Virchows Archiv

, Volume 460, Issue 1, pp 47–60 | Cite as

MHC class II expression in pancreatic tumors: a link to intratumoral inflammation

  • Matthias M. Gaida
  • Thilo Welsch
  • Esther Herpel
  • Darjus F. Tschaharganeh
  • Lars Fischer
  • Peter Schirmacher
  • G. Maria Hänsch
  • Frank Bergmann
Original Article

Abstract

Major histocompatibility complex class II antigens (MHC class II) are constitutively expressed by professional antigen presenting cells and present antigenic peptides to specific CD4+ T lymphocytes. MHC class II expression, however, can also be induced on epithelial cells and in a variety of solid tumors. We tested MHC class II expression on tissue samples derived from patients with pancreatic ductal adenocarcinoma (PDAC) and pancreatic endocrine tumors (PET). Immunohistochemistry revealed MHC class II expression in 86 of 112 (76.8%) PDAC samples and in 30 of 43 (70.0%) PET samples. In PDAC and PET, MHC class II expression correlated significantly with severity and activity of intratumoral inflammation, as well as with the infiltration of CD4+ T lymphocytes. High MHC class II expression significantly correlated with a better histological grade of differentiation in PDAC. In vitro MHC class II expression could be induced on PDAC tumor cell lines by interferon-γ. These cells were then able to present the staphylococci enterotoxin B superantigen to T lymphocytes, which resulted in T cell proliferation. Our findings suggest that MHC class II expression on pancreatic tumor cells is induced by the intratumoral inflammatory reaction in pancreatic tumors.

Keywords

Pancreatic cancer Pancreatic endocrine tumor MHC class II Intratumoral inflammation T cell proliferation 

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive neoplasm, ranking fourth in cancer-related deaths in western countries [1]. Tumors are significantly influenced by the cross-talk among the tumor cells and by their microenvironment, particularly the tissue-resident cells such as fibroblasts, or the tumor-infiltrating lymphocytes and the inflammatory cells [2]. The intratumoral inflammation in the PDAC microenvironment is often distinct, and multiple interactions between tumor cells and cells of the immune system have been previously described, but the impact on tumor progression is not yet understood [3, 4, 5]. Systemically increased C-reactive protein concentrations, a marker for an active inflammatory constitution, correlates with shorter survival of PDAC patients [6]. Moreover, an intensive interplay between pancreatic tumor cells and inflammatory cells has been reported [3]. Cytokines released by tumor infiltrating lymphocytes were shown to enhance the invasion of pancreatic cancer cells [4, 7], while neutrophil-derived lipocalin provided an anti-tumoral and anti-metastatic effect in PDAC [8].

Key findings include the fact that PDAC cells are specifically recognized by autologous T lymphocytes [9] and that chemokines or their receptors, such as CXCL14 or CXCL16 and its corresponding receptor CXCR6, are found in PDAC that could mediate interactions with inflammatory cells in particular [10, 11, 12, 13]. Moreover, major histocompatibility antigen (MHC) class I and class II are found on a variety of tumor cells, including pancreatic adenocarcinoma cells [14], and it has been assumed that they provide the immunologic recognition structure of the tumor by presenting either an “altered self” or a “non-self antigen” to T lymphocytes [15, 16].

The constitutive MHC class II expression is restricted to professional antigen presenting cells, such as dendritic cells, monocytes, and B lymphocytes. MHC class II expression, however, can be induced in non-neoplastic tissue under inflammatory conditions, particularly on epithelial cells like keratinocytes or intestinal mucosa [17, 18]. MHC class II antigens are also expressed by human malignancies, e.g. in Ewing sarcoma [15], malignant melanoma [19], renal cell carcinoma [20], colon carcinoma [21], breast cancer [22], and squamous cell carcinoma of the larynx [22, 23]. MHC class II expression has been associated with better tumor differentiation and a better prognosis, e.g. in colon, breast and larynx carcinoma [21, 22, 23, 24].

Expression of MHC class II in PDAC tissue was indicated by two preliminary studies, comprising only a small number of patients [14, 25], on isolated cells of primary PDAC [9], and on pancreatic adenocarcinoma cell lines [14]. These first studies reported that approximately 30% of PDAC tissues revealed MHC class II positivity [14, 25]. In the present study, we examined MHC class II expression in a large series of 112 PDAC patients and for comparison additionally of 43 patients with PET and assessed the correlation with clinical and pathological parameters focussing the intratumoral inflammation. To gain insight into the functional role of MHC class II expression in tumors, we tested whether pancreatic cancer cells that were induced to express MHC class II molecules were able to activate T lymphocytes using staphylococcus enterotoxin B as a model superantigen [26]. We found a strong correlation between MHC class II expression by pancreatic tumor cells and the intratumoral inflammatory reaction, particularly with respect to the infiltration of CD4+ T lymphocytes. Moreover, the in vitro data provided evidence for the propensity of MHC class II positive tumor cells to induce the proliferation of T lymphocytes. Taken together, our data suggest that the proinflammatory microenvironment induces the MHC class II expression on the tumor, and the tumor cell, in turn, might participate in a localized immune reaction.

Material and methods

Patients

PDAC tumor tissue samples were obtained from 112 patients (46 female, 66 male; age range, 39–85 years; mean, 64.9 years; median, 66.0 years). In 84 patients the tumors were located in the pancreatic head, 7 in the body, 13 in the body and tail, and 8 were in the tail. The tissue specimens were formalin-fixed and paraffin-embedded.

Based on haematoxylin and eosin (H&E) staining, the diagnosis of PDAC and the tumor stage were established according the criteria recommended by the World Health Organization (WHO) (2010) [27] and the UICC criteria (2009) [28]. Pathological examination revealed a pT3 stage in 110 patients and pT1 and pT2 stages in one case each. Regional lymph node metastases (pN1) occurred in 98 patients and metastases to other organs (liver and/or non-regional lymph nodes) were seen in 13 cases (pM1). Four PDAC were well-differentiated (G1), 75 moderately (G2) and 33 poorly differentiated (G3). Follow-up information was available for 104 patients: 61 patients died from the cancer within 25–1,187 days after the operation (mean, 427 days; median, 347 days), 37 patients were alive after a follow-up of 15–1,044 days (mean, 551 days; median, 663 days), and six patients died of non-cancer-related disease and were thus excluded from further analysis (Table 1).
Table 1

Clinical and pathological parameters of PDAC patients

Patients (n = 112)

PDAC

Gender (F/M) (n = 112)

46, 66

Age [years]

39–85 (mean, 64.9; median, 66.0)

Location of tumor

Head, 84

Body, 7

Body and tail, 13

Tail, 8

pT

pT1, 1

pT2, 1

pT3, 110

pT4, 0

pN

pN0, 14

pN1, 98

pM

pM0, 99

pM1, 13

G

G1, 4

G2, 75

G3, 33

Survival (n = 104 patients)

Death of disease, 61 patients; 25–1,187 days (mean, 427; median, 347)

Alive, 37 patients; 15–1,044 days (mean, 551; median, 663)

