Forschungsprojekt am INB

Development of a multivalent therapeutic HPV-16 and HPV-18 -specific DNA vaccine against cervical cancer

The two human papillomaviruses (HPV) types 16 and 18 are associated with the development of cervical cancer (cc) in about 70 % of all cases

Worldwide approximately 370,000 cases of cc are being diagnosed each year and almost 200,000 deaths are attributed to this disease [Pisani, P. et al., Int J of Cancer 83 (1999) 18-29]. In Third World countries cc is one of the major cause of cancer-related deaths. About 80 % of women dying from this disease originate from those low-budget countries where screening programs for early detection and the medical infrastructure for treatment are not available. As a consequence of cytologic screening programs the mortality was reduced by 70% in the US during the last 50 years [Piver, MS., Handbook of gynecologic oncology, Boston, Little, Brown (1996)]. But even under optimal circumstances of medical care about 40% of cc patients dye of this disease [Gatta, G. et al., Eur J Cancer 34 (1998) 2218-2225].
Based on the nucleotide sequence of the L1 major structural protein (see fig. 1) more than 130 different HPV types have as yet been identified [de Villiers, E.M., Clin Dermatol 15 (1997) 199-206], 14 of which were found to be associated with cc ("high-risk types"). The high-risk types 16 and 18 are responsible for about 70% of the cases [Bosch, FX. et al., J Natl Cancer Inst 87 (1995) 796-802].

Fig.1: HPV-assembly. Five L1 main structure proteins (monomers) are assembling to one pentamer (caspomere). Finally, 72 capsomeres are building one virus-like particle (VLP) (according to Lutz Gissmann, modified).

The both oncoproteins E6 and E7 of the high-risk HPVs are excellent targets for immunotherapy 

Since E6 and E7 are exclusively and consistently expressed within the HPV-infected tumor cells they represent specific targets for a tumor-specific immune therapy. Thereby the risk of auto-immune reactions following the application of an HPV E6/E7 vaccine is considered to be very low since no significant homology between the viral and cellular proteins is known. On the other side the transforming activity of the high-risk HPV types has been assigned to the oncoproteins E6 and E7 [Pecoraro, G. 
et al., Proc Natl Acad Sci USA 86 (1989) 563-567]. They interfere with the cell cycle control by interaction with the cellular tumor suppressor gene products p53 and pRB [Munger, K. et al., Cancer Surv 12 (1992) 197-217 and Dyson, N. et al., Science 243 (1989) 934-937]. 
One very interesting approach within the development of therapeutic HPV vaccines are chimeric virus-like-particles (CVLPs). These particles contains E7 sequences fused to a truncated L1-protein [Müller, M. et al., Virology 234 (1) (1997) 93-111] and are very immunogenic in the mouse model, able to induce L1/E7-specific cellular immune response [Müller, M. et al., Virology 234 (1) (1997) 93-111 and Schäfer, K. et al., Int J Cancer 81(6) (1999) 881-8]. Problematical for the effect of this vaccine is a potential pre-existing humoral anti-HPV immunitiy [Da Silva, D. M. et al., Virology 290(2) (2001) 350-60] and the fact that not all natural occurring E7-epitopes were offered. In a first clinical trial no significant clinical response in humans were observed (
Due to the fact that the CVLP production is quiet expensive, capsomeres were investigated in mice. In fact, HPV vaccines are particularly needed in developing countries, and subcapsid particles (pentameric capsomeres) may be an economically advantageous alternative to VLPs especially in medically underfunded areas, since they can be readily purified after expression in Escherichia coli [Chen, Y. 
et al., J. Mol. Biol. 307 (2001) 173–182 and Zhang, W. et al., Virology 243 (1998) 423–431] and are very stable [McCarthy, M. P. et al., J. Virol. 72 (1998) 32–41]. We were able to show for the first time that capsomeres are able to induce a specific cellular immune response [Öhlschläger, P. et al., Vaccine 24 (2006) 2880-2893]. Moreover, also peptides [Steller, M.A. et al., Clin Cancer Res 4(9) (1998) 2103-9 and Muderspach, L. et al., Clin Cancer Res 6(9) (2000) 3406-16], HPV-recombinant viruses [Fiander, A.N. et al., Int J Gynecol Cancer 16(3) (2006) 1075-81] and in vitro manipulated (loaded with HPV proteins/peptides or DNA-transfected) dendritic cells (DCs) were performed [Ferrara, A. and Nonn, M. et al., J Cancer Res Clin Oncol 129(9) (2003) 521-30].

