• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jun 28, 2011; 108(26): 10656–10661.
Published online Jun 13, 2011. doi:  10.1073/pnas.1100354108
PMCID: PMC3127933
Immunology

HLA-E expression by gynecological cancers restrains tumor-infiltrating CD8+ T lymphocytes

Abstract

HLA-E is a nonclassical HLA class I molecule, which differs from classical HLA molecules by its nonpolymorphic, conserved nature. Expression and function of HLA-E in normal tissues and solid tumors is not fully understood. We investigated HLA-E protein expression on tissue sections of 420 ovarian and cervical cancers and found equal or higher levels than normal counterpart epithelia in 80% of the tumors. Expression was strongly associated with components of the antigen presentation pathway, e.g., transporter associated with antigen processing (TAP), endoplasmic reticulum aminopeptide (ERAP), β2 microglobulin (β2m), HLA classes I and II, and for ovarian cancer with tumor infiltrating CD8+ T lymphocytes (CTLs). This association argues against the idea that HLA-E would compensate for the loss of classical HLA in tumors. In situ detection of HLA-E interacting receptors revealed a very low infiltrate of natural killer (NK) cells, but up to 50% of intraepithelial CTLs expressed the inhibiting CD94/NKG2A receptor. In cervical cancer, HLA-E expression did not alter the prognostic effect of CTLs, most likely due to very high infiltrating CTL numbers in this virus-induced tumor. Overall survival of ovarian cancer patients, however, was strongly influenced by HLA-E, because the beneficial effect of high CTL infiltration was completely neutralized in the subpopulation with strong HLA-E expression. Interestingly, these results indicate that CTL infiltration in ovarian cancer is associated with better survival only when HLA-E expression is low and that intratumoral CTLs are inhibited by CD94/NKG2A receptors on CTLs in the tumor microenvironment.

Keywords: cervical carcinoma, NK receptors, ovarian carcinoma, Tumormilieu, immune surveillance

HLA-E is a nonclassical major histocompatibility complex (MHC) class I molecule that is almost nonpolymorphic, in contrast to its classical class I counterparts HLA-A, -B, and -C (1, 2). There are two HLA-E subtypes, which differ by only one amino acid (35). This coding variation is located outside the peptide binding groove, and both HLA-E variants are indeed indistinguishable in their structure and peptide binding features (35). HLA-E normally presents a very limited variety of peptides, derived from signal peptide sequences of classical MHC class I. However, infections and transformation can mediate the presentation of alternative peptides (1, 2). Surface expression of HLA-E is largely dependent on β2 microglobulin (β2m) as well as transporter associated with antigen processing (TAP) and tapasin (6, 7).

HLA-E/peptide complexes are ligands of the CD94 receptor in conjunction with the inhibitory NKG2A or the stimulatory NKG2C molecule, which are expressed on the majority of natural killer (NK) cells and some activated CD8+ T lymphocytes (CTLs) (8, 9). The presentation of “self” signal peptides by HLA-E enables these lymphocytes to gauge the overall MHC class I expression on the surface of target cells. Engagement of CD94/NKG2A receptors on lymphocytes reduces the reactivity of NK cells or CTLs, which protects against excessive immune-mediated tissue damage, for instance during infections (10, 11). On the other hand, CD94/NKG2A expression can be induced in response to cytokines such as IL-15 (12) and transforming growth factor β (TGF-β) (13). These cytokines are frequently present in the tumor microenvironment, suggesting that CD94/NKG2A inhibitory receptors play a role in immune escape by tumor cells.

With the availability of specific antibodies to detect HLA-E in native conformation (clone 3D12; ref 6) or as denatured protein (clone MEM-E/02; ref 14), we recently investigated the normal expression of HLA-E on human tissues in collaboration with the Human Proteome Resource program (HPR at www.proteinatlas.org). Expression on blood cells, endothelium, melanocytes, and intestinal epithelial cells was confirmed (6, 1517). Furthermore, we observed a pattern of tissue staining that was very similar to that of classical MHC class I.

Here, we determined the expression of HLA-E in 150 cervical and 270 ovarian cancer samples and analyzed its association with clinical and immunological parameters. Our results indicate that HLA-E is frequently overexpressed in these tumor types and positively associated with expression patterns of antigen processing components, classical HLA molecules, and immune cell infiltrate. In situ analysis of the interacting receptors of HLA-E, i.e., the inhibitory CD94/NKG2A and the activating CD94/NKG2C, revealed a frequent expression of the inhibitory receptor on intraepithelial CD8+ T cells. NK cells, the predominant cell type expressing CD94/NKG2A and CD94/NKG2C, were hardly found in both tumor types. Importantly, the beneficial prognostic effect of infiltrating CTLs in ovarian cancer was neutralized by high expression of HLA-E, indicating that HLA-E hampers activity of antitumor CTLs in the tumor microenvironment.

