HLA antigen expression on cultured human arterial endothelial cells

Negative regulation of HLA-DR expression on endothelial cells by anti-blood group A/B antibody ligation and mTOR inhibition

Donor-specific antibody (DSA), particularly against HLA class II, is a major cause of chronic antibody-mediated rejection (CAMR) after transplantation, although ABO-incompatible kidney transplantation has recently demonstrated favorable graft outcomes. The condition of no injury even in the presence of anti-donor antibody has been referred to as "accommodation", which would be one of the key factors for successful long-term graft survival. The purpose of this study was to analyze the beneficial effect of anti-blood group A/B antibody ligation on endothelial cells against HLA-DR antibody-mediated, complement-dependent cytotoxicity (CDC). Blood group A/B-expressing endothelial cells EA.hy926 or Human Umbilical Vein Endothelia Cells (HUVEC) were incubated with IFNγ in the presence or absence of anti-blood group A/B antibody or mTOR inhibitor (mTOR-i) for 48h. The effects on signaling pathway, HLA expression, complement regulatory factors, and CDC were investigated. Expression of HLA-DR on EA.hy926 or HUVEC were successfully elicited by IFNγ treatment, although little or no expression was observed in quiescent cells. Pre-incubation with anti-blood group A/B antibody had resistance to HLA-DR antibody-mediated CDC against IFNγ-treated cells in a concentration-dependent manner. This finding was ascribed to decreased expression of HLA-DR by post-translational regulation and increased expression of CD55/59, which was related to ERK and mTOR pathway inhibition. mTOR-i also inhibited HLA-DR expression by itself. Furthermore, the combination of mTOR-I and anti-blood group A/B ligation had an additive effect in preventing HLA-DR antibody-mediated CDC. Anti-blood group A/B antibody might play a preventive role in CAMR. Inhibition of the ERK and mTOR pathways may contribute to the development of a novel treatment in the maintenance period after transplantation.

Donor-specific antibody levels and three generations of crossmatches to predict antibody-mediated rejection in kidney transplantation

BACKGROUND

This study evaluated the prognostic impact of pretransplant donor-specific anti-human leukocyte antigen antibodies (DSA) detected by single-antigen beads and compared the three generations of crossmatch (XM) tests in kidney transplantation.

METHODS

Thirty-seven T-cell complement-dependent cytotoxicity crossmatch (CXM) negative living donor kidney recipients with a retrospectively positive antihuman leukocyte antigen antibody screening assay were included. A single-antigen bead test, a flow cytometry XM, and a Luminex XM (LXM) were retrospectively performed, and the results were correlated with the occurrence of antibody-mediated rejections (AMRs) and graft function.

RESULTS

We found that (1) pretransplant DSA against class I (DSA-I), but not against class II, are predictive for AMR, resulting in a sensitivity of 75% and a specificity of 90% at a level of 900 mean fluorescence intensity (MFI); (2) with increasing strength of DSA-I, the sensitivity for AMR is decreasing to 50% and the specificity is increasing to 100% at 5200 MFI; (3) the LXM for class I, but not for class II, provides a higher accuracy than the flow cytometry XM and the B-cell CXM. The specificity of all XMs is increased greatly in combination with DSA-I values more than or equal to 900 MFI.

CONCLUSIONS

In sensitized recipients, the best prediction of AMR and consecutively reduced graft function is delivered by DSA-I alone at high strength or by DSA-I at low strength in combination with the LXM or CXM.

Vascular Endothelial Cells Evade Apoptosis Triggered by Human Leukocyte Antigen-DR Ligation Mediated by Allospecific Antibodies

BACKGROUND

Human leukocyte antigen (HLA)-DR ligation mediates cell death of antigen-presenting cells (APC), including mature B cells, macrophages, and dendritic cells. This study investigates the apoptotic effects of HLA class II ligation mediated by anti-HLA antibodies on activated human vascular graft endothelial cells (ECs).

METHODS

HLA class II expression was examined by flow cytometry using a panel of HLA-typed vascular ECs isolated from transplant donors and compared with that of B lymphocytes. The apoptotic effects of anti-HLA-DR monoclonal antibodies (mAbs) were investigated using viability assays, DNA content analysis, and annexin-V labeling. Intracellular signaling pathways mediated by HLA-DR ligation on ECs were examined by Western blotting.