Non-cancer-related death, 6 patients

The study furthermore comprised 43 samples of primary PET (21 females, 22 males; age range, 13–85 years; mean, 56.0 years; median, 60 years). In 23 cases, the tumors were located in the pancreatic head, four were in the body, five were in the body and tail, and 11 cases were found in the tail. Using the criteria established by the World Health Organization of 2004 [29] and of 2010 [30] and the criteria of the European Neuroendocrine Tumor Society (ENETS) [31], diagnosis and tumor stage were established. The PET were classified as benign well-differentiated endocrine tumors (n = 6), well-differentiated endocrine tumors of uncertain behavior (n = 8), well-differentiated endocrine carcinoma (n = 25) and poorly differentiated endocrine carcinoma (n = 4). According to the WHO classification of 2010, 35 cases were classified as neuroendocrine tumor and eight as neuroendocrine carcinoma. No case revealed a pT4, 28 cases a pT3, six cases a pT2 and nine cases a pT1 stage. Regional lymph node metastases (pN1) were detected in 20 cases. Distant organ metastases (liver and/or non-regional lymph nodes) were observed in nine cases (pM1). According to their mitotic activity and proliferative activity, 21 cases were graded as G1, 14 cases as G2, and 8 as G3. Immunohistochemistry revealed positivity for the neuroendocrine markers synaptophysin and chromogranin A in all 43 cases. All 43 cases were immunohistochemically evaluated for hormone expression. Immunohistochemical stains showed expression of insulin in 12 cases, gastrin in ten cases, glucagon, pancreatic polypeptide and somatostatin in three cases, each, and serotonin in one case. Of these tumors, functional activity was clinically determined in 16 patients (37% of patients, 11 insulinomas, 4 gastrinomas, one glucagonoma). Twenty-seven cases clinically revealed no functional activity (63% of patients). Four patients had a hereditary syndrome (MEN 1 syndrome). Follow-up information was available for 37 patients: eight patients died of their tumors within 1–1,258 days postoperatively (mean, 477 days; median, 260 days) and 29 patients were alive after a follow-up of 47 to 4,980 days (mean, 1,480 days; median, 1,466 days) (Table 2).
Table 2

Clinical and pathological parameters of PET patients

Patients (n = 43)

PET

Gender (F/M) (n = 43)

21, 22

Age [years]

13–85 (mean, 56.0; median, 60.0)

Location of tumor

Head, 23

Body, 4

Body and tail, 5

Tail, 11

Diagnosis (WHO 2004)

Well-differentiated endocrine tumor, 6

Well-differentiated endocrine tumor of uncertain behaviour, 8

Well-differentiated endocrine carcinoma, 25

Poorly differentiated endocrine carcinoma, 4

Diagnosis (WHO 2010)

Neuroendocrine tumor, 35

Neuroendocrine carcinoma, 8

pT

pT1, 9

pT2, 6

pT3, 28

pT4, 0

pN

pN0, 10

pN1, 20

pNx, 13

pM

pM0, 34

pM1, 9

G

G1, 21

G2, 14

G3, 8

Endocrine markers

Chromogranin A, 43

Synaptophysin, 43

Immunohistochemical hormone expression

Insulin, 12 (27.9% of 43 cases)

Gastrin, 10 (23.2% of 43 cases)

Glucagon, 3 (7.0% of 43 cases)

Pancreatic polypeptide, 3 (7.0% of 43 cases)

Somatostatin, 3 (7.0% of 43 cases)

Serotonin, 1 (2.3% of 43 cases)

Clinically determined functional activity

Insulinoma, 11 (25.6% of 43 cases)

Gastrinoma, 4 (9.3% of 43 cases)

Glucagonoma, 1 (2.3% of 43 cases)

Hereditary syndrome

4

Survival (n = 37 patients)

Death of disease, 8 patients; 1–1,258 days (mean, 477; median, 260)

Alive, 29 patients; 47–4,980 days (mean, 1,480; median, 1,466)

The study was approved by the Ethics Committee of the University of Heidelberg and written informed consent was obtained from the patients.

Immunohistochemistry

Paraffin-embedded tissue sections (4 μm) were used for the immunohistochemical analyses. Immunostaining was performed as previously described [32], using the avidin–biotin complex method. Prior to antibody incubation, heat pre-treatment in an antigen retrieval solution (DAKO; pH 9.0) was performed. Primary antibodies included a mouse monoclonal antibody to MHC class II (Abcam, Cambridge, UK; diluted 1:250) and a mouse monoclonal antibody to CD4 (Novocastra, Newcastle, UK; diluted 1:10). MHC class II staining was performed on tissue microarrays from 112 PDAC samples, 43 PET samples and ten normal pancreas samples. To validate these immunohistochemical results obtained from the microarrays, 16 and 26 of the cases were additionally stained for MHC class II and CD4, respectively, using whole tissue tumor sections, revealing comparable results.

Scoring of inflammation, CD4+ T lymphocyte infiltration and MHC class II expression

The severity of the inflammation was evaluated microscopically on whole tumor sections, using a previously reported scoring system [33]. Briefly, the severity was determined as absent (score, 0), mild (score, 1), moderate (score, 2) or severe (score, 3), depending on the accumulation of inflammatory cells (lymphocytes, plasma cells, macrophages) and the formation of lymph follicles. The activity of inflammation was semiquantitatively scored as absent (score, 0), mild (score, 1) or moderate to severe (score, 2), depending on the density of neutrophil granulocytes.

The immunohistochemical MHC class II expression (distribution) of the tumor cells was determined semiquantitatively as score, 0 for 0%; score, 1 for 1–5%; score, 2 for 5–25%; score, 3 for 25–50%; score, 4 for 50–75% and score, 5 for 75–100% positive tumor cells according to a previously reported scale for the immunohistochemical evaluation of the expression of MHC molecules [15, 34, 35]. This scoring system includes also a semiquantification of the staining intensity, as grouped in no staining (score, 0), weak staining intensity (score, 1), moderate staining intensity (score, 2) or strong staining intensity (score, 3). According to Berghuis et al. we used an immunoreactivity score (in accordance to the Allred-Score), in which the summation of both scores was composed [15].

CD4+ T lymphocytes have been described to interact with MHC class II molecules [36]. To test the correlation between the MHC class II expression on PDAC and PET and the infiltration of CD4+ lymphocytes, the CD4+ T lymphocytes were counted in ten representative high power fields (HPF) of each case in 112 PDAC and in 21 PET cases.

TMA specimen and the corresponding whole tissue sections were compared, and the ratio between the mean number of CD4+ infiltrating cells in sections with high MHC class II expression and those without any expression was calculated. The ratio (CD4+ lymphocytes in high MHC class II expressing tumors/CD4+ lymphocytes in non-MHC class II expressing tumors) was 4.0 in whole tissue sections and 3.8 in TMA; thus a valid tool for the quantification of CD4+ lymphocytes was generated. Next, we compared the expression pattern and intensity of MHC class II with the number of infiltrating CD4+ T lymphocytes. Furthermore, we correlated the infiltrate of CD4+ lymphocytes in tumors with MHC class II expression (irrespective of the quantity of positively staining cells) versus tumors without any expression. Moreover, we selected 13 cases without MHC class II expression and 13 cases with notably high MHC class II expression (more than 75% positive cells, moderate to strong staining intensity) to compare the infiltration of CD4+ T lymphocytes.

