volumes of 5.5 M guanidinium hydrochloride, 25 
mM tris-HCl, 10 mM EDTA, pH 6. The solution was dialyzed twice against 5 1 
2 M urea, 4 mM EDTA, 0.1 M NH 4 C1, 20 mM sodium borate (pH 8.3) in the 
presence of an appropriate redox shuffle (oxidized glutathione/reduced 
glutathione; cystine/cysteine). The refolded protein was then applied to an ion 
exchange chromatography. The protein was stored in an appropriate buffer with 
a pH above 7 in the presence of 2-10 mM DTT to keep the cysteine residues in 



-114- 



a reduced form. Prior to coupling of the protein with the alphavirus particles, 
DTT was removed by passage of the protein solution through a Sephadex G-25 
gel filtration column. 

EXAMPLE 31 

Construction of a HBcAg devoid of free cysteine residues and containing 
an inserted lysine residue 
[0371] A Hepatitis core Antigen (HBcAg), referred to herein as HBcAg-lys-2cys- 

Mut, devoid of cysteine residues at positions corresponding to 48 and 107 in SEQ 
ID NO: 134 and containing an inserted lysine residue was constructed using the 
following methods. 

[0372] The two mutations were introduced by first separately amplifying three 

fragments of the HBcAg-Lys gene prepared as described above in Example 23 
with the following PCR primer combinations. PCR methods essentially as 
described in Example 1 and conventional cloning techniques were used to prepare 
the HBcAg-lys-2cys-Mut gene. 

[0373] In brief, the following primers were used to prepare fragment 1 ' 

Primer 1 : EcoRIHBcAg(s) 

CCGGAATTCATGGACATTGACCCTTATAAAG (SEQ ID NO: 148) 
Primer 2: 48as 

GTGCAGTATGGTGAGGTGAGGAATGCTCAGGAGACTC (SEQ ID 
NO: 149) 



[0374] 



The following primers were used to prepare fragment 2: 
Primer 3 : 48s 

GSGTCTCCTGAGCATTCCTCACCTCACCATACTGCAC (SEQ ID NO: 1 50) 



-115- 



Primer 4: 107as 

CTTCCAAAAGTGAGGGAAGAAATGTGAAACCAC (SEQ ID NO: 151) 

[0375] The following primers were used to prepare fragment 3 . 

Primer 5: HBcAgl49hind-as 

CGCGTCCCAAGCTTCTAAACAACAGTAGTCTCCGGAAGCGTTGATAG 
(SEQ ID NO: 152) 

Primer 6: 107s 

GTGGTTTCACATTTCTTCCCTCACTTTTGGAAG (SEQ ID NO: 153) 

[0376] Fragments 1 and 2 were then combined with PCR primers 

EcoRIHBcAg(s) and 107as to give fragment 4. Fragment 4 and fragment 3 were 
then combined with primers EcoRJHBcAg(s) and HBcAgl49hind-as to produce 
the full length gene. The full length gene was then digested with the EcoRI 
(GAATTC) and Hindlll (AAGCTT) enzymes and cloned into the pKK vector 
(Pharmacia) cut at the same restriction sites. The amino acid sequence of the 
HBcAg-Lys-2cys-Mut polypeptide is MDIDPYKEFGATVELLSFL 
PSDFFPSVRDLLDTASALYREALESPEHSSPHHTALRQAILCWGELMTL 
ATWVGTNLEDGGKGGSRDLVVSYVNTNMGLKIRQLLWFHISSLTFGR 
ETVLEYLVSFGVWIRTPPAYRPPNAPILSTLPETTVV (SEQ ID NO: 186). 

EXAMPLE 32 

Blockage of free cysteine residues of a HBcAg followed by cross-linking 
[0377] The free cysteine residues of the HBcAg-Lys prepared as described above 

in Example 23 were blocked using lodacetamide. The blocked HBcAg-Lys was 
then cross-linked to the FLAG peptide with the hetero-bifunctional cross-linker 
m-maleimidonbenzoyl-N-hydroxysuccinimide ester (Sulfo-MBS). 
[0378] The methods used to block the free cysteine residues and cross-link the 

HBcAg-Lys are as follows. HBcAg-Lys (550 ug/ml) was reacted for 1 5 minutes 



-116- 



at room temperature with Iodacetamide (Fluka Chemie, Brugg, Switzerland) at 
a concentration of 50 mM in phosphate buffered saline (PBS) (50 mM sodium 
phosphate, 150 mM sodium chloride), pH 7.2, in a total volume of 1 ml. The so 
modified HBcAg-Lys was then reacted immediately with Sulfo-MBS (Pierce) at 
a concentration of 530 uM directly in the reaction mixture of step 1 for 1 hour at 
room temperature. The reaction mixture was then cooled on ice, and dialyzed 
against 1000 volumes of PBS pH 7.2. The dialyzed reaction mixture was finally 
reacted with 300 uM of the FLAG peptide (CGGDYKDDDDK (SEQ ID 
NO: 147)) containing an N-terminal free cysteine for coupling to the activated 
HBcAg-Lys, and loaded on SDS-PAGE for analysis. 
[0379] As shown in Figure 8, the resulting patterns of bands on the SDS-PAGE 

gel showed a clear additional band migrating slower than the control HBcAg-Lys 
derivatized with the cross-linker, but not reacted with the FLAG peptide. 
Reactions done under the same conditions without prior derivatization of the 
cysteines with Iodacetamide led to complete cross-linking of monomers of the 
HBcAg-Lys to higher molecular weight species. 

EXAMPLE 33 

Isolation of Type- 1 pili and chemical coupling of FLAG peptide to Type-1 pili 

of Escherichia coli using a heterobifunctional cross-linker 
A. Introduction 

[0380] Bacterial pili or fimbriae are filamentous surface organelles produced by 

a wide range of bacteria. These organelles mediate the attachment of bacteria to 
surface receptors of host cells and are required for the establishment of many 
bacterial infections like cystitis, pyelonephritis, new born meningitis and diarrhea. 

[0381] Pili can be divided in different classes with respect to their receptor 

specificity (agglutination of blood cells from different species), their assembly 
pathway (extracellular nucleation, general secretion, chaperone/usher, alternate 
chaperone) and their morphological properties (thick, rigid pili; thin, flexible pili; 
atypical structures including capsule; curli; etc). Examples of thick, rigid pili 



-in- 



forming a right handed helix that are assembled via the so called chaperone/usher 
pathway and mediate adhesion to host glycoproteins include Type-1 pili, P-pili, 
S-pili, FIC-pili, and 987P-pili). The most prominent and best characterized 
members of this class of pili are P-pili and Type-1 pili (for reviews on adhesive 
structures, their assembly and the associated diseases see Soto, G. E. & Hultgren, 
S. J., J. Bacteriol. 757:1059-1071 (1999); Bullitt & Makowski, Biophys. J. 
74:623-632 (1998); Hung, D. L. & Hultgren, S. I, J. Struct, Biol. 724:201-220 
(1998)). 

