FEBS Letters 579 (2005) 586–590 FEBS 29184

Hypothesis Mimotope-hormesis and mortalin/grp75/mthsp70: a new hypothesis on how infectious disease-associated mimicry may explain low cancer burden in developing nations

Custer C. Deocarisa,b,c, Kazunari Tairaa,b, Sunil C. Kaula, Renu Wadhwaa,* a Cell Proliferation Research Team, Gene Function Research Center, National Institute of Advanced Industrial Science & Technology (AIST), 1-1-1 Higashi, Tsukuba Science City 305-8562, Japan b Department of Chemistry and Biotechnology, School of Engineering, University of Tokyo, Hongo, Tokyo 113-8656, Japan c Atomic Research Division, Philippine Nuclear Research Institute (PNRI), Commonwealth Avenue, Quezon City 1101, Philippines

Received 24 November 2004; revised 25 November 2004; accepted 29 November 2004

Available online 24 December 2004

Edited by Beat Imhof

cal systems and has accumulated moderate experimental Abstract It is generally observed that countries with heavy infectious burden show lower cancer incidence as compared to evidence in the recent past is the concept of hormesis. It is more affluent nations. With the emerging paradigm on microbial broadly defined as a beneficial or positive effect of low doses heat shock proteins (hsps) as molecular link between infections of an otherwise harmful agent or stressor. The most widely and autoimmune diseases, we posit a new hypothesis, the ‘‘mimo- known is radiation hormesis that is commonly implicated in tope-hormesis’’, on the immunologic impact of infections on re- the observed decreased cancer risk from low-dose irradiation. gional cancer prevention. According to this, assaults of Anti-aging and life-prolonging effects of a wide variety of the infection during early adulthood could fortify the so-called stressors, such as irradiation, caloric restriction, heat to evoke more potent defenses against late-onset diseases, such shock, pro-oxidants, aldehydes and hypergravity, have been as cancer, via . Interestingly, both experimental dissected from cellular to organismic levels [6–12]. Corre- and clinical data support the beneficial role of autoimmunity in spondingly, we put forward here a hypothesis on mimotope- long-term cancer survivors. We illustrate this by a comprehensive in silico mimotope (epitope mimicry) analysis of human infec- hormesis that may provide a rational and integrative molecular tious pathogens against mortalin (mthsp70/PB74/GRP75), a link to the inverse trend of infectious disease burden and type of hsp70 protein involved in the control of cell proliferation, cancer. immortalization and tumorigenesis. Ó 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. 2. Hormesis hypothesis and tumor vaccinology

Keywords: Mortalin; Autoimmunity; Heat shock protein; The thrust of mimotope-hormesis, in part, relies on the initi- Mimotope; Cancer ation and long-term survival of a T and pool capable of mounting protective immune responses. In early adulthood, individuals may be challenged by an enormous variety of pathogens, which activate high-affinity repertoires that may 1. Introduction be cross-reactive to tumor [13]. From another perspec- tive, ‘‘beneficial’’ autoimmunity may particularly be exploited Although specific infectious agents (e.g., HPV in cervical as an effective preventive or prophylactic measure for a large cancer, Helicobacter pylori in gastric cancer, Schistosoma variety of old age cancers. This is in contrast with the proposed japonicum with bladder cancer, etc.) that target cellular onco- immunotherapies that are expensive and require repeated genes and tumor suppressors, induce chronic inflammation administration for longer durations. Hence, identification of and result in production of reactive oxygen species have been mimotopes (epitope mimetics) by combinatorial epitope dis- associated with malignancies [1,2], many regions of the world, covery with peptide libraries, or by recently developed in silico particularly the developing countries, in spite of having high strategies, as demonstrated in this study and their use as alter- infectious disease burden appear to show an inverse trend in natives to natural antigens that are not readily available for terms of cancer incidence [3–5]. The reason for this is not yet serological testing and development, are now a most well understood and will likely rely on individual genetic challenging avenue in tumor research [13–18]. make-up, cancer type, infection history and a host of environ- As , for example, five candidate mimotopes of Her- mental factors. 