Identification of two HLA-A*0201 immunogenic epitopes of lactate dehydrogenase C (LDHC): potential novel targets for cancer immunotherapy

One of the major challenges in cancer immunotherapy remains the persistence of high affinity T cells specifically targeting tumor-associated antigens within an immunosuppressive tumor microenvironment. A good candidate target for immunotherapy should confer a high tumor selectivity with minimal adverse events. In addition, the target should preferably play a pivotal role in promoting tumor development and progression and/or impairing anti-tumor immunity, hence increasing the success rate of the therapeutic intervention to eradicate the tumor. Based on these criteria, the cancer testis antigen (CTA) lactate dehydrogenase C (LDHC) could be considered a novel promising immunotherapeutic target.

LDHC belongs to the lactate dehydrogenase family that catalyzes the interconversion of pyruvate and l-lactate and plays important roles in aerobic glycolysis [1]. Lactate dehydrogenase isozymes exist as homo- or hetero-tetramers composed of two major subunits, LDH-M and LDH-H that are encoded by LDHA and LDHB, respectively. Different combinations of these subunits assemble into 5 distinct isozymes with different tissue specificity; LDH1/LDHB (4H), LDH2 (3H1M), LDH3 (2H2M), LDH4 (1H3M) and LDH5/LDHA (4M). While LDHA is predominantly expressed in skeletal muscle and preferentially converts pyruvate to lactate, LDHB is mainly expressed in the heart and brain where it catalyzes the interconversion of lactate to pyruvate. LDHC, encoded by the LDHC gene, assembles into a homotetramer of LDHC subunits, also known as the LDHC or LDHX isoform [2]. Gene evolution models indicate that LDHC arose from gene duplication of the LDHA gene in mammals with 75% sequence homology with LDHA and 70% with LDHB [2]. LDHC expression is restricted to mature testis and spermatozoa, with low expression in oocytes and early embryos [3]. LDHC deficiency has been linked to male infertility, partly caused by diminished spermatozoa motility, whereas female null mice are fertile [4, 5]. Hence, the role of LDHC in spermatogenesis, oogenesis, fertility and early development remains unclear.

Although LDHC expression is tightly controlled and suppressed in normal somatic tissues, it is re-expressed in various malignant tissues, making its expression highly tumor specific [6]. Furthermore, increased LDHC expression has been associated with poor prognosis in renal cell carcinoma [7]. Very little data are available on the role of LDHC in cancer. Based on the observations of LDHA- and LDHB-mediated cancer progression, we can speculate that LDHC could be involved in metabolic reprograming of cancer cells. It is well established that growing tumors can bypass oxidative phosphorylation in favor of aerobic glycolysis to support their increasing metabolic need, which involves metabolic enzymes such as lactate dehydrogenases [8]. Indeed, dysregulation of LDHA and LDHB expression has been observed in tumors with increased glycolysis [9]. Hence, altered expression of LDHC could be involved in maintaining an alternative energy source by contributing to the metabolic switch in cancer cells. In addition, increased LDHA and decreased LDHB expressions facilitate tumor formation and progression through remodeling of the tumor microenvironment, increasing proliferation, and inducing epithelial-to-mesenchymal transition, cell migration and invasion, and angiogenesis [10‐20]. In line with this, two studies to date demonstrate that enhanced expression of LDHC induces epithelial-to-mesenchymal transition, matrix metalloproteinase-9 (MMP9) expression and promotes cancer cell migration and invasion [7, 21].

Targeting LDHC could be a promising novel approach for cancer immunotherapy. First, given its restricted expression profile, it is likely that LDHC-specific immune-based interventions will result in the generation of LDHC-specific T cells with high affinity and low off-target effects. Moreover, targeting LDHC would not only inhibit LDHC-mediated cancer progression and specifically eradicate LDHC positive tumor cells, but could also induce reversal of the acidic tumor microenvironment, thereby releasing anti-tumor immunity. It is important to note that lactate and the concomitant tumor acidity negatively influence the anti-tumor immune response by skewing the immune cell compartment towards an immunosuppressive environment [22‐24]. More specifically, LDHA has been found to promote upregulation of PD-L1 on tumor cells, impeding effector T cell activity [25]. Furthermore, elevated serum LDHA levels are associated with tumor burden as well as poor clinical outcome to PD-1 and CTLA-4 immune checkpoint blockade therapy [26]. Therefore, targeting key players of lactate metabolism including LDHC could aid to re-establish anti-tumor immunity and is a largely unexplored area of research.

