Heat Shock Proteins as Mediators of Aggregation-Induced "Danger" Signals: Implications of the Slow Evolutionary Fine-Tuning of Sequences for the Antigenicity of Cancer Cells

By Donald R. Forsdyke

An address given at the First International Conference on Heat-Shock Proteins in the Immune Response, Farmington, CT. October 1998.

Published as a "Perspective" in Cell Stress & Chaperones (1999) 4, 205-210    Copyright permission from Harcourt Brace & Co.

Fine tuning on an evolutionary time-scale
The crowded cytosol
Intracellular self/not-self discrimination
Non-specific distraction
Distractions of RNAs and proteins
Qualitative fine-tuning of molecules
Individual-specific destruction of tumor cells
Implications for therapy

End Note (August 2008)





Video of the 1998 conference presentation of this paper, with debate between Srivastava and Forsdyke (2015 You Tube version; 53 min)  (Vimeo version here)


    Organisms "tune" to their environment through adaptations which confer a selective advantage. However, in complex systems, a primary change of positive adaptive value might have multiple minor secondary effects, usually of negative adaptive value, which could invoke further counter-adaptations. This "fine-tuning", a "debugging", mainly at the intracellular level, would appear an evolutionary burden detracting from the positive nature of the primary change.

    However, if the primary mutation is in a potential oncogene, secondary, short-term, effects may include the recruitment, in an apparently random manner, of unmutated non-oncogene products into the antigenic repertoire of the cancer cell. This "danger" signal, provided by the co-aggregation of oncogene and non-oncogene products, would be mediated by inducible heat-shock proteins (Hsps), and lead to display of corresponding MHC-peptide complexes.

    It was argued previously that T cells specific for peptides from most "self" intracellular antigens are not eliminated during T cell "education", and so would be available for subsequent immune activation by the corresponding peptides. These considerations might explain:

  • why cancer specific antigens have been so elusive,
  • why cancer antigenicity is often individual specific, and
  • why therapeutic approaches involving complexes of peptides with Hsps may be successful.


    The antigens recognized by cytotoxic T cells are often neither cancer-specific, nor cancer type-specific. Rather, they are specific to the individual with cancer. Antigens specific for cancers in general, or for particular cancer types, are hard to find and do not appear to serve as useful primary targets of immune attack (Srivastava et al 1998). Thus, a cancer attributed to a mutation in oncogene A (A -> A') in one individual may be attacked immunologically with respect to a particular set of antigens (corresponding to genes B, C, D). In another individual the same oncogene may mutate (A -> A''), yet the immune response will be towards another set of antigens (corresponding to genes E, F, G). In a third individual the mutation (A -> A''') may be associated with yet another set of antigens (corresponding to genes H, I, J).

    This apparently random recruitment of non-oncogenes has led to the proposal that the cancer phenotype predisposes to mutation genes which are not likely to be related functionally to the genes whose primary mutation results in cancer. These non-oncogenes, by virtue of the mutations they contain, become marked as "non-self", and hence are recognized by an immune system considered to have been "educated" to recognize "not-self", but not "self" (Srivastava 1993, 1996). However, if random in time, conferring no proliferative advantage, and occurring in different cells, it is hard to see how a number of independent mutations in non-oncogenes, in addition to the primary oncogenic mutation A' (i.e. A' + B', A' + C', A' + D'), will be collected together in one clonal line (i.e. A' + B' + C' + D'; see APPENDIX). Furthermore, it appears that the genes associated in this way with A' are, in fact, usually not mutated (Srivastava 1996; Srivastava et al 1998).

    The requirement that genes must mutate to "not-self" in order to be able to trigger immune defences would not be necessary if:

Thus, instead of describing a potentially antigenic phenotype as A' + B' + C' + D', we would write A' + B + C + D.

   To understand how this might come about, we will here consider previous arguments against the widely-held assumption that self/not-self discrimination is strictly an extracellular phenomenon. It is shown that once this assumption, although currently resting on little direct evidence, is made, the fact of individual-specific cancer immunity becomes intelligible, and recently proposed therapeutic approaches appear more plausible.

