 |  |
| ||||
Copyright © 1999 Blackwell Science Ltd Enterobacterial infection modulates major histocompatibility complex class I expression on mononuclear cells Correspondence: Dr Kaisa Granfors, National Public Health Institute, Department in Turku, Kiinamyllynkatu 13, FIN-20520 Turku, Finland. Received November 16, 1998; Revised March 3, 1999; Accepted March 3, 1999.  This article has been cited by other articles in PMC. | |||||||||||||
Abstract Major histocompatibility complex (MHC) class I expression is reduced in several viral infections, but it is not known whether the same happens during infections caused by intracellular enterobacteria. In this study, the expression of MHC class I antigens on peripheral blood mononuclear cells (PBMC) from 16 patients with Salmonella, Yersinia, or Klebsiella infection was investigated. During or after the acute infection, the expression of MHC class I antigens was markedly decreased in eight patients, all with genotype HLA-B27, and six out of eight with reactive arthritis (ReA). A significant decrease of monomorphic MHC class I was found in three patients, of HLA-B27 in eight (P < 0·05) and of HLA-A2 in two. However, patients negative for the HLA-B27 genotype, or healthy HLA-B27-positive individuals, did not have a significant decrease of MHC class I antigens. During the decreased expression on the cell surface, intracellular retention of MHC class I antigens was observed, whereas HLA-B27 mRNA levels did not vary significantly. This is the first evidence that enterobacterial infection may down-regulate expression of MHC class I molecules in vivo and that down-regulation is predominant in patients with the HLA-B27 genotype. | |||||||||||||
INTRODUCTION The major histocompatibility complex (MHC) class I antigens present endogenous and exogenous peptides in complexes that are recognized by the appropriate cytotoxic T lymphocytes (CTL). In general, the cells presenting foreign peptides are subjected to CTL-mediated lysis whereas defective cell surface expression of the MHC class I antigens leads to natural killer (NK) cell-mediated lysis.1 The expression of MHC I molecules is known to be affected by pathological events such as viral infection and malignant transformation. Several viruses have evolved strategies to reduce class I expression in the host, e.g. adenovirus type 2 encodes an endoplasmic reticulum (ER) resident protein, E19, which associates with class I molecules and prevents their transportation from the ER to the cell surface.2 Other large DNA viruses, herpes simplex virus 1 (HSV-1) and HSV-2, also cause reduction of cell-surface expression of MHC class I in early and later stages of the infection.3 These and other immune evasion mechanisms have been recently reviewed.4 Regulation of expression of MHC class I antigens during bacterial infection is not well characterized. However, it is of special interest as development of reactive arthritis (ReA), a well-known complication of certain bacterial infections, is strongly associated with a MHC class I antigen, human leucocyte antigen (HLA)-B27.5 These infections are gastrointestinal and urogenital infections caused by Salmonellae, Yersiniae, Shigellae, Campylobacter and Chlamydiae. The pathogenesis of ReA, and mechanisms by which HLA-B27 is involved in disease susceptibility, are unknown.5–7 Recently, evidence has been presented that HLA-B27 directly modulates the interaction between host cells and ReA-triggering microbes.8–10 Our results show that the expression of certain HLA-B27 epitopes on monocytes is down-regulated during Salmonella and Yersinia infection in vitro,11 but it is not known whether down-regulation, analogous to in vitro findings, also occurs in vivo. However, there is accumulating evidence that several, also non-MHC, genes are indeed differentially regulated in HLA-B27-positive subjects during infection.12–15 This study investigated whether enterobacterial infection results in disturbances in the expression of MHC class I antigens, especially HLA-B27, on peripheral blood mononuclear cells (PBMC). The cell surface and intracellular expression of these antigens, and their transcription, were analysed. | |||||||||||||
MATERIALS AND METHODS Patients Peripheral blood follow-up samples were collected from seven patients (P1–P7) with acute Salmonella-triggered ReA and from one patient (P8) with acute Yersinia-triggered ReA (Table 1). All except two patients had typical clinical signs of acute infection, e.g. fever, gastrointestinal pain and diarrhoea. One patient without clear symptoms of infection had a positive stool culture for S. typhimurium and the other had a high level of antibodies to Salmonella in serum. All the patients had a typical clinical course of ReA. The laboratory diagnosis of Salmonella infection was based on positive stool culture in five of seven patients and solely on elevated serum antibodies (high concentrations of IgM, IgG and IgA antibodies) to Salmonella16 in two patients. The laboratory diagnosis of Yersinia infection was based on a positive stool culture for Y. enterocolitica O:3. All the patients were treated with non-steroidal anti-inflammatory drugs (NSAIDs), and patient 5 also with ciprofloxacin, oral cortisone and sulfasalazine.