2/neu, a gene upregulated in most breast cancers, have been A provocative and legitimate hypothesis that has stemmed identified by phage display using the humanized monoclonal from the non-linearity between dose and response (for exam- Herceptin. The strong immunological response to ple, low-dose stimulation and high-dose inhibition) in biologi- these peptides makes them suitable candidates for formulation of a breast cancer vaccine [17]. In another study, mimotopes of *Corresponding author. Fax: +81 29 861 2900. a shared tumor-associated epitope in cutaneous E-mail address: [email protected] (R. Wadhwa). lymphoma were tested for their capacities to induce clinical

0014-5793/$30.00 Ó 2004 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2004.11.108 C.C. Deocaris et al. / FEBS Letters 579 (2005) 586–590 587 and immunological responses in cancer patients. These cross- [36]. On one end, insulin-dependent diabetes mellitus in NOD reactive peptides were derived using a combinatorial peptide li- mice was inhibited by immunization with a cross-reactive epi- brary without any prior knowledge of the corresponding natu- tope of hsp60, termed as p277. This epitope activated a pool of ral tumor-associated . Vaccination with these CD4+ T-cell clones underscoring the potential of hsps for pep- mimotopes together with helper T cell-inducing antigens led tide therapy while highlighting the complexity of immune re- to complete tumor remission in the two patients tested. After sponse to hsps [37]. For hsp70, reactive T cells and each booster vaccination, enhanced frequencies of mimotope- were detected in serum and cerebrospinal fluid of specific CD8+ T cells were detected in the peripheral blood multiple sclerosis patients [38], while paradoxically, a bolus of the patients, and the CTL proved to be cytotoxic and of recombinant hsp70 attenuated some experimental autoim- tumoricidal when tested in vitro. These data provided the first mune diseases [39]. Whilst, other studies failed to provide evi- indication of clinical efficacy of mimotopes in cancer patients dence that the titer levels exceeded those found in normal [16]. Notably, this phenomenon of epitope mimicry (mimot- subjects [40]. A robust phage display strategy was able to opy) has been proposed as an initiator of tumor-directed identify another hsp70 member, GRP78, as major tumor- auto- observed among long-time survivors [19]. associated antigen in serum of prostate cancer patients [41]. Mimotopes for heat shock proteins have been shown to trig- Mortalin/PBP74/GRP75 (Accession No. P38646) is a unique ger beneficial autoimmune responses [20,21]. This is consistent member of the hsp70 family and is associated with cell prolif- with the observation that average mortality rate of cancer pa- eration, tumorigenesis and immortalization [42–44]. It also tients with multiple sclerosis, an autoimmune attack against functions in antigen processing, stress response, cell prolifera- ‘‘self’’ hsp70, is significantly lower than that of ordinary cancer tion, differentiation and tumorigenesis [43–47]. Mortalin is patients [22]. In this paper, we associate mimotopy as an both overexpressed and differentially localized (pancytoplas- important component of an immunologic hormetic phenome- mic in normal to perinuclear in immortal cells) in all human non. We demonstrate this by showing by an in silico epitope tumor-derived and in vitro immortalized cells analyzed so far analysis of mortalin and hsps from known human pathogens. [48,49]. In a comparative proteome profile of the cell surface of tumors, mortalin is also found to be one of the more abun- dant residents along with some other chaperones and known 3. Hsps as potential therapeutics and trigger factors of cell surface markers, e.g., immunoglobulin and laminin recep- autoimmune diseases tors [50]. These evidences (overexpression, altered localization and protein clients) are indicators for mortalinÕs potential neo- Heat shock proteins are ancient families of chaperones for immunogenicity during