[0382] Type-1 pili are long, filamentous polymeric protein structures on the 

surface of E. coli They possess adhesive properties that allow for binding to 
mannose-containing receptors present on the surface of certain host tissues 
Type-1 pili can be expressed by 70-80% of all E. coli isolates and a single E. coli 
cell can bear up to 500 pili. Type- pili reach a length of typically 0.2 to 2 [iM with 
an average number of 1000 protein subunits that associate to a right-handed helix 
with 3.125 subunits per turn with a diameter of 6 to 7 nm and a central hole of 2.0 
to 2.5 nm 

[0383] The main Type-1 pilus component, FimA, which represents 98% of the 

total pilus protein, is a 15.8 kDa protein. The minor pilus components FimF, 
FimG and FimH are incorporated at the tip and in regular distances along the pilus 
shaft (Klemm, P & Krogfelt, K. A., "Type I fimbriae of Escherichia coli," in: 
Fimbriae. Klemm, P. (ed.), CRC Press Inc., (1994) pp. 9-26). FimH, a 29.1 kDa 
protein, was shown to be the mannose-binding adhesin of Type-1 pili (Krogfelt, 
K. A., et al, Infect. Immun. 55:1995-1998 (1990); Klemm, P , et al, Mol. 
Microbiol. 4:553-560 (1990); Hanson, M. S. & Brinton, C. C. J., Nature 17.265- 
268 (1988)), and its incorporation is probably facilitated by FimG and FimF 
(Klemm, P. & Christiansen, G, Mol. Gen. Genetics 208 439-445 (1987), Russell, 
P W. & Orndorff, P. E., J. Bacteriol. 774:5923-5935 (1992)). Recently, it was 
shown that FimH might also form a thin tip-fibrillum at the end of the pili (Jones, 
C. H., et al, Proc. Nat. Acad. Sci. USA 92:2081-2085 (1995)). The order of 
major and minor components in the individual mature pili is very similar, indicating 



-118- 



a highly ordered assembly process (Soto, G. E. & Hultgren, S. J., J. Bacteriol. 
757:1059-1071 (1999)). 
[0384] P-pili of E. coli are of very similar architecture, have a diameter of 6. 8 nm, 

an axial hole of 1 .5 nm and 3.28 subunits per turn (Bullitt & Makowski, Biophys. 
J. 74:623-632 (1998)). The 16.6 kDa PapA is the main component of this pilus 
type and shows 36% sequence identity and 59% similarity to FimA (see Table 1). 
As in Type-1 pili the 36.0 kDa P-pilus adhesin PapG and specialized adapter 
proteins make up only a tiny fraction of total pilus protein. The most obvious 
difference to Type-1 pili is the absence of the adhesin as an integral part of the 
pilus rod, and its exclusive localization in the tip fibrillium that is connected to the 
pilus rod via specialized adapter proteins that Type-1 pili lack (Hultgren, S. J , et 
al, Cell 73:887-901 (1993)). 

[0385] Table 1 . Similarity and identity between several structural pilus 

proteins of Type-1 and P-pili (in percent). The adhesins 
were omitted. 



Similarity 





FimA 


PapA 


FimI 


FimF 


FimG 


PapE 


PapK 


PapH 


Papl 


FimA 




59 


57 


56 


44 


50 


44 


46 


46 


PapA 


36 




49 


48 


41 


45 


49 


49 


47 


FimI 


35 


31 




56 


46 


40 


47 


48 


48 


FimF 


34 


26 


30 




40 


47 


43 


49 


48 


FimG 


28 


28 


28 


26 




39 


39 


41 


45 


PapE 


25 


23 


18 


28 


22 




43 


47 


54 


PapK 


24 


29 


25 


28 


22 


18 




49 


53 


PapH 


22 


26 


22 


22 


23 


24 


23 




41 


PapF 


18 


22 


22 


24 


28 


27 


26 


21 





[0386] Type- 1 pili are extraordinary stable hetero-oligomeric complexes. Neither 

SDS-treatment nor protease digestions, boiling or addition of denaturing agents 
can dissociate Type-1 pili into their individual protein components. The 
combination of different methods like incubation at 100°C at pH 1 .8 was initially 
found to allow for the depolymerization and separation of the components 



-119- 



(Eshdat, Y., etaL, J. Bacteriol. 745:308-314 (1981); Brinton, C.C. J., Trans, N. 
Y.Acad. Sci. 27:1003-1054 (1965); Hanson, A. S., etaL, J. Bacteriol. , 770:3350- 
3358 (1988); Klemm, P. & Krogfelt, K. A., "Type I fimbriae of Escherichia coli," 
in Fimbriae. Klemm, P. (ed.), CRC Press Inc., (1994) pp. 9-26). Interestingly, 
Type-1 pili show a tendency to break at positions where FimH is incorporated 
upon mechanical agitation, resulting in fragments that present a FimH adhesin at 
their tips This was interpreted as a mechanism of the bacterium to shorten pili to 
an effective length under mechanical stress (Klemm, P. & Krogfelt, K. A., "Type 
I fimbriae of Escherichia coli," in: Fimbriae. Klemm, P (ed.), CRC Press Inc., 
(1994) pp. 9-26). Despite their extraordinary stability, Type-1 pili have been 
shown to unravel partially in the presence of 50% glycerol; they lose their helical 
structure and form an extended and flexible, 2 nm wide protein chain (Abraham, 
S. N., etaL, J. Bacteriol. 774:5145-5148 (1992)). 

[0387] P-pili and Type-1 pili are encoded by single gene clusters on the E. coli 

chromosome of approximately 10 kb (Klemm, P. & Krogfelt, K. A., "Type I 
fimbriae of Escherichia coli," in: Fimbriae. Klemm, P. (ed.), CRC Press Inc., 
(1994) pp. 9-26; Orndorff, P. E. & Falkow, S., J. Bacteriol. 760:61-66 (1984)). 
A total of nine genes are found in the Type- 1 pilus gene cluster, and 1 1 genes in 
the P-pilus cluster (Hultgren, S. J., etaL, Adv. Prot. Chem. 44:99-123 (1993)). 
Both clusters are organized quite similarly. 

[0388] The first two fim-genes>,fimB and fimE, code for recombinases involved 

in the regulation of pilus expression (McClain, M. S., et al, J. Bacteriol. 
773:5308-53 14 (1991)). The main structural pilus protein is encoded by the next 
gene of the cluster, fimA (Klemm, P , Euro. J. Bwchem. 743:395-400 (1984); 
Orndorff, P. E. & Falkow, S., J. Bacteriol. 160:61-66 (1984); Orndorff, P. E. & 
Falkow, S., J. Bacteriol. 1 62A54-AS1 (1985)). The exact role of fiml is unclear 
It has been reported to be incorporated in the pilus as well (Klemm, P. & Krogfelt, 
K. A., "Type I fimbriae of Escherichia coli" in: Fimbriae. Klemm, P. (ed.), CRC 
Press Inc., (1994) pp. 9-26). The adjacent fimC codes not for a structural 
component of the mature pilus, but for a so-called pilus chaperone that is essential 



-120- 



for the pilus assembly (Klemm, P., Res. Microbiol. 743:831-838 (1992), Jones, 
C. H., etai, Proc. Nat. Acad Set USA 90:8397-8401 (1993)). 

[0389] The assembly platform in the outer bacterial membrane to which the 

mature pilus is anchored is encoded by fimD (Klemm, P. & Christiansen, G., Mol. 
Gen, Genetics 220.334-338 (1990)). The three minor components of the Type-1 
pili, FimF, FimG and FimH are encoded by the last three genes of the cluster 
(Klemm, P. & Christiansen, G.,Mo/. Gen. Genetics 208:439-445 (1987)). Apart 
from fimB and fimE, all genes encode precursor proteins for secretion into the 
periplasm via the sec-pathway. 

[0390] The similarities between different pili following the chaperone/usher 

pathway are not restricted to their morphological properties. Their genes are also 
arranged in a very similar manner Generally the gene for the main structural 
subunit is found directly downstream of the regulatory elements at the beginning 
of the gene cluster, followed by a gene for an additional structural subunit (fiml 
in the case of Type-1 pili and papH in the case of P-pili). PapH was shown and 
Fiml is supposed to terminate pilus assembly (Hultgren, S. J., etal, Cell 73 887- 
901 (1993)). The two proteins that guide the process of pilus formation, namely 
the specialized pilus chaperone and the outer membrane assembly platform, are 
located adjacently downstream. At the end of the clusters a variable number of 
minor pilus components including the adhesins are encoded. The similarities in 
morphological structure, sequence (see Table 1), genetic organization and 
regulation indicate a close evolutionary relationship and a similar assembly process 
for these cell organelles. 