Three to five peripheral blood follow-up samples from seven patients with Salmonella infection without ReA were collected from two small outbreaks caused by S. enteritidis.17 The patients diagnosed with S. enteritidis infection were informed about this study and volunteers spontaneously contacted the research group. All these patients had a typical clinical course of Salmonella infection including high fever (>38·5°), gastrointestinal pain, diarrhoea and a positive stool culture for S. enteritidis. Five follow-up samples from one HLA-B27-positive patient with sinuitis caused by Klebsiella pneumoniae (culture positive) were also studied during the 1-year study period. The symptoms persisted for up to 2 months and resolved after treatment with ciprofloxacin and corticosteroids. Peripheral blood samples were collected, during the 1-year study period, from two HLA-B27-positive healthy subjects in our laboratory for use as controls for studying HLA-B27 baseline expression. In addition, samples from seven healthy subjects were used as controls for expression of cell-surface adhesion and activation molecules on PBMC. Monoclonal antibodies (mAbs) and flow cytometry Monoclonal antibodies used in this study are listed in Table 2. PBMC were isolated using Ficoll–Paque (Pharmacia LKB Biotechnology AB, Uppsala, Sweden) gradient centrifugation, as previously described.18 Briefly, PBMC were incubated with saturating concentrations (10 μg/ml) of primary mAb at +4° for 15 min. Cells were then washed twice and incubated with fluorescein isothiocyanate (FITC)-labelled F(ab′)2 fragments of goat antimouse IgG (1:200 v/v) at +4° for 15 min (Sigma Chemical Company, St Louis, MO).11 Cell surface expression was analysed by fluorescence-activated cell sorter (FACScan®, Becton-Dickinson Immunocytometry Systems, Mountain View, CA). Lymphocytes and monocytes were gated according to scatter characteristics, and 5000 cells were collected per cell population.
To evaluate and present the results of immunofluorescence staining, binding of the negative control was subtracted from each value before plotting. A sample was considered to be positive if the expression of specific mAb was >10 mean fluorescence intensity (MFI) units after subtraction of MFI from the negative control. Immunofluorescence staining and confocal microscopy PBMC were fixed in suspension with 4% formaldehyde in phosphate-buffered saline (PBS) for 10 min, permeabilized with 0·1% saponin in PBS for 10 min, and then blocked in PBS supplemented with 2% bovine serum albumin (BSA) and 0·1% saponin for 10 min. The primary antibodies (w6/32, HC-10 and 3G6) were overlaid and incubated at +4° for 30 min. After two washes with the same buffer, PBMC were incubated with FITC-labelled F(ab′)2 fragments of goat antimouse IgG (1:200 v/v) at +4° for 15 min. After a further two washes, PBMC were centrifuged on slides and mounted with 90% glycerol in PBS containing 1 mg/ml phenylenediamine (Sigma Chemical Company). Corresponding samples without saponin were stained simultaneously. The samples were analysed using confocal microscopy (Leitz, Heidelberg, Germany) with 488 nm excitation wavelength.Reverse transcription–polymerase chain reaction (RT–PCR) Expression of HLA-B27 mRNA was analysed by semiquantitative RT–PCR. Details of RNA isolation, treatment of extracted RNA with DNase I, cDNA synthesis and PCR have been previously described.19 HLA-B27 primers and β-actin primers19 were as described earlier. RT–PCR assays were first optimized, and therefore five different cycles, from 20 to 35, were assessed to ensure that PCR products were detected before they reached saturation. For all samples tested, 25 cycles were used for HLA-B27 and 22 cycles for β-actin. The differences in RNA isolation and cDNA synthesis of HLA-B27 were normalized to a housekeeping gene, β-actin. All samples were tested at least twice.PCR for the HLA-B27 gene and microlymphocytotoxicity (MLCT) testing for MHC class I antigens All individuals in this study were tested for HLA-B27 by PCR as previously described.21 All the PCR tests were repeated at least three times. All eight patients with ReA and two without ReA were positive for HLA-B27. Two of the healthy control subjects were also positive for HLA-B27.Patients 1–5 were also typed by standard MLCT using antisera against HLA-A, -B, and -C antigens (Lymphotype ABC-72; Biotest AG, Dreieich, Germany) at the Department of Medical Microbiology, University of Turku, Turku, Finland. The typing results were as follows: HLA-A2, 24; B27, 60, w4, w6; Cw1, w3 (P1), HLA-A3, 32; B7, 27, w4, w6; Cw2, w7 (P2), HLA-A1, 2; B8, 27, w4, w6; Cw1, w7 (P3), HLA-A2, 3; B27, w4; Cw1 (P4) and HLA-A2, 28; B7, 27, w4, w6; Cw1, w7 (P5). Statistical analysis Statistical significance was evaluated using the Mann–Whitney U-test. | |||||||||||||