Cytofluorometry and induction of MHC class II expression on pancreatic cancer cell lines

The pancreatic tumor cell lines Capan-1, MiaPaca-2, BxPC-3, Panc-1, SU8686, AspC1, (ATCC, Rockville, MD, USA), T3M4 and Colo-357 (R.S. Metzgar Duke University, Durham, NC, USA) were cultivated in RPMI 1640, supplemented with 10 % fetal calf serum, 1% l-glutamine and 1% penicillin/streptomycin (all obtained from Invitrogen, Karlsruhe, Germany). For induction of MHC class II expression, the cells were seeded into 6-well plates (NUNC, Roskilde, Denmark; 1 × 105/ml), and cultivated with interferon-γ (Serotec, Düsseldorf, Germany; 100 U/ml) for 24 and 48 h, respectively. Next, the cells were washed, removed from the plates by treatment with EDTA/Trypsin (Invitrogen) and suspended in phosphate-buffered saline (PBS) containing 1% bovine serum albumin and 0.1% sodium azide. To detect MHC class II surface expression, 5 μl PE-labelled antibody (anti-HLA-DR-DQ-DP; Biozol, Eching, Germany) was added. Mouse IgG-PE (BD Pharmingen, Heidelberg, Germany) was used as an isotype control. After 30 min, the cells were washed, resuspended in 1% paraformaldehyde in PBS and antibody binding was measured by FACScalibur® (Becton and Dickinson, Heidelberg, Germany).

MHC class II-mediated superantigen presentation of pancreatic tumor cells to CD4+ T lymphocytes

The pancreatic cell lines Colo-357 and SU8686 were cultivated in the presence of interferon-γ to induce MHC class II expression as described above and then irradiated for 14 min (60 Gy) to prevent their proliferation. T lymphocytes were isolated from the peripheral blood of a healthy donor according to established methods. In brief, heparinized blood was layered onto PolymorphPrep® (Axis-Shield, Oslo, Norway), and the mononuclear cell fraction was recovered according to the protocol of the supplier. T lymphocytes were separated from the mononuclear cell fraction by magnetic beads separation using anti-CD3 beads (MACS Miltenyi Biotec, Bergisch Gladbach, Germany). The T lymphocytes were adjusted to 1 × 106/ml in RPMI supplemented with 10% fetal calf serum, 1% HEPES, 1% l-glutamine, and 1% penicillin/streptomycin (all obtained from Invitrogen) and 100 μl were placed into a round-bottom 96-well culture plate. The irradiated pancreatic cells (100 μl in the concentrations indicated in the respective experiment) were added with or without staphylococci enterotoxin B (SEB; 2 ng/well; Sigma-Aldrich, Munich, Germany). After 48 h, 3H-thymindine was added (1 μCi/well) for another 24 h and incorporation into the cells of radioactivity was measured. For comparison, pancreatic cancer cells without previous interferon-γ stimulation were used, as were T lymphocytes alone and pancreatic cancer cells alone with or without SEB. The T lymphocyte proliferation was determined as mean value of five replicas, and the differences between groups were determined using ANOVA.

Statistical analysis

For statistical analysis of survival, the non-parametric logrank test was performed. Correlation of MHC class II expression with clinical and pathological parameters was performed using Spearman’s rho analysis. Correlation of MHC class II staining with the density of CD4+ T lymphocytes was calculated with the Spearman’s rho analysis and the Mann–Whitney U test, respectively. For the array analysis, the staining results were grouped as described above. Significance levels were defined as p < 0.05. The statistical analyses were carried out with the SPSS software version 18.0 for Windows (SPSS Inc., Chicago, USA). Graphs were made using OriginPro7.5 software (Additive Software, Friedrichsdorf, Germany).

Results

Correlation of MHC class II expression with the severity and activity of intratumoral inflammatory reaction in PDAC and PET

Intratumoral inflammation was assessed by analysing and quantifying the infiltrating leucocytes, and the activity of inflammatory reaction was judged by the presence of neutrophils as indicators of an acute reaction. Inflammatory activity was seen in 108 of the 112 PDAC cases, while in PET—though inflammation was seen in 32 of 43 cases—only 22 cases revealed an inflammatory activity (data summarized in Fig. 1 and Table 3).
Fig. 1

a PDAC tumor tissue revealing no significant (score, 0) inflammation. b PDAC tissue revealing a severe (score, 3) intratumoral inflammatory reaction. c PDAC specimen with densely, mixed inflammatory infiltration dominated by neutrophil granulocytes, as indicator for a severe inflammatory activity (score, 2). d High infiltrate of CD4+ T lymphocytes

Table 3

Severity and activity of intratumoral inflammation in PDAC and PET

Number of cases

Severity of inflammation

Activity of inflammation

PDAC (n = 112)

Total: 112/112 (100.0%)

Total: 108/112 (96.4%)

Score 3: 20/112 (17.8%)

Score 2: 51/112 (45.5%)

Score 2: 80/112 (71.4%)

Score 1: 57/112 (50.9%)

Score 1: 12/112 (10.7%)

 

Score 0: 0/112

Score 0: 4 /112 (3.6%)

PET (n = 43)

Total: 32/43 (74.4%)

Total: 22/32 (68.8%)

Score 3: 0/43

Score 2: 2/32 (6.3%)

Score 2: 5/43 (11.6%)

Score 1: 20/32 (62.5%)

Score 1: 27/43 (62.8%)

 

Score 0: 11/43 (25.6%)

Score 0: 10/32 (31.2%)

As shown in Fig. 2 and Table 4, MHC class II antigen expression on tumor cells was seen in 86 of the 112 PDAC cases (76.8%); of these six cases revealed 1–5% positive tumor cells (score, 1), five cases 5–25% positive tumor cells (score, 2), 44 cases 25–50% positive tumor cells (score, 3), 12 cases 50–75% positive tumor cells (score, 4) and 19 cases more than 75% positive tumor cells (score, 5). The staining intensity was predominately weak to moderate and significantly correlated with the distribution of MHC class II expression (p < 0.0001). In PET, 30 of the 43 cases (70.0%) revealed MHC class II expression; of these two cases revealed 1–5% positive tumor cells (score, 1), one case 5–25% positive tumor cells (score, 2), 16 cases 25–50% positive tumor cells (score, 3), five cases 50–75% positive tumor cells (score, 4) and six cases more than 75% positive tumor cells (score, 5). In both types of tumor cells, an immunohistochemical staining of MHC class II on the cellular membrane, as well as a cytoplasmatic immunopositivity was detected [37]. Antigen-presenting inflammatory cells served as internal positive control. Additionally, stained normal pancreas tissue revealed no MHC class II expression in acini, ducts and islets of normal pancreas tissue.
Fig. 2

Immunohistochemical analyses of MHC class II expression on PDAC cells and endocrine tumor cells revealed different staining intensities, classified as, weak staining intensity (score, 1), moderate staining intensity (score, 2) or strong staining intensity (score, 3) or no staining (score, 0). a Exemplarily PDAC specimen with weak (score, 1), b moderate (score, 2), c strong staining intensity (score, 3). d No staining (score, 0). e Endocrine tumor cells with moderate staining (score, 2), f without staining (score, 0). Immune cells served in each case as an internal positive control

Table 4

MHC class II expression in PDAC and PET with percentage of immunoreactive tumor cells

Number of cases

MHC class II expression

MHC class II intensity

Immunoreactivity score

PDAC (n = 112)

Total, 86/112 (76.8%)

Strong, 9/86 (10.5%)

0, 26/112 (23.2%)

>75%, 19/86 (22.1%)

Moderate, 32/86 (37.2%)

2, 5/112 (4.5%)

50–75%, 12/86 (13.9%)

Weak, 45/86 (52.3%)

3, 5/112 (4.5%)

25–50%, 44/86 (51.2%)

 

4, 29/112 (25.9%)

5–25%, 5/86 (5.8%)

 

5, 22/112 (19.7%)

1–5%, 6/86 (7.0%)

 

6, 8/112 (7.1%)

  

7, 8/112 (7.1%)

  

8, 9/112 (8.0%)

PET (n = 43)

Total, 30/43 (70.0%)

Strong, 0/30 (0.0%)

0, 13/43 (30.0%)

>75%, 6/30 (20.0%)

Moderate, 13/30 (43.3%)

2, 8/43 (18.7%)

50–75%, 5/30 (16.7%)

Weak, 17/30 (56.7%)

3, 4/43 (9.3%)

25–50%, 16/30 (53.3%)

 

4, 1/43 (2.3%)

5–25%, 1/30 (3.3%)

 

5, 4/43 (9.3%)

1–5%, 2/30 (6.7%)

 

6, 8/43 (18.7%)

  

7, 5/43 (11.6%)

  

8, 0/43 (0%)

MHC class II antigen expression on PDAC cancer cells showed a significant positive correlation with the severity of the inflammation (p = 0.001) as well as with the activity of the inflammation (p = 0.007). Furthermore, the MHC class II staining intensity correlated positively with the severity (p < 0.001) and the activity of the inflammation (p = 0.002).