[0391] Bacteria producing Type-1 pili show a so-called phase-variation. Either 

the bacteria are fully piliated or bald. This is achieved by an inversion of a 3 14 bp 
genomic DNA fragment containing the fimA promoter, thereby inducing an "all 
on" or "all off' expression of the pilus genes (McClain, M. S.,et al, J. Bacteriol. 
773:5308-53 14 (1991)). The coupling of the expression of the other structural 
pilus genes to fimA expression is achieved by a still unknown mechanism. 



-121- 



However, a wide range of studies elucidated the mechanism that influences the 
switching between the two phenotypes. 

[0392] The first two genes of the Type-1 pilus cluster, fimB and fimE encode 

recombinases that recognize 9 bp DNA segments of dyad symmetry that flank the 
invertable fimA promoter. Whereas FimB switches pilation "on", FimE turns the 
promoter in the "off' orientation. The up- or down-regulation of either fimB or 
fimE expression therefore controls the position of the so-called "//'m- switch" 
(McClain, M. S , etal, J. Bacteriol. 773:5308-5314 (1991); Blomfield, I. C, et 
al, J. Bacteriol. 773:5298-5307 (1991)). 

[0393] The two regulatory proteins fimB and fimE are transcribed from distinct 

promoters and their transcription was shown to be influenced by a wide range of 
different factors including the integration host factor (IHF) (Blomfield, I. C, et 
al, Mol. Microbiol. 23:105-1X1 (1997)) and the leucine-responsive regulatory 
protein (LRP) (Blomfield, I. C, etal, J. Bacteriol. 775:27-36 (1993); Gaily, D. 
L., etal, J. Bacteriol. 775:6186-6193 (1993); Gaily, D. L., etal, Microbiol. 
27:725-738 (1996); Roesch, R. L. & Blomfield, I. C, Mol. Microbiol, 27:751-761 
(1998)) Mutations in the former lock the bacteria either in "on" or "off" phase, 
whereas LRP mutants switch with a reduced frequency. In addition, an effect of 
leuXorv pilus biogenesis has been shown. This gene is located in the vicinity of 
the /wrc-genes on the chromosome and codes for the minor leucine tRNA species 
for the UUG codon. Whereas fimB contains five UUG codons, fimE contains 
only two, and enhanced leuX transcription might favor FimB over FimE 
expression (Burghoff, R. L., et al, Infect. Immun. 67:1293-1300 (1993); 
Newman, J. V , etal, FEMS Microbiol. Lett. 722:281-287 (1994); Ritter, A., et 
al, Mol. Microbial, 25:871-882 (1997)). 

[0394] Furthermore, temperature, medium composition and other environmental 

factors were shown to influence the activity of FimB and FimE. Finally, a 
spontaneous, statistical switching of the fimA promoter has been reported. The 
frequency of this spontaneous switching is approximately 10" 3 per generation 
(Eisenstein, B. I., Science 274:337-339 (1981); Abraham, S. M., etal, Proc. Nat. 



-122- 



Acad. Sci, USA 52:5724-5727 (1985)), but is strongly influenced by the above 
mentioned factors. 

[0395] The genes fiml mdfimC are also transcribed from the fimA promoter, but 

directly downstream of fimA a DNA segment with a strong tendency to form 
secondary structure was identified which probably represents a partial 
transcription terminator (Klemm, P , Euro. J. Biochem. 143:395-400 (1984)); and 
is therefore supposed to severely reduce fiml and fimC transcription. At the 3 ' 
end offimC an additional promoter controls the fimD transcription; at the 3' end 
of fimD the last known fim promoter is located that regulates the levels of FimF, 
FimG, and FimH. Thus, all of the minor Type-1 pili proteins are transcribed as a 
single mRNA (Klemm, P. & Krogfelt, K. A., "Type I fimbriae of Escherichia 
colif in: Fimbriae Klemm, P. (ed ), CRC Press Inc., (1994) pp. 9-26). This 
ensures a 1 : 1 : 1 stochiometry on mRNA-level, which is probably maintained on the 
protein level 

[0396] In the case of P-pili additional regulatory mechanisms were found when 

the half-life of mRNA was determined for different P-pilus genes. The mRNA for 
papA was extraordinarily long-lived, whereas the mRNA for papB, a regulatory 
pilus protein, was encoded by short-lived mRN A (Naureckiene, S. & Uhlin. B. E., 
Mol. Microbiol. 27:55-68 (1996); Nilsson, P., eta!., J. Bacterial. 775:683-690 
(1996)). 

[0397] In the case of Type-1 pili, the gene for the Type-1 pilus chaperone FimC 

starts with a GTG instead of an ATG codon, leading to a reduced translation 
efficiency. Finally, analysis of the fimH gene revealed a tendency of the fimH 
mRNA to form a stem-loop, which might severely hamper translation. In 
summary, bacterial pilus biogenesis is regulated by a wide range of different 
mechanisms acting on all levels of protein biosynthesis. 

[0398] Periplasmic pilus proteins are generally synthesized as precursors, 

containing a N-terminal signal-sequence that allows translocation across the inner 
membrane via the Sec-apparatus. After translocation the precursors are normally 
cleaved by signal-peptidase I. Structural Type-1 pilus subunits normally contain 



-123- 



disulfide bonds, their formation is catalyzed by DsbA and possibly DsbC and 
DsbG gene products. 

[0399] The Type- 1 pilus chaperone FimC lacks cysteine residues. In contrast, the 

chaperone of P-pili, PapD, is the only member of the pilus chaperone family that 
contains a disulfide bond, and the dependence of P-pili on DsbA has been shown 
explicitly (Jacob-Dubuisson, F , etal, Proc. Nat. Acad. Sci. USA 91: 1 1552-1 1 556 
(1994)). PapD does not accumulate in the periplasm of a.AdsbA strain, indicating 
that the disturbance of the P-pilus assembly machinery is caused by the absence 
of the chaperone (Jacob-Dubuisson, F., et al, Proc. Nat. Acad. Sci. USA 
97:1 1552-1 1556 (1994)). This is in accordance with the finding that Type-1 pili 
are still assembled in aAdsbA strain, albeit to reduced level (Hultgren, S. J., et al., 
"Bacterial Adhesion and Their Assembly", in: Escherichia coli and Salmonella, 
Neidhardt, F. C. (ed.) ASM Press, (1996) pp. 2730-2756). 

[0400] Type- 1 pili as well as P-pili are to 98% made of a single or main structural 

subunit termed FimA and PapA, respectively. Both proteins have a size of ~1 5.5 
kDa. The additional minor components encoded in the pilus gene clusters are very 
similar (see Table 1 ). The similarities in sequence and size of the subunits with the 
exception of the adhesins suggest that all share an identical folding motif, and 
differ only with respect to their affinity towards each other. Especially the N- and 
C-terminal regions of these proteins are well conserved and supposed to play an 
important role in chaperone/subunit interactions as well as in subunit/subunit 
interactions within the pilus (Soto, G. E. & Hultgren, S. J., J. Bacteriol. 
181 : 1 059- 1 07 1 ( 1 999)) Interestingly, the conserved N-terminal segment can be 
found in the middle of the pilus adhesins, indicating a two-domain organization of 
the adhesins where the proposed C-terminal domain, starting with the conserved 
motif, corresponds to a structural pilus subunit whereas the N-terminal domain 
was shown to be responsible for recognition of host cell receptors (Hultgren, S. 
J., etal, Proc. Nat. Acad. Sci. USA £6:4357-4361 (1989), Haslam, D. B., etal, 
Mol. Microbiol. 74:399-409 (1994), Soto, G. E., etal, EMBO J. 77:6155-6167 
(1998)). The different subunits were also shown to influence the morphological 



-124- 



properties of the pili. The removal of several genes was reported to reduce the 
number of Type-1 or P-pili or to increase their length, (fimH, papG, papK, fimF, 
fimG) (Russell, P. W. & Orndorff, P. E., J. Bacteriol. 774:5923-5935 (1992); 
Jacob-Dubuisson, R„ et al, EMBO J. 72:837-847 (1993); Soto, G. E. & 
Hultgren, S. J., J. Bacteriol. 18 1 1059-1 071 (1999)); combination of the gene 
deletions amplified these effects or led to a total loss of pilation (Jacob-Dubuisson, 
R., etal, EMBO J. 72:837-847 (1993)). 