RESULTS Transient down-regulation of HLA-B27 expression in patients with ReA Samples from seven patients who developed ReA after Salmonella infection and were positive for HLA-B27 by PCR were studied for the cell surface expression of different HLA-B27 epitopes. During the acute phase of Salmonella infection (1–4 weeks) all seven patients with ReA were positive for HLA-B27 with mAb HLA-ABC-m3. Examination of monocytes after the acute phase of infection revealed a significant decrease of cell-surface HLA-B27 in five patients (P < 0·05, Fig. 1a, a representative patient is shown in Fig. 2a). In all of these patients, the decrease of HLA-B27 epitopes was transient and after recovery the expression of HLA-B27 increased to the level found in the acute phase (Fig. 1a,2a). Patients 4–8 were studied in more detail with four mAbs recognizing different epitopes of HLA-B27. A decrease of HLA-B27 on monocytes was detected with all these mAbs but a decrease of the epitope recognized by mAb FD705 was detected 2 weeks later on monocytes from patient 5 (P5 is shown in Fig. 2a). Patient 5 was also positive for HLA-B7 (as detected with a mAb specific for HLA-B7) and its expression followed that of HLA-B27. In the patient with Yersinia-triggered ReA (P8) the expression of HLA-B27 on monocytes decreased 3 weeks after onset of the ReA, as detected by B27 M1, Me-1, FD705 and HLA-B27 Kit mAbs. In contrast, the expression of B27 M2 increased simultaneously, suggesting a role for presented peptides (Fig. 2b).
Lymphocytes were analysed separately from the same samples. Despite their weaker baseline expression in general, they showed a decrease similar to monocytes after the acute infection (data not shown). Transient decrease of monomorphic MHC class I and HLA-A2 in patients with ReA In three of eight patients with ReA (P4, P5 and P8) a major decrease was also observed in expression of the monomorphic epitope of MHC class I molecules (w6/32) studied at the same time as HLA-B27 (Fig. 1). Four patients (P1–4) had a moderate decrease of MHC class I and P3 had a slight increase of MHC class I after 6 months. P4 and P5 were also HLA-A2 positive, and the expression of HLA-A2 was markedly decreased in these patients in parallel with HLA-B27 (P5 is shown in Fig. 2a).Expression of MHC class I antigens in patients with Salmonella or Klebsiella infection without ReA To study whether Salmonella infection also caused a decrease of monomorphic MHC class I molecules in patients without ReA, monocytes from seven patients were analysed three to five times during a 6-month period. One of these patients was positive for the HLA-B27 gene. This patient had a significant down-regulation of HLA-B27 in the acute phase, at 2 weeks after the onset of the infection, but at 4 weeks the expression was already at a high, constant level as detected in four follow-up samples up to 2 years (Fig. 3a). Three patients were positive for HLA-A2, but no decrease was found on their monocytes (data not shown). A moderate decrease of monomorphic MHC class I was found in one of seven and a slight decrease in two of seven patients during the acute phase (P > 0·05, Fig. 3b). One HLA-B27-positive patient with chronic sinuitis caused by Klebsiella was followed-up for HLA-B27 expression to study further whether a decrease of MHC class I is associated with infection, HLA-B27 or ReA. This patient had a markedly decreased expression of HLA-B27 epitopes, which was detected with five different mAbs in four follow-up samples up to 4 months after the first symptoms, compared to expression after recovery (data not shown).