RESULTS

Even with optimal stimulation, the expression of HLA-DR on interferon (IFN)-gamma-treated ECs was quantitatively lower (3-5-fold) than that on B cells. Whereas anti-HLA-DR monomorphic mAbs induced apoptosis of B cells (approximately 22%), no significant apoptosis of IFN-gamma-activated (DR-positive) ECs ( < 5%), collected from the same donor, was observed under the same conditions. Similarly, specific polymorphic anti-HLA-DR11 or -DR16 antibodies were unable to induce EC apoptosis. Nevertheless, antibody-binding to HLA-DR on ECs is sufficient to induce intracellular signaling, as evident in the modulation of tyrosine phosphorylation and protein kinase (PK)C-alpha/beta and PKB/Akt activation. Our results suggest that HLA-DR ligation induces both common and divergent signaling events in ECs and B cells.

CONCLUSION

Collectively, our data suggest that, in contrast with professional APC, graft ECs evade apoptosis mediated by HLA-DR ligation, not as a result of moderate HLA-DR expression but rather as a result of a specific signaling pathway.

Non-HLA-type endothelial cell reactive alloantibodies in pre-transplant sera of kidney recipients trigger apoptosis

The vascular endothelium of transplanted organs represents an important target for allograft-directed immune responses. Although HLA antigens expressed on graft endothelial cells (EC) can become targets of the host immune response, the role of other, non-HLA-encoded EC antigens has been proposed but is still unclear. The aim of this study was to investigate the presence of and to characterize anti-EC antibodies (AECA) in 57 kidney transplant recipients according to their HLA-immunization status. Flow cytometry in pretransplant sera was used to detect AECA reactive with surface antigens on ABO and HLA-typed primary cultures of arterial ECs, stimulated or not with tumor necrosis factor-alpha (TNFalpha) or interferon-gamma (IFNgamma). FACS analysis revealed the presence of AECA in 47% of HLA-sensitized (PRA = 10%: mostly IgG) vs. 16.0% in nonsensitized patients (PRA < 10%) (p < 0.02). No significant correlation was found between the presence of AECA and acute rejection occurrence and graft outcome. Non-HLA reactive AECA are directed against TNFalpha- and IFNgamma-inducible membrane molecule(s), and react with two predominant antigens of approximately 35 kDa and approximately 50 kDa expressed on ECs but not on B cells. Binding of AECA decreases in vitro EC viability by 50-60% by promoting EC apoptosis, as demonstrated by DNA fragmentation assays.

Is presensitization relevant to liver transplantation outcome?

The impact of anti-HLA antibodies and crossmatch (CM) on liver transplantation (LT) outcome is still controversial. In this retrospective study we analyzed LT outcome according to pretransplant pre-formed anti-HLA antibodies and CM status. Serum anti-HLA antibodies were screened by ELISA assay, utilizing One Lambda antigen tray-mixed (LAT-M). CMs were performed by the complement dependent cytotoxicity test using Dithiotreitol treated sera. Anti-HLA antibodies were studied in 80 recipients; 56/80 had positive LAT-M tests (PLAT-M), whereas the remaining 24 recipients tested negative for both classes I and II (NLAT-M). Rejection episodes were more frequent in PLAT-M compared with NLAT-M group in post-LT intervals of <1 week (p = 0.05), 1 week-3 months (p = 0.035), and 3-12 months (p = 0.076). Graft and patient survival rates were better, albeit not significantly, in the NLAT-M compared with PLAT-M recipients. CM status was investigated in 62/80 recipients, 18/62 recipients had positive CM (PCM), and 44 had negative CM (NCM). Five of 18 PCM recipients (28%) experienced early graft loss compared with 1/44 (2%) with NCM (p = 0.006). Rejection episodes were more frequent within first 3 months post-LT in PCM recipients compared with NCM (p = 0.015). One-year graft survival rate was better in NCM, compared with PCM recipients (graft loss of 2/44 vs 5/18). NCM PLAT-M had a higher incidence of rejection episodes compared with the NCM NLAT-M group (p = 0.031). The presence of anti-HLA antibodies suggests a deleterious effect on LT outcome, and was associated with an increased incidence of early graft loss and rejection episodes.