In non-fimbrial adhesive cell organelles also assembled via 
chaperones/usher systems such as Myf fimbriae and CS3 pili, the conserved C- 
terminal region is different. This indirectly proves the importance of these C- 
terminal subunit segments for quaternary interactions (Hultgren, S. J., et al., 
"Bacterial Adhesion and Their Assembly", in: Escherichia coli and Salmonella, 
Neidhardt, F. C. (ed ) ASM Press, (1996) pp. 2730-2756). 

Gene deletion studies proved that removal of the pilus chaperones leads 
to a total loss of piliation in P-pili and Type-1 pili (Lindberg, F , et al, J. 
Bacteriol. 777:6052-6058 (1989), Klemm, P., Res. Microbiol 743:831-838 
(1992); Jones, C. H., etal, Proc. Nat. Acad Sci. USA 90:8397-8401 (1993)). 
Periplasmic extracts of a AfimC strain showed the accumulation of the main 
subunit FimA, but no pili could be detected (Klemm, P., Res. Microbiol. 143:83 1- 
838 (1992)). Attempts to over-express individual P-pilus subunits failed and only 
proteolytically degraded forms could be detected in the absence of PapD; in 
addition, the P-pilus adhesin was purified with the inner membrane fraction in the 
absence of the chaperone (Lindberg, F., et al, J. Bacteriol. 777 6052-6058 
(1989)) However, co-expression of the structural pilus proteins and their 
chaperone allowed the detection of chaperone/subunit complexes from the 
periplasm in the case of the FimC/FimH complex as well as in the case of different 
Pap-proteins including the adhesin PapG and the main subunit PapA (Tewari, R., 
etal, J. Biol. Chem. 265:3009-3015 (1993); Lindberg, F., etal, J. Bacteriol. 
777:6052-6058 (1989)). The affinity of chaperone/subunit complexes towards 
their assembly platform has also been investigated in vitro and was found to differ 



-125- 



strongly (Dodsonef a/., /Voc. Natl. Acad. Sci. USA 90:3670-3674(1993)). From 
these results the following functions were suggested for the pilus chaperones. 
[0403] They are assumed to recognize unfolded pilus subunits, prevent their 

aggregation and to provide a "folding template" that guides the formation of a 
native structure. 

[0404J The folded subunits, which after folding display surfaces that allow 

subunit/subunit interactions, are then expected to be shielded from interacting with 
other subunits, and to be kept in a monomeric, assembly-competent state. 

[0405] Finally, the pilus chaperones are supposed to allow a triggered release of 

the subunits at the outer membrane assembly location, and, by doing so with 
different efficiency, influence the composition and order of the mature pili (see 
also the separate section below). 

[0406] After subunit release at the outer membrane, the chaperone is free for 

another round of substrate binding, folding assistance, subunit transport through 
the periplasm and specific delivery to the assembly site Since the periplasm lacks 
energy sources, like ATP, the whole pilus assembly process must be 
thermodynamically driven (Jacob-Dubuisson, F.,etaL, Proc. Nat. Acad. Sci. USA 
97: 1 1552-1 1556 (1994)). The wide range of different functions attributed to the 
pilus chaperones would implicate an extremely fine tuned cascade of steps. 

[0407] Several findings, however, are not readily explained with the model of 

pilus chaperone function outlined above One example is the existence of 
multimeric chaperone/subunit complexes (Striker, R T., et ai, J. Biol. Chem. 
269: 12233-12239 (1994)), where one chaperone binds subunit dimers or trimers. 
It is difficult to imagine a folding template that can be "double-booked". The 
studies on the molecular details of chaperone/subunit interaction (see below) 
partially supported the functions summarized above, but also raised new 
questions. 

[0408] All 3 1 periplasmic chaperones identified by genetic studies or sequence 

analysis so far are proteins of approximately 25 kDa with conspicuously high pi 
values around 10. Ten of these chaperones assist the assembly of rod-like pili, 



-126- 



four are involved in the formation of thin pili, ten are important for the biogenesis 
of atypically thin structures (including capsule-like structures) and two adhesive 
structures have not been determined so far (Holmgren, A., et al, EMBO J. 
77:1617-1622 (1992); Bond, A., et al, J. Mol. Evolution 44:299-309 (1997); 
Smyth, C. J., etal, FEMSImmun. Med Microbiol. 16: 127-139 (1996); Hung, D. 
L. & Hultgren, S. J., J. Struct, Biol 724:201-220 (1998)). The pairwise sequence 
identity between these chaperones and PapD ranges from 25 to 56%, indicating 
an identical overall fold (Hung, D. L , etal, EMBO J. 75:3792-3805 (1996)). 

The first studies on the mechanism of chaperone/substrate recognition was 
based on the observation that the C-termini of all known pilus chaperones are 
extremely similar. Synthetic peptides corresponding to the C-termini of the P- 
pilus proteins were shown to bind to PapD in ELISA assays (Kuehn, M. J., et al, 
Science 262: 1234-1241 (1993)). Most importantly, the X-ray structures of two 
complexes were solved in which PapD was co-crystallized with 19-residue 
peptides corresponding to the C-termini of either the adhesin PapG or the minor 
pilus component PapK (Kuehn, M. J., et al, Science 262:1234-1241 (1993); 
Soto, G. E., et al, EMBO J. 77:6155-6167 (1998)). Both peptides bound in an 
extended conformation to a p-strand in the N-terminal chaperone domain that is 
oriented towards the inter-domain cleft, thereby extending a p-sheet by an 
additional strand. The C-terminal carboxylate groups of the peptides were 
anchored via hydrogen-bonds to Arg8 and Lysll2, these two residues are 
invariant in the family of pilus chaperones. Mutagenesis studies confirmed their 
importance since their exchange against alanine resulted in accumulation of non- 
functional pilus chaperone in the periplasm (Slonim, L. N., et al, EMBO J. 
77 4747-4756 (1992)). The crystal structure of PapD indicates that neither Arg8 
nor Lysll2 is involved in stabilization of the chaperone, but completely solvent 
exposed (Holmgren, A. & Branden, C I., Nature 342 248-251 (1989)). On the 
substrate side the exchange of C-terminal PapA residues was reported to abolish 
P-pilus formation, and similar experiments on the conserved C-terminal segment 
of the P-pilus adhesin PapG prevented its incorporation into the P-pilus (Hultgren, 



-127- 



S. J., et al., "Bacterial Adhesion and Their Assembly", in: Escherichia coli and 
Salmonella, Neidhardt, F. C. (ed.) ASM Press, (1996) pp. 2730-2756). All 
evidence therefore indicated pilus subunit recognition via the C-terminal segments 
of the subunits. 

[0410] A more recent study on C-terminal amino acid exchanges of the P-pilus 

adhesin PapG gave a more detailed picture. A range of amino acid substitutions 
at the positions -2, -4, -6, and -8 relative to the C-terminus were tolerated, but 
changed pilus stability (Soto, G. K, etal, EMBO J. 77:6155-6167 (1998)). 

[0411] Still, certain problems arise when this model is examined more closely. 