carcinogenesis. While it shares 70–80% the folding/unfolding, degradation/stabilization, translocation homology with other known hsp70s from microbial and para- of proteins during various stresses, such as heat shock, ER- sitic infections, we deem mortalin as a highly attractive candi- stress, glucose deprivation, etc. Originally recognized for their date in exploring the phenomenon of epitope mimicry role in resistance to stress, they have received a lot of attention (mimotopy) by infectious agents and in formulating a model from immunologists for their ability to stimulate the adaptive on how such molecular events could possibly translate to tu- and innate immune system. Tumor cells subjected to a non- mor regression, or cancer prevention at a population level. lethal heat shock stress are unable to form tumors in syngenic mice. This is attributed to the enhancement of T-cell-mediated response with the expression of the inducible hsp70 that func- 4. In silico identification of tumor-associated hsp70 mimotopes tions in antigen processing and presentation by MHC I [23]. Furthermore, immunogenicity of hsps is believed to be en- We identified using an HLA-DR ligand prediction hanced from the antigenic peptides with which they associate software TEPITOPE obtained from Dr. Hammer and co- with and, hence, the use of HSP-peptide complexes has been workers [51,52]. Peptide strings strongly binding to at least proposed as tumor vaccines [24]. Hsp70 purified from malig- three of the eight HLA-DR alleles (DRB1*0101, DRB1 nant and virally infected cells could transfer and deliver anti- *0301, DRB1*0401, DRB1*0701, DRB1*0801, DRB1*1101, genic peptides to APCs to elicit peptide-specific immunity DRB1*1501 and DRB5*0101) were selected as promiscuous [25]. The conserved nature of hsp70 proteins and their inherent epitopes based on the reported criteria from previous work immunogenicity, alone or in tandem with another protein, on TEPITOPE-based vaccines [53,54]. These MHC alleles ac- prompted that they may provide the link between infection count for more than 90% of the HLA class II isotypes on anti- and autoimmunity by acting as autoantigens with which, gen presenting cells [55]. Prediction threshold was set at 1% molecular mimicry could play a dominant role [24–27]. such that our identified epitope would belong to the top 1% As auto-antigens on the other hand, reactivity to heat shock best binders based on a pool of 8000 peptide frames from a proteins may explain associations between infectious agents randomized library of peptide sequences [56]. and autoimmunity that include b-hemolytic streptococci and We also obtained hsp70 protein sequences of 76 bacteria, 29 rheumatic fever; Trypanosoma cruzi and ChagasÕ disease; mycoses and 11 parasites of medical relevance [57] available Borellia burgdorfii and Lyme arthritis; Rubella and type 1 dia- from the NCBI (http://www.ncbi.nlm.nih.gov/) and EXPASY betes [28–30]. Certain strains of rats, for example, developed (http://us.expasy.org/) websites and aligned the sequences with severe polyarthritis that resembled rheumatoid arthritis in hu- Clustal W to identify regions possessing 100% homology with mans after immunization with heat-killed Mycobacterium mortalin epitopes. Such homologous sequences were desig- tuberculosis and its hsp60 epitopes [31]. Correlations of anti- nated as mimotopes. Five immunodominant clusters of mort- hsp60 and disease severity have been demon- alin were identified and of these, VLLLDVTPLS and strated for patients with atherosclerosis [32], psoriasis [33], LGQFLIGPPAPR appeared promiscuous. Out of 116 protein cystic fibrosis [34], Kawasaki disease [35] and BehcetÕs dieases sequences examined, only 14 pathogen-derived hsp70s showed 588 C.C. Deocaris et al. / FEBS Letters 579 (2005) 586–590

Table 1 Mortalin mimotopes from human pathogens

Cross-reactivity, in this case, corresponds to pathogensÕ proteins sharing 100% homologies with mortalin epitopes. Quantitative evaluation of promiscuity of the two mimotopes is shown at the right. Bars indicate the threshold setting at which peptide is predicted as ligand for each listed HLA-DR allele. Peptides displaying affinity for two out of the five designated HLA-DRs (see Part 2) at 1% are designated as highly promiscuous [54–62].