Major histocompatibility complex class II-dependent antigen presentation by human intestinal endothelial cells

BACKGROUND & AIMS

In the normal gut, human intestinal microvascular endothelial cells (HIMECs) express major histocompatibility complex (MHC) class II molecules. Enhanced expression is found in chronic inflammation. We examined the cytokine regulation of MHC class II molecules and the associated invariant chain (Ii) in HIMECs and investigated whether such cells can process and present a complex protein antigen to T cells.

METHODS

Enzyme-linked immunosorbent assay, flow cytometry, immunoelectron microscopy, as well as T-cell activation assay with HIMECs and HLA-DR-restricted T-cell clones were employed.

RESULTS

In unstimulated HIMEC monolayers, HLA-DR, -DP, and -DQ and Ii were undetectable at the protein level, but interferon gamma (IFN-gamma) (100 U/mL) induced expression that peaked for DR after 2-3 days, for DP after 4-6 days, for DQ after 10-12 days, and for Ii after 2-3 days. Tumor necrosis factor alpha had no effect alone but enhanced class II expression in combination with IFN-gamma, most notably for DQ and DP. HLA-DR3-restricted and Mycobacterium tuberculosis heat shock 65-kilodalton-specific T-cell clones were activated to produce IFN-gamma in response to relevant antigen presented by IFN-gamma-treated HIMECs. This response was inhibited by blocking monoclonal antibody to HLA-DR and by chloroquine when compared to professional antigen-presenting cells, HIMECs activated T-cell clones quite efficiently.

CONCLUSIONS

These data suggest that microvascular endothelial cells can present complex protein antigens in the human gut.

Anti-HLA antibody ligation to HLA class I molecules expressed by endothelial cells stimulates tyrosine phosphorylation, inositol phosphate generation, and proliferation

The major threat to long-term survival of solid organ allografts is chronic rejection. Progressive narrowing and ultimate luminal occlusion of the arteries and arterioles of the transplanted organ are the hallmarks of the disease. The mechanism of chronic rejection is poorly understood, but it is suspected that the associated vascular changes are a result of anti-HLA antibody-mediated injury to the endothelium. We have postulated that anti-HLA antibodies initiate chronic rejection by binding to class I molecules on the endothelium and transducing signals that result in endothelial cell activation and proliferation. Our data demonstrate that anti-HLA class I antibodies transduce signals in endothelial cells stimulating increased tyrosine phosphorylation of intracellular proteins. Antibody binding to class I antigens also leads to the generation of inositol phosphate and endothelial cell proliferation. These results indicate that anti-HLA antibodies can deliver functionally important signals to endothelial cells, a finding that may be fundamental to an understanding of the mechanisms of chronic rejection.

Malignant lymphoma of the uterus. Report of seven cases with immunohistochemical study

BACKGROUND

Uterine lymphoma is a rare disease; therefore, information regarding histologic type, immunophenotype of tumor cells, and etiologic factors are limited.

METHODS

Seven patients with uterine lymphoma, three from the corpus and four from the cervix, were collected by a nationwide study in Japan. Selection of cases was preferentially made from the "Annual of Pathologic Autopsy Cases in Japan."

RESULTS

All cases with cervical lymphoma presented with vaginal bleeding. Abdominal pain or backache was observed in patients with corpus lymphoma. The age ranges of patients with corpus and cervical lymphomas were 46-78 years of age (mean, 63 years) and 30-71 years of age (mean, 53 years), respectively. Information about clinical staging was available for six patients; two patients with Stage I, three patients with Stage II, and one patient with Stage III. A definite diagnosis of uterine lymphoma was made by biopsy in all cases. Total hysterectomy, with or without bilateral salpingo-oophorectomy, was carried out in three patients, and tumor resection was carried out in one patient. Adjuvant therapy was given in six cases. Follow-up showed that five patients died due to tumor within 1 year of treatment. Histologically, all cases were non-Hodgkin lymphoma showing a diffuse pattern of proliferation. All but one were diffuse large cell type. Immunohistochemistry revealed the tumor cells in all cases were of B-cell nature. Expression of HLA-DR antigen was evaluable in four cases, of these three showed an increased expression on the vascular endothelium.