Adhesive bacterial structures not assembled to rigid, rod-like pili lack the 
conserved C-terminal segments (Hultgren, S. J., et al., "Bacterial Adhesion and 
Their Assembly", in: Escherichia coli and Salmonella, Neidhardt, F. C. (ed.) 
ASM Press, (1996) pp 2730-2756), even though they are also dependent on the 
presence of related pilus chaperones. This indicates a different general role for the 
C-terminal segments of pilus subunits, namely the mediation of quaternary 
interactions in the mature pilus Moreover, the attempt to solve the structure of 
a C-terminal peptide in complex with the chaperone by NMR was severely 
hampered by the weak binding of the peptide to the chaperone (Walse, B , et al, 
FEBS Lett. 4 1 '2:115-120 (1997)); whereas an essential contribution of the C- 
terminal segments for chaperone recognition implies relatively high affinity 
interactions. 

[0412] An additional problem arises if the variability between the different 

subunits are taken into account. Even though the C-terminal segments are 
conserved, a wide range of conservative substitutions is found. For example, 15 
out of 1 9 amino acid residues differ between the two peptides co-crystallized with 
PapD (Soto, G. E., et al, EMBO J. 77:6155-6167 (1998)). This has been 
explained by the kind of interaction between chaperone and substrate, that occurs 
mainly via backbone interactions and not specifically via side-chain interactions. 
Then again, the specificity of the chaperone for certain substrates is not readily 
explained. On the contrary to the former argument, the conserved residues have 



-128- 



been taken as a proof for the specificity (Hultgren, S J., et al , "Bacterial 
Adhesion and Their Assembly", in: Escherichia coh and Salmonella, Neidhardt, 
F. C. (ed.) ASM Press, (1996) pp. 2730-2756). 

[0413] The outer membrane assembly platform, also termed "usher" in the 

literature, is formed by homo-oligomers of FimD or PapC, in the case of Type- 1 
and P-pili, respectively (Klemm, P. & Christiansen, G., Mol. Gen, Genetics 
220:334-338 (1990); Thanassi, D. G., etal, Proc. Nat. Acad Sei. USA 95:3146- 
3151 (1998)). Studies on the elongation of Type-1 fimbriae by electron 
microscopy demonstrated an elongation of the pilus from the base (Lowe, M. A., 
etal, J. Bacteriol. 169:\Sl-\63 (1987)). In contrast to the secretion of unfolded 
subunits into the periplasmic space, the fully folded proteins have to be 
translocated through the outer membrane, possibly in an oligomeric form 
(Thanassi, D. G., etal, Proc. Nat. Acad. Sei. USA 95:3 146-3 151 (1998)). This 
requires first a membrane pore wide enough to allow the passage and second a 
transport mechanism that is thermodynamically driven (Jacob-Dubuisson, F., et 
al, J. Biol. Chem. 269:12447-12455 (1994)). 

[0414] FimD expression alone was shown to have a deleterious effect on bacterial 

growth, the co-expression of pilus subunits could restore normal growth behavior 
(Klemm, P. & Christiansen, G.,M>/. Gen, Genetics 220:334-338 (1990)). Based 
on this it can be concluded that the ushers probably form pores that are completely 
filled by the pilus. Electron microscopy on membrane vesicles in which PapC had 
been incorporated confirmed a pore-forming structure with an inner diameter of 
2 nm (Thanassi, D. G , etal, Proc. Nat. Acad. Sei. USA 95:3146-3151 (1998)). 
Since the inner diameter of the pore is too small to allow the passage of a pilus 
rod, it has been suggested that the helical arrangement of the mature pilus is 
formed at the outside of the bacterial surface. The finding that glycerol leads to 
unraveling of pili which then form a protein chain of approximately 2 nm is in 
good agreement with this hypothesis, since an extended chain of subunits might 
be formed in the pore as a first step (Abraham, S. N., et al, J. Bacteriol. 
774:5145-5148 (1992); Thanassi, D. G, et al, Proc. Nat. Acad. Sei. USA 



-129- 



95:3 146-3 151 (1998)). The formation ofthe helical pilus rod at the outside of the 
bacterial membrane might then be the driving force responsible for translocation 
of the growing pilus through the membrane. 
[0415] It has also been demonstrated that the usher proteins of Type- 1 and P-pili 

form ternary complexes with chaperone/subunit complexes with different affinities 
(Dodson, K. W„ etal, Proc. Nat. Acad. Sci. USA 90:3670-3674 (1993); Saulino, 
E. T., etal., EMBO J. 77:2177-2185 (1998)). This was interpreted as "kinetic 
partitioning" that allows a defined order of pilus proteins in the pilus. Moreover, 
it has been suggested that structural proteins might present a binding surface only 
compatible with one other type of pilus protein; this would be another mechanism 
to achieve a highly defined order of subunits in the mature pilus (Saulino, E. T., 
etal, EMBO J. 77:2177-2185 (1998)). 

B. Production of Type- 1 pili from Escherichia coli 
[0416] E. coli strain W3110 was spread on LB (10 g/L tryptone, 5 g/L yeast 

extract, 5 g/L NaCl, pH 7.5, 1 % agar (w/v)) plates and incubated at 37°C 
overnight. A single colony was then used to inoculate 5 ml of LB starter culture 
(10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, pH 7.5). After incubation for 
24 hours under conditions that favor bacteria that produce Type-1 pili (37°C, 
without agitation) 5 shaker flasks containing 1 liter LB were inoculated with one 
milliliter of the starter culture. The bacterial cultures were then incubated for 
additional 48 to 72 hours at 37°C without agitation. Bacteria were then harvested 
by centrifugation (5000 rpm, 4°C, 10 minutes) and the resulting pellet was 
resuspended in 250 milliliters of 10 mM Tris/HCl, pH 7.5 Pili were detached 
from the bacteria by 5 minutes agitation in a conventional mixer at 17.000 rpm. 
After centrifugation for 10 minutes at 10,000 rpm at 4°C the pili containing 
supernatant was collected and 1 M MgC12 was added to a final concentration of 
100 mM. The solution was kept at 4°C for 1 hour, and the precipitated pili were 
then pelleted by centrifugation (10,000 rpm, 20 minutes, 4°C). The pellet was 



-130- 



then resuspended in 10 mM HEPES, pH 7.5, and the pilus solution was then 
clarified by a final centrifugation step to remove residual cell debris. 

C. Coupling of FLAG to purified Type-1 pili of E. coli using m- 
Maleimidonbenzoyl-N-hydroxysulfosuccinimide ester (sulfo-MBS) 

[0417] 600 ul of a 95% pure solution of bacterial Type-1 pili (2 mg/ml) were 

incubated for 30 minutes at room temperature with the heterobifunctional 
cross-linker sulfo-MBS (0.5 mM). Thereafter, the mixture was dialyzed overnight 
against 1 liter of 50 mM Phosphate buffer (pH 7.2) with 1 50 mM NaCl to remove 
free sulfo-MBS. Then 500 ul of the derivatized pili (2 mg/ml) were mixed with 
0.5 mM FLAG peptide (containing an amino-terminal Cysteine) in the presence 
of 10 mM EDTA to prevent metal-catalyzed sufhydryloxidation The non- 
coupled peptide was removed by size-exclusion-chromatography. 

[0418] Figure 9 depicts an analysis of coupling of the FLAG peptide to type-1 

bacterial pili by SDS-P AGE. Lane 1 shows the unreacted pili subunit FimA. Lane 
3 shows the purified reaction mixture of the pili with the FLAG peptide. The 
upper band corresponds to the coupled product, while the lower band corresponds 
to the unreached subunit. 

EXAMPLE 34 
Construction of an expression plasmid for 
the expression of Type- 1 pili of Escherichia coli 
[0419] The DNA sequence disclosed in GenBank Accession No. U14003, 

the entire disclosure of which is incorporated herein by reference, contains all of 
the Escherichia coli genes necessary for the production of type-1 pili from 
nucleotide number 233947 to nucleotide number 240543 (the fim gene cluster). 
This part of the sequences contains the sequences for the genes fim A, fim\,fimC, 
fimD, fimV, fimG, and JirriH. Three different PCRs were employed for the 
amplification of this part of the E. coli genome and subsequent cloning into 
pUC19 (GenBank Accession Nos. L09137 and X02514) as described below. 