DNA-based therapeutic vaccination could be an ideal supplement to existing therapies 

As compared to protein- or peptide-based vaccines a DNA vaccine has remarkable advantages making it of potential interest for Third World countries. Its production costs are relatively low and predictable. DNA is stable and does not require refrigeration for storage. There are no unwanted immune reactions against other components of the vaccine as it is observed in case of vector based-vaccines thus DNA vaccines can be used for repeated boosting [Liu, M. A., Nat Med 4 (1998) 515].

There is a principle concern that integration of plasmid DNA could lead to an induction of oncogenes or inactivation of tumor suppressor genes. However, in mouse experiments it was shown that even under the most unfavorable conditions the mutation rate is unmeasurable, i.e. at least 3000 times below the frequency of spontaneous mutations [Martin, T. et al., Hum Gene Ther 10 (1999) 759-768 and Nichols, W.W. et al., Ann N Y Acad Sci 772 (1995) 30-39]. 
Clinical studies in humans demonstrated the absence of severe side effects after DNA immunization. In clinical trials mostly HIV genes are being tested and complete lack of severe side effects was published [MacGregor, R.R. et al., The J of Inf Dis 178 (1998) 92-100]. Presence of DNA-specific antibodies was not reported. In contrast a humoral immune response after DNA immunization was found in a mouse model [Mor, G. et al., Hum Gene Ther 8 (1997) 293-300]: The number of anti-DNA IgG secreting B cells increased by two- to three fold shortly after vaccination but no symptoms of autoimmunity were detected [Katsumi, A. et al., Hum Gene Ther 5 (1994) 1335-1339 and Mor, G. et al., Hum Gene Ther 8 (1997) 293-300 and Xiang, Z.Q. et al., Virol 209 (1995) 569-579 and Gilkeson, G.S. et al., J Immunol 161 (1998) 3890-3895].
Regarding DNA vaccination against HPV-related diseases Zycos announced the completion of a phase 2b clinical trial. The vaccine ("ZYC101a") consists of several minigenes encapsidated into Zycos' proprietary microparticles that code for HPV-specific peptides. The packaging is assumed to facilitate the transport of the DNA to the pAPCs and lymphoid tissues. One hundred and sixty-one patients with biopsy-proven high-grade cervical dysplasia were included into this randomized, double blind and placebo-controlled trial. The vaccine was reported safe and well tolerated. Regression of the lesion was observed in 43% of the vaccine, but only 27% of the placebo group. 

Following infection of plasmids expression of the encoded antigen was reported to last for up to 19 months [Wolff, J.A. et al., Hum Mol Genet 1 (1992) 363-369]. Professional antigen-presenting cells are either transfected directly and express the antigen themselves [Condon, C. et al., Nat Med 2 (1996) 1122-1128 and Iwasaki, A. et al., Vaccine 17 (1999) 2081-2088 and Casares, S. et al., J Exp Med 186 (1997) 1481-1486] or they take up the antigen by phagocytosis following expression and secretion by lysis of other cells (e.g. myoblasts) [Fu, T.M. et al., Mol Med 3 (1997) 362-371 and Doe, B. et al., Proc Natl Acad Sci USA 93 (1996) 8578-8583]. Transfer of influenza nucleoprotein-expressing myoblasts into mice protected the animals from a viral challenge. Thus it was concluded that antigen uptake by non-professional APCs and subsequent phagocytosis by pAPCs is existent [Ulmer, J. B. et al., Immunology 89 (1996) 59-67].
Although the number of DNA-specific IgG secreting B cells was shown to rise 2-3 fold shortly after DNA vaccination [Mor, G. et al., Hum Gene Ther 8 (1997) 293-300] this increase was not accompanied by symptoms of an autoimmune disease in a mouse model [Katsumi, A. et al., Hum Gene Ther 5 (1994) 1335-1339 and Mor, G. et al., Hum Gene Ther 8 (1997) 293-300 and Xiang, Z.Q. et al., Virol 209 (1995) 569-579 and Gilkeson, G.S. et al., J Immunol 161 (1998) 3890-3895].

Development of an artificial HPV-16 E7 gene (HPV-16 E7SH) providing all naturally occurring epitopes but lacking transforming properties 

