Geneticists have recognized for some time that many genes exhibit pleiotropy, meaning that one mutation can manifest in two or more distinguishable phenotypic effects. In a fascinating study recently published in Science [2014 Jan 10;343(6167):152-7. doi:10.1126/science.1246886], Joseph et al. offer evidence for an example of pleiotropy in which the distinct phenotypic effects associated with mutation of the POLR3A gene, which encodes a subunit (RPC1) of RNA polymerase III, are associated with two different diseases: one or another form of cancer and an autoimmune disease (scleroderma).

Specifically, the authors set out to test the hypothesis that the known association between scleroderma in patients with antibodies to RPC1 and a variety of cancers was based on somatic mutations in the POLR3A gene eliciting immune responses that ultimately contribute to the vasculopathy and tissue fibrosis associated with that specific autoimmune condition. They found somatic POLR3A mutations or loss of heterozygosity (LOH) for the POLR3A gene in tumors from six of eight patients having both scleroderma and antibodies to RPC1. In eight patients with cancer and scleroderma but with antibodies to molecules (TOP1, CENPB) other than RPC1, no somatic mutations were found in the genes encoding these apparent autoantigens (or in the POLR3A gene).

Interestingly, in the eight patients with antibodies to RPC1 who were studied, the cancer and scleroderma were diagnosed within 0.3 to 2.5 years of each other, with five cancer diagnoses preceding identification of scleroderma and three following. Thus it may be plausible that even in those patients for whom the diagnosis of scleroderma preceded the diagnosis of cancer, a mutation of the POLR3A gene in a subclinical tumor could have precipitated the immune response to RPC1 that contributed to the scleroderma. In contrast, for seven of the eight patients with antibodies to TOP1 or CENPB, the diagnosis of scleroderma preceded that for cancer by a median interval of 14.2 years.

Also of potential relevance, for four of the five patients with antibodies to RPC1 and POLR3A LOH, the LOH was subclonal (i.e, only some of the tumor cells were missing a copy of the POLR3A gene). This pattern of LOH suggests immune-mediated selection (termed immunoediting) against a mutant allele of the genes. Thus, tumor cells with POLR3A LOH generated by ‘random’ genetic events could have evaded immune attack and elimination with a higher probability than tumor cells expressing a presumably mutated form of the gene product, RPC1. If this mechanism is operative, as the authors note, some incipient tumors with mutations in the POLR3A gene might be completely eliminated before detection, potentially accounting for patients with scleroderma and antibodies to RPC1 who do not also have cancer.

Investigation of the immune responses to the RPC1 molecules in three scleroderma patients with cancer revealed that the serum antibodies did not distinguish between wild-type and mutant RPC1. Therefore, the investigators focused on cell-mediated immune responses to RPC1. They were able to demonstrate responses by CD4+ T cells to mutant RPC1-derived peptides in two of the three patients tested. Analysis of the T-cell receptor (TCR) genes of RPC1-specific responding CD4+ T cells revealed that T cells responding to mutant peptides were in general different from those responding to wild-type peptides. In addition, some TCR amino acid sequences were encoded by multiple nucleotide sequences suggesting antigen-driven selection of the T cells.

The presence of reactivity of CD4+ T cells in some of the patients with RPC1 antibodies to peptides derived from both wild-type and mutant is consistent with the previously described phenomenon of determinant (or epitope) spreading in autoimmune disease (Lehmann et al., 1993). Determinant spreading refers to the situation where immune responses to an initial autoantigen generate an inflammatory milieu that facilitates the development of autoimmune responses against additional antigens (or additional epitopes on the original autoantigen) from the same tissue. The authors speculate that it is possible that the RPC1-specific autoimmune B cells contribute to these responses against wild-type RPC1 by serving as antigen presenting cells.

Additional questions are prompted by this study. Why should missense mutations in the POLR3A gene lead to scleroderma and not some other autoimmune condition? Are there other genes for which mutation in the setting of cancer could lead to scleroderma or another autoimmune disease? Joseph et al. note that there are other examples of temporal clustering of autoimmune conditions (myositis, vasculitis, systemic lupus erythematosus) and cancer. It will be of interest to know if subsequent research reveals particular mutated genes that could account for these associations.

The authors also point out that paraneoplastic syndromes have been shown in some instances to involve autoimmune responses, but the targets of these responses are non-mutated proteins produced by the tumor cells. Other studies have documented that immune responses to tumor antigens encoded by mutated genes do not typically lead to autoimmune responses targeting the corresponding wild-type proteins. Why these responses do not lead to epitope spreading and clinical autoimmunity is not clear.

In summary, mutations that occur in malignant cells, whether contributing to tumorigenesis or not, may elicit immune responses to the altered-self protein(s) that both select against tumor cells with the mutation, leading to, for example, LOH, and initiate autoimmunity of clinical relevance. Thus, in this setting, selection favoring genetic instability in tumor cells can lead to mutation that then leads to antigen-driven selection of the CD4+ T cells that is as or more likely to be damaging as protective.


Joseph CG, Darrah E, Shah AA, Skora AD, Casciola-Rosen LA, Wigley FM, Boin F,Fava A, Thoburn C, Kinde I, Jiao Y, Papadopoulos N, Kinzler KW, Vogelstein B, Rosen A. Association of the autoimmune disease scleroderma with an immunologic response to cancer. Science. 2014 Jan 10;343(6167):152-7. doi: 10.1126/science.1246886. Epub 2013 Dec 5. PubMed PMID: 24310608.

Lehmann PV, Sercarz EE, Forsthuber T, Dayan CM, Gammon G. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol Today. 1993 May;14(5):203-8. Review. PubMed PMID: 7686009.