-131- 



[0420] The PCR template was prepared by mixing 1 0 ml of a glycerol stock of the 

E. coli strain W3110 with 90 ml of water and boiling of the mixture for 10 
minutes at 95 °C, subsequent centrifugation for 10 minutes at 14,000 rpm in a 
bench top centrifuge and collection of the supernatant. 

[0421] Ten ml of the supernatant were then mixed with 50 pmol of a PCR primer 

one and 50 pmol of a PCR primer two as defined below. Then 5 ml of a 1 OX PCR 
buffer, 0.5 ml of Taq-DNA-Polymerase and water up to a total of 50 ml were 
added. All PCRs were carried out according to the following scheme: 94°C for 
2 minutes, then 30 cycles of 20 seconds at 94°C, 30 seconds at 55°C, and 2 
minutes at 72°C. The PCR products were then purified by 1% agarose gel- 
electrophoresis. 

[0422] Oligonucleotides with the following sequences with were used to amplify 

the sequence from nucleotide number 233947 to nucleotide number 235863, 
comprising the Jim A, Jiml, and JimC genes: TAGATGATTACGCCAAGC 
TTATAATAGAAATAGTTTTTTGAAAGGAAAGCAGCATG (SEQ ID 
NO: 196) and GTCAAAGGCCTTGTCGACGTTATTCCATTACGCCCGTC 
ATTTTGG (SEQ ID NO. 197) 

[0423] These two oligonucleotides also contained flanking sequences that allowed 

for cloning of the amplification product into puc 1 9 via the restriction sites Hindlll 
and Sail. The resulting plasmid was termed pFIMAIC (SEQ ID NO: 198). 

[0424] Oligonucleotides with the following sequences with were used to amplify 

the sequence from nucleotide number 235654 to nucleotide number 238666, 
comprising the JimD gene: AAGATCTTAAGCTAAGCTTGAATTCTC 
TGACGCTGATTAACC (SEQ ID NO: 199) and ACGTAAAGCATTTCT 
AGACCGCGGATAGTAATCGTGCTATC (SEQ ID NO.200). 

[0425] These two oligonucleotides also contained flanking sequences that allowed 

for cloning of the amplification product into puc 1 9 via the restriction sites Hindlll 
and Xbal, the resulting plasmid was termed pFIMD (SEQ ID NO:201). 

[0426] Oligonucleotides with the following sequences with were used to amplify 

the sequence from nucleotide number 238575 nucleotide number 240543, 



-132- 



comprising the fiiriF, fimG, and fimH gene: AATTACGTGAGCA 
AGCTT ATGAGAAAC AAACCTTTTT ATC (SEQ ID NO: 202) and GACTAAG 
GCCTTTCTAGATTATTGATAAACAAAAGTCACGC (SEQ ID NO 203). 

[0427] These two oligonucleotides also contained flanking sequences that allowed 

for cloning of the amplification product into pucl 9 via the restriction sites HindiU 
and Xbal; the resulting plasmid was termed pFIMFGH. (SEQ ID NO:204). 

[0428] The following cloning procedures were subsequently carried out to 

generate a plasmid containing all the above-mentioned//'m-genes: pFIMAIC was 
digested EcoBl and HindUl (2237-3982), pFIMD was digested EcoRI and Sstll 
(2267-5276), pFIMFGH was digested Sstll and HindlU (2327-2231). The 
fragments were then ligated and the resulting plasmid, containing all the/iw-genes 
necessary for pilus formation, was termed pFIMAICDFGH (SEQ ID NO:205). 



EXAMPLE 35 

Construction of an expression plasmid for Escherichia coli type-1 pili that lacks 
the adhesion FimH 

[0429] The plasmid pFIMAICDFGH (SEQ ID NO:205) was digested with Kpnl, 

after which a fragment consisting of nucleotide numbers 8895-8509 was isolated 
by 0.7% agarose gel electrophoresis and circularized by self-ligation. The resulting 
plasmid was termed pFIMAICDFG (SEQ ID NO: 206), lacks the fimH gene and 
can be used for the production of FIMH-free type-1 pili. 



EXAMPLE 36 

Expression of type-1 pili using the plasmid pFIMAICDFGH 
[0430] E. coli strain W3110 was transformed with pFIMAICDFGH (SEQ ID 

NO.205) and spread on LB (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, pH 
7.5, 1 % agar (w/v)) plates containing 100 ug/ml ampicillin and incubated at 37 °C 
overnight. A single colony was then used to inoculate 50 ml of LB-glucose starter 



-133- 



culture (10 g/L tryptone, 5 g/L yeast extract, 1% (w/v) glucose, 5 g/L NaCl, pH 
7.5, lOOmg/ml ampicillin). After incubation for 12-16 hours at 37°C at 1 50 rpm, 
a 5 liter shaker flasks containing 2 liter LB-glucose was inoculated with 20 
milliliter of the starter culture The bacterial cultures were then incubated for 
additional 24 hours at 37 °C with agitation (150 rpm). Bacteria were then 
harvested by centrifugation (5000 rpm, 4°C, 10 minutes) and the resulting pellet 
was resuspended in 250 milliliters of 10 mM Tris/HCl, pH 8. Pili were detached 
from the bacteria by agitation in a conventional mixer at 1 7,000 rpm for 5 minutes. 
After centrifugation for 10 minutes at 10,000 rpm, 1 hour, 4°C the supernatant 
containing pili was collected and 1 M MgCl 2 was added to a final concentration 
of 100 mM. The solution was kept at 4 °C for 1 hour, and precipitated pili were 
then pelleted by centrifugation (10,000 rpm, 20 minutes, 4°C). The pellet was 
then resuspended in 10 mM HEPES, 30 mM EDTA, pH 7.5, for 30 minutes at 
room temperature, and the pilus solution was then clarified by a final 
centrifugation step to remove residual cell debris. The preparation was then 
dialyzed against 20 mM HEPES, pH 7 4. 

EXAMPLE 37 
Activation of HBcAg-Lys with SPDP 
[0431] HBcAg-Lys at a concentration of 15 uM was reacted with SPDP at a 

concentration of 456 uM SPDP for 60 minutes at room temperature, resulting in 
a thirty-fold excess of cross-linker over capsid subunit. The reaction mixture was 
subsequently loaded on SDS-PAGE for analysis, as shown in Fig. 10. The gel 
shows that the monomer subunits are cross-linked to dimers and higher-order 
polymers during the reaction. 

EXAMPLE 38 

Multimerization of HBcAg-Lys Upon Reaction With Sulfo-MBS 
[0432] HBcAg-Lys at a concentration of 1 1 8 uM was reacted with 20 mM Sulfo- 

MBS for 30 minutes at room temperature. As shown in Fig. 11, analysis of the 



-134- 



reaction mixture by SDS-PAGE revealed that the HBcAg-Lys monomers 
internally cross-linked to multimers, as reflected in the absence of a band 
corresponding to the subunit monomer after cross-linking. 

EXAMPLE 39 
Conjugation of HBcAg-Lys-2cys Mut to the FLAG Peptide 
[0433] HBcAg-Lys-2cys-Mut at a concentration of 80 uM was reacted with sulfa- 

MBS at a concentration of 8.8 mM for 30 minutes at room temperature, resulting 
in a 1 10-fold excess of cross-linker over capsid subunit. The reaction mixture was 
precipitated two times with 50% ammoniumsulfate and resuspended in 20 mM 
Hepes, 150 mM NaCl, pH 7.4, in a volume equivalent to the reaction volume 
before precipitation FLAG peptide containing an N-terminal cysteine was added 
at a concentration of 1.6 mM and the reaction was allowed to proceed for four 
hours at room temperature. The reaction mixture was subsequently loaded on 
SDS-PAGE for analysis, and the coupling products are shown in Fig. 12. 