Results

HLA-E Expression in Gynecological Cancers.

Two cohorts of gynecological tumor tissue were evaluated: 270 ovarian and 150 cervical cancers. Table S1 summarizes clinicopathological characteristics and survival data of the two cohorts. Notably, ovarian cancer is generally diagnosed at a much later stage than cervical cancer. Hence, this cohort consists mainly of high stage tumors with an average disease-specific survival of only 3.5 y, whereas cervical cancer patients live on average 14 y after diagnosis.

We first determined the expression of HLA-E on nonmalignant ovarian and cervical tissue. For ovarian tissue, we selected seven pre- and postmenopausal samples with intact, nonmalignant ovarian epithelium. Also, we stained nine cervical sections containing normal ectocervical squamous epithelium and endocervical glands (Fig. 1). These are the structures that give rise to ovarian and cervical cancer, respectively. The ovarian epithelium showed a weak positive staining in both pre- and postmenopausal samples, whereas ovarian stroma was negative (Fig. 1 A and B). Both cervical epithelia stained negative to weak positive for HLA-E; stroma was negative (Fig. 1 E and F). The endothelium of blood vessels was highly positive for HLA-E as well as resident leukocytes, in line with previous reports (15).

Fig. 1.
HLA-E expression in ovarian and cervical cancer. Paraffin-embedded tissue sections were stained with the monoclonal HLA-E–specific antibody MEM-E/02. (A and B) Two examples of the one-layer epithelial cells of normal ovaries. No difference was ...

Next, we assessed HLA-E expression on ovarian cancer (n = 270) and cervical cancer (n = 150) confined in tissue microarrays (TMAs) using a validated specific antibody. Examples of negative- and positive-staining tumors are depicted for ovarian cancer (Fig. 1 C and D, respectively) and cervical cancer (Fig. 1 G and H, respectively). Staining of HLA-E on tumor cells was scored for intensity and percentage surface area, as previously described (18), giving a range from 0 to 8.0. For both tumor types the median score was 6.0, with a range between 0 and 7.75. On the basis of the mean intensity score of normal epithelium (score 1 on a scale of 0–3), ovarian tumors and cervical tumors expressed equal or higher levels of HLA-E in 89% and 83% of the tumors, indicating that expression of HLA-E is mostly conserved in these tumors.

Associations Between HLA-E, Clinicopathologic, and Immunologic Factors.

To assess whether HLA-E expression was preferentially associated with certain patient groups, we determined the relationship between HLA-E expression and well-known clinicopathologic factors. To this end, the gradual scores of HLA-E expression were dichotomized on the basis of the lowest quartile. For ovarian cancer, there was no relationship between HLA-E expression and histology, stage, grade, or presence of residual tumor after debulking surgery (Table S2). Similarly, HLA-E expression in cervical cancer was not related to histology, stage, infiltration depth, tumor size, human papillomavirus (HPV) infection, lymph node positivity, or vascular invasion (Table S2).

We previously collected data from these cohorts of tumor samples describing immune cell infiltration and HLA-related molecules (1922). We determined associations between HLA-E and components of the antigen processing machinery, using the same cutoff values as previously described for these molecules (1922). The proteasome subunit LMP7, peptide transporter heterodimer TAP1, and the endoplasmic reticulum aminopeptidase (ERAP) (23) were associated with increased HLA-E expression in ovarian cancer (Table 1). In cervical cancer, the TAP1 and TAP2 transporters were the only components that associated with HLA-E expression. These results suggest that antigen processing components contribute to the protein expression of HLA-E.

Table 1.
Relationship of HLA-E expression in ovarian and cervical cancer with immunological characteristics

Next, the association with classical HLA class I molecules (HLA-A, HLA-B/C, and β2m) and HLA class II molecules (HLA-DP/DQ/DR) was analyzed (Table 1). Induction of HLA class II molecules is observed in a majority of these cancers and can be mediated by cytokines such as IFN-γ, similar to HLA-E (2426). This revealed a clear association with HLA-E expression, especially in ovarian cancer. This indicates that HLA-E is present in tumors with strong classical HLA expression and contrasts with the idea that HLA-E expression would compensate for loss of classical HLA molecules in cancer. In contrast, high expression of classical HLA class I promotes the stabilization of HLA-E through the delivery of leader sequences, which bind to the groove of HLA-E (2, 6, 27).