Expression of HLA-B27 on monocytes from healthy subjects To study baseline variability of HLA-B27 expression on monocytes from healthy subjects, cell surface expression of HLA-B27 from two healthy individuals was analysed three times during the 1-year study period. The expression of HLA-B27 on monocytes from a healthy person did not vary significantly (data from a representative subject is shown in Fig. 3c).Expression of adhesion/co-stimulatory and MHC class II molecules To study specificity of the changes observed in the expression of MHC class I antigens, other cell surface molecules were analysed from the same samples (Fig. 4). No significant changes in the expression of CD18, CD44, intracellular adhesion molecule-1 (ICAM-1), and in the expression of MHC class II antigens, (HLA)-DQ, HLA-DP and HLA-DR, were detected during the follow-up. However, in both groups of patients, the expression of several antigens, especially CD18 integrins and ICAM-1, was higher compared with healthy controls.
The decrease of HLA-B27 expression is post-transcriptional To investigate whether the decrease of HLA-B27 was pre- or post-transcriptionally regulated, the follow-up samples from four patients (P2, P3, P6 and P7) with significant changes in cell surface expression were analysed for HLA-B27 mRNA expression by semiquantitative RT–PCR. However, there were no marked differences between the follow-up samples (Fig. 5).
Intracellular expression of MHC class I antigens To investigate the mechanism of down-regulation of MHC class I expression on the cell surface, cell surface and intracellular expression of these antigens were analysed by confocal microscopy. Intracellular staining of follow-up samples with a mAb (w6/32) recognizing a folded MHC peptide complex was in agreement with the results obtained by flow cytometry. Staining with a mAb (HC-10) that recognized a free MHC heavy chain revealed increased intracellular retention during the acute phase (1 week), and particularly during the amelioration of ReA (6 months) (Fig. 6). After recovery (16 months), only cell surface staining of folded MHC class I was detected (Fig. 6).
| |||||||||||||
DISCUSSION Our results demonstrate for the first time that Salmonella, Yersinia and Klebsiella infection may cause a decrease in the expression of MHC class I antigens, especially HLA-B27, on PBMC in vivo (Table 3). During or after the acute phase of Salmonella infection, the expression of HLA-B27 on PBMC was significantly decreased in five of seven patients with ReA and in one patient without ReA. This phenomenon was detected with six mAb against HLA-B27 that recognized slightly different epitopes of the molecule. There was also a significant decrease of HLA-B27 expression during Yersinia-triggered ReA.
The simultaneous and significant decrease of HLA-A2 and monomorphic class I epitopes in the patients with ReA, but not in the patients without, suggests that there is a general underlying phenomenon linked to the HLA-B27 genotype, and that it is not restricted only to HLA-B27 antigens. Although the decrease of HLA-B27 antigens was striking, a significant variation in HLA-B27 mRNA levels was not found, but instead a marked intracellular retention of MHC molecules. This might be a result of retention in the ER or defective transport to the cell surface.21 On the other hand, the expression of co-stimulatory/adhesion and MHC class II molecules did not significantly decrease, indicating that a decrease in MHC class I expression is not caused by general deactivation of the cells or experimental artefacts. It is also possible that NSAID treatment could cause a decrease of MHC class I antigens. However, some of the patients without ReA were also treated with NSAID and showed no significant decrease of MHC class I expression. We also followed-up patients without ReA to study whether a decrease of MHC class I molecules is a general phenomenon related to infection, HLA-B27 or ReA. Only one patient with Salmonella infection without ReA was HLA-B27 positive and he had a significant down-regulation of HLA-B27 during the acute phase, but not later during the follow-up. Interestingly, one HLA-B27-positive patient with sinuitis caused by Klebsiella also had a significant decrease of HLA-B27 during acute infection and amelioration of symptoms. These results