EXAMPLE 40 
Conjugation of Pili to the p33 Peptide 
[0434] A solution of 1 ml pili at a concentration of 1.5 mg/ml (concentration of 

the subunit) was reacted with 750 ul of a 100 mM Sulfo-MBS solution in 20 mM 
Hepes, pH 7.4, for 45 minutes at room temperature. The reaction mixture was 
desalted over a Sephadex G25 column equilibrated with 20 mM Hepes, pH 7.4. 
Fractions containing pili protein were pooled after analysis by dot blot stained with 
amidoblack, and 0.6 ul of a solution of 100 mM p33 peptide 
(CGGKAVYNFATM, SEQ ID NO' 175), containing an N-terminal cysteine, in 
DMSO was added to 1 00 ul of the desalted activated pili and reaction allowed to 
proceed for four hours at room temperature. The reaction mixture was 
subsequently analyzed by SDS-PAGE, as shown in Fig 13. 



-135- 



EXAMPLE 41 
Expression of HBcAg-Lys-2cys-Mut 
[0435] The plasmid coding for HBcAg-Lys-2cys-Mut was transformed into E. 

coli K802. A single colony was inoculated into 50 ml LB containing 100 mg/ml 
ampicillin. The next day, the overnight culture was diluted into 2 L LB medium 
containing 100 mg/ml ampicillin and grown until ID 600 = 0.6 at 37°C. Cells were 
induced with 1 mM IPTG, and grown for another 4 hours at 37°C. The cells 
were then harvested, and the pellet resuspended in 5 ml of 10 mM NajHPC^, 03 
mM NaCl, 10 mM EDTA, 0.25% Tween, pH 7.0. Cells were then disrupted by 
sonification, and ammoniumsulfate was added to a concentration of 20%. The 
pellet was resuspended in 3 ml PBS buffer, and loaded onto a Sephacryl S-400 
column. The protein peak containing the capsid protein corresponding to the size 
of assembled capsid was collected and loaded onto a hydoxyapatite column for 
subsequent purification. The protein was eluted in the paththrough fraction. 

EXAMPLE 42 

Coupling of DP 178c peptide, immunization of mice and determination 
of the IgG subtypes 

[0436] DP 178c peptide is a fragment of the gp41 protein of HIV virus (Kilby, 

J.M. etal, 'Nature Medicine 4- 1302-07 (1998)); Wild, C. et al., Aids Res. Hum. 
Retroviruses 9: 1051-53 (1993)). 
A. Coupling of DP 1 78c to Pili 

[0437] A solution of 3 ml Pili (2.5 mg/ml) produced as described in Example 33 

B was reacted with 500 jA of a 100 mM Sulfo-MBS solution for 45 minutes at 
RT. The reaction mixture was desalted on a Sephadex G25 column equilibrated 
with 20 mM hepes pH 7 4, and fractions containing pili were pooled. An aliquot 
of 750 fu.1 of the activated pili was diluted in 750 jj\ DMSO, and 2-5 {A of a 100 
mM DP 178c solution in DMSO was added. The reaction was left to react 4 hours 
at RT, and glucose was added to the reaction mixture to give a final concentration 
of 0.2%. This solution was then dialyzed against 20 mM Hepes, 0.1% glucose, 



-136- 



pH 7.4. The dialyzed coupled pili were centrifliged and loaded on SDS-PAGE for 
analysis The result of the coupling reaction is depicted on Figure 14 A. The 
sequence of the DPI 78c peptide (fragment of the HIV gp41 protein) is 
C YTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWF (SEQ ID No : 1 76). 

B. Immunization of mice and IgG subtype determination 
[0438] 80 jug of Pili-DP178c was injected in saline intravenously into 

female Balb/c mice. These mice were boosted with the same amount of vaccine 
on day 14 and bled on day 24. DP178-specific IgG in serum was determined on 
day 24 in a DP 178 peptide specific ELISA (DP 178c peptide was conjugated to 
Ribonuclease A using the cross-linker SPDP). In Figure 14B, average results 
from two mice are shown as optical densities obtained with a 1 :50 dilution of the 
serum. 

EXAMPLE 43 
Expression and purification of GRA2 polypeptide 
[0439] Gra2 is an antigen of Toxoplasma Gondii. The 59 c-terminal amino acids 

acids of GRA2 with a c-terminal linker of 6 amino acids (GSGGCG, SEQ ID No. 
1 77) were cloned into the pGEX-2T vector (Pharmacia, 27-4801 -0 1). Expression 
and purification of the GST- fusion protein was carried out as described in the 
instructions. GST was cleaved from GRA2 with thrombin while the fusion protein 
was bound to glutathione-sepharose-beads and the reaction stopped after 20 min. 
with 1 mM PMSF. The sepharose beads were then pelleted by centrifugation and 
the supernatant containing the GRA2-polypeptide was collected. The solution 
was then concentrated 10-fold with a Ultrafree-4 centrifugal filter-5K (Millipore, 
UFV4BCC25). To reduce disulfide bonds which might eventually have formed, 
the solution was treated with 20 mM DTT 1 h on ice. DTT was removed by 
loading the protein solution on a PD10 column (Pharmacia). Protein 
concentration was determined by the Lowry test and concentration of free 
cysteines in an Ellmann's test. The protein was subsequently analyzed by SDS- 



-137- 



PAGE. The GRA2 protein can however not be detected by Commassie staining. 
A yield of 9 mg GRA2 was obtained from an 8 L culture. The GRA2 amino acid 
sequence isKEAAGRGMVT VGKKLANVES DRSTTTTQAPDSPNGLAETE 
VPVEPQQRAA HVPVPDFSQGSGGCG (SEQ ID No. 178) 

EXAMPLE 44 
Coupling of GPvA2 to Pili 

A. Coupling of GRA2 to Pili. 

[0440] 6 ml of a 2.5 mg/ml Pili protein solution (produced as described in 

Example 33 B) were reacted with a 50 fold molar excess of Sulfo-MBS, and 
desalted over a PD10 column (Pharmacia). 1.5 ml of the reaction mixture were 
loaded on one column, 1 ml was added and the first 1.5 ml were collected. 
Fractions containing Pili were identified on a dot blot stained with amidoblack. 
A 300 peg/ml solution of GRA2 was concentrated 100 fold, and 100 (A were 
reacted with 1 .2 ml of the desalted activated Pili solution for 4 hours at RT. The 
reaction mixture was then dialyzed against 21 of a 20 mM Hepes, 1 50 mM NaCl, 
pH 7.2 overnight. Figure 15A shows an analysis of the coupling reaction. 

B. Immunization of mice with Pili-GRA2 and IgG subtype determination. 
[0441] Mice, were immunized with 50 4g of Pili-GRA2 and boosted on day 

14,vith the same amount of vaccine. Serum samples we're taken on day 0,6,14 
and 21 after the first immunization. GRA2 specific IgG in serum was determined 
on day 2 1 in a GRA2 specific ELIS A. Results of two individual mice in each 
group are shown in Figure 1 5B. The titer was determined as the dilution of sera 
resulting in half-maximal optical density (OD 50 ). 



-138- 



EXAMPLE 45 
Coupling of B2- and D2-peptide to Pili 
[0442] D2 and B2 peptides are sequences from the OmpC protein of Salmonella 

typhi It is an outer membrane porin. High level of antiporin antibodies have been 
detected in the sera of patients with typhoid fever (Arocklasamy, A. and 
Krishnaswamy, S., FEBS Letters 453. 380-82 (1999)). 

A. Coupling of B2- or D2-peptides of the ompC protein of Salmonella typhi 
to Pili 

[0443] 6 ml of a 2.5 mg/ml Pili protein solution (produced as described in . 