Furthermore, expression of HLA-E was correlated with the presence of T cells. The degree of infiltration of CTLs and regulatory T lymphocytes (Tregs) was recently reported by our groups for these two cohorts (20, 21). The number of tumor-infiltrating CTLs was positively correlated to HLA-E expression in ovarian cancer, but not in cervical cancer (Table 1). We previously found that the ratio between CTL and Treg is predictive of clinical outcome in cervical cancer instead of the CTL counts as such (21). For the current study, we examined the relation between the CTL/Treg ratio and HLA-E expression, but these two parameters were not associated (P = 0.343, Table 1).

In, conclusion, HLA-E expression in ovarian and cervical cancers is positively associated with other components of HLA-mediated antigen presentation—indicative of a well-functioning processing and presentation pathway—and the influx of T cells. These associations are especially prominent in ovarian cancer.

Intratumoral CTLs Express HLA-E Engaging Receptors.

The receptors for HLA-E, i.e., CD94/NKG2A and CD94/NKG2C, are predominantly expressed on NK cells. We therefore assessed the presence of these innate immune cells in our cohort of ovarian and cervical cancers using antibodies against the NK-associated markers CD56 and CD57, and the NK-specific marker NKp46 (28). In ovarian cancer, only 14% of the samples contained detectable NK cells, and the number of cells was very low in these tumors (less than 7/mm2). Cervical cancers also largely lacked infiltrating NK cells, and stainings with an anti-NKp46 antibody corroborated our previous results where we scored CD3CD57+ cells (21). Clinicopathologic factors or HLA-E expression did not differ between tumors with or without NK cells.

Besides, on NK cells, the inhibiting heterodimer CD94/NKG2A and the activating CD94/NKG2C are also expressed on a small subset of CTLs (2). We hypothesized that HLA-E in cancers might serve as ligand for these receptors on intratumoral CTLs. We applied eight-color flow cytometry analysis on fresh surgical samples, which were mechanically dissected to single cell suspensions (Fig. 2). Gating on CD3+CD4+ T cells and CD3+CD8+ cytotoxic T cells visualized the expression of CD94, NKG2A, and NKG2C receptors on these T-cell subsets (Fig. 2A). Importantly, a high frequency of the tumor-infiltrating CTLs displayed the inhibiting NKG2A chain, but not the activating NKG2C chain (Fig. 2A). Nearly all NKG2A+ CTLs also coexpressed the partner CD94 (overall 98%). In contrast, CD4+ T cells were largely devoid of these HLA-E interacting receptors. The ovarian cancers contained very low numbers of CD4+ T cells, leading to seemingly high frequencies of receptor-positive subsets. Interestingly, large populations of CD4CD8 T cells were observed in the samples of ovarian cancer and a high percentage of these cells were positive for CD94/NKG2A. These cells are currently the subject of further investigation. When five cervical cancer and four ovarian cancer samples were analyzed, up to 50% of CTLs were CD94/NKG2A+ with a median of 12% (Fig. 2B). The frequency of CD94/NKG2A+ CTLs in age-matched normal blood was found to be around 3%, indicating that this inhibiting HLA-E binding receptor is enriched at the site of the tumor. To substantiate this finding and to analyze the localization of these CD94/NKG2A+ CTLs, we performed triple stainings on cryosections of cervical cancer using fluorescently labeled antibodies to CD3, CD94, and NKG2A (Fig. 3). Most T cells resided in stoma areas and not within tumor nests, in line with our previous findings (20, 21). Strikingly, CD94/NKG2A expression was found on only 6% of the stoma T cells, whereas 48% of intraepithelial T cells displayed this inhibiting receptor (SD, 9 and 32%, respectively; P = 0.0032, Student's t test). Together, these data implied that the frequency of tumor-interacting T cells expressing CD94/NKG2A (Fig. 3) is much higher than anticipated on the basis of the total pool of T cells in the resected tumor sample (Fig. 2B).

Fig. 2.
Flow cytometry analysis of CD94, NKG2A, and NKG2C expression on T cells. Dissociated tissues from fresh tumor samples from surgery were stained with fluorescently labeled antibodies against CD14, CD56, CD3, CD4, CD8, CD94, NKG2A, and NKG2C. (A) Eight-color ...
Fig. 3.
Triple fluorescence staining of cervical cancer detecting intraepithelial CD94+NKG2A+ T cells. Immunofluorescent staining of T cells (CD3+ in blue) expressing NKG2A (in green) and CD94 (in red). These two pictures of different cervical tumor samples are ...

Expression of HLA-E Neutralizes Survival Benefit of Infiltrating CTLs in Ovarian Cancer.