suggest that down-regulation of MHC class I alleles is predominant in the patients with the HLA-B27 genotype, with or without ReA, during or after enterobacterial infections. Down-regulation of HLA-B27 seems to occur later in patients with ReA, compared to patients without ReA, which might reflect abnormal elimination of the infection in the patients with ReA. Notably, one patient with a chronic Salmonella-and Yersinia-triggered ReA did not have any decrease of HLA-B27 during or after two relapses (unpublished observation). However, studies with an increased number of patients are needed to verify these findings. Recently, we found a marked decrease of HLA-B27, but not other MHC class I antigens, on monocytes that had ingested Yersinia or Salmonella bacteria in vitro.11 In that study, the decrease of HLA-B27 cell surface expression was most apparent on monocytes fed with Yersinia or Salmonella, intermediate with enteroinvasive Escherichia coli while no decrease was found with Streptococci. It might be caused by changes in the peptide repertoire presented on the cell surface.22 In another in vitro study, Shigella infection did not down-regulate the expression of HLA-B27 but induced clustering of HLA-B27 molecules during macropinosytosis and alterations in the peptide presentation by HLA-B27.23 The decrease in expression of HLA-B27 epitopes has probably been a reason for discrepancies found in serological typing for HLA-B27.21 We recently reported two patients with Yersinia-triggered ReA and Reiter’s disease who were negative for HLA-B27 by MLCT during the acute phase of the disease but were definitively positive 20 years later by MLCT, flow cytometry and PCR.21 Otherwise, down-regulation of MHC antigens during bacterial infection in vivo has not been previously reported but transient loss or masking of HLA-B27 epitopes has earlier been suspected in patients with ankylosing spondylitis. However, false negative typing results can be avoided by using techniques based on detection of the HLA-B27 gene.21 The changes we observed, decrease or loss of HLA-B27, might also modify T-cell recognition in these patients.26 In a recent study, the number of HLA-B27-restricted peripheral blood CTLs was low or undetectable during the acute phase of ReA but increased at the time of disease remission.27 However, the down-regulation of HLA-A2 and monomorphic MHC class I suggests that this phenomenon is not restricted only to HLA-B27. MHC class I down-regulation is a frequent event associated with viral infection and tumour invasion.1 It is probably an escape mechanism disturbing immunogenic peptide presentation to T cells.1 These viruses are mostly large DNA viruses, resembling intracellular bacteria in many aspects. In viral infections, the down-regulation of MHC class I alleles includes typically all class I alleles.28 Therefore, the decrease of MHC class I antigens after Salmonella or Yersinia infection might be analogous to viral infections. The postinfectious decrease of MHC class I epitopes on the cell surface indicates that in certain conditions enterobacterial infection may also cause a decrease of MHC molecules, especially in patients positive for HLA-B27, with or without ReA. The decrease also affects serological MHC class I typing. However, the biological background and significance of the phenomenon remain to be elucidated. | |||||||||||||
Acknowledgments This study was supported by grants from The Academy of Finland, the Sigrid Jusélius Foundation and the European Commission Biomed 2 Programme. We are grateful to Mrs Tiina Lähde, Ms Tuula Lehtonen, Mr Erkki Nieminen and Mrs Susanna Reponen for their excellent technical assistance. | |||||||||||||