Using the immunoreactivity score, a positive correlation between severity (p < 0.001) and activity (p = 0.001) and the MHC class II expression was seen. Again, PET yielded essentially similar results, a positive correlation of MHC class II antigen expression with the severity of the intratumoral inflammation (p < 0.001) and with the activity of the intratumoral inflammation (p < 0.001) was seen; MHC class II staining intensity correlated with the severity (p = 0.003) and the activity of the intratumoral inflammation (p < 0.001). The immunoreactivity score revealed a significant positive correlation between the MHC class II expression and the severity (p < 0.001) and the activity (p < 0.001).

Among the infiltrating leucocytes, CD4+ lymphocytes were detected. In PDAC, the expression pattern of MHC class II as well as the staining intensity correlated significantly with the number of infiltrating CD4+ T lymphocytes (p < 0.001 for each parameter). Similar results were detected for PET tissues, in which MHC class II dispersion (p = 0.003) and staining intensity (p < 0.001) correlated significantly with the number of infiltrating CD4+ T lymphocytes. Scored by the immunoreactivity score, the MHC class II expression correlated significantly with the number CD4+ T lymphocytes in PDAC (p < 0.001) and PET (p = 0.007). In addition, we analysed the density of CD4+ T lymphocytes in MHC class II negative tumors versus tumors expressing MHC class II, irrespective of the staining pattern and the intensity. Here, the density of CD4+ T lymphocytes was higher in tissues containing MHC class II-positive tumor cells (irrespective of focal or diffuse staining pattern) compared to tissues with tumor cells negative for MHC class II (p < 0.001 for PDAC; p < 0.01 for PET). In PDAC tissue without MHC class II expression, the number of CD4+ cells ranged from zero to six cells per 10 HPF (mean, 1.5; median, 1), in samples revealing MHC class II expression, the number of CD4+ lymphocytes ranged from 0 to 97 cells per 10 HPF (mean, 11.4; median, 5). In PET without MHC class II expression, the number of CD4+ lymphocytes ranged between zero and one cell per 10 HPF (mean, 0.6; median, 1), in samples revealing MHC class II expression, the number of CD4+ lymphocytes ranged from zero to eight cells per 10 HPF (mean, 2.5; median, 2). In whole tissue sections of PDAC, CD4+ T lymphocytes could be detected in all 26 investigated cases. In cancer tissue without MHC class II expression, the density of CD4+ T lymphocytes ranged from five to 90 CD4+ T lymphocytes per 10 HPF (mean, 24.6; median, 14). Cases revealing a high MHC class II expression (more than 75% positive cells, moderate to strong staining intensity) showed a significantly higher density of CD4+ T lymphocytes (p < 0.0002), ranging from six to 201 CD4+ T lymphocytes per 10 HPF (mean, 102.7; median, 104) (data summarized in Figs. 3, 4 and 5).
Fig. 3

In PDAC a significantly higher density of CD4+ T lymphocytes was found in tissues revealing MHC class II expression versus no expression (PDAC: p < 0.0001). In PDAC tissue without MHC class II expression, the number of CD4+ cells ranged from 0 to 6 cells per 10 HPF (mean, 1.5; median, 1); in samples revealing MHC class II expression, the number of CD4+ lymphocytes ranged from 0 to 97 cells per 10 HPF (mean, 11.4; median, 5)

Fig. 4

In PET, a significantly higher density of CD4+ T lymphocytes in tissues revealing MHC class II expression versus no expression could be shown (p < 0.01). In PET without MHC class II expression, the number of CD4+ lymphocytes ranged between 0 and 1 cell per 10 HPF (mean, 0.6; median, 1); in PET revealing MHC class II expression, the number of CD4+ lymphocytes ranged from 0 to 8 cells per 10 HPF (mean, 2.5; median, 2)

Fig. 5

All 26 investigated PDAC specimen contained CD4+ T lymphocytes. In cancer specimen without MHC class II expression (left bar), CD4+ T lymphocytes ranged from 5 to 90 cells per 10 HPF (mean, 24.6; median, 14). Cases revealing high MHC class II expression showed a significantly higher density of CD4+ T lymphocytes (p < 0.0002), ranging from 6 to 201 cells per 10 HPF (mean, 102.7; median, 104)

Correlation of MHC class II expression in PDAC and PET with clinical and pathological parameters

The expression of MHC class II in PDAC (distribution, intensity and the immunoreactivity score) showed a significantly negative correlation with the tumor grade (p = 0.001), but not with the local tumor stage (pT), the presence of regional lymph node metastases (pN), distant metastases (pM) or patient gender. No correlation between the survival of the patients and MHC class II expression (p = 0.70) or intensity (p = 0.88) was found. In PET, the distribution of the MHC class II (p = 0.042) and the calculated immunoreactivity score (p = 0.027) significantly correlated with the local tumor stage (pT).

MHC class II expression parameters (distribution, intensity and the immunoreactivity score) revealed no significant correlation between the diagnostic tumor classification (WHO 2004 and 2010 and ENETS), histopathological grading, hormone expression of insulin or glucagon, functional activity, presence of a hereditary syndrome (MEN 1 syndrome), or gender of the patients. The endocrine carcinomas among the collective showed no correlation between the MHC class II expression and the presence or absence of regional lymph node (pN) or distant organ metastases (pM). No correlation with patient survival could be demonstrated with respect to distribution (p = 0.91) or intensity (p = 0.40).