Fine tuning on an evolutionary time-scale

   Most mutations are deleterious. Natural species appear so well adapted to their contemporary environments that change is generally for the worse, and is selected against. So well-adapted are many natural plant species that we designate them as "weeds". They grow rapidly and form seed even in nutritionally unfavourable conditions. On the other hand, domestic plants, relative late-comers on the evolutionary scene, usually grow more slowly, even in nutritionally favourable conditions. Such observations suggest that speciation is but the beginning of a much slower process of adaptive fine-tuning. Following initial speciation events, a multiplicity of secondary accommodations might then be necessary. Even if the environment were constant, in biological evolution the asymptotic approach to perfect adaptation might be of long duration.

    This is something quite familiar to writers of computer software. A small change in one part of a program may have unforeseen consequences for other parts of the program. The laborious "debugging" of programs largely involves the hunting down, correction, or accommodation to, such secondary effects. Thus, Hunter (1993) noted:

"Small steps in genotype space can have large consequences in phenotype space. ... As computer programmers well know, small changes in a complex system often lead to far-reaching and destructive consequences (and computer programmers make these small changes by design, and with the hope of improving the code!).

    Just as perfect adaptation may be approached very slowly, so may perfect deadaptation. Darwin (1859) was much puzzled by apparently non-functional organs (e.g. nipples in males) which are designated as "vestigial" because they resemble an organ which we see functional in some members of the same species (e.g. females), or in other species. The organ is often smaller than expected if it were functional. It appears to play no role in the economy of the organism. There seems to be no selection pressure for its retention, and it is on a path towards disappearance, but its progress along that path is very slow.

The crowded cytosol

    We consider here evolutionary fine-tuning at the level of intracellular molecules. The processes within a cell which give it the attribute "live" are based on interactions between molecules (proteins, nucleic acids, and other molecular species, both large and small). For example, interactions of specific tRNA molecules with specific proteins, are required to load the tRNAs with appropriate amino acids for protein synthesis. Subsequently, interactions occur between tRNA anticodon loops and specific codons on mRNAs. These interactions are relatively weak (non-covalent), as opposed to the strong (covalent) interactions between the atoms which unite to form molecules. From consideration of the volume available and the large number and variety of molecules dissolved in the cytosol, it is apparent that the cytosol is a very crowded place (Fulton 1982). This has important functional implications, both quantitative and qualitative, for non-covalent interaction between molecules; in particular, interactions which significantly increase system entropy may be promoted as temperature increases (Lauffer 1975; Reich et al 1995).

Intracellular self/not-self discrimination

    Since cytotoxic T cells play a major role in the elimination of cancer cells, for present purposes we will deal with just this aspect of the immune response. Immunologically competent cells are generated through processes of positive and negative selection (which generate expanded clones of specificity for "near self" or "altered self"), and MHC restriction (Burnet 1959; Lederberg, 1959; Zinkernagel and Doherty 1974; Forsdyke 1973a,b; 1975). T cells recognize MHC proteinslympho01.gif (729 bytes) complexed with peptide fragments from intracellular proteins. Negative selection is widely perceived as necessary to eliminate, in advance, T cells with the potential to react with MHC proteins complexed with peptides derived from "self" proteins.

    A major problem is that this negative selection might so deplete the T cell repertoire that recognition of MHC complexes with "not-self" (foreign) peptides might be impaired (Mitchison 1981;Vidovik and Matzinger 1988; Forsdyke 1995a). If it were possible for cells to evolve mechanisms for distinguishing self from not-self proteins intracellularly, in a non-immunological manner, prior to complexing their peptides with MHC proteins, this would:

This would preserve the repertoire of T cells able to react both against peptides derived from foreign (e.g. virus-derived) proteins, and against peptides derived from normal proteins (should a circumstance arise such that it would be advantageous for cells to display some of these peptides).