REFERENCES 1. Garrido, F; Ruiz-Cabello, F; Cabrera, T, et al. Implications for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol Today. 1997;18:89. [PubMed] 2. Andersson, M; Pääbo, S; Nilsson, T, et al. Impaired intracellular transport of class I MHC antigens as a possible means for adenoviruses to evade immune surveillance. Cell. 1985;43:215. [PubMed] 3. Hill, AB; Barnett, BC; McMichael, AJ, et al. HLA class I molecules are not transported to the cell surface in cells infected with herpes simplex virus types 1 and 2. J Immunol. 1994;152:2736. [PubMed] 4. Ploegh, HL. Viral strategies of immune evasion. Science. 1998;280:248. [PubMed] 5. Sieper, J; Braun, J. Pathogenesis of spondylarthropathies. Persistent antigen, autoimmunity, or both? Arthritis Rheum. 1995;38:1547. [PubMed] 6. Feltkamp, TEW; Khan, MA; Lopez De Castro, JA. The pathogenetic role of HLA-B27. Immunol Today. 1996;17:5. [PubMed] 7. López De Castro, JA. The pathogenetic role of HLA-B27 in chronic arthritis. Curr Opin Immunol. 1998;10:59. [PubMed] 8. Kapasi, K; Inman, RD. HLA-B27 expression modulates Gram-negative bacterial invasion into transfected L cells. J Immunol. 1992;148:3554. [PubMed] 9. Virtala, M; Kirveskari, J; Granfors, K. HLA-B27 modulates the survival of Salmonella enteritidis in transfected L cells, possibly by impaired nitric oxide production. Infect Immun. 1997;65:4236. [PubMed] 10. Laitio, P; Virtala, M; Salmi, M, et al. HLA-B27 modulates intracellular survival of Salmonella enteritidis in human monocytic cells. Eur J Immunol. 1997;27:1331. [PubMed] 11. Wuorela, M; Jalkanen, S; Kirveskari, J, et al. Yersinia enterocolitica serotype O:3 alters the expression of serologic HLA-B27 epitopes on human monocytes. Infect Immun. 1997;65:2060. [PubMed] 12. Repo, H; Jäättelä, M; Leirisalo-Repo, M, et al. Production of tumour necrosis factor and interleukin 1 by monocytes of patients with previous Yersinia arthritis. Clin Exp Immunol. 1988;72:410. [PubMed] 13. Pazmany, L; Rowland-Jones, S; Huet, S, et al. Genetic modulation of antigen presentation by HLA-B27 molecules. J Exp Med. 1992;175:361. [PubMed] 14. Ikawa, T; Ikeda, M; Yamaguchi, A, et al. Expression of arthritis-causing HLA-B27 on HeLa cells promotes induction of c-fos in response to in vitro invasion by Salmonella typhimurium. J Clin Invest. 1998;101:263. [PubMed] 15. Yin, Z; Braun, J; Neure, L, et al. Crucial role of interleukin-10/interleukin-12 balance in the regulation of the type 2 T helper cytokine response in reactive arthritis. Arthritis Rheum. 1997;40:1788. [PubMed] 16. Isomäki, O; Vuento, R; Granfors, K. Serological diagnosis of salmonella infections by enzyme immunoassay. Lancet. 1989;i:1411. 17. Kirveskari, J; Jalkanen, S; Mäki-Ikola, O, et al. Increased synovial endothelium binding and transendothelial migration of mononuclear cells during Salmonella infection. Arthritis Rheum. 1998;41:1054. [PubMed] 18. Wuorela, M; Jalkanen, S; Toivanen, P, et al. Yersinia lipopolysaccharide is modified by human monocytes. Infect Immun. 1993;61:5261. [PubMed] 19. Tanaka, J; Imamura, M; Kasai, M, et al. Cytokine gene expression in peripheral blood mononuclear cells during graft-versus-host disease after allogeneic bone marrow transplantation. Br J Haematol. 1993;85:558. [PubMed] 20. Dominguez, O; Coto, E; Martinez-Naves, E, et al. Molecular typing of HLA-B27 alleles. Immunogenetics. 1992;36:277. [PubMed] 21. Kirveskari, J; Kellner, H; Wuorela, M, et al. False-negative serological HLA-B27 typing results may be due to altered antigenic epitopes and can be detected by polymerase chain reaction. Br J Rheumatol. 1997;36:185. [PubMed] 22. Wang, J; Yu, DTY; Fukazawa, T, et al. A monoclonal antibody that recognizes HLA-B27 in the context of peptides. J Immunol. 1994;152:1197. [PubMed] 23. Boisgérault, F; Mounier, J; Tieng, V, et al. Alteration of HLA-B27 peptide presentation after infection of transfected murine L cells by Shigella flexneri. Infect Immun. 1998;66:4484. [PubMed] 24. Amor, B; Kahan, A; Georgiadis, AE. Transient loss of HLA B27. Lancet. 1978;i:284. [PubMed] 25. Neumuller, J; Fischer, M; Eberl, R. Failure of the serological determination of HLA-B27 due to antigen masking in patients with ankylosing spondylitis. Rheumatol Int. 1993;13:163. [PubMed] 26. Ugrinovic, S; Mertz, A; Wu, P, et al. A single nonamer from the Yersinia 60-kDa heat shock protein is the target of HLA-B27-restricted CTL response in Yersinia-induced reactive arthritis. J Immunol. 1997;159:5715. [PubMed] 27. Ackermann, B; Gohlke, F; Herr, W, et al. Quantification of TNF-α and IFN-γ producing Yersinia urease β-subunit specific CD8+ and CD4+ T cells using an ELISPOT assay – a study in reactive arthritis. Arthritis Rheum. 1996;39:S163. 28. Kleijnen, MF; Huppa, JB; Lucin, P, et al. A mouse cytomegalovirus glycoprotein, gp34, forms a complex with folded class I MHC molecules in the ER which is not retained but is transported to the cell surface. EMBO J. 1997;16:685. [PubMed] | |||||||||||||