MHC class II-positive PDAC tumor cells present superantigen to T lymphocytes and mediate T cell activation

MHC class II surface expression was evaluated on pancreatic cancer cells lines Capan-1, MiaPaca-2, BxPC-3, Panc-1, SU8686, AspC1, T3M4 and Colo-357. As determined by cytofluorometry, cultured cells were negative for MHC class II. Following cultivation with interferon-γ, MHC class II surface expression could be induced to high levels (21.4% to 44.4%), measured as the percent of cells expressing MHC class II on the following cell lines: SU8686 (44.4%), AspC-1 (43.7%), T3M4 (29.9%), Capan-1 (24.5%), BxPC3 (21.4%) and not at all or to a lesser extent on Panc-1 (4.8%) Colo-357 (4.0%) and MiaPaca-2 (1.4%) (data for SU8686 are shown in Fig. 6; data for all cells summarised in Table 5).
Fig. 6

MHC class II expression on the pancreatic tumor cell line SU8686 as determined by cytofluorometry. a Untreated cells do not express MHC class II (left panel); b following culture with interferon-γ for 48 h, MHC class II expression was seen (right panel). The dotted line indicates the antibody to HLA-DR-DQ-DP and the thick line indicates the isotype control

Table 5

Induction of MHC class II expression on ductal pancreatic tumor cell lines (48 h), measured as the percent of cells expressing MHC class II

Cell line

MHC class II expression

Isotype (IgG) control

Culture without interferon-γ

Following culture with interferon-γ

Culture without interferon-γ

Following culture with interferon-γ

AspC-1

4.1%

43.7%

3.9%

4.4%

BxPC3

4.0%

21.4%

4.2%

6.2%

Capan1

0.3%

24.5%

0.4%

0.9%

Colo-357

0.4%

4.0%

0.5%

1.0%

MiaPaca

0.5%

1.4%

0.7%

1.3%

SU8686

2.8%

44.4%

2.2%

6.9%

T3M4

0.2%

29.9%

1.0%

0.4%

Panc-1

2.0%

4.8%

3.1%

3.7%

To assess the functionality of the MHC class II antigens, two of the tumor cell lines, one expressing high levels of MHC class II antigens (SU8686; 44.4% MHC class II expressing cells after incubation with interferon-γ), and one expressing little to no MHC class II antigens (Colo-357; 4.0% MHC class II expressing cells after incubation with interferon-γ), as determined by FACS analysis, were chosen and cultivated with interferon-γ for 48 h. The tumor cells were irradiated to prevent their proliferation and then co-cultivated with isolated T cells and staphylococci enterotoxin B (SEB), a well-established “super” antigen. Proliferation of T lymphocytes was then measured after 36 h. Co-cultivation of T lymphocytes with SEB and SU8686, pre-incubated with interferon-γ, resulted in profound T lymphocyte proliferation, compared to T cells co-cultivated with SU8686 without prior interferon-γ exposure, or to T cells cultivated in the absence of SU8686 or SEB (Figs. 7 and 8). Colo-357 that only acquired little MHC class II induced little to no proliferation of the T lymphocytes.
Fig. 7

Induction of T lymphocyte proliferation by the interferon-γ-treated pancreas tumor cell line SU8686 and staphylococci enterotoxin B (SEB). SU8686—either pre-treated with interferon-γ or not—were cultivated with T cells of a healthy donor in absence or presence of SEB. After 3 days, proliferation was measured as incorporation of 3H-thymidine

Fig. 8

Two concentrations of interferon-γ-treated SU8686 were used, SEB and T cells of a healthy donor. Values are given as counts per minute (cpm); shown is the mean ± SD of 5 replica)

Discussion

In the present study, we focussed on rather specialized interaction molecules, the MHC class II complex (HLA-DR, DP, DQ) because as specialized recognition molecules they could provide a link between the local inflammation and the specific immune response. Under physiological conditions, constitutive expression of MHC class II molecules is restricted to professional antigen presenting cells, including dendritic cells, monocytes and B lymphocytes. MHC class II molecules bind processed antigenic peptides and present them to the antigen-specific T cell receptors on CD4+ T lymphocytes. The ensuing antigen-specific T cell activation is dependent on appropriate co-stimulatory signals and can be modified by the cytokines within the microenvironment [38, 39]. MHC class II expression, however, is not limited to antigen presenting cells. Under inflammatory conditions, polymorphonuclear neutrophils [26] and epithelial cells [40] acquire MHC class II molecules. Moreover, a variety of tumor cells express MHC class II [21, 22, 23, 24]. The data for pancreatic ductal adenocarcinoma are rather limited. In two studies MHC class II expression in three out of eight patients [14] and in 11 out of 37 patients [25] was described. According to another study, expression of MHC class II could not be demonstrated by cytofluorometry in untreated, primary PDAC cells of 19 patients [9], while others found MHC class II on cultured PDAC lines [14] that could be induced by interferon-γ [14, 25]. In our series, MHC class II expression was detected in the majority of PDAC (76.8%) and of PET (70.0%). MHC class II expression correlated positively with the severity and the activity of the inflammatory reaction, and notably, with the infiltration of CD4+ T lymphocytes. Since the latter produce interferon-γ, the major effector cytokine for induction of MHC class II antigens in professional and non-professional antigen presenting cells [41], these data suggest that the infiltrating CD4+ T lymphocytes induce the MHC class II expression on the tumor cells. The fact that infiltrating T lymphocytes or PDAC tissue-derived tumor-reactive T lymphocytes release interferon-γ has been shown before [3, 42, 43]; moreover, we found that interferon-γ induced surface expression of MHC class II antigen on pancreatic tumor cell lines. Additionally, stained normal pancreas tissue revealed no MHC class II expression, reflecting the MHC class II expression as an effect induced by the inflammatory microenvironment in the tumor.

The fact that the cell lines varied with regard to MHC class II expression could reflect their differentiation status: well or moderately differentiated cell lines such as Capan-1, Su8686 T3M4 or AsPC1 acquired more MHC class II after interferon-γ stimulation compared to the poorly differentiated cell lines Panc-1 or Colo-357. These data are in accordance to our histological findings, in which high levels of MHC class II expression coincided with a better histological differentiation grade. De-differentiation of the tumors may probably cause a loss of MHC class II antigen expression of tumor cells. In our study, the presence or absence of lymph node or distant metastases did not correlate with MHC class II expression in the PDAC collective, nor did the pT stage, the latter obviously due to the fact that 110 patients revealed a pT3 status. In PET, pT stage correlated significantly with the expression of MHC class II expression by endocrine tumor cells.

MHC class II expression did not correlate with lymph node or organ metastases, corresponding to the findings of Monti et al., who tested pancreatic cancer cell lines either cultivated from the primary tumors or from metastases revealing no differences of MHC class II expression [44].

So far, the functional role of MHC class II molecules on tissue cells is poorly understood. Except for the special situation in the thymus, in which thymus epithelial cells present self-antigens and thereby contribute to the shaping of the T cell repertoire, only in vitro data are available showing that MHC class II on tissue cells such as keratinocytes [45], tubular epithelial cells [46] or synovial fibroblasts [47] can present superantigen to T cells, thereby inducing their proliferation. The special situation of superantigens, such as, the staphylococcus enterotoxin (SE) B used as reliable model superantigen in our study, is that they bind to the non-variable region of the MHC class II molecule without prior processing and are recognised by a large number of CD4+ T lymphocytes in a MHC class II-dependent but unrestricted and antigen-unspecific manner. Binding results in the activation of T cells, apparent as proliferation [48]. Under our experimental conditions, we found that pancreatic tumor cell lines induced to express MHC class II antigens were able to trigger T cell proliferation when SEB was present.

The role of activated T cells, particularly that of CD4+ T lymphocytes, is currently under intense investigation and pro- or antitumorigenic functions have been described, depending on the cytokine environment [49, 50, 51]. Pancreatic tumor cells are specifically recognized by autologous T lymphocytes [9]. Based on these observations and our results, MHC class II antigens on pancreatic tumor cells might serve as a means to present tumor-associated antigens. According to the classical dogma, MHC class II molecules present processed exogenous antigens as opposed to self-antigens that are presented by MHC class I. This strict segregation, however, no longer holds true, and multiple alternative antigen-presenting pathways involving MHC class II antigens have been described [52]. In that, MHC class II expression on pancreatic tumor cells could participate in the CD4+ T lymphocyte-mediated arrest of carcinogenesis. However, no correlation between MHC class II expression and patient survival was seen in our study, which is quite in contrast to data on patients with colorectal cancer, in which MHC class II expression correlated with a better survival [21]. The generally dismal prognosis of PDAC patients with a 5-year survival rate of less than 5% assembling all components like aggressive and invasive tumor growth, early metastasis and resistance to radiation and chemotherapy [53], overbalances probably the effects of the MHC class II-mediated immune reaction and of the surrounding inflammatory infiltrate in PDAC. For PET, correlations with MHC class II molecules and survival should be performed in a larger series including longer follow-up periods to obtain reliable results. Nevertheless, the inflammatory infiltrate, influencing the tumor cell growth, tumor cell migration or tumor neoangiogenesis [10, 12, 54], might also be an important factor in PET biology. As previously shown for malignant melanoma, even scarce inflammatory infiltrates may significantly affect the tumor biology and the resulting prognosis of the patients [55].