    A quantitative implication of the crowded state of the cytosol is that proteins close to the limit of their solubility tend to aggregate (Lauffer 1975; Jarrett and Lansbury 1993). This is frequently seen when proteins are artificially expressed in a foreign cell. The proteins aggregate as highly insoluble "inclusion bodies", a process which is decreased by growth at low temperatures, suggesting a significant entropy-driven component (Schein 1989). Similar intracellular bodies are seen in various pathological states (Russell 1890; Ordway et al 1997). It is shown elsewhere how this provides a possible basis for intracellular self/not-self discrimination, which would be promoted by systemic pyrexia (Forsdyke 1985, 1991, 1992, 1994a,b, 1995a,b).

To be successful an intracellular pathogen must restrain (fine-tune) the rate of production of pathogen-specific proteins to avoid inadvertent aggregation of these proteins, which would provide an alarm triggering their selective display as peptides with MHC protein at the cell surface. On the other hand, genes encoding self-proteins have been fine-tuned over evolutionary time to their companion genes, so that cytosolic concentrations of their products do not exceed thresholds above which aggregation and MHC display would occur.

    Any tendency to denature and aggregate would normally be prevented by specialized, constitutively synthesized, molecular chaperones, which include Hsps. Under normal conditions the latter would even appear able to mask minor mutations which might impair structure and hence function (Rutherford and Lindquist 1998). However, when confronted with an intracellular pathogen this function becomes a liability. The cell enters "alarm" or "danger" mode (Matzinger 1994; Stark et al 1998) and decreases the concentration of normal proteins (thus decreasing their auto-aggregatability), while increasing the concentration, and changing the quality, of Hsps (Menoret et al 1995). The primary role of Hsps is now to facilitate degradation of aggregated pathogen proteins and the complexing of their peptides with MHC proteins for display at the cell surface (Nagasawa et al 1992; Li and Srivastava 1993; Srivastava et al 1994). Sometimes, mode of operation, but not quality, changes. Thus, the widely conserved Hsp DepG (HtrA) switches mode from chaperone to proteolytic at high temperatures (Spiess et al 1999).

Non-specific distraction

    The crowded state of the cytosol also has qualitative implications, which are most pertinent to the present issue. A metaphor may be helpful. Imagine you are seeking someone in a crowded place.

crowd2.gif (25481 bytes)

You scan those around you, then push your way to a new part of the crowd, and scan again. From time to time you may note people who look a bit like the person you are seeking. You look again to check, and then continue your search. Every person who looks a bit like the person you are seeking is a distractor who marginally slows down your rate of search. In future, to speed up the search process it would be nice if someone would line up every one in the room and eliminate in advance the potential distractors. In the absence of this fine-tuning of the crowd, your search is less efficient.

    Although electron-micrographs show that the crowded cytosol is structured, in practice the molecules are relatively disordered and seek order (the interaction of one molecule with another) through a search which requires stereospecific (lock and key-type) recognition. (Paradoxically, there may be an increase of system disorder due to the release of structured water as part of the interaction; Lauffer 1975). The price of attempting this in a crowded environment must be a large number of chance interactions of varying degrees of specificity with many other molecules. Each of these interactions is non-productive in terms of the proper functioning of the cell, and might cause a slowing of the overall search process and a decrease in the efficiency of cell function.

Distractions of RNAs and proteins

    That the cytosol is not effectively structured to prevent incidental interactions is evident from the ease with which it is possible to interfere experimentally with cellular processes by injecting or expressing molecules in cells. An "antisense" RNA introduced into a cell can interact with the corresponding complementary "sense" RNA and block its function (Izant and Weintraub 1984). Thus mRNAs are available for mRNA-mRNA interactions (Melton, 1985), as well as for the normal mRNA-tRNA interactions. These:

Recent studies suggest that an RNA molecule moving from one part of a cell to another does indeed have to cope with a multiplicity of distracting interactions with other RNAs. By "purine-loading" their loops, RNAs appear to avoid such interactions (Cristillo et al 1998; Bell et al 1998; Bell and Forsdyke 1999a,b).

    One way protein molecules might evade distracting interactions is by possession of a similar charge. Molecules with the same charge would tend to repel each other, so that if a majority of molecules were similarly charged, there might be a selection pressure favouring retention of this charge and a decrease in the number of oppositely charged molecules. Consistent with this, most cytosolic proteins are negatively charged, and it is found that the evolutionary rate of change in charge is much less than predicted from the observed rate of accepted amino acid mutations (McConkey 1982).