We wondered whether the observed expression of HLA-E and CD94/NKG2A in the tumor site would translate into survival differences in the context of CTL infiltration. In ovarian cancer, HLA-E expression on its own did not affect survival (Table 2). We previously demonstrated (20) that high CTL counts do predict improved survival in ovarian cancer [hazard ratio (HR) 0.71, Table 2 and Fig. 4A]. We hypothesized that, due to the presence of CD94/NKG2A on infiltrating CTL, HLA-E high tumors might resist CTL mediated lysis. To this end, we performed survival analysis for CTL infiltration stratified by HLA-E expression. Indeed, the prognostic benefit of CD8+ T cells was strongly present in the stratum with low HLA-E expression (HR 0.53, P = 0.001, Table 2 and Fig. 4B). This hazard ratio was much lower than that of the whole population, without HLA-E stratification. Strikingly, patients with high HLA-E expression, representing 75% of our cohort, completely lost the benefit of infiltrating CTLs (HR 0.97, P = 0.816, Table 2 and Fig. 4C). These data indicate that the minor subpopulation of patients with low HLA-E expression on their tumors benefits from infiltrating CTLs and, moreover, that expression of HLA-E neutralizes the survival benefit of ovarian cancers with high numbers of CTLs.

Table 2.
Cox regression survival analysis
Fig. 4.
Kaplan–Meier survival curves of ovarian cancer. Overall survival in months of 249 patients with ovarian cancer for whom two or more cores were available is plotted. (A) Infiltrating CD8+ T cells were counted and stratified in two groups with a ...

In cervical cancer, we observed a decreased risk of death associated with high HLA-E expression in univariate analysis. However, HLA-E expression was not an independent predictor of death in multivariate analysis (Table 2). We previously reported that infiltrating CTL frequency is not an independent predictive survival factor (P = 0.879, Table 2) (21). Stratified analysis of CTL infiltration based on HLA-E expression did not affect these results. When repeating these analyses for disease-free survival, similar results were obtained.

A notable difference between ovarian and cervical cancer is the number of intratumoral CTLs, as cervical cancers are infiltrated with at least three times more CTLs (median 95.3 ± 221.6/mm2; ovarian cancer, 28.3 ± 120/mm2; P < 0.001), suggesting that the virus-positive cervical cancers are relatively overloaded with infiltrating CTLs. When we repeated the stratified analysis in the subpopulation of cervical cancer with CTL counts comparable to ovarian cancer, HLA-E expression seemed to have the same impact as in ovarian cancer. However, the numbers of cervical cancer with such low numbers of CTLs were insufficient for proper statistical analysis. We are currently further evaluating the differences between CTL numbers in several tumor types.

In conclusion, HLA-E is regularly expressed in ovarian and cervical cancer, often concurrently with classical MHC molecules. Instead of inhibiting NK cells, which are hardly present in these tumor types, the main role of HLA-E seems to be the inhibition of infiltrating CD8+ CTLs. This effect translates into survival differences in ovarian cancer, which contains fewer CTLs and might therefore be more affected by a decrease of CTLs below a certain threshold.

Discussion

In the current study, we determined the clinical and immunological relevance of HLA-E expression in ovarian and cervical cancer. Knowledge on the expression of HLA-E in these two cancer types was limited to small cohorts, and here we show that 89.4% of ovarian cancers and 83.7% of cervical cancers display higher levels compared with their normal epithelial counterparts. Total lack of HLA-E is rare in these tumors. Importantly, HLA-E protein expression was strongly associated with expression of classical HLA molecules (class I and class II) and components of the antigen processing machinery (immunoproteasome, peptide transporter TAP, trimming enzyme ERAP, and chaperone Erp57) (Table 1). This association implies that tumor expression of HLA-E is regulated in a comparable fashion to classical HLA and that its presence on tumors is not a defense mechanism against NK cell-mediated lysis in classical class I-negative tumors, as sometimes suggested in the literature (26, 2931). Instead, our data argue that HLA-E expression arises in the setting of an intact antigen processing apparatus and, in ovarian cancer, abundant CTL infiltration. A positive association between classical and nonclassical HLA expression has recently also been reported for a large cohort of breast cancers (32) and is moreover anticipated on the basis of the stabilization of HLA-E by leader peptides derived from classical HLA molecules (2, 33).

Traditionally, interaction with NK cells via receptors CD94/NKG2A and CD94/NKG2C was considered the main purpose of HLA-E. The presence of infiltrating NK cells in ovarian and cervical cancers was previously reported by several groups (3439). Detection of NK cells in tumor samples has predominantly been performed with antibodies against CD56 and CD57, whereas these molecules can also be found on T lymphocytes. We carefully analyzed NK infiltration by inclusion of the CD3-specific T lymphocyte marker or using the really specific molecule NKp46, which is not expressed on T lymphocytes (28). Our data reveal that NK cells hardly infiltrate ovarian and cervical cancers, in line with the general impression in solid tumors (40), in contrast to leukemias, where NK cell responses have been connected to better survival (41).