CONCLUSIONS

Uterine lymphoma comprises exclusively B-cell type.

Cytomegalovirus- and interferon-related effects on human endothelial cells. Cytomegalovirus infection reduces upregulation of HLA class II antigen expression after treatment with interferon-gamma

Cultured human umbilical vein endothelial cells (HUVEs) were infected with human cytomegalovirus (HCMV) strain AD169. Up to 50% HUVEs proved to be positive for HCMV early nuclear antigens 24 hours after inoculation with virus. Following infection kinetics of surface expression of HLA class I and II, intercellular adhesion molecule (ICAM-1) and endothelial lymphocyte adhesion molecule (ELAM-1) on HUVEs were investigated by means of flow cytometry. A slight increase in HLA class I expression was observed, whereas expression of HLA class II (DR, DP, DQ) antigens was not induced by infection with HCMV. Furthermore, when compared with uninfected cells treated with interferon-gamma (IFN-gamma), reduced enhancement of HLA-DR expression was conspicuous in HCMV-infected cells treated with IFN-gamma. There is evidence that only a portion of HUVE is affected in its ability to upregulate HLA class II antigens. While expression of ICAM-1 was found to be enhanced between 8 and 20 hours after infection with a maximum at 12 hours after infection, no modulation of ELAM-1 was seen.

Involvement of HLA class II molecules in acquisition of staphylococcal enterotoxin A-binding activity and accessory cell activity in activation of human T cells by related toxins in vascular endothelial cells

Human umbilical vascular endothelial cells (HUVEC) express HLA class II molecules upon stimulation with recombinant human interferon-gamma (IFN-gamma). Staphylococcal enterotoxin (SE) A (SEA)-binding assay using [125I]-SEA showed the presence of specific SEA binding in HUVEC stimulated with IFN-gamma but not in unstimulated HUVEC. Levels of HLA class II expression and SEA-binding increased as the IFN-gamma concentration and the period of stimulation were increased. Binding of [125I]-SEA to the IFN-gamma-stimulated HUVEC was reduced markedly by an anti-DR/DP MoAb. T cells produced IL-2 upon stimulation with a group of SEs (SEA, SEB, SEC, SED and SEE) in the presence HUVEC stimulated with IFN-gamma but not in the presence of control HUVEC. The level of accessory cell activity in the IFN-gamma-stimulated HUVEC was related to the level of HLA class II expression and SEA-binding activity. Antibodies to HLA class II molecules almost completely inhibited the response. These results indicate that HLA class II molecules are directly involved in the acquisition of these activities in HUVEC.

Cultivation of human dermal microvascular endothelial cells in vitro: immunocytochemical and ultrastructural characterization and effect of treatment with three synthetic retinoids

A new reliable and reproducible technique to culture endothelial cells from the small vessels and capillaries of human skin is introduced, and proliferation and differentiation of the growing cells are characterized. The endothelial origin of the culture cells was confirmed by light- and electron microscopy and by labelling with Ulex europaeus Agglutinin I and an antibody against Factor VIII-related antigen. Further immunocytochemical characterization showed that 92-100% of the cells were positive for beta 2-microglobulin and the entire cell population expressed vimentin, whereas cytokeratins, desmin, HLA-DR antigen, Leu 6 and S 100 protein, could not be detected. As vascular endothelium is a common site of inflammation and retinoids have been shown to be of good clinical efficacy in some chronic inflammatory skin diseases, we investigated the influence of etretinate, etretin and isotretinoin on proliferation and antigen expression of our culture cells. All retinoids applied inhibited proliferation of endothelial cells in a dose- and time-dependent manner whereas they induced neither HLA-DR nor intercellular adhesion molecule-1 (ICAM-1). Furthermore, none of the retinoids applied influenced the gamma-interferon-induced expression of these surface antigens on endothelial cells. Our results suggest that the action of retinoids in skin inflammation is not mediated by modulation of HLA-DR or ICAM-1. The cell culture technique described here is an interesting and reliable model for studying the influence of drugs on endothelial cell growth and differentiation in vitro.