The question arises whether the expression of MHC class II antigens on tumor cells might have some benefit for patients with PDAC or PET. A major problem in the host defence and by analogy of immunologic-based therapies is the “tumor-immune-escape” [56], which—at least in part—is due to the downregulation of recognition structures on the tumor cells, including MHC molecules [14]. In this context, pancreatic ductal adenocarcinomas [25] as well as pancreatic endocrine tumor cells [57], are characterized to reveal a loss of MHC class I molecules, probably to escape a CD8+ T cell-mediated cytotoxic reaction. A direct relationship between the expression profiles of MHC class I and class II molecules could not be demonstrated in a study for PDAC, though both molecules are inducible by interferon-γ [25]. Consequently, in various clinical trials, e.g. on malignant melanoma, interferon was applied to induce MHC class I and MHC class II molecules and as consequence levels of melanoma-specific CD4+ T lymphocytes increased [58, 59]. Since MHC class II can be induced also on pancreatic cancer cells, an interferon-γ-based therapy is worth considering, especially in better differentiated cancers. On the other hand, a note of caution is imperative: MHC class molecules can also induce antigen-specific anergy of T lymphocytes, e.g. when co-stimulatory signals on the target cells are missing [60]. In conclusion, our findings suggest that MHC class II expression on pancreatic tumor cells is induced by the intratumoral inflammatory reaction in pancreatic tumors and might play an important role in the interaction of the inflammatory cells with pancreatic tumors.

Notes

Acknowledgements

We thank Prof. Dr. Ulrich Abel, Department of Medical Biometry, University of Heidelberg for the professional evaluation of the biostatistics. We thank Mrs. Birgit Prior, Institute for Immunology, University of Heidelberg and Mrs. Sarah Messnard, Institute of Pathology, University of Heidelberg for their excellent technical support.

Conflict of interest statement

None of the authors declare a conflict of interest.