In addition to NK cells, the inhibiting receptor CD94/NKG2A and activating receptor CD94/NKG2C are expressed by minor populations of CD8+ T cells. Although this subset is generally very scarce in peripheral blood mononuclear cells (PBMCs) of healthy subjects (~4%) (Fig. 2) (32, 42), the frequency of CD94/NKG2A expressing CD8+ T cells is much higher in tumor infiltrating lymphocytes, as shown in our study and by others (43, 44).

Interestingly, the immunosuppressive cytokine TGF-β, which is regularly detected in ovarian and cervical cancer (4547), seems to induce this inhibiting receptor on T cells (44). Several studies have shown that the inhibiting receptor CD94/NKG2A dampens the incoming activation signals of T cells by recruitment of phosphatases like SHP-1 to the signal transducing synaps, resulting in decreased effector functions (1, 44, 48). Strikingly, the activating receptor CD94/NKG2C was absent on tumor-infiltrating T cells (Fig. 2), whereas it is expressed in other inflammatory situations (4951). This implies that expression of the NKG2 chains is differentially and independently regulated and that NKG2A is selectively up-regulated in tumors.

Protein expression of HLA-E was previously analyzed on cultured cancer cell lines and small cohorts of surgical specimen of some cancer types (16, 26, 5254). HLA-E expression was correlated with increased infiltration of CD8+ CTLs in glioblastoma (53) and decreased infiltration of NK cells as well as a worse progression-free survival in colorectal cancer (26). In cervical cancer, HLA-E expression seemed to gradually increase from cervical intraepithelial neoplasia (CIN) I to invasive cervical cancer (54). Intriguingly, we and others (55) found no associations with tumor stage or grade. We have to note, however, that our cervical cancer cohort represented early stage patients with relatively highly differentiated tumors, whereas the ovarian cancer cohort consisted of mostly late stage, high grade tumors. The expression pattern and frequency of HLA-E was quite similar in our two studied cancer types as well as its positive association with antigen presenting molecules. The effect on survival, however, was clearly different. High HLA-E expression in ovarian cancer appeared to neutralize the beneficial effect of CTL infiltration. These results are in line with the in vitro data by Malmberg et al. (56), who demonstrated that HLA-E on freshly isolated ovarian cancer cells was up-regulated by IFN-γ treatment, resulting in a CD94/NKG2A-mediated resistance to CTL lysis. However, in cervical cancer, HLA-E did not influence the prognostic effects of CTLs or the CTL/Treg ratio. This difference might be explained by the significantly higher numbers of infiltrating CTLs in cervical cancer. At least three times more intratumoral CTLs can be found in this tumor type (20, 21, 57), which is most likely the result of the presence of viral antigens from HPV and an active inflammatory response.

In conclusion, our results suggest that HLA-E expression in ovarian and cervical cancer is the result of a smoldering inflammatory response. This emerging concept (58) entails the presence of an inflammatory milieu that can either promote tumor progression or antitumor activity. The inhibiting impact of HLA-E in cervical cancer is limited, due to beneficial signs of inflammation such as high CTL infiltrate, strong viral antigens, and stimulating HLA ligands (MICA and classical HLA). In ovarian cancer, the presence of HLA-E is able to neutralize the protective role of the relatively scarce intratumoral CTLs (1921, 59, 60).

Materials and Methods

Patient Material.

Ovarian cancers were selected from primary surgery by a gynecological oncologist from the University Medical Center Groningen between May 1985 and June 2006 and paraffin embedded in a TMA (n = 270). Cervical cancers were taken from radical hysterectomy with complete pelvic lymphadenectomy in the Leiden University Medical Center from 1985 to 1999, without previous radiotherapy or chemotherapy. Tissues were paraffin embedded in a TMA (n = 150). Further information is provided in SI Materials and Methods.

Immunohistochemistry.

TMA sections were stained with mouse monoclonal antibodies recognizing HLA-E (clone MEM-E/02; Abcam; ab2216). To detect NK cells, paraffin sections were stained with anti-CD56 (clone 1B6, Monosan) and anti-NKp46 (polyclonal AF1850; R&D Systems). Simultaneous detection of CD3, CD94, and NKG2A was performed by three color fluorescence staining on 10 cryosections of cervical carcinomas using anti-CD3 (mouse IgG1; Dako; clone F7.2.38), anti-CD94 (mouse IgG2a; Abcam; clone ab61974), and anti-NKG2A (mouse IgG2b; Immunotech; clone Z199). Second step antibodies were Alexa fluorochrome-labeled goat antimouse isotype-specific antibodies. Details are given in SI Materials and Methods.