In the present study, we generated several T cell responses against LDHC using either in vitro stimulation of T cells with synthetic peptides or priming of T cells with autologous peptide-pulsed dendritic cells. Using both approaches, we found that several peptide pools and individual peptides could elicit a cellular immune response, as determined by IFN-γ secretion. More in-depth analysis of the responses in HLA-A*0201 healthy donors enabled us to identify two HLA-A*0201-restricted immunogenic epitopes, LDHC⁴¹⁻⁵⁵(LKDLADELALVDVAL) and LDHC^288−303 (LSIPCVLGRNGVSDV), that could possibly be targeted by adoptive T cell therapy. Using different breast cancer cell lines, we demonstrated that LDHC⁴¹⁻⁵⁵and LDHC^288−303 specific T cells were capable of recognizing and eradicating HLA-A*0201 positive/LDHC positive tumor cells, while no specific cytolytic activity was detected against HLA-A*0201 negative/LDHC positive tumor cells. Interestingly, we found that reduction of LDHC expression in the HLA-A*0201 positive cancer cells attenuated the T cell responses against LDHC, suggesting a plausible threshold of LDHC expression to elicit immune reactivity. To conclude, we demonstrate for the first time that LDHC exhibits immunogenicity and our findings warrant further study into the potential of LDHC as a novel therapeutic target for cancer immunotherapy.

Cancer testis antigens are gaining interest as targets for adoptive T cell therapy with numerous preclinical studies and clinical trials focusing on NY-ESO-1, MAGE-A3 and PRAME [30‐32]. To date, there are no published data on the immunogenicity and targetability of the cancer testis antigen LDHC for cancer immunotherapy except for a Master’s thesis that is deposited in the public domain [33]. LDHC expression has been detected in different tumor types at varying degrees with frequencies up to 100% in lung adenocarcinoma, 83% in cervical cancer, 76% in high-grade serous ovarian carcinoma (HGSC), 44% in melanoma, and 35% in breast cancer [6, 33, 34]. As aforementioned, to the best of our knowledge, the immunogenicity of LDHC has been investigated in one study only that demonstrated the presence of LDHC-peptide reactive T cells in the ascites of three of five patients with high-grade serous ovarian carcinoma [33]. Following expansion of LDHC-reactive T cells of 2 out of 3 patients; one patient displayed specific T cell responses against one peptide pool. Further analysis revealed that these responses were elicited by the 11-mer YTSWAIGLSVM peptide, corresponding to peptide p62. However, p62-specific T cells were not able to recognize autologous ascites, autologous B cells transfected with LDHC or tumor cell lines with endogenous LDHC. Of note, the 11-mer peptide identified in their study was also included in our peptide pool 7 as P62. In our study, we did not observe any strong T cell responses against P62, as determined by IFN-γ release, using in vitro peptide stimulation of T cells isolated from 14 healthy donors, rather than patients. HLA prediction analysis of the P62 peptide sequence (KGYTSWAIGLSVMDL) did not reveal strong nonamer or decamer binders for the reported HLA haplotype of the HGSC patient (HLA-A*02, HLA-B*08, HLA-B*057, HLA-C*06, and HLA-C*07) which may in part explain the lack of endogenous LDHC-reactive T cell responses in the study. In addition, it is unclear whether the LDHC-derived peptide is processed and presented as a cognate peptide and to what extent LDHC is expressed in the patient’s tumor cells and the tumor cell lines. Furthermore, it is likely that pre-existing high-affinity LDHC-specific T cells are partially exhausted or are only present in the tumor microenvironment and not in the ascites of patients.