Relevance of a positive crossmatch in liver transplantation

We studied 27 liver transplants in 24 patients performed between November 1984 and January 1988. We investigated retrospectively the importance of donor reactive HLA class I and class II and of non-HLA antibodies for graft survival in these patients. In order to determine the specificity and class of the antibodies, we used monoclonal antibodies to HLA-A, -B, -C and DR and DQ antigens to block cytotoxicity of sera and the reagent dithiothreitol to characterize the immunoglobulin class. We found that humoral immunity to HLA antigens in liver-grafted patients, demonstrable as the presence of cytotoxic antibodies reactive with donor splenic T and/or B cells in the pretransplantation period, is associated with significantly lower graft survival as compared with patients without demonstrable preformed HLA antibodies (P = 0.01). In addition we found that a substantial proportion of patients had donor-reactive cytotoxic antibodies which were not HLA specific. Thus, our study shows that HLA immunity can influence liver allograft survival, and that it is useful to have patient cytotoxic antibodies characterized with regard to HLA reactivity prior to transplantation.

Effects of rIFN alpha, beta, and gamma on the morphology, proliferation, and cell surface antigen expression of human dermal microvascular endothelial cells in vitro

The influence of recombinant human interferon alpha 2a (rIFN alpha), recombinant human interferon beta 1 (rIFN beta), and recombinant human interferon gamma (rIFN gamma) on human dermal microvascular endothelial cells (HDMEC) cultured in vitro was studied in various rIFN concentrations (0.1 IU/ml-10(4) IU/ml) over 2, 3, 4, 6, 8, and 10 d. Cell morphology and ultrastructure, cell proliferation, expression of class II alloantigens (HLA-DR and HLA-DQ), and intercellular adhesion molecule-1 (ICAM-1) were investigated using an in vitro technique established in our laboratory. All rIFN tested induced alterations of typical HDMEC morphology; the cells became spindle-shaped and fibroblastoid, although they maintained their endothelial cell marker expression. Also, all IFN dose- and time-dependently inhibited the proliferation of HDMEC in vitro (rIFN alpha greater than beta greater than gamma), whereby rIFN alpha exerted the strongest growth-inhibitory effect. Alkaline phosphatase anti-alkaline phosphatase (APAAP) immunocytochemistry of the cultured cells showed dose- and time-dependent stimulation of ICAM-1 and class II antigen expression only by rIFN gamma (HLA-DR greater than HLA-DQ), rIFN alpha and beta did not exert any immunomodulatory activity on HDMEC in vitro. These results indicate that HDMEC are an important target for the action of IFN. Besides growth inhibition, it seems that rIFN gamma in particular may be involved in the modulation of leucocyte adhesion and trafficking by altering the immunophenotype of the endothelial cell population.

Kinetics of class II MHC expression on cytotoxic T cells generated by skin allograft

In the present work the kinetics of class II MHC expression on OX8+ lymphocytes generated by skin allograft and its relationship to the lytic activity were studied. Mononuclear cells from the spleen of LEW (RT1(1) rats bearing BN (RT1n) skin graft for 3, 5 or 7 days were sorted out by sequential immune affinity using columns of Degalan-V26 beads treated with anti-rat or anti-mouse Ig. After depletion of B cells, T cells were precoated with W3/25 MoAb (anti-CD4 equivalent) and sorted out using an anti-mouse Ig column. The W3/25-/OX8+ cells (CD8 equivalent) were then coated with OX4 MoAb (anti-RT1.B) or murine A.TH anti-A.TL alloantiserum (anti I-E, cross-reacts with RT1.D) and were passed through a new anti-mouse Ig column in order to obtain the four subpopulations, RT1.B+, .B-, .D+ and .D-. Their specific lytic activity against BN Con A-stimulated cells increased from the 3rd to the 7th d after the skin graft. The lytic activity observed on the 3rd and 5th d was associated with all four subpopulations analyzed. In contrast, on the 7th d, the lytic activity was concentrated in the RT1.B+ subpopulation. These results, associated with the increase in the number of OX8+/RT1.B+ cells along with days after graft, suggest that RT1.B expression is not essential but is associated with the effectiveness of the cytotoxic activity. It is also possible that RT1.B expression is a marker of cytotoxic T-cell differentiation.