Flow Cytometry Analyses.

Fresh ovarian and cervical cancer specimens were dissected in small fragments with surgical blades and passed through a cell strainer to obtain single cell suspensions. CD94, NKG2A, and NKG2C expression on T cells was analyzed by eight-color flow cytometry as described in SI Materials and Methods.

Statistics.

The statistical tests are described in the legends in Figs. 14 and in SI Materials and Methods.

Supplementary Material

Supporting Information:

Acknowledgments

The authors thank Claudia Cunha Oliveira for critical reading of the manuscript. Financial support was received from the Dutch Cancer Society (UL 2007-3897; RUG 2007-3919).

Footnotes

The authors declare no conflict of interest.

*This Direct Submission article had a prearranged editor.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1100354108/-/DCSupplemental.

References

1. Rodgers JR, Cook RG. MHC class Ib molecules bridge innate and acquired immunity. Nat Rev Immunol. 2005;5:459–471. [PubMed]
2. van Hall T, Oliveira CC, Joosten SA, Ottenhoff TH. The other Janus face of Qa-1 and HLA-E: Diverse peptide repertoires in times of stress. Microbes Infect. 2010;12:910–918. [PubMed]
3. Grimsley C, et al. Definitive high resolution typing of HLA-E allelic polymorphisms: Identifying potential errors in existing allele data. Tissue Antigens. 2002;60:206–212. [PubMed]
4. Strong RK, et al. HLA-E allelic variants. Correlating differential expression, peptide affinities, crystal structures, and thermal stabilities. J Biol Chem. 2003;278:5082–5090. [PubMed]
5. O'Callaghan CA, et al. Structural features impose tight peptide binding specificity in the nonclassical MHC molecule HLA-E. Mol Cell. 1998;1:531–541. [PubMed]
6. Lee N, Goodlett DR, Ishitani A, Marquardt H, Geraghty DE. HLA-E surface expression depends on binding of TAP-dependent peptides derived from certain HLA class I signal sequences. J Immunol. 1998;160:4951–4960. [PubMed]
7. Braud VM, Allan DS, Wilson D, McMichael AJ. TAP- and tapasin-dependent HLA-E surface expression correlates with the binding of an MHC class I leader peptide. Curr Biol. 1998;8:1–10. [PubMed]
8. Braud VM, et al. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature. 1998;391:795–799. [PubMed]
9. Speiser DE, et al. CD28-negative cytolytic effector T cells frequently express NK receptors and are present at variable proportions in circulating lymphocytes from healthy donors and melanoma patients. Eur J Immunol. 1999;29:1990–1999. [PubMed]
10. Zhou J, Matsuoka M, Cantor H, Homer R, Enelow RI. Cutting edge: Engagement of NKG2A on CD8+ effector T cells limits immunopathology in influenza pneumonia. J Immunol. 2008;180:25–29. [PubMed]
11. Hu D, et al. Analysis of regulatory CD8 T cells in Qa-1-deficient mice. Nat Immunol. 2004;5:516–523. [PubMed]
12. Mingari MC, et al. HLA class I-specific inhibitory receptors in human T lymphocytes: Interleukin 15-induced expression of CD94/NKG2A in superantigen- or alloantigen-activated CD8+ T cells. Proc Natl Acad Sci USA. 1998;95:1172–1177. [PMC free article] [PubMed]
13. Bertone S, et al. Transforming growth factor-beta-induced expression of CD94/NKG2A inhibitory receptors in human T lymphocytes. Eur J Immunol. 1999;29:23–29. [PubMed]
14. Menier C, et al. Characterization of monoclonal antibodies recognizing HLA-G or HLA-E: New tools to analyze the expression of nonclassical HLA class I molecules. Hum Immunol. 2003;64:315–326. [PubMed]
15. Coupel S, et al. Expression and release of soluble HLA-E is an immunoregulatory feature of endothelial cell activation. Blood. 2007;109:2806–2814. [PubMed]
16. Derré L, et al. Expression and release of HLA-E by melanoma cells and melanocytes: Potential impact on the response of cytotoxic effector cells. J Immunol. 2006;177:3100–3107. [PubMed]
17. Perera L, et al. Expression of nonclassical class I molecules by intestinal epithelial cells. Inflamm Bowel Dis. 2007;13:298–307. [PubMed]
18. Ruiter DJ, et al. Quality control of immunohistochemical evaluation of tumour-associated plasminogen activators and related components. European BIOMED-1 Concerted Action on Clinical Relevance of Proteases in Tumour Invasion and Metastasis. Eur J Cancer. 1998;34:1334–1340. [PubMed]