Example 33 B) were reacted with a 50 fold molar excess of Sulfo-MBS, and 
desalted over a PD10 column (Pharmacia). 1.5 ml of the reaction mixture were 
loaded on one column, 1 ml was added, and the first 1.5 ml were collected. 
Fractions containing Pili were identified on a dot blot stained with amidoblack. 
An aliquot of 5 /A of a 100 mM solution of peptide was reacted with 2.6 ml of the 
desalted activated Pili solution for 4 hours at RT. The reaction mixture was then 
dialyzed against 21 of a 20 mM Hepes, 150 mM NaCl, pH 7.2 overnight. Figure 
16A shows an analysis of the coupling reaction. The sequence of the D2 peptide 
is CGG TSN GSN PST SYG FAN (SEQ ID No. 179). The sequence of the B2 
peptide is CGG DIS NGY GAS YGD NDI (SEQ ID No. 180). 



B Immunization of mice with Pili-B2 and IgG subtype determination. 
[0444] Mice were immunized interaperitoneally in female Balb/c mice with 50 ,ug 

of Pili-B2 in saline and boosted on day 14 with the same amount of vaccine, and 
bled on day 33. B2-peptide specific IgG in serum was determined on day 33 in 
a B2-specific ELISA (B2 peptide was conjugated to Ribonuclease A with the 
cross-linker SPDP) Average of the results of two individual mice are shown in 
Figure 16B. 



-139- 



EXAMPLE 46 

[0445] The muTNFa peptide, comprising amino acids 22-3 3 of TNFa protein was 

coupled to Pili as described in Example 42, except that no glucose was 
addedduring the final dialysis step, where the reaction solution was dialyzed 
against 20 mM Hepes, pH 7.4 only. Two Balb/c female mice, 8 days of age were 
immunized intravenously with 100 fu,g of Pili-muTNFa each. These mice were 
boosted at day 14 with the same amount of vaccine, and bled on day 20. IgG 
specific for native TNFa protein in serum was detected at day 20 in an ELISA. 
As a control, preimmune sera of two mice were assayed for binding to TNFa 
protein. See Figure 17. The sequence of the muTNFa peptide was 
CGGVEEQLEWLSQR (SEQ ID No. 181). 



EXAMPLE 47 

A.Preparation of bacterial type-1 pili coupled to TNF peptides 
[0446] Two peptides comprising murine TNFa sequences were designed. 

Peptide 3' murine TNFa II (3 '-TNFa II) was SSQNSSDKPVAHVVANHGVGGC 
(SEQ ID No 182). Peptide 5' murine TNFa II (5' TNFa II) was 
CSSQNSSDKPVAHVVANHGV (SEQ ID No. 183). The peptides 5 '-TNFa II 
and 3 '-TNFa II were coupled to bacterial type-1 pili as follows. An aliquot of 1 
ml of aPili solution (2.5 mg/ml) was reacted with 503 /A of a 100 mM Sulfo-NMS 
solution for 45 minutes at RT The reaction mixture was desalted over a desalting 
column previously saturated with Pili protein and equilibrated in 20 mM Hepes, 
pH 7.4. The fractions containing protein were pooled. Art aliquot of 1 ml of 
desalted Pili was mixed with 1.56 /A of peptide (100 mM in DMSO), and the 
reaction left to proceed for 4 hours at RT. The reaction solution was then 
dialyzed overnight against 20 mM Hepes, 1 50 mM NaCl, pH 7.4 in the cold. See 
Figure 18 A. 



-140- 



B . Immunization and detection of antibodies specific for native TNFa and the 
3' TNFII and 5' TNFII peptides 
[0447] Balb/c mice were vaccinated intraperitoneally with 3 0 //g protein in saline, 

on day 0, 14 and 33. IgG antibodies specific for native TNFa protein (Fig. 18B) 
and for the 3' TNFII and 5' TNFII peptides (Fig. 1 8C) were measured in a specific 
ELISA. 

1 . Native TNFa ELISA 

[0448] 2 ££g/ml native TNFa protein was coated on ELISA plates. Sera were 

added at different dilutions and bound IgG was detected with a horseradish 
peroxidase-conjugated anti-murine IgG antibody. Results from four individual 
mice are shown on day 2 1 and day 43 . 

2. Anti peptide ELISA 

[0449] IgG antibodies specific for the 3' TNFII and 5' TNFII peptides 

were measured in a specific ELISA 1 0 ug/ml Ribonuclease A coupled to 3' TNFII 
or 5'TNFII peptide was coated on ELISA plates. Sera were added at different 
dilutions and bound IgG was detected with a horseradish peroxidase-conjugated 
anti-murine IgG antibody. Results from four individual mice are shown on day 2 1 . 

C. Analysis of sera from mice immunized under B.: IgG subtype 
determination 

[0450] Sera from the immunized mice described under B. were taken on 

day 50. Antibodies specific for the TNF peptides described under A. were 
measured in a specific ELISA on day 50. RNAse coupled to the corresponding 
TNF peptide was coated on ELISA plates at a concentration of 10 ug/ml. Sera 
were added at different dilutions and bound antibody was detected with horse 
radish peroxidase-conjugated anti-murine antibodies. See Figure 18D. 



-141- 



EXAMPLE 48 

Coupling of Pili to M2 peptide, immunization of mice, and IgG subtype 
determination 

[0451] M2 peptide was coupled to pili as described in Example 47. The peptide 

was reacted at a fivefold molar excess with the activated Pili. Female Balb/c mice 
were injected with 50 //g Pili-M2 in saline subcutaneously. Mice were boosted 
with the same amount of vaccine on day 14 and bled on day 27, M2 specific IgG 
in serum was determined on day 27 in a M2-specific ELISA (peptide conjugated 
to Ribonuclease A with the cross-linker SPDP for coating). See Figures 19A and 
19B. 



EXAMPLE 49 

Immunization of mice with HbcAg-Lys-2cys-Mut coupled to the Flag peptide, 
and IgG subtype determination 
[0452] Flag peptide (SEQ ID NO: 147) was coupled to HBcAg-Lys-2cvs-Mut as 

described in Example 39. Two Balb/c mice were vaccinated intravenously with 
50 jj,g HBc-Ag-Lys-2cys-Mut -Flag. On day 14 mice were boosted with the same 
amount of vaccine and bled on day 40, Flag-specific antibodies (Flag peptide was 
conjugated to Ribonuclease A with the cross-linker SPDP for coating) in serum 
were measured on day 40 in a specific ELISA. ELISA plates were coated with 
1 0 //g /ml RNAse coupled to Flag peptide and serum was added at a 1 :40 dilution. 
Bound antibodies were detected with peroxidase conjugate isotype-specific IgG. 
Results from the two mice are shown as ELISA titers in Figure 20. 



EXAMPLE 50 
Purification of Type- 1 Pili of Eschericia coli 
[0453] Isolated Type-1 pili of Eschericia coli prepared as described in Example 

33B were precipitated with ammonium sulfate, added to a final concentration of 



-142- 



0.5 M, at 4° C for 30 minutes The pili were then pelleted by centrifugation at 
20,000 rpm for 15 min at 4°C and the pellet was resuspended in 25 ml of 20 mM 
HEPES buffer, pH 7.3. The precipitation step was repeated once, and the final 
sample was resuspended in 9 ml of 20 mM HEPES, pH 7.3 and finally dialyzed 
against the same buffer to remove residual ammonium sulfate. The pili were 
subsequently purified on an SR-400 size exclusion chromatography column (20 
mM HEPES, pH 7.3) and the pili containing fractions were collected and pooled. 
[0454] All patents and publications referred to herein are expressly incorporated 

by reference. 

[0455] The entire disclosure of U.S. Application No. 09/449,631, filed 

November 30, 1999, is herein incorporated by reference. All publications and 
patents mentioned hereinabove are hereby incorporated in their entireties by 
reference.