Qualitative fine-tuning of molecules

    Consider a situation where organisms in which molecule A interacts efficiently with molecule B are much better adapted to their environment than are other members of the species where A and B interact less efficiently. Having arranged the structure of molecule A such that it can better react with molecule B, the "hand" of evolution would have achieved a significant advance. Such evolutionary "tuning" would occur quite rapidly because of the distinct selective advantage it would confer. However, one would expect that concomitant with and following this, there would be an on-going process of adaptation of other molecules of the cytosol ("fine-tuning").

    If a molecule C, which had no functional interest in A, became, by virtue of the change in A, slightly reactive with A, then a mutation in C which would not interfere with the normal functioning of C, but would decrease its reactivity for A, might be marginally advantageous. It might take thousands of generations for many such marginally advantageous mutations to recombine together to create a genome conferring a significant increase in the efficiency of cell function (resulting in a distinct selective advantage).   

     By the same line of reasoning, it can be seen that a deleterious mutation in B (B -> B'), decreasing its attraction to A, might at the same time confer on it various degrees of attractiveness for molecules D, E and F. This might confer no functional advantage, and might even impair, the functioning of D, E and F. On the other hand, there is a circumstance where reaction with D, E and F might actually be turned to the advantage of the individual organism. B might be a potential oncogene and the deleterious mutation (B') resulting in decreased affinity for A might have "thrown" one of the critical switches leading to cancer. In this circumstance, concomitant B'-dependent changes in D, E and F might increase the probability that the primary change in B would be detected.

Individual-specific destruction of tumor cells

    For present purposes we consider normal cells which have become potentially tumorogenic because of a mutation in a single oncogene. (In practice, a number of such switches involving different oncogenes might need to be thrown in order for cells to manifest the full tumor phenotype. Each of these oncogenes might recruit distinct sets of normal genes into the antigenic repertoire, as discussed below for the individual case of oncogene B'.)

    Cytotoxic T lymphocytes should destroy cells which have become cancer cells because of a mutation in a gene B (B -> B') which might positively affecting cell proliferation, or negatively affecting cell destruction (i.e. it is a potential oncogene). The mutation might affect protein B so as to mark it as potentially "not-self" (like a virus protein). If reactions of the mutated protein B' with unmutated proteins D, E and F, resulted in their also being marked as "non-self", then cytotoxic T cells might be able to target peptides from B', D, E and F, rather than just from B'. The potential T cell response would be amplified.

    This recruitment of D, E and F would depend on the fact that the primary mutation (B -> B') had occurred in somatic cells and, within the life-span on the individual, it would not be possible to fine-tune the cytosol for non-reactivity of cytosolic proteins with B' (as could occur over evolutionary time if a mutation were of positive or negative adaptive value and had affected germ line cells).

    The mutation (B') would be non-adaptive, since it would impair homeostatic mechanisms maintaining the size of the cell population concerned. This detrimental effect would be likely to impact the individual organism with the mutation, decreasing its lifespan, and hence impairing, directly or indirectly, its chances of successful reproduction. Thus, there would be a selection pressure supporting the evolution of mechanisms for the elimination of the original tumor cell and its progeny. Cancer would become largely a disease of old age, when reproductive success is less at stake.

    The concept of each gene fine-tuning the concentration of its cytosolic product close to the solubility limit imposed by the concentrations of the other proteins with which it has been travelling over evolutionary time, lends itself well to this problem (Forsdyke 1995a,b). Normally protein D, for example, would not react with protein B. When protein B mutates to B' thus conferring a potential cancer phenotype, B' might happen to acquire sufficient affinity for D as to form heteroaggregates. Thus D, an unmutated "self" protein, would be inadvertently labelled as "not-self", and processed to peptides. These would be displayed at the cell surface complexed with MHC proteins. By virtue of the fact that T cells with the potential to react against D-derived self peptides would not have been eliminated in advance, they would assist the elimination of cells containing B'.