References

  1. 1.
    Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ (2009) Cancer statistics, 2009. CA Cancer J Clin 59(4):225–249PubMedCrossRefGoogle Scholar
  2. 2.
    Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100(1):57–70PubMedCrossRefGoogle Scholar
  3. 3.
    Kleeff J, Beckhove P, Esposito I, Herzig S, Huber PE, Lohr JM, Friess H (2007) Pancreatic cancer microenvironment. Int J Cancer 121(4):699–705PubMedCrossRefGoogle Scholar
  4. 4.
    Welsch T, Kleeff J, Friess H (2007) Molecular pathogenesis of pancreatic cancer: advances and challenges. Curr Mol Med 7(5):504–521PubMedCrossRefGoogle Scholar
  5. 5.
    Greer JB, Whitcomb DC (2009) Inflammation and pancreatic cancer: an evidence-based review. Curr Opin Pharmacol 9(4):411–418PubMedCrossRefGoogle Scholar
  6. 6.
    Tingstedt B, Johansson P, Andersson B, Andersson R (2007) Predictive factors in pancreatic ductal adenocarcinoma: role of the inflammatory response. Scand J Gastroenterol 42(6):754–759PubMedCrossRefGoogle Scholar
  7. 7.
    Esposito I, Menicagli M, Funel N, Bergmann F, Boggi U, Mosca F, Bevilacqua G, Campani D (2004) Inflammatory cells contribute to the generation of an angiogenic phenotype in pancreatic ductal adenocarcinoma. J Clin Pathol 57(6):630–636PubMedCrossRefGoogle Scholar
  8. 8.
    Bolignano D, Donato V, Lacquaniti A, Fazio MR, Bono C, Coppolino G, Buemi M (2010) Neutrophil gelatinase-associated lipocalin (NGAL) in human neoplasias: a new protein enters the scene. Cancer Lett 288(1):10–16PubMedCrossRefGoogle Scholar
  9. 9.
    Schmitz-Winnenthal FH, Escobedo LV, Beckhove P, Schirrmacher V, Bucur M, Ziouta Y, Volk C, Schmied B, Koch M, Antolovic D, Weitz J, Buchler MW, Z'Graggen K (2006) Specific immune recognition of pancreatic carcinoma by patient-derived CD4 and CD8 T cells and its improvement by interferon-gamma. Int J Oncol 28(6):1419–1428PubMedGoogle Scholar
  10. 10.
    Wente MN, Gaida MM, Mayer C, Michalski CW, Haag N, Giese T, Felix K, Bergmann F, Giese NA, Friess H (2008) Expression and potential function of the CXC chemokine CXCL16 in pancreatic ductal adenocarcinoma. Int J Oncol 33(2):297–308PubMedGoogle Scholar
  11. 11.
    Gaida MM, Gunther F, Wagner C, Friess H, Giese NA, Schmidt J, Hansch GM, Wente MN (2008) Expression of the CXCR6 on polymorphonuclear neutrophils in pancreatic carcinoma and in acute, localized bacterial infections. Clin Exp Immunol 154(2):216–223PubMedCrossRefGoogle Scholar
  12. 12.
    Wente MN, Mayer C, Gaida MM, Michalski CW, Giese T, Bergmann F, Giese NA, Buchler MW, Friess H (2008) CXCL14 expression and potential function in pancreatic cancer. Cancer Lett 259(2):209–217PubMedCrossRefGoogle Scholar
  13. 13.
    Koshiba T, Hosotani R, Miyamoto Y, Ida J, Tsuji S, Nakajima S, Kawaguchi M, Kobayashi H, Doi R, Hori T, Fujii N, Imamura M (2000) Expression of stromal cell-derived factor 1 and CXCR4 ligand receptor system in pancreatic cancer: a possible role for tumor progression. Clin Cancer Res 6(9):3530–3535PubMedGoogle Scholar
  14. 14.
    Scupoli MT, Sartoris S, Tosi G, Ennas MG, Nicolis M, Cestari T, Zamboni G, Martignoni G, Lemoine NR, Scarpa A, Accolla RS (1996) Expression of MHC class I and class II antigens in pancreatic adenocarcinomas. Tissue Antigens 48(4 Pt 1):301–311PubMedCrossRefGoogle Scholar
  15. 15.
    Berghuis D, de Hooge AS, Santos SJ, Horst D, Wiertz EJ, van Eggermond MC, van den Elsen PJ, Taminiau AH, Ottaviano L, Schaefer KL, Dirksen U, Hooijberg E, Mulder A, Melief CJ, Egeler RM, Schilham MW, Jordanova ES, Hogendoorn PC, Lankester AC (2009) Reduced human leukocyte antigen expression in advanced-stage Ewing sarcoma: implications for immune recognition. J Pathol 218(2):222–231PubMedCrossRefGoogle Scholar
  16. 16.
    Townsend A, Bodmer H (1989) Antigen recognition by class I-restricted T lymphocytes. Annu Rev Immunol 7:601–624PubMedCrossRefGoogle Scholar
  17. 17.
    Albanesi C, Cavani A, Girolomoni G (1998) Interferon-gamma-stimulated human keratinocytes express the genes necessary for the production of peptide-loaded MHC class II molecules. J Invest Dermatol 110(2):138–142PubMedCrossRefGoogle Scholar
  18. 18.
    Fais S, Capobianchi MR, Marcheggiano A, Iannoni C, Pallone F (1992) MHC class II antigens on the epithelial cells of the human gastrointestinal tract. Gastroenterology 102(1):377–378PubMedGoogle Scholar
  19. 19.
    Ruiter DJ, Bergman W, Welvaart K, Scheffer E, van Vloten WA, Russo C, Ferrone S (1984) Immunohistochemical analysis of malignant melanomas and nevocellular nevi with monoclonal antibodies to distinct monomorphic determinants of HLA antigens. Cancer Res 44(9):3930–3935PubMedGoogle Scholar
  20. 20.
    Dengjel J, Nastke MD, Gouttefangeas C, Gitsioudis G, Schoor O, Altenberend F, Muller M, Kramer B, Missiou A, Sauter M, Hennenlotter J, Wernet D, Stenzl A, Rammensee HG, Klingel K, Stevanovic S (2006) Unexpected abundance of HLA class II presented peptides in primary renal cell carcinomas. Clin Cancer Res 12(14 Pt 1):4163–4170PubMedCrossRefGoogle Scholar
  21. 21.
    Walsh MD, Dent OF, Young JP, Wright CM, Barker MA, Leggett BA, Bokey L, Chapuis PH, Jass JR, Macdonald GA (2009) HLA-DR expression is associated with better prognosis in sporadic Australian clinicopathological Stage C colorectal cancers. Int J Cancer 125(5):1231–1237PubMedCrossRefGoogle Scholar
  22. 22.
    Concha A, Esteban F, Cabrera T, Ruiz-Cabello F, Garrido F (1991) Tumor aggressiveness and MHC class I and II antigens in laryngeal and breast cancer. Semin Cancer Biol 2(1):47–54PubMedGoogle Scholar
  23. 23.
    Esteban F, Concha A, Huelin C, Perez-Ayala M, Pedrinaci S, Ruiz-Cabello F, Garrido F (1989) Histocompatibility antigens in primary and metastatic squamous cell carcinoma of the larynx. Int J Cancer 43(3):436–442PubMedCrossRefGoogle Scholar
  24. 24.
    Andersen SN, Rognum TO, Lund E, Meling GI, Hauge S (1993) Strong HLA-DR expression in large bowel carcinomas is associated with good prognosis. Br J Cancer 68(1):80–85PubMedCrossRefGoogle Scholar
  25. 25.
    Pandha H, Rigg A, John J, Lemoine N (2007) Loss of expression of antigen-presenting molecules in human pancreatic cancer and pancreatic cancer cell lines. Clin Exp Immunol 148(1):127–135PubMedCrossRefGoogle Scholar
  26. 26.
    Radsak M, Iking-Konert C, Stegmaier S, Andrassy K, Hansch GM (2000) Polymorphonuclear neutrophils as accessory cells for T-cell activation: major histocompatibility complex class II restricted antigen-dependent induction of T-cell proliferation. Immunology 101(4):521–530PubMedCrossRefGoogle Scholar
  27. 27.
    Bosman FT, Carneiro F, Hruban RH, Theise ND (eds) (2010) World Health Organisation classification of tumours. Pathology and genetics of tumours of the digestive system. Ductal adenocarcinoma of the pancreas. IARC Press, LyonGoogle Scholar
  28. 28.
    Sobin LH, Gospodarowicz MK, Wittekind C (2009) TNM classification of malignant tumours, 7th edition. Wiley, OxfordGoogle Scholar
  29. 29.
    Heitz PU, Komminoth P, Perren A (2004) Tumours of endocrine organs: pathology and genetics World Health Organisation classification of tumors. IARC press, LyonGoogle Scholar
  30. 30.
    Bosman FT, Carneiro F, Hruban RH, Theise ND (2010) World Health Organisation classification of tumours. Pathology and genetics of tumours of the digestive system. Neuroendocrine neoplasms of the pancreas. IARC press, LyonGoogle Scholar
  31. 31.
    Rindi G, Kloppel G, Alhman H, Caplin M, Couvelard A, de Herder WW, Erikssson B, Falchetti A, Falconi M, Komminoth P, Korner M, Lopes JM, McNicol AM, Nilsson O, Perren A, Scarpa A, Scoazec JY, Wiedenmann B (2006) TNM staging of foregut (neuro)endocrine tumors: a consensus proposal including a grading system. Virchows Arch 449(4):395–401PubMedCrossRefGoogle Scholar
  32. 32.