19. Leffers N, et al. Down-regulation of proteasomal subunit MB1 is an independent predictor of improved survival in ovarian cancer. Gynecol Oncol. 2009;113:256–263. [PubMed]
20. Leffers N, et al. Prognostic significance of tumor-infiltrating T-lymphocytes in primary and metastatic lesions of advanced stage ovarian cancer. Cancer Immunol Immunother. 2009;58:449–459. [PubMed]
21. Jordanova ES, et al. Human leukocyte antigen class I, MHC class I chain-related molecule A, and CD8+/regulatory T-cell ratio: Which variable determines survival of cervical cancer patients? Clin Cancer Res. 2008;14:2028–2035. [PubMed]
22. Mehta AM, Jordanova ES, Kenter GG, Ferrone S, Fleuren GJ. Association of antigen processing machinery and HLA class I defects with clinicopathological outcome in cervical carcinoma. Cancer Immunol Immunother. 2008;57:197–206. [PMC free article] [PubMed]
23. Kloetzel PM, Ossendorp F. Proteasome and peptidase function in MHC-class-I-mediated antigen presentation. Curr Opin Immunol. 2004;16:76–81. [PubMed]
24. Satoh A, et al. Epigenetic inactivation of class II transactivator (CIITA) is associated with the absence of interferon-gamma-induced HLA-DR expression in colorectal and gastric cancer cells. Oncogene. 2004;23:8876–8886. [PubMed]
25. Wright KL, Ting JP. Epigenetic regulation of MHC-II and CIITA genes. Trends Immunol. 2006;27:405–412. [PubMed]
26. Levy EM, et al. Human leukocyte antigen-E protein is overexpressed in primary human colorectal cancer. Int J Oncol. 2008;32:633–641. [PubMed]
27. Braud V, Jones EY, McMichael A. The human major histocompatibility complex class Ib molecule HLA-E binds signal sequence-derived peptides with primary anchor residues at positions 2 and 9. Eur J Immunol. 1997;27:1164–1169. [PubMed]
28. Walzer T, et al. Identification, activation, and selective in vivo ablation of mouse NK cells via NKp46. Proc Natl Acad Sci USA. 2007;104:3384–3389. [PMC free article] [PubMed]
29. Dutta N, Majumder D, Gupta A, Mazumder DN, Banerjee S. Analysis of human lymphocyte antigen class I expression in gastric cancer by reverse transcriptase-polymerase chain reaction. Hum Immunol. 2005;66:164–169. [PubMed]
30. Dutta N, Gupta A, Mazumder DN, Banerjee S. Down-regulation of locus-specific human lymphocyte antigen class I expression in Epstein-Barr virus-associated gastric cancer: Implication for viral-induced immune evasion. Cancer. 2006;106:1685–1693. [PubMed]
31. Bianchini M, et al. Comparative study of gene expression by cDNA microarray in human colorectal cancer tissues and normal mucosa. Int J Oncol. 2006;29:83–94. [PubMed]
32. de Kruijf EM, et al. HLA-E and HLA-G expression in classical HLA class I-negative tumors is of prognostic value for clinical outcome of early breast cancer patients. J Immunol. 2010;185:7452–7459. [PubMed]
33. Sullivan LC, Clements CS, Rossjohn J, Brooks AG. The major histocompatibility complex class Ib molecule HLA-E at the interface between innate and adaptive immunity. Tissue Antigens. 2008;72:415–424. [PubMed]
34. Li K, et al. Clinical significance of the NKG2D ligands, MICA/B and ULBP2 in ovarian cancer: High expression of ULBP2 is an indicator of poor prognosis. Cancer Immunol Immunother. 2009;58:641–652. [PubMed]
35. Liu M, et al. Classification using hierarchical clustering of tumor-infiltrating immune cells identifies poor prognostic ovarian cancers with high levels of COX expression. Mod Pathol. 2009;22:373–384. [PubMed]
36. Dong HP, et al. NK- and B-cell infiltration correlates with worse outcome in metastatic ovarian carcinoma. Am J Clin Pathol. 2006;125:451–458. [PubMed]
37. Garcia-Iglesias T, et al. Low NKp30, NKp46 and NKG2D expression and reduced cytotoxic activity on NK cells in cervical cancer and precursor lesions. BMC Cancer. 2009;9:186. [PMC free article] [PubMed]
38. Textor S, et al. Activating NK cell receptor ligands are differentially expressed during progression to cervical cancer. Int J Cancer. 2008;123:2343–2353. [PubMed]