Implications for therapy

   In this light, the prospect for general tumor-specific immunotherapies does not appear good. Rather, the harnessing of immunological mechanisms for defence against tumors becomes a highly individualized process. Fortunately, in the course of their preparation for MHC display it is likely that peptides complex transiently with various carrier proteins, including Hsps. There is encouraging evidence that immunization with autogenous peptide-Hsp complexes will elicit the individual-specific immunity against unmutated self antigens which is necessary in order to secure tumor elimination by cytotoxic T cells (Tamura et al 1997; Srivastava et al 1998). Consideration of likely T cell receptor affinity/avidity (Houghton 1994), suggests that this might require in vitro expansion of autologous T cell clones by appropriate concentrations of peptide-Hsp complexes, and then infusion of such custom-activated clones into the patient donor (D. R. Forsdyke, unpublished NCI grant application in Queen's University Archives, 1977; Cheever et al 1977; Zeh et al 1999).

colorbar.gif (2362 bytes)


Bell, SJ, Chow, YC, Ho, JYK and Forsdyke, DR (1998) Correlation of Chi orientation with transcription indicates a fundamental relationship between recombination and transcription. Gene 216, 285-292.

Bell, SJ and Forsdyke, DR (1999a) Accounting units in DNA. J. Theor. Biol. 197, 51-61.

Bell, SJ and Forsdyke, DR (1999b) Deviations from Chargaff's second rule correlate with direction of transcription. J. Theor. Biol. 197, 63-76.

Burnet, FM (1959) The Clonal Selection Theory of Acquired Immunity. Cambridge University Press, Cambridge.

Cantor, CR and Schimmel, PR (1980) Biophysical Chemistry. Freeman, W. H., San Francisco, 1183-1264.

Cheever, M., Kempf, RA and Fefer, A (1977) Tumor neutralization, immunotherapy, and chemoimmunotherapy of a Friend leukaemia with cells secondarily sensitized in vitro. J. Immunol. 119, 714-719.

Cristillo, AD, Lillicrap, TP and Forsdyke, DR (1998) Purine-loading of EBNA-1 mRNA avoids sense-antisense "collisions". FASEB. J. 12, A1453. Abstract # 828.

Darwin, CR (1859) The Origin of Species by Means of Natural Selection or the Preservation of Favoured Races in the Struggle for Life. John Murray, London.

Eguchi, Y, Itoh, T. and Tomizawa, J (1991). Antisense RNA. Annu. Rev. Biochem. 60, 631-652.

Forsdyke, DR (1973a) Serum factors affecting the incorporation of [3H]thymidine by lymphocytes stimulated by antigen. I. Serum concentration. Immunology 25, 583-596.

Forsdyke, DR (1973b) Serum factors affecting the incorporation of [3H]thymidine by lymphocytes stimulated by antigen. II. Evidence for a role of complement from studies with heated serum. Immunology 25, 597-612.

Forsdyke, DR (1975) Further implications of a theory of immunity. J. Theor. Biol. 52, 187-198.

Forsdyke DR (1985) Heat shock proteins defend against intracellular pathogens: a non- immunological basis for self/not-self discrimination. J. Theor. Biol. 115, 471-473.

Forsdyke DR (1991) Early evolution of MHC polymorphism. J. Theor. Biol. 150, 451-456.

Forsdyke DR (1992) Two signal model of self/not-self immune discrimination: an update. J. Theor. Biol. 154, 109-118.

Forsdyke DR (1994a) The heat-shock response and the molecular basis of genetic dominance. J. Theor. Biol. 167, 1-5.

Forsdyke DR (1994b) Relationship of X chromosome dosage compensation to intracellular self/not-self discrimination: a resolution of Muller's paradox. J. Theor. Biol. 167, 7-12.

Forsdyke, DR (1995a) Entropy-driven protein self-aggregation as the basis for self/not- self discrimination in the crowded cytosol. J. Biol. Sys. 3, 273-287.

Forsdyke, DR (1995b) Fine tuning of intracellular protein concentrations, a collective protein function involved in aneuploid lethality, sex-determination and speciation? J. Theor. Biol. 172, 335-345.