100% homology to mortalin epitopes (Table 1) with Two probable strategies for a ‘‘beneficial’’ autoimmunity by VLLLDVTPLS appearing to be most cross-reactive. It is also exogenous cross-reactive antigens may be staged: first, epitope observed that number of non-promiscuous epitopes, or pep- mimicry can prime a second pool of precursor tides that do not bind to a wide range of human MHCs, do harboring high affinity towards immunodominant domains not fall in the highly conserved regions (data not shown). of the tumor and this, consequently, is expected to escape Interestingly, infectious agents that bear mortalin mimo- selective pressures since they pose as primordial armamentaria topes are highly prevalent in developing and agricultural re- against the microbial/parasitic invader [13]; second, a ‘‘double gions and are often associated with non-hygienic whammy’’ may seem to happen, whereby cross-reactive anti- environments and poor food manufacturing standards, like bodies target a rogue cell at the very early stage of carcinogen- Escherichia coli, Salmonella and Shigella. Cross-reactivity be- esis, as exemplified by mortalin, a cognate for cellular tween cancer cells, i.e., guinea pig line 10 hepatocarcinoma cell, immortalization. Thus, immunologic memory for the promis- and chronic infectious agents has been reported in Brucella cuous cross-reactive determinants of mortalin may be gener- abortus as early as 1976 [58]. Leptospirosis is worldwide in dis- ated within a population based on frequent re-stimulation by tribution, but is more prevalent in regions having warmer cli- subsequent onslaughts by these microbes. mates and frequent floods. Caveat, associations drawn based on the characteristics and epidemiology of the pathogens may be crude as cancer is a highly complex disease that brings 6. Possible rebuttal against the model forth interactions of individual genetic susceptibility and envi- ronmental factors. Moreover, we do not discount the possibil- In an independent way, infection may also cause modulation ity that cancer diagnostic methods and tumor registries may of the Th1 and Th2 balance and concurrent activation of non- not be properly in place and statistics may be underestimated. specific ‘‘innate’’ immune response that have suppressive ef- fects on tumor growth. These have been shown from an earlier work on anti-tumor effects of lipopolysaccharide preparations 5. The mimotope-hormesis model for mortalin from Shigella sonnei and Bordetella pertussis [59]. Truly, as components of the MHC II pathway, the Th1 response in- From an immunologic standpoint, in some clinically ad- duced by acute infections, e.g., from toxoplasma and Listeria vanced cases, anti-tumor antibodies that are detected come monocytogenes, but not the Th2 response from S. japonicum, from a circulating pool of low avidity T and B cells targeted has been demonstrated to inhibit implanted B16 tumors in against cryptic determinants. More often, such autoimmune mice [60]. Activation of with tumoricidal acitiv- reaction may not be sufficient for tumor regression to happen. ity, though not dependent on MHC recognition, can be C.C. Deocaris et al. / FEBS Letters 579 (2005) 586–590 589 achieved by a variety of infections from intracellular [4] Vijh, A.K. (2004) Observations on the proposed relationship microorganism, such as mycobacteria, Listeria and Toxo- between infection burden and early malignancy in developing plasma; endotoxins, soluble products from bacteria and by countries (e.g., India). Med. Hypotheses 62, 233–236. [5] Murray, C.J. and Lopez, A.D. (1997) Mortality by cause for eight the principal Th1 IFNc. Natural killer cells that can regions of the world: global burden of disease study. Lancet 349, lyse a number of tumor cells, are similarly activated by cyto- 1269–1276. kines IFNc, IL-2 and IL-2 during acute inflammatory pro- [6] Kato, H., Schull, W.J., Awa, A., Akiyama, M. and Otake, M. cesses [61]. However, such innate immunity, in contrast to T (1987) Dose-response analyses among atomic bomb survivors exposed to low-level radiation. Health Phys. 52, 645–652. and B cells, does not confer immunologic memory and may [7] Yakovlev, A., Tsodikov, A.D. and Bass, L. (1993) A stochastic not impact as much during the long carcinogenesis time frames. model of hormesis. Math. Biosci. 