In our study, we found a wide range of T cell responses against synthetic 15-mer LDHC-derived peptides using T cells from 14 healthy donors. Further analysis of four HLA-A*02 healthy individuals demonstrated the immunogenic potential of two individual peptides, P11 (LDHC⁴¹⁻⁵⁵) and P73 (LDHC^288−303), and the functional activity of the respective-primed T cells against breast cancer cell lines. HLA binding prediction and peptide-T2 cell experiments support the HLA-A*0201 specificity of both peptides. LDHC⁴¹⁻⁵⁵ and LDHC^288−303-primed T cells exhibited increased IFN-γ secretion and cytolytic activity against HLA-A*0201 breast cancer cell lines with endogenous LDHC expression. In contrast, no specific T cell responses were observed against a HLA-A*0201 negative breast cancer cell line. Depletion assays demonstrated a predominant CD8 + T cell response against both peptides. Moreover, our results suggest that there is a plausible threshold of LDHC expression to elicit an anti-tumor immunity since we observed attenuated T cell responses against the HLA-A*0201 breast cancer cell line following LDHC silencing. More specifically, we observed reduced IFN-γ secretion and a complete lack of cytolytic activity of T cells in co-culture with low LDHC expressing cancer cells, which was confirmed, in a second breast cancer cell line model. Furthermore, LDHA expression is not altered in LDHC high versus LDHC low expressing cancer cells (unpublished data), suggesting that the difference in peptide-specific T cell responses in LDHC high versus LDHC low expressing cells might not be affected by cross-reactivity with LDHA cognate peptides. However, future studies are required to study cross-reactivity with LDHA and LDHB in more detail and to ensure LDHC specificity of the peptide-induced cytotoxic T cell responses. Our findings are in line with previous observations of a recognition threshold for expression of the cancer testis antigen PRAME and of different signaling thresholds for CD8⁺ T cell IFN-γ secretion and acquisition of cytolytic activity [35, 36]. The potential existence of an expression threshold for an effective anti-tumor response has important implications for LDHC-specific immunotherapy. In accordance, current efforts are directed towards increasing the expression of CTAs, including NY-ESO-1 and PRAME, through combination treatment of demethylating agents and histone deacetylase inhibitors [30, 32]. Using this combination treatment, both the intra-tumor heterogeneous expression of CTAs and the expression threshold could be addressed. Interestingly, we found that the majority of P11 (LDHC⁴¹⁻⁵⁵) and P73 (LDHC^288−303) specific T cells displayed an effector memory phenotype. This is of importance since an effective long-term anti-tumor response requires multiple T cell subpopulations, including memory and effector cells. We found an increase in the number of CD4⁺ and CD8⁺T effector memory cells after priming with the immunogenic LDHC-derived peptides P11 (LDHC⁴¹⁻⁵⁵) and P73 (LDHC^288−303), and after co-culture with HLA-A*0201/LDHC positive breast cancer cells, suggesting that the CD8⁺ T effector memory cell population was responsible for the cancer cell killing in our in vitro model. Likewise, it has been reported that the majority of redirected T cells against NY-ESO-1 is comprised of CD8⁺ effector memory cells [37]. However, it might be beneficial to also expand the smaller population of LDHC-specific T central memory cells to more readily sustain in vitro proliferation and in vivo persistence after antigen re-encounter [38, 39]. Indeed, secondary activation or re-stimulation of NY-ESO-1-specific T central memory cells in vitro induced differentiation into functional effector T cells which may be able to generate an anti-tumor immune response against minimal residual disease [37].

To conclude, this is the first study to induce T cell responses against numerous LDHC-derived peptides and identified two HLA-A*0201 restricted LDHC-specific peptides, P11 (LDHC⁴¹⁻⁵⁵) and P73 (LDHC^288−303), with immunogenic potential. Moreover, we were able to demonstrate the functional activity of the peptide-specific CD8 + T cells against breast cancer cell lines with endogenous LDHC expression, albeit with a constraint of an antigen threshold. Given the expression of LDHC in breast tumors, future studies are needed to evaluate the presence of pre-existing LDHC⁴¹⁻⁵⁵ and LDHC^288−303-reactive T cells in the peripheral blood of breast cancer patients.