Fulton, AB (1982) How crowded is the cytoplasm? Cell 30, 345-347.

Houghton, AN (1994) Cancer antigens: immune recognition of self and altered self. J. Exp. Med. 180, 1-4.

Hunter, L (1993) Molecular biology for computer scientists. In Artificial Intelligence and Molecular Biology, pp. 1-46. Edited by L. Hunter. MIT Press, Boston.

Izant, JG and Weintraub, H (1984). Inhibition of thymidine kinase gene expression by anti-sense RNA: a molecular approach to genetic analysis. Cell 36, 1007-1015.

Jarrett, JT and Lansbury, PJ (1993) Seeding "one-dimensional crystallization" of amyloid: a pathogenic mechanism in Alzheimer's disease and amyloid? Cell 73, 1055-1058.

Lauffer, MA (1975) Entropy-driven Processes in Biology. Springer-Verlag, New York, NY.

Lederberg, J (1959) Do antigens bear instructions for antibody specificity, or do they select cell lines that arise by mutation? Science 129, 1649-1653.

Li, Z and Srivastava PK (1993) Tumor rejection antigen Gp96/Grp94 is an ATPase: implications for protein folding and antigen presentation, EMBO. J. 12, 3143-3151.

Matzinger, P (1994) Tolerance, danger, and the extended family. Annu. Rev. Immunol. 12, 991-1045.

McConkey, EH (1982) Molecular evolution, intracellular organization, and the quinary structure of proteins. Proc. Natl. Acad. Sci. USA 79, 3236-3240.

Melton, DA (1985) Injected anti-sense RNAs specifically block messenger RNA translation in vivo. Proc. Natl. Acad. Sci. USA 82, 144-148.

Menoret, A, Patry, Y, Burg, C, and Le Pendu, J (1995) Co-segregation of tumor immunogenicity with expression of inducible but not constitutive hsp70 in rat colon carcinomas. J. Immunol. 155, 740-747.

Mitchison, NA (1981) Information transfer between the minor antigen and T cell receptor repertoires. Scand. J. Immunol. 14, 631-635.

Nagasawa, H, Oka, M, Maeda, K, Jian-Cuo, C, Hisaeda, H, Ito, Y, Good, RA and Himeno, K (1992) Induction of heat shock proteins closely correlates with protection against Toxoplasma gondii infection. Proc. Natl. Acad. Sci. USA 89, 3155-3158.

Ordway, JM, Tallaksen-Greene, S, Gutekunst, C-A, Bernstein, EM, Cearley, JA, Wiener, HW, Dure, LS, Lindsey, R, Hersch, SM, Jope, RS, Albin, RL and Detloff, PJ (1997) Ectopically expressed CAG repeats cause intranuclear inclusions and a progressive late onset neurological phenotype in the mouse. Cell 91, 753-763.

Reich, Z, Wachtel, EJ and Minsky, A (1995) In vivo quantitative characterization of intermolecular interactions. J. Biol. Chem. 270, 7045-7046.

Russell, W (1890) A characteristic organism of cancer. Brit. Med. J. 2, 1356-1360.

Rutherford, SL and Lindquist, S (1998) Hsp90 as a capacitator for morphological evolution. Nature 396, 336-342.

Schein, CH (1989) Production of soluble recombinant proteins in bacteria, Biotechnol. 7, 1141-1149.

Schild, H, Rtzschke, O, Kalbacher, H and Rammensee, HG (1990) Limit of T-cell tolerance to self proteins by peptide presentation. Science 247, 1587-1589.

Spiess, C, Beil, A and Ehrmann, M (1999) A temperature-dependent switch from chaperone to protease in a widely conserved heat shock protein. Cell 97, 339-347.

Srivastava, PK (1993) Peptide-binding heat shock proteins in the endoplasmic reticulum: role in immune response to cancer and in antigen presentation. Adv. Cancer. Res. 62, 153-177.

Srivastava, PK (1996) Do human cancers express shared protective antigens? or the necessity of remembrance of things past. Sem. Immunol. 8, 295-302.

