Identification of Huntingtin Interacting Protein 1 Related (HIP1R) as a Novel Lymphoma Autoantigen Using Antigen Array Profiling

We screened lymphoma patients’ serum samples against protein/antigen arrays, containing over 10,000 different human proteins, to identify disease-associated auto-IgG antibody interactions. This approach identified a humoral immune response to the HIP1R antigen in 40% of patients’ sera (4/6 follicular lymphomas, 3/5 germinal centre [GC]-derived diffuse large B-cell lymphomas [DLBCL], 2/4 transformed DLBCL, 1/5 mantle cell lymphomas, 1/5 non-GC DLBCL and 1/5 peripheral T-cell lymphomas). HIP1R has also previously been identified as an autoantigen in colon cancer using the SEREX technique. HIP1R is one of two members of the Huntingtin interacting protein 1 family. Both HIP1R and HIP1 share significant sequence homology and are cytoplasmic proteins that interact with inositol lipids, clathrin and actin. Interestingly, both proteins can stabilize pools of receptor tyrosine kinases by inhibiting their trafficking to the lysosome for degradation, which may mediate affects on cell growth and transformation. HIP1 has been more extensively characterized in cancer and was originally implicated in hematological malignancy when it was identified as a platelet-derived growth factor β receptor fusion partner in chronic myelomonocytic leukemia. HIP1 expression is elevated in lymphomas and can be induced by activated NF-κB while transgenic overexpression of HIP1 is associated with the development of lymphoid neoplasms. Interestingly, the increased severity of a murine Hip1/Hip1r double-knockout phenotype indicates that Hip1r can partially functionally compensate for the loss of Hip1, suggesting they have overlapping roles in vivo. Autoantibodies to HIP1 have been reported in 59% of human lymphoma sera, compared to 31% of healthy controls and showed higher titre in patients who had undergone remission than those that relapsed (Bradley et al. 2007. Cancer Res 67:8923). However, the region of HIP1 tested for this serum reactivity shares high sequence homology with HIP1R, and thus this previous study is likely to have identified autoantibodies to both HIP1 and HIP1R. In contrast the region of HIP1R on our antigen array does not share homology with HIP1, suggesting that HIP1R expression is also altered in lymphomas. Quantitative RT-PCR was performed for both HIP1 and HIP1R on a panel of lymphoma cell lines, normal tissues and purified B- and T-cell populations. Overall HIP1R was more widely expressed and at higher levels than HIP1 in both normal lymphocytes and lymphoma cell lines. The particularly high level expression of HIP1R in Burkitt lymphoma cell lines and increased expression in GC- compared to non-GC DLBCL is consistent with the high frequency of autoantibodies in the sera from patients with GC-derived lymphomas. Two commercial monoclonal antibodies were successfully validated for the detection of HIP1R, although only the BD Transduction Laboratories’ antibody is raised to a region that lacks any significant stretches of homology with HIP1. Western blotting experiments have confirmed the expression of the HIP1R protein in lymphoma cell lines and showed a good correlation with mRNA expression levels. The BD Transduction Laboratories anti-HIP1R antibody recognises a formalin resistant epitope and studies are already underway to elucidate HIP1R expression patterns in lymphoma biopsies.