116, 197–219. Included among the candidate mimotopes was a hsp70 pep- [8] Kristensen, T.N., Sorensen, J.G. and Loeschcke, V. (2003) Mild tide from S. japonicum, a major disease agent in Africa, Egypt, heat stress at a young age in Drosophila melanogaster leads to the southern parts of Europe, and Japan. Paradoxically, to- increased Hsp70 synthesis after stress exposure later in life. J. Genet. 82, 89–94. gether with Opisthorchis and Clonorchis, it belongs to a major [9] Verbeke, P., Clark, B.F. and Rattan, S.I. (2000) Modulating genera of parasites that are known or suspected cancer risk cellular aging in vitro: hormetic effects of repeated mild heat stress factors. Oncogenic characteristics of schistosoma operative on protein oxidation and glycation. Exp. Gerontol. 35, 787–794. during bladder carcinogenesis have not been completely eluci- [10] Hercus, M.J., Loeschcke, V. and Rattan, S.I. (2003) Lifespan extension of Drosophila melanogaster through hormesis by dated. It is believed that the parasite involves the production of repeated mild heat stress. Biogerontol. 4, 149–156. excessive reactive oxygen species via inflammation, elevated [11] Rattan, S.I. (2004) Mechanisms of hormesis through mild heat bladder cell proliferation, genetic instability, and interaction stress on human cells. Ann. NY Acad. Sci. 1019, 554–558. with other carcinogens and co-infection with viral agents [12] Luckey, T.D. (1999) Nurture with ionizing radiation: a provoc- [62]. Recently, schistosomal infection has also been shown to ative hypothesis. Nutr. Cancer 34, 1–11. [13] Sioud, M. (2002) How does autoimmunity cause tumor regres- influence DNA methylation in key cancer genes [63]. In the sion. A potential mechanism involving cross-reaction through parlance of hormesis, Schistosoma, along with other known epitope mimicry. Mol. Med. 8, 115–119. carcinogenic infectious agents, EBV, HPV, H. pylori among [14] Xu, L., Jin, B.Q. and Fan, D.M. (2003) Selection and identifica- others, cannot be categorized as ‘‘mild stress’’. Long-term hor- tion of mimic epitopes for gastric cancer-associated antigen MG7 Ag. Mol. Cancer Ther. 2, 301–306. metic effects of any form of mimotope-associated tumor auto- [15] Linnemann, T., Tumenjargal, S., Gellrich, S., Wiesmuller, K., immunity, however, may exert its benefits in all but the types Kaltoft, K., Sterry, W. and Walden, P. (2001) Mimotopes for of tumors that are known to be targets of these microbes as tumor-specific T lymphocytes in human cancer determined a direct result of their pathogenicity. with combinatorial peptide libraries. Eur. J. Immunol. 31, 156– 165. [16] Tumenjargal, S., Gellrich, S., Linnemann, T., Muche, J.M., Lukowsky, A., Audring, H., Wiesmuller, K.H., Sterry, W. and 7. Concluding remarks Walden, P. (2003) Anti-tumor immune responses and tumor regression induced with mimotopes of a tumor-associated T cell Emerging evidence supports the role of hsps and their epitope epitope. Eur. J. Immunol. 33, 3175–3185. [17] Riemer, A.B., Klinger, M., Wagner, S., Bernhaus, A., Mazzucch- mimics (mimotopes) in priming immune responses, both in elli, L., Pehamberger, H., Scheiner, O., Zielinski, C.C. and Jensen- improving host defenses and in auto-aggression. Along such Jarolim, E. (2004) Generation of Peptide mimics of the epitope line, this article proposes novel attributes of mortalin/GRP75/ recognized by trastuzumab on the oncogenic protein Her-2/neu. J. mthsp70 as an immunomodulator expounding beyond its Immunol. 