Srivastava, PK, Menoret, A, Basu, S, Binder, RJ. and McQuade, KL (1998) Heat shock proteins come of age: primitive functions acquire new roles in an adaptive world. Immunity 8, 657-665.

Srivastava, PK, Udono, H, Blachere, NE and Li, Z (1994) Heat shock proteins transfer peptides during antigen processing and CTL priming. Immunogenetics 39, 93-98.

Stark, GR, Kerr, IM, Williams, BRG, Silverman, RH and Schreiber, RD (1998) How cells respond to interferons. Annu. Rev. Biochem. 67, 227-264.

Tamura, Y, Peng, P, Liu, K, Daou, M, Srivastava, PK (1997) Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 278, 117-120.

Vidovic, D and Matzinger, P (1988) Unresponsiveness to foreign antigen can be caused by self-tolerance. Nature 336, 222-225.

Zeh, HJ, Perry-Lalley, D, Dudley, ME, Rosenberg, SA and Yang, JC (1999) High avidity CTLs for two self-antigens demonstrate superior in vitro and in vivo antitumor efficiency. J. Immunol. 162, 989-994.

Zinkernagel, RM and Doherty, PC (1974) Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248, 701-702.

colorbar.gif (2362 bytes)


Random recruitment of non-oncogenes because the cancer phenotype predisposes to mutation?

    Assume a primary mutation activating a potential oncogene A (A -> A') in one cell which thus acquires a cancer phenotype with a predisposition to mutate non-oncogenes. We will designate the cell as A' to reflect the difference of its product from normal cells (A). The cell divides, so that there are now two identical potential cell lines, A'1 and A'2, with the predisposition. To simplify, we will have them both mutate in the first generation, rather in random subsequent generations. A'1 mutates non-oncogene B (B -> B'). A'2 mutates non-oncogene C (C -> C'). These two lines, A'1B' and A'2C', now divide and proliferate as independent clonal lines. There is no line with the genotype A'B'C'. A'1B' and A'2C', by virtue of mutations resulting in "not-self" proteins, will be attacked independently by cytotoxic T cells, but the chance of successful attack is much less than in the case of a clonal line able to accumulate non-oncogene mutations in every cell.

    To achieve this, one of the lines, say A'1B', by virtue of the mutation B' would have to acquire a growth advantage. Given some turn-over of each population, then eventually A'1B' would become the dominant cancer cell population. Given the predisposition to mutation, a further mutation in non-oncogene D (D -> D') could then occur, so that now non-oncogene mutations would be present in one cell (A'1B'D').

    There would now be two lineages A'1B', A'1B'D'. If the latter had a growth advantage, it would become the dominant cancer cell population. This cycle would continue in order to further accumulate mutations in one clonal line. However, as the line accumulated mutations, hence making it more susceptible to T-cell attack, it would be preferentially selected against by the host, thus the growth advantage which allowed it to accumulate mutations in one line would have become a hazard. The scenario seems unlikely, particularly since the usual situation is that B and D are not mutated, yet peptides from their products become targets of attack by cytotoxic T cells.

End Note (August 2008)

Bacterial Inclusion Bodies. Foreign proteins expressed in bacteria may aggregate and be seen as 'inclusion bodies' (Schein, 1989). Rather than being amorphous, bacterial inclusion bodies are now considered as structured, containing "amyloid-like" segments that favour an orderly aggregation process. Thus "amyloid formation is an omnipresent process both in eukaryotes and prokaryotes" and "amino acid sequences evolve to avoid the amyloid conformation." (Wang et al. 2008. PLOS Biology 6, issue 8, e195). The well known acceleration of inclusion body formation in bacteria grown at higher temperatures is reflected in the observation that "the apparent growth of fibrils from inclusion bodies can be accelerated by higher temperature, ... but is not present in a sample stored for 10 months at 4C." The authors also note that inclusion bodies are:
"Often enriched in beta-sheet structure, bind amyloid-tropic dyes, have a seeding capacity reminiscent of amyloids, form homogeneous aggregates without cross-aggregation, and display aggregation propensities strongly affected by mutations."