For safety reasons at least a functional oncogene can not be applied to humans. Therefore efforts were made to inactivate the oncogenic properties of the HPV-16 E7. Some investigators have introduced point mutations into the sites of the E7-oncogene that are associated with transforming potential [Shi, W. et al., J Virol 73 (9) (1999) 7877–81 and Smahel, M. et al., Virology 281(2) (2001) 231–8], whereas others have used HLA-restricted singular epitopes [Doan, T. et al., Cancer Res 60(11) (2000) 2810–5 and Velders, M.P. et al., J Immunol 166(9) (2001) 5366–73]. These approaches, however, may lead to an unwanted loss of naturally occurring epitopes that is potentially associated with a decrease in vaccine-efficacy. Our aim was to supply all potential naturally occurring T cell epitopes, covering the broad range of MHC restriction. In consequence, prior knowledge of the patient´s HLA-haplotype is not required. Therefore, a more potent immune response may be induced, involving all occurring HLA-restriction elements in the vaccine. We have generated an artificial HPV-16 E7-gene (HPV-16 E7SH) with a rearranged primary sequence inducing specific immunity in vivo in mice and after in vitro immunization of human lymphocytes and, therefore, holds promise for a therapeutic HPV-vaccine [Öhlschläger, P. et al., Vaccine 24 (2006) 2880-2893]. The E7 wildtype gene (E7WT) was dissected exactly at the positions that are critical for transforming properties of the protein (pRB-binding site, C-X-X-C motifs). The resulting 4 fragments were subsequently arranged in a different order as a "core-element". In order to maintain all putative CTL epitopes located at the original junctions of the fragments an "appendix" was added to the core-element that consists of the sequences encompassing the original junctions. The codons in the core-element were adapted for the optimal use in human cells (almost identical to mouse cells). This feature was introduced in order to minimize the risk of intra- or intermolecular recombination within the E7SH genes that might eventually reconstitute the E7WT gene (see fig. 2).

Fig. 2: Generation of the HPV-16 E7SH gene. The HPV-16 E7 wildtype gene (E7WT) was dissected at the positions corresponding to the pRB binding site (nt 72/73) and in between the two Cys-X-X-Cys motifs (nt 177/178 and 276/277). The resulting four fragments a, b, c and d were rearranged ("shuffled“) forming the core-element with the sequence a, d, c, b. To avoid loss of putative CTL epitopes at the junctions a-b, b-c and c-d, these sequences (3 x 27 nucleotides = 3 x 9 amino acids) were added forming the appendix. A Kozak sequence was added in front of the gene to enhance the translation. 

The artificial HPV-16 E7SH gene is immunogenic in the murine model as well as in humans 

The E7SH genes proved to be very immunogenic in mice as measured by IFN-
γ Elispot-assays (testing for cytotoxic T-cells, CTLs) and by 51Cr-release assay [Öhlschläger, P. et al., Vaccine 24 (2006) 2880-2893]. In tumor regression experiments a significant tumor regression was observed mediated by HPV-16 E7SH (see fig. 3). As expected, in tumor formation assays (NIH 3T3-based soft-agar transformation assay) no transformation activity was observed [Öhlschläger, P. et al., Vaccine 24 (2006) 2880-2893]. 
A more advanced version of this artificial HPV-16 E7SH gene is now ready for a clinical trial phase I which is currently prepared.

Fig. 3: Tumor regression experiment. Animals received HPV-16 E7 wildtype expressing tumor cells subcutaneously into the right flank. After the establishment of small tumors (3-5 mm2 in diameter) animals were treated with HPV-16 E7SH DNA or with empty plasmid (control). In the left an animal treated with E7SH DNA showing total tumor regression is given. Right one animal of the control group developed a large tumor.

Enhancement of the of immunogenicity of the HPV-16 E7SH DNA-based immunotherapy by delayed co-application of nucleotide-encoded adjuvant genes 

Tumors consist often of highly selected cells, circumventing effective immune response. For example, in the case of cc in the mouse model was shown, that the exposure to E7 in a non-inflammatory epithelium leads to the induction of peripheral tolerance to E7 in the CTL population (Doan T, J Virol., 73, 6166-6170, 1999). One further challenge for a DNA-base therapeutic tumor vaccine is the lower immunogenicity of DNA vaccines in general compared to protein-based vaccines. In order to create a low-cost and easily to produce DNA vaccine which is highly immunogenic, all possible mechanisms to enhance potency has to be investigated. 