173, 394–401. known chaperone functions and proposes it as a candidate can- [18] Sherev, T., Wiesmuller, K.H. and Walden, P. (2003) Mimotopes of tumor-associated T-cell epitopes for cancer vaccines deter- cer vaccine. Supported by the data on epitopes shared by mort- mined with combinatorial peptide libraries. Mol. Biotechnol. 25, alin (enriched in immortal and cancerous cells), and hsp70 53–61. cognates in human pathogens prevalent in developing countries, [19] Hansen, M.H., Ostenstad, B. and Sioud, M. (2001) Identification we put forward a hypothesis that the phenomenon of epitope of immunogenic antigens using a phage-displayed cDNA library from an invasive ductal breast carcinoma tumour. Int. J. Oncol. mimicry may play a role in regional cancer prevention and ex- 19, 1303–1309. plain the inverse trend in cancer and infection in developing [20] Kaufmann, S.H. (1991) Heat shock proteins and autoimmunity: countries. In this way, the ‘‘mimotope hormesis hypothesis’’ facts or fiction? Curr. Biol. 1, 359–361. provides a molecular Ôglass-half-fullÕ view that perhaps havoc [21] Millar, D.G., Garza, K.M., Odermatt, B., Elford, A.R., Ono, N., wrought by these infections may, in a way, spare populations Li, Z. and Ohashi, P.S. (2003) Hsp70 promotes antigen-presenting cell function and converts T-cell tolerance to autoimmunity in of miseries of cancer in the poorer regions of the world. vivo. Nat. Med. 9, 1469–1476. [22] Palo, J., Duchesne, J. and Wilkstrom, J. (1977) Malignant diseases among patients with multiple sclerosis. J. Neurol. 216, References 217–222. [23] Clark, P.R. and Menoret, A. (2001) The inducible Hsp70 as a [1] Kuper, H., Adami, H.O. and Trichopoulos, D. (2000) Infections marker of tumor immunogenicity. Cell Stress Chaperones 6, 121– as a major preventable cause of human cancer. J. Intern. Med. 125. 248, 171–183. [24] Wang, X.Y., Kaneko, Y., Repasky, E. and Subjeck, J.R. (2000) [2] Ye, W., Held, M., Lagergren, J., Engstrand, L., Blot, W.J., Heat shock proteins and cancer immunotherapy. Immunol. McLaughlin, J.K. and Nyren, O. (2004) Helicobacter pylori Invest. 29, 131–137. infection and gastric atrophy: risk of adenocarcinoma and [25] Todryk, S.M., Gough, M.J. and Pockley, A.G. (2003) Facets of squamous-cell carcinoma of the esophagus and adenocarcinoma heat shock protein 70 show immunotherapeutic potential. Immu- of the gastric cardia. J. Natl. Cancer Inst. 96, 388–396. nologist 110, 1–9. [3] Vijh, A.K. (2004) High infectious burden, low cancer incidence, and [26] Massa, C., Guiducci, C., Arioli, I., Parenza, M., Colombo, M.P. early malignancy in developing countries: a molecular hypothesis in and Melani, C. (2004) Enhanced efficacy of tumor cell vaccines term of the role of nitric oxide. Med. Hypotheses 63, 208–210. transfected with secretable hsp70. Cancer Res. 64, 1502–1508. 590 C.C. Deocaris et al. / FEBS Letters 579 (2005) 586–590

[27] Pockley, A.G. (2003) Heat shock proteins as regulators of the [46] Sadekova, S., Lehnert, S. and Chow, T.Y. (1997) Induction of immune response. Lancet 362, 469–476. PBP74/mortalin/Grp75, a member of the hsp70 family, by low [28] Weigl, E., Kopecek, P., Raska, M. and Hradilova, S. (1999) Heat doses of ionizing radiation: a possible role in induced radioresis- shock proteins in immune reactions. Folia Microbiol. 44, 561– tance. Int. J. Radiat. Biol. 72, 653–660. 566. [47] Dahlseid, J.N., Lill, R., Green, J.M., Xu, X., Qiu, Y. and Pierce, [29] Smiley, J.D. and Hoffman, W.L. (1991) The role of infections in S.K. (1994) PBP74, a new member of the mammalian 70-kDa heat the rheumatic diseases: molecular mimicry between bacterial and shock protein family, is a mitochondrial protein. Mol. Biol. Cell 5, human stress proteins? Am. J. Med. Sci. 301, 138–149. 