End Note (July 2016)

Neoantigens Sometimes Similar. Although the details of this scheme were modified in later work (Forsdyke 2001), the basic idea of rendering immunogenic certain individual-specific neoantigens, has stood the test of time. New work now shows that some of these neoantigens may be common to several patients (Munson et al. 2016. PNAS, "Identification of shared TCR sequences from T cells in human breast cancer using emulsin RT-PCR"). Thus, more general immunotherapeutic approaches to cancer treatment, rather than individual-specific approaches, become feasible.

End Note (Dec 2016)

Non-Specialist Adaptive Mutations.

That there are so many adaptive mutations was described as "puzzling" by Enard et al. (2016) in an eLife paper. They suggest this might relate to the spectrum of non-specialist proteins that, in addition to their conventionally acknowledged functions, have evolved to interact with viral proteins. This line of thought could include mutated self-proteins as in cancer.

Enard, Cai, Gwennap & Petrov (2016) Viruses are a dominant driver of protein adaptation in mammals. eLife 5:e12469.


End Note (April 2018) Antisense RNA Network

Antisense RNA Network. Another Defence System of Animals

Our studies of purine-loading (see Cristillo et al. 1998 abstract) led to the idea of what we came to call "antibody RNA" within cells. Interestingly, Dexing Fang wrote a paper in Chinese with an English abstract posted in Europe PMC and copied here without correction:  Dexing Abstract  [01 Jan 1994, 21(2):178-181]:

"On the basis of the nature of nucleic scids and recent rescarch achievements or findings about the macromolecules. such as DNA-replication-repressor RNA. transcription-factor RNA, extracellular “communicator RNA”, ribozyme, gene shears, RNA editing, anti-virus and anti-tumor activities of antisense RNA, a new hypothesis, antisense RNA network, is advanced. i.e. There are many kinds of small antisense RNAs and their complementary antiantisense RNAs from genomic DNA within the organism. Because of self-modification (or other mechanism). the antisense RNAs and the anti-antisense RNAs base-pair. but do not reanneal or hybridize with each other. This antisense RNA network, on the one hand, participates in regulating the expression of certain genes in particular tissues at particular time. keeps relative balance of various functional activities. On the other hand, the network plays an important role in specifically recognizing and eliminating the nucleic acids mutated within the body or invaded into the body from the outside."

What may be a full English translation of this paper was published at the Bionet.immunology website on 10th October 1999 with a comment from Morten Lindow, Copenhagen. Here Feng declared: "The antisense RNA network is to nucleic acids what the immune network is to proteins (antigens)": see Bionet Immunology

Feng D (1994) Antisense RNA network: a new hypothesis. Biochem Biophys Prog 21:178-181.


End Note (May 2018) Tumour Infiltrating Lymphocyte Specificities

There is more evidence for possible recruitment of cytotoxic T cells, designated "bystanders" into the response against cancer cells. Simoni et al. (2018) "show that CD8+ TILs are not all specific for tumour antigens, but can include bystander CD8+ TILs that are specific for cancer-unrelated epitopes." Furthermore, "our results demonstrate that human CD8+ TILs can not only be specific for tumour antigens but may also contain a population that recognizes cancer-unrelated epitopes, characterized by an absence of CD39 expression."

Simoni Y et al. (2018) Nature online. Bystander CD8 T cells are abundant and phenotypically distinct in human tumour infiltrates.

Video presentation and discussion of above paper: Pramod Srivastava and Donald Forsdyke (2015 You Tube version; 53 min)  (Vimeo version here)

Next: Chaperone Mode-Switches and Double-Stranded RNA (2000) (Click Here)

Go to: Role of Polymorphism in Intracellular Self/Not-Self Discrimination (2001) (Click Here)

Go to: Immunity as a Function of the Unicellular State (2002) (Click Here)

Return to: Intracellular Self/Not-Self Discrimination (Click Here)

Return to: Theoretical Immunology Page (Click Here)

Return to: HomePage (Click Here)

colorbar.gif (2362 bytes)

This page was last edited on 18 May 2018 by Donald Forsdyke.