One way to do so is the application of cytokines, which recruits dendritic cells (DCs) to the application site (in most cases the muscle). The ability of DCs to present antigens und to induce cellular immunity depends on their degree of activation. During conversion of immature to mature DCs, a number of changes are observed, e.g. increased expression of MHC and co-stimulatory molecules and secretion of cytokines and chemokines for attraction and expansion of T-cells. Through the use of a combination of adjuvant genes / CpG enriched vector, probably recruitment (e.g. by MIP- 1α), maturation (e.g. by CpG motifs) and activation (e.g. by GM-CSF) of DCs should be concerted resulting in an enhanced DNA vaccine potency. 
That co-injection of DNA encoding for adjuvant cytokines is in principle a valuable strategy to increase the immunogenicity of a DNA vaccine was demonstrated for example for IL-2 and IFN- γ [Chow, Y.H. et al., J Immunol 160 (1998) 1320-1329]. Both cytokines stimulate the Th1 immune response and thus the generation of CTLs. IL-12 expressed by B cells and macrophages is of particular importance since it favors a Th1-response [Guler, L. et al., J of Immunol (1999) 1339-1347; Heinzel, F. P. et al., J Exp Med 177 (1993) 1505-1509; Afonso, L. C. et al., Science 263 (1994) 235-237]. Granulocyte-macrophage colony-stimulating factor (GM-CSF) has been shown to enhance immune responses in different animal models after DNA vaccination [Weiss, W.R. et al., J Immunol 161 (1998) 2325-2332 and Ahlers, J.D. et al., Proc Natl Acad Sci USA 99 (2002) 13020-13025]. GM-CSF particular promotes the differentiation and activation of macrophages and DCs [Morrissey, P.J. et al., J Immunol 139 (1987) 1113-1119 and Caux, C. et al., Nature 360 (1992) 258-261]. Because GM-CSF protein has a short half time in plasma (0.9 to 2.5 h) [Stute, N. et al., Blood 79 (1992) 2849-2854], a single injection of recombinant protein is therefore insufficient. To achieve a sustained GM-CSF availability many studies have used the GM-CSF plasmid DNA approach [Sedegah, M. et al., Genes Immun 5 (2004) 553-561]. GM-CSF was also used as a successful adjuvant in tumor therapy studies [Dou, J. et al., Immunol Invest 35 (2006) 227-237]. MIP-1α (macrophage inflammatory protein 1 α) which binds to CC chemokine receptor 5 (CCR5) on immature DCs, is able to recruit these pAPCs to the site of inoculation, resulting in enhanced induction of the cellular as well as the humoral immune response [McKay, P.F. et al., Eur J Immunol 34 (2004) 1011-1020].

Because migration of DCs to the site of application and DC maturation are influenced by different chemokines it is necessary to find an optimal combination of different adjuvant genes. Moreover, in the last few years the timing of co-administration of the adjuvant genes was shown to be crucial (Kusakabe, K. et al., The J of Immunol 164 (2000) 3102-3111 Barouch, al., The J of Immunol 161 (1998) 1875-1882). We have investigated the optimal combination and the the time point of application of the gene-encoded antigen (HPV-16 E7SH) and adjuvant genes (IL-2, IL-12, IFN-γ, GM-CSF, MIP-1α). Thereby, the gene-encoded MIP-1α applied five days prior to E7SH-immunization combined with IFN-γ or IL-12 (3 days) or IL-2 (5 days) post immunization lead to a significantly enhanced tumor response which was clearly associated with granzyme B secretion and target cells lysis. Our results suggest that a conditioning application and combination with adjuvant genes may be a promising strategy to enhance synergistically immune responses by DNA immunization for the treatment of cervical cancer (Öhlschläger, P. et al., Int J Cancer 125 (2009) 189-98). 

Enhancement of the immunogenicity of the HPV-16 E7SH DNA vaccine by a Prime-Boost strategy with HPV-recombinant influenza viruses 

Recombinant viruses are able to transfer a foreign gene with a very high efficiency into cells. Thereby, influenza viruses are very safe due to the fact that no DNA intermediates were observed during the whole replication cycle. Therefore an integration of the recombinant genes into the host genome is impossible. Moreover the dsRNA genome of influenza provides a danger signal to immature DCs resulting in maturation [Cella, M. et al., J Exp Med 189 (5) (1999) 821-9]. Within this study we use a highly attenuated virus which self-limits its own replication (see fig. 4).

Fig. 4: Generation of HPV-16 E7SH recombinant avian influenza viruses. The plasmid containing the HPV-16 E7SH gene flanked by a mutated overactive influenza virus promoter, packaging signals as well as RNA polymerase I control sequences is transfected into cells. These cells were infected with an avian influenza helper virus. The heterologous HPV-16 E7SH gene is transcribed by the cellular RNA polymerase I in “virus RNA” (-ssRNA). These molecules are subsequently amplified by the viral RNA-polymerase. All amplified genomic RNA molecules are packaged into new assembled virus particles (according to Strobel, I., et al., 2000). 

Due to the overactive promoter of the recombinant gene the heterologous fragment is after 2-3 passages in canine kidney cells over presented. As a consequence the recombinant fragment replaces the essential influenza RNA segments within the particles and the viruses becomes defective by attenuation [Neumann, G. et al., J Virol 74 (1) (2000) 547-51]. Here, we found an increased CTL-response und stronger tumor-regression in BL/6 animals in the DNA prime / flu boost group compared to the DNA only treated animals. 

FH Aachen Campus Jülich

Institut für Nano- und Biotechnologien

Prof. Dr. Peter Öhlschläger
Heinrich-Mußmann-Str. 1

52428 Jülich

T  +49.241.6009 53835

F  +49.241.6009 53834