The variable clinical response seen with most cancer immunotherapy suggests that there is a large interindividual variation in immunologic response to tumors. One of the key functional parameters of an immune response is the local production of cytokines. As a method to survey the immune status of tumor-infiltrating cells, we have investigated the constitutive expression of cytokine mRNA in biopsies from epithelial ovarian carcinomas by using a PCR-assisted mRNA amplification assay. Using a set of cytokine-specific primers for 10 different cytokines, we have found selective expression of interleukin 10 (IL-10), granulocyte-macrophage colony-stimulating factor, and interferon gamma mRNA in ovarian tumor tissue as compared to normal ovaries and ovarian tumor cell lines. Such differences could not be explained by the extent of T-cell infiltration, since comparing samples with the same intensity of T-cell receptor (TCR) constant region alpha-chain product from the tumor and normal biopsies demonstrated different cytokine patterns. No IL-2 gene expression was detected in the tumor biopsies. IL-2 mRNA, however, became expressed after stimulation of the tumor-derived cells via the CD3 molecule but not after growth in recombinant IL-2 alone. Using the same methodology, we also analyzed the TCR variable region beta-chain gene repertoire. No restriction or biased expression of these genes was observed.

Freshly isolated tumor-infiltrating lymphocytes (TIL) are often functionally deficient. Since one of the key functional parameters of an immune response is the local production of cytokines, we studied the expression of cytokine genes in freshly isolated renal cancer tissue. Using a PCR-assisted mRNA amplification assay, the constitutive expression of mRNA for 10 different cytokines was assessed in renal cancer tissue. We compared the cytokine mRNA expression in freshly isolated samples of renal carcinomas, renal cancer cell lines established from the tumor samples, peripheral blood mononuclear cells (PBMC) and non-tumor kidney tissue isolated from the same patients. IL-10 mRNA expression was detected only in tumor samples, while renal cancer lines, PBMC and non-tumorous kidney tissues were devoid of this cytokine. One-third of the tumor samples but none of the normal kidney samples also expressed G-CSF mRNA. IL-6, TNF-alpha and IFN-gamma mRNA were expressed non-selectively in tumors, PBMC and normal rental tissue. Expression of IL-2, IL-3 and IL-4 mRNA was not detected in any of the tissues analyzed. Established renal cancer lines exhibited expression of IL-1 alpha, IL-6, TNF-alpha and GM-CSF. Culture of tumor-derived T cells with anti-CD3 monoclonal antibody (MAb) resulted in expression of IL-2, IL-3 and IL-4 mRNA. In contrast, none of these cytokines was detected in culture with recombinant human IL-2 alone. Since IL-10 is known to suppress antigen presentation, these findings have important implications for the possible in vivo role of IL-10 as a suppressor of local anti-tumor response.

Certain T cell antigen receptor V gene products in man have been shown by us and others to display a reproducible bias for preferential expression in CD4+ or CD8+ T cell subsets. In order to investigate whether such a skewed representation of V gene segments is also present at the J gene segment level, we tested the relative J beta gene usage by V beta 5.1 + T cells, as this V beta gene is biased towards CD4+ T cell expression in virtually all individuals. To analyze the usage of the 13 J beta gene segments, we developed a new approach using V beta 5.1 and C beta specific oligonucleotides as 5' and 3' primers respectively for polymerase chain reaction (PCR) amplification of cDNA derived from CD4+ or CD8+ peripheral blood lymphocyte (PBL) T cells. The PCR products were visualized for reactivity with individual J beta 1.1-1.6 and J beta 2.1-2.7 32P-labelled oligonucleotide probes using autoradiography and quantitative gel-scanning. Eleven normal blood donors provided the PBL T cells. The results showed that in every individual's V beta 5.1+ T cell populations (CD4 and CD8), all V beta/J beta combinations were used although at varying but reproducible levels for each J beta gene. Thus, no discernible disallowance of combinations existed. Moreover, we could show that six of 13 J beta genes were unequally expressed when compared in pairs with regard to expression in CD4+ and CD8+ T cell subsets.(ABSTRACT TRUNCATED AT 250 WORDS)

Sjögren's syndrome (SS) is a chronic autoimmune disorder characterized by lymphocytic infiltration of the salivary and lacrimal glands. These patients have a markedly increased frequency of developing non-Hodgkin's lymphoma in their salivary glands and cervical lymph nodes. Translocations of proto-oncogene bcl-2 t(14;18) were observed in five of seven SS-associated lymphomas by Southern blot analysis. Using primers specific for chromosomes 14 and 18, translocation of the proto-oncogene bcl-2 was detected by polymerase chain reaction (PCR) in all five lymphomas positive by Southern blot analysis. Among SS patients lacking clinical evidence of coexistent lymphoma, no bcl-2 translocations were detected in 50 consecutive salivary gland biopsies. Of particular interest, pre-lymphoma biopsies were available from the seven SS patients who subsequently developed lymphoma and these DNA samples lacked detectable t(14;18) translocations even though they exhibited oligoclonal rearrangements of their immunoglobulin genes. We conclude that the great sensitivity of PCR can help us in detecting early onset of lymphoma in SS patients and aid in understanding the transition from autoimmunity to lymphoma.