39. Papadopoulos N, et al. Gains and losses of CD8, CD20 and CD56 expression in tumor stroma-infiltrating lymphocytes compared with tumor-associated lymphocytes from ascitic fluid and lymphocytes from tumor draining lymph nodes in serous papillary ovarian carcinoma patients. Eur J Gynaecol Oncol. 2002;23:533–536. [PubMed]
40. Waldhauer I, Steinle A. NK cells and cancer immunosurveillance. Oncogene. 2008;27:5932–5943. [PubMed]
41. Velardi A, Ruggeri L, Mancusi A, Aversa F, Christiansen FT. Natural killer cell allorecognition of missing self in allogeneic hematopoietic transplantation: A tool for immunotherapy of leukemia. Curr Opin Immunol. 2009;21:525–530. [PubMed]
42. Tilburgs T, et al. Expression of NK cell receptors on decidual T cells in human pregnancy. J Reprod Immunol. 2009;80:22–32. [PubMed]
43. Chang WC, et al. Expression of inhibitory natural killer receptors on tumor-infiltrating CD8+ T lymphocyte lineage in human endometrial carcinoma. Int J Gynecol Cancer. 2005;15:1073–1080. [PubMed]
44. Sheu BC, et al. Up-regulation of inhibitory natural killer receptors CD94/NKG2A with suppressed intracellular perforin expression of tumor-infiltrating CD8+ T lymphocytes in human cervical carcinoma. Cancer Res. 2005;65:2921–2929. [PubMed]
45. Henriksen R, et al. Expression and prognostic significance of TGF-beta isotypes, latent TGF-beta 1 binding protein, TGF-beta type I and type II receptors, and endoglin in normal ovary and ovarian neoplasms. Lab Invest. 1995;73:213–220. [PubMed]
46. Santin AD, et al. Increased levels of interleukin-10 and transforming growth factor-beta in the plasma and ascitic fluid of patients with advanced ovarian cancer. BJOG. 2001;108:804–808. [PubMed]
47. Hazelbag S, Gorter A, Kenter GG, van den Broek L, Fleuren G. Transforming growth factor-beta1 induces tumor stroma and reduces tumor infiltrate in cervical cancer. Hum Pathol. 2002;33:1193–1199. [PubMed]
48. Lanier LL. NK cell recognition. Annu Rev Immunol. 2005;23:225–274. [PubMed]
49. Gumá M, et al. The CD94/NKG2C killer lectin-like receptor constitutes an alternative activation pathway for a subset of CD8+ T cells. Eur J Immunol. 2005;35:2071–2080. [PubMed]
50. Meresse B, et al. Reprogramming of CTLs into natural killer-like cells in celiac disease. J Exp Med. 2006;203:1343–1355. [PMC free article] [PubMed]
51. van Stijn A, et al. Human cytomegalovirus infection induces a rapid and sustained change in the expression of NK cell receptors on CD8+ T cells. J Immunol. 2008;180:4550–4560. [PubMed]
52. Marín R, et al. Analysis of HLA-E expression in human tumors. Immunogenetics. 2003;54:767–775. [PubMed]
53. Mittelbronn M, et al. Elevated HLA-E levels in human glioblastomas but not in grade I to III astrocytomas correlate with infiltrating CD8+ cells. J Neuroimmunol. 2007;189:50–58. [PubMed]
54. Gonçalves MA, et al. Classical and non-classical HLA molecules and p16(INK4a) expression in precursors lesions and invasive cervical cancer. Eur J Obstet Gynecol Reprod Biol. 2008;141:70–74. [PubMed]
55. Hanak L, et al. Expression pattern of HLA class I antigens in renal cell carcinoma and primary cell line cultures: Methodological implications for immunotherapy. Med Sci Monit. 2009;15:CR638–CR643. [PubMed]
56. Malmberg KJ, et al. IFN-gamma protects short-term ovarian carcinoma cell lines from CTL lysis via a CD94/NKG2A-dependent mechanism. J Clin Invest. 2002;110:1515–1523. [PMC free article] [PubMed]
57. Cannistra SA. Cancer of the ovary. N Engl J Med. 2004;351:2519–2529. [PubMed]
58. Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A. Cancer-related inflammation, the seventh hallmark of cancer: Links to genetic instability. Carcinogenesis. 2009;30:1073–1081. [PubMed]
59. Sato E, et al. Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc Natl Acad Sci USA. 2005;102:18538–18543. [PMC free article] [PubMed]
60. Karim R, et al. Tumor-expressed B7-H1 and B7-DC in relation to PD-1+ T-cell infiltration and survival of patients with cervical carcinoma. Clin Cancer Res. 2009;15:6341–6347. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...