    Bergmann F, Breinig M, Hopfner M, Rieker RJ, Fischer L, Kohler C, Esposito I, Kleeff J, Herpel E, Ehemann V, Friess H, Schirmacher P, Kern MA (2009) Expression pattern and functional relevance of epidermal growth factor receptor and cyclooxygenase-2: novel chemotherapeutic targets in pancreatic endocrine tumors? Am J Gastroenterol 104(1):171–181PubMedCrossRefGoogle Scholar
  33. 33.
    Ceyhan GO, Bergmann F, Kadihasanoglu M, Erkan M, Park W, Hinz U, Giese T, Muller MW, Buchler MW, Giese NA, Friess H (2007) The neurotrophic factor artemin influences the extent of neural damage and growth in chronic pancreatitis. Gut 56(4):534–544PubMedCrossRefGoogle Scholar
  34. 34.
    Bijen CB, Bantema-Joppe EJ, de Jong RA, Leffers N, Mourits MJ, Eggink HF, van der Zee AG, Hollema H, de Bock GH, Nijman HW (2010) The prognostic role of classical and nonclassical MHC class I expression in endometrial cancer. Int J Cancer 126(6):1417–1427PubMedGoogle Scholar
  35. 35.
    Rolland P, Deen S, Scott I, Durrant L, Spendlove I (2007) Human leukocyte antigen class I antigen expression is an independent prognostic factor in ovarian cancer. Clin Cancer Res 13(12):3591–3596PubMedCrossRefGoogle Scholar
  36. 36.
    Itano AA, Jenkins MK (2003) Antigen presentation to naive CD4 T cells in the lymph node. Nat Immunol 4(8):733–739PubMedCrossRefGoogle Scholar
  37. 37.
    Rocha N, Neefjes J (2008) MHC class II molecules on the move for successful antigen presentation. EMBO J 27(1):1–5PubMedCrossRefGoogle Scholar
  38. 38.
    Radfar S, Wang Y, Khong HT (2009) Activated CD4+ T cells dramatically enhance chemotherapeutic tumor responses in vitro and in vivo. J Immunol 183(10):6800–6807PubMedCrossRefGoogle Scholar
  39. 39.
    DeNardo DG, Barreto JB, Andreu P, Vasquez L, Tawfik D, Kolhatkar N, Coussens LM (2009) CD4(+) T cells regulate pulmonary metastasis of mammary carcinomas by enhancing protumor properties of macrophages. Cancer Cell 16(2):91–102PubMedCrossRefGoogle Scholar
  40. 40.
    Han DC, Huang GT, Lin LM, Warner NA, Gim JS, Jewett A (2003) Expression of MHC Class II, CD70, CD80, CD86 and pro-inflammatory cytokines is differentially regulated in oral epithelial cells following bacterial challenge. Oral Microbiol Immunol 18(6):350–358PubMedCrossRefGoogle Scholar
  41. 41.
    Giacomini P, Tecce R, Gambari R, Sacchi A, Fisher PB, Natali PG (1988) Recombinant human IFN-gamma, but not IFN-alpha or IFN-beta, enhances MHC- and non-MHC-encoded glycoproteins by a protein synthesis-dependent mechanism. J Immunol 140(9):3073–3081PubMedGoogle Scholar
  42. 42.
    Tang KF, Chan SH, Loh KS, Chong SM, Wang D, Yeoh KH, Hu H (1999) Increased production of interferon-gamma by tumour infiltrating T lymphocytes in nasopharyngeal carcinoma: indicative of an activated status. Cancer Lett 140(1–2):93–98PubMedCrossRefGoogle Scholar
  43. 43.
    Schmitz-Winnenthal FH, Volk C, Z'Graggen K, Galindo L, Nummer D, Ziouta Y, Bucur M, Weitz J, Schirrmacher V, Buchler MW, Beckhove P (2005) High frequencies of functional tumor-reactive T cells in bone marrow and blood of pancreatic cancer patients. Cancer Res 65(21):10079–10087PubMedCrossRefGoogle Scholar
  44. 44.
    Monti P, Marchesi F, Reni M, Mercalli A, Sordi V, Zerbi A, Balzano G, Di Carlo V, Allavena P, Piemonti L (2004) A comprehensive in vitro characterization of pancreatic ductal carcinoma cell line biological behavior and its correlation with the structural and genetic profile. Virchows Arch 445(3):236–247PubMedCrossRefGoogle Scholar
  45. 45.
    Nickoloff BJ, Mitra RS, Green J, Shimizu Y, Thompson C, Turka LA (1993) Activated keratinocytes present bacterial-derived superantigens to T lymphocytes: relevance to psoriasis. J Dermatol Sci 6(2):127–133PubMedCrossRefGoogle Scholar
  46. 46.
    Schulz H, Karau A, Filsinger S, Schoels M, Kabelitz D, Richter R, Hansch GM (1998) Tubular epithelial cells as accessory cells for superantigen-induced T cell activation. Exp Nephrol 6(1):67–73PubMedCrossRefGoogle Scholar
  47. 47.
    Kraft M, Filsinger S, Kramer KL, Kabelitz D, Hansch GM, Schoels M (1995) Synovial fibroblasts as accessory cells for staphylococcal enterotoxin-mediated T-cell activation. Immunology 85(3):461–466PubMedGoogle Scholar
  48. 48.
    Fanger NA, Liu C, Guyre PM, Wardwell K, O'Neil J, Guo TL, Christian TP, Mudzinski SP, Gosselin EJ (1997) Activation of human T cells by major histocompatability complex class II expressing neutrophils: proliferation in the presence of superantigen, but not tetanus toxoid. Blood 89(11):4128–4135PubMedGoogle Scholar
  49. 49.
    Muller-Hermelink N, Braumuller H, Pichler B, Wieder T, Mailhammer R, Schaak K, Ghoreschi K, Yazdi A, Haubner R, Sander CA, Mocikat R, Schwaiger M, Forster I, Huss R, Weber WA, Kneilling M, Rocken M (2008) TNFR1 signaling and IFN-gamma signaling determine whether T cells induce tumor dormancy or promote multistage carcinogenesis. Cancer Cell 13(6):507–518PubMedCrossRefGoogle Scholar
  50. 50.
    Marzo AL, Kinnear BF, Lake RA, Frelinger JJ, Collins EJ, Robinson BW, Scott B (2000) Tumor-specific CD4+ T cells have a major "post-licensing" role in CTL mediated anti-tumor immunity. J Immunol 165(11):6047–6055PubMedGoogle Scholar
  51. 51.
    Fukunaga A, Miyamoto M, Cho Y, Murakami S, Kawarada Y, Oshikiri T, Kato K, Kurokawa T, Suzuoki M, Nakakubo Y, Hiraoka K, Itoh T, Morikawa T, Okushiba S, Kondo S, Katoh H (2004) CD8+ tumor-infiltrating lymphocytes together with CD4+ tumor-infiltrating lymphocytes and dendritic cells improve the prognosis of patients with pancreatic adenocarcinoma. Pancreas 28(1):e26–e31PubMedCrossRefGoogle Scholar
  52. 52.
    Lindner R, Unanue ER (1996) Distinct antigen MHC class II complexes generated by separate processing pathways. EMBO J 15(24):6910–6920PubMedGoogle Scholar
  53. 53.
    Li D, Xie K, Wolff R, Abbruzzese JL (2004) Pancreatic cancer. Lancet 363(9414):1049–1057PubMedCrossRefGoogle Scholar
  54. 54.
    Strieter RM, Burdick MD, Mestas J, Gomperts B, Keane MP, Belperio JA (2006) Cancer CXC chemokine networks and tumour angiogenesis. Eur J Cancer 42(6):768–778PubMedCrossRefGoogle Scholar
  55. 55.
    Clemente CG, Mihm MC Jr, Bufalino R, Zurrida S, Collini P, Cascinelli N (1996) Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer 77(7):1303–1310PubMedCrossRefGoogle Scholar
  56. 56.
    von Bernstorff W, Voss M, Freichel S, Schmid A, Vogel I, Johnk C, Henne-Bruns D, Kremer B, Kalthoff H (2001) Systemic and local immunosuppression in pancreatic cancer patients. Clin Cancer Res 7(3 Suppl):925s–932sGoogle Scholar
  57. 57.
    Ye X, Kralli A, Ge R, Ricciardi RP, Knowles BB (1994) Down-regulation of MHC class I antigen in insulinoma cells controlled by the R1 element of the H-2 enhancer. Oncogene 9(4):1195–1204PubMedGoogle Scholar
  58. 58.
    Lotem M, Machlenkin A, Hamburger T, Nissan A, Kadouri L, Frankenburg S, Gimmon Z, Elias O, David IB, Kuznetz A, Shiloni E, Peretz T (2009) Autologous melanoma vaccine induces antitumor and self-reactive immune responses that affect patient survival and depend on MHC class II expression on vaccine cells. Clin Cancer Res 15(15):4968–4977PubMedCrossRefGoogle Scholar
  59. 59.
    Propper DJ, Chao D, Braybrooke JP, Bahl P, Thavasu P, Balkwill F, Turley H, Dobbs N, Gatter K, Talbot DC, Harris AL, Ganesan TS (2003) Low-dose IFN-gamma induces tumor MHC expression in metastatic malignant melanoma. Clin Cancer Res 9(1):84–92PubMedGoogle Scholar
  60. 60.
    Schwartz RH (1996) Models of T cell anergy: is there a common molecular mechanism? J Exp Med 184(1):1–8PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Matthias M. Gaida
    • 1
  • Thilo Welsch
    • 2
  • Esther Herpel
    • 1
  • Darjus F. Tschaharganeh
    • 1
  • Lars Fischer
    • 2
  • Peter Schirmacher
    • 1
  • G. Maria Hänsch
    • 3
  • Frank Bergmann
    • 1
  1. 1.Institute of PathologyUniversity of HeidelbergHeidelbergGermany
  2. 2.Department of General SurgeryUniversity of HeidelbergHeidelbergGermany
  3. 3.Institute of ImmunologyUniversity of HeidelbergHeidelbergGermany

Personalised recommendations