1265–1275. [30] Benoist, C. and Mathis, D. (2001) Autoimmunity provoked by [48] Wadhwa, R., Pereira-Smith, O.M., Reddel, R.R., Sugimoto, Y., infection: how good is the case for T cell epitope mimicry? Nat. Mitsui, Y. and Kaul, S.C. (1995) Correlation between comple- Immunol. 2, 797–801. mentation group for immortality and the cellular distribution of [31] Karopoulos, C., Rowley, M.J., Handley, C.J. and Strugnell, R.A. mortalin. Exp. Cell Res. 216, 101–106. (1995) Antibody reactivity to mycobacterial 65 kDa heat shock [49] Ran, Q., Wadhwa, R., Kawai, R., Kaul, S.C., Sifers, R.N., Bick, protein: relevance to autoimmunity. J. Autoimmun. 8, 235–248. R.J., Smith, J.R. and Pereira-Smith, O.M. (2000) Extramitochon- [32] Zhu, J., Katz, R.J., Quyyumi, A.A., Canos, D.A., Rott, D., drial localization of mortalin/mthsp70/PBP74/GRP75. Biochem. Csako, G., Zalles-Ganley, A., Ogunmakinwa, J., Wasserman, Biophys. Res. Commun. 275, 174–179. A.G. and Epstein, S.E. (2004) Association of serum antibodies to [50] Shin, B.K., Wang, H., Yim, A.M., Le Naour, F., Brichory, F., heat-shock protein 65 with coronary calcification levels: sugges- Jang, J.H., Zhao, R., Puravs, E., Tra, J., Michael, C.W., Misek, tion of pathogen-triggered autoimmunity in early atherosclerosis. D.E. and Hanash, S.M. (2003) Global profiling of the cell surface Circulation 109, 36–41. proteome of cancer cells uncovers an abundance of proteins with [33] Rambukkana, A., Das, P.K., Witkamp, L., Yong, S., Meinardi, chaperone function. J. Biol. Chem. 278, 7607–7616. M.M. and Bos, J.D. (1993) Antibodies to mycobacterial 65-kDa [51] Bian, H., Reidhaar-Olson, J.F. and Hammer, J. (2003) The use of heat shock protein and other immunodominant antigens in bioinformatics for identifying class II-restricted T-cell epitopes. patients with psoriasis. J. Invest. Dermatol. 100, 87–92. Methods 29, 299–309. [34] Jensen, P., Johansen, H.K., Carmi, P., Hoiby, N. and Cohen, I.R. [52] Sturniolo, T., Bono, E., Ding, J., Raddrizzani, L., Tuereci, O., (2001) Autoantibodies to pancreatic hsp60 precede the develop- Sahin, U., Braxenthaler, M., Gallazzi, F., Protti, M.P., Sinigaglia, ment of glucose intolerance in patients with cystic fibrosis. J. F. and Hammer, J. (1999) Generation of tissue-specific and Autoimmun. 17, 165–172. promiscuous HLA ligand databases using DNA microarrays and [35] Yokota, S., Tsubaki, K., Kuriyama, T., Shimizu, H., Ibe, M., virtual HLA class II matrices. Nat. Biotechnol. 17, 555–561. Mitsuda, T., Aihara, Y., Kosuge, K. and Nomaguchi, H. (1993) [53] Cochlovius, B., Stassar, M., Christ, O., Raddrizzani, L., Hammer, Presence in Kawasaki disease of antibodies to mycobacterial heat- J., Mytilineos, I. and Zoller, M. (2000) In vitro and in vivo shock protein HSP65 and autoantibodies to epitopes of human induction of a Th cell response toward peptides of the melanoma- HSP65 cognate antigen. Clin. Immunol. Immunopathol. 67, 163– associated glycoprotein 100 protein selected by the TEPITOPE 170. program. J. Immunol. 165, 4731–4741. [36] Lehner, T. (1997) The role of heat shock protein, microbial and [54] Consogno, G., Manici, S., Facchinetti, V., Bachi, A., Hammer, J., autoimmune agents in the aetiology of BehcetÕs disease. Int. Rev. Conti-Fine, B.M., Rugarli, C., Traversari, C. and Protti, M.P. Immunol. 