Analysis of Class I Major Histocompatibility Complex Gene Transcription in Human Tumors Caused by Human Papillomavirus Infection

Abstract

Oncoproteins from high-risk human papillomaviruses (HPV) downregulate the transcription of the class I major histocompatibility complex (MHC-I) antigen presentation apparatus in tissue culture model systems. This could allow infected or transformed cells to evade the adaptive immune response. Using data from over 800 human cervical and head & neck tumors from The Cancer Genome Atlas (TCGA), we determined the impact of HPV status on the mRNA expression of all six MHC-I heavy chain genes, and the β2 microglobulin light chain. Unexpectedly, these genes were all expressed at high levels in HPV positive (HPV+) cancers compared with normal control tissues. Indeed, many of these genes were expressed at significantly enhanced levels in HPV+ tumors. Similarly, the transcript levels of several other components of the MHC-I peptide-loading complex were also high in HPV+ cancers. The coordinated expression of high mRNA levels of the MHC-I antigen presentation apparatus could be a consequence of the higher intratumoral levels of interferon γ in HPV+ carcinomas, which correlate with signatures of increased infiltration by T- and NK-cells. These data, which were obtained from both cervical and oral tumors in large human cohorts, indicates that HPV oncoproteins do not efficiently suppress the transcription of the antigen presentation apparatus in human tumors.

Keywords: MHC-I; antigen presentation; cervical carcinoma; head & human papillomavirus; immune evasion; major histocompatibility complex; neck carcinoma.

Figure 1

Expression of classical major histocompatibility complex class I (MHC-I) heavy chain genes in head & neck or cervical carcinomas stratified by human papillomavirus (HPV) status. Normalized RNA-seq data for the indicated MHC-I heavy chain genes was extracted from The Cancer Genome Atlas (TCGA) database for the head & neck cancer (HNSC) (A) and cervical carcinoma (CESC) (B) cohorts for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; ns—not significant; red brackets indicate a comparison that did not achieve significance with a power value <0.8.

Figure 2

Expression of non-classical MHC-I heavy chain genes in head & neck or cervical carcinomas stratified by HPV status. Normalized RNA-seq data for the indicated MHC-I heavy chain genes was extracted from the TCGA database for the HNSC (A) and CESC (B) cohorts for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; ns—not significant; red brackets indicate a comparison that did not achieve significance with a power value <0.8.

Figure 3

Expression of the MHC-I light chain and other genes involved in MHC-I dependent antigen presentation in head & neck carcinomas stratified by HPV status. Normalized RNA-seq data for the indicated genes involved in MHC-I dependent antigen presentation was extracted from the TCGA database for the HNSC cohort for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; ns—not significant.

Figure 4

Expression of the MHC-I light chain and other genes involved in MHC-I dependent antigen presentation in cervical carcinomas stratified by HPV status. Normalized RNA-seq data for the indicated genes involved in MHC-I dependent antigen presentation was extracted from the TCGA database for the CESC cohort for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; ns—not significant; red brackets indicate a comparison that did not achieve significance with a power value <0.8.

Figure 5

Detection of tumor infiltrating T-cells and NK-cells in head & neck and cervical carcinomas stratified by HPV status. Normalized RNA-seq data for genes indicative of tumor-infiltrating T cells (CD3D, CD3E and CD3G) or NK cells (FCGR3A) was extracted from the TCGA database for the HNSC (A) and CESC (B) cohorts for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; ns—not significant; red brackets indicate a comparison that did not achieve significance with a power value <0.8.

Figure 6

Detection of IFN-γ mRNA in head & neck and cervical carcinomas stratified by HPV status. Normalized RNA-seq data for the IFN-γ gene produced by tumor-infiltrating lymphocytes was extracted from the TCGA database for the HNSC (A) and CESC (B) cohorts for HPV+, HPV−, and normal control tissues. Numbers in brackets refer to the number of samples included in each analysis. ** p ≤ 0.01; **** p ≤ 0.0001; ns—not significant.

Figure 7

Comparison of the somatic mutation frequency in genes involved in MHC-I dependent antigen presentation in head & neck and cervical carcinomas stratified by HPV status. The presence or absence of somatic mutations in selected genes of the tumor samples in the TCGA HNSC and CESC cohorts was obtained from the Broad Genome Data Analysis Center’s Firehose server. The frequency of mutations in individual genes or groups of genes in HPV+ versus HPV− samples was calculated to determine whether they were influenced by HPV status. Analysis of TP53 and CDKN2A (left panels) revealed a significantly reduced frequency of somatic mutations in HPV+ versus HPV− samples in both the TCGA HNSC (A) and CESC (B) cohorts. No significant decrease was observed for somatic mutation frequency in the MHC-I heavy chain genes, or a collective grouping of all genes involved in MHC-I dependent antigen presentation between HPV+ and HPV− samples in either the TCGA HNSC (A) or CESC (B) cohorts. Numbers in brackets refer to the number of samples included in each analysis. * p ≤ 0.05; ns—not significant.