14, 21–32. (2003) Identification of immunodominant regions among promis- [37] Raz, I., Elias, D., Avron, A., Tamir, M., Metzger, M. and Cohen, cuous HLA-DR-restricted CD4+ T-cell epitopes on the tumor I.R. (2001) Beta-cell function in new-onset type 1 diabetes and antigen MAGE-3. Blood 101, 1038–1044. immunomodulation with a heat-shock protein peptide (Dia- [55] Meyer, A.L., Trollmo, C., Crawford, F., Marrack, P., Steere, Pep277): a randomised, double-blind, phase II trial. Lancet 358, A.C., Huber, B.T., Kappler, J. and Hafler, D.A. (2000) Direct 1749–1753. enumeration of Borrelia-reactive CD4 T cells ex vivo by using [38] Salvetti, M., Buttinelli, C., Ristori, G., Carbonari, M., Cherchi, MHC class II tetramers. Proc. Natl. Acad. Sci. USA 97, 11433– M., Fiorelli, M., Grasso, M.G., Toma, L. and Pozzilli, C. (1992) 11438. T- reactivity to the recombinant mycobacterial 65- [56] Kwok, W.W., Gebe, J.A., Liu, A., Agar, S., Ptacek, N., Hammer, and 70-kDa heat shock proteins in multiple sclerosis. J. Autoim- J., Koelle, D.M. and Nepom, G.T. (2001) Rapid epitope mun. 5, 691–702. identification from complex class-II-restricted T-cell antigens. [39] Wendling, U., Paul, L., van der Zee, R., Prakken, B., Singh, M. Trend Immunol. 22, 583–588. and van Eden, W. (2000) A conserved mycobacterial heat shock [57] Murray, P.R., Rosenthal, K.S., Kobayashi, G.S. and Pfaller, protein (hsp) 70 sequence prevents adjuvant arthritis upon nasal M.A. (1998) Medical Microbiology, 3rd ed, Mosby-Year Book, administration and induces IL-10-producing T cells that cross Inc., St. Louis, MO. react with the mammalian self-hsp70 homologue. J. Immunol. [58] Minden, P., Sharpton, T.R. and McClatchy, J.K. (1976) Shared 164, 2711–2717. antigens between human malignant melanoma cells and Myco- [40] Zugel, U. and Kaufmann, S.H.E. (1999) Role of heat shock bacterium bovis (BCG). J. Immunol. 116, 1407–1414. proteins in protection from pathogenesis of infectious diseases. [59] Shevliagin, V., Snegireva, A.E., Shaposhnikova, G.M., Lapaeva, Clin. Microbiol. Rev. 12, 19–39. I.A. and Petrukhin, V.G. (1989) The effect of gram-negative [41] Mintz, P.J., Kim, J., Do, K.A., Wang, X., Zinner, R.G., bacteria lipopolysaccharides on the development of Rauscher Cristofanilli, M., Arap, M.A., Hong, W.K., Troncoso, P., leukosis in BALB/c mice. Zh. Mikrobiol. Epidemiol. Immunobiol. Logothetis, C.J., Pasqualini, R. and Arap, W. (2003) Fingerprint- 4, 14–17. ing the circulating repertoire of antibodies from cancer patients. [60] Rankin, E.B., Yu, D., Jiang, J., Shen, H., Pearce, E.J., Golds- Nat. Biotechnol. 21, 57–63. chmidt, M.H., Levy, D.E., Golovkina, T.V., Hunter, C.A. and [42] Wadhwa, R., Kaul, S.C., Ikawa, Y. and Sugimoto, Y. (1993) Thomas-Tikhonenko, A. (2003) An essential role of Th1 Identification of a novel member of mouse hsp70 family. Its responses and interferon gamma in infection-mediated suppres- association with cellular mortal phenotype. J. Biol. Chem. 268, sion of neoplastic growth. Cancer Biol. Ther. 2, 687–693. 6615–6621. [61] Kohchi, C., Inagawa, H., Hino, M., Oda, M., Nakata, K., [43] Wadhwa, R., Taira, K. and Kaul, S.C. (2002) An Hsp70 family Yoshida, A., Hori, H., Terada, H., Makino, K., Takiguchi, K. chaperone, mortalin/mthsp70/PBP74/Grp75: what, when, and and Soma, G. (2004) Utilization of macrophages in anticancer where? Cell Stress Chaperones 7, 309–316. therapy: the network theory. Anticancer Res. 24, [44] Kaul, S.C., Taira, K., Pereira-Smith, O.M. and Wadhwa, R. 3311–3320. (2002) Mortalin: present and prospective. Exp. Gerontol. 37, [62] Abdel-Rahim, A.Y. (2001) Parasitic infections and hepatic 1157–1164. neoplasia. Dig. Dis. 19, 288–291. [45] Domanico, S.Z., DeNagel, D.C., Dahlseid, J.N., Green, J.M. and [63] Gutierrez, M.I., Siraj, A.K., Khaled, H., Koon, N., El-Rifai, W. Pierce, S.K. (1993) Cloning of the gene encoding peptide-binding and Bhatia, K. (2004) CpG island methylation in Schistosoma- protein 74 shows that it is a new member of the heat shock protein and non-Schistosoma-associated bladder cancer. Mod. Pathol. 17, 70 family. Mol. Cell. Biol. 13, 3598–3610. 1268–1274.