- Email: [email protected]
MHC class II regulation in vivo in the mouse kidney
Microbes and Infection, 1, 1999, 903−912
MHC class II regulation in vivo in the mouse kidney Tasha N. Simsa, Philip F. Hallorana,b* b
a Department of Medical Microbiology and Immunology, University of Alberta, #303, 8249–114 Street, Edmonton, Alberta T6G 2R8, Canada Department of Medicine, Division of Nephrology and Immunology, University of Alberta, #303, 8249-114 Street, Edmonton, Alberta T6G 2R8, Canada
1. Introduction The expression of MHC class II products is relevant to T-cell development and to T-cell activation in host defence against infection, in autoimmune disease, and in graft rejection. In addition, peripheral T cells require continuous ‘tonic’ stimulation through their antigen receptors to survive [46], indicating a role for MHC expression in vivo in maintaining the peripheral T-cell population [4, 37]. MHC genes are ‘housekeeping’ genes, expressed or inducible in many tissues. MHC molecules are expressed at a basal level in the healthy host and are increased during infections, tissue injury, autoimmune disease, and graft rejection [2, 5, 22, 28, 34, 40, 42]. The basal level of MHC expression in each cell reflects a variety of influences, including constitutive expression and induction by exogenous signals such as cytokines and extracellular matrix, effects of tissue injury and repair, and changing cell populations. For example, class I expression is often considered constitutive but, in mouse, class I in the basal state is partly dependent on IFN-γ [18, 19]. Thus basal IFN-γ production can contribute to basal MHC expression in nonlymphoid tissue. Inducible class II expression is regulated at the level of transcription by the class II transactivator (CIITA) [7, 47, 55, 57, 58], which is itself likely regulated transcriptionally. Short-term changes in class II expression in vivo are preceded by changes in CIITA. Other transcription factors bound to the class II promoter presumably determine the degree of transcription of individual class II genes, making CIITA more an on-off switch for the whole class II antigen presentation system than a fine tuner of expression of each class II gene. Cell-surface expression of class II molecules requires sequential interaction with other molecules in the antigen presentation pathway, namely the invariant chain (Ii), and the peptide loader HLA-DM (H2-M), the availability of which are also regulated at least in part by CIITA [6, 12, 13, 36, 43, 50]. Thus, class II expression in nonmarrow-derived tissues is potentially regulated at multiple levels.
* Correspondence and reprints Microbes and Infection 1999, 903-912
2. MHC class II regulation in mouse kidney Our laboratory studies class II regulation in vivo in nonlymphoid tissues as a possible regulator of immune recognition during organ-specific disease states and graft rejection. We use the mouse kidney as a model because of its relevance to human kidney transplantation and kidney disease, and because the kidney has little contamination from other tissues (i.e., lymphoid tissue). We use a variety of gene-disrupted (‘knockout’) mice [including mice lacking IFN-γ (GKO), IFN-γ receptor a-chains (GRKO), or the transcription factor IRF-1 (IRF-1KO)] to examine the roles of IFN-γ and its signalling pathways in various states. We have developed a battery of standard stimuli that alter renal MHC expression: rejection of a kidney allograft, rejection of a P815 ascites tumour allograft, oxazaloneinduced skin inflammation, and bacterial lipopolysaccharide (LPS) of Salmonella minnesota (25 µg i.p.) [31, 32, 47]. In addition, we induce renal tissue injury by unilateral ischemia (produced by cross-clamping the vascular pedicle of the right kidney for 60 min to produce reversible acute tubular necrosis) and by administering nephrotoxins [18, 20, 21, 52, 55]. Using these stimuli, we have identified in mouse kidneys three states of class II expression: basal, systemically induced, and local injury induced (table I). Changes in class II expression can be mediated by altered gene expression or by population changes, underscoring the complexity of studying class II regulation in vivo. 2.1. The basal state
In the basal state in humans, class II is expressed in the lymphoid organs but is also found in nonlymphoid organs on some endothelial and epithelial cells and on interstitial dendritic-like cells (IDCs) [11, 17]. (We use the term IDC to describe the interstitial class-II-positive population, including macrophages and true dendritic cells). Human kidney expresses class II on the IDCs with low but variable expression on the tubules, vascular endothelium, and glomeruli [11, 16]. In the basal state, mice do not express class II on parenchymal or endothelial cells [45]; only IDCs express class II [26]. These IDCs are presumably 903
Forum in Immunology
Sims and Halloran
Table 1. Summary of approximate MHC class II molecule fold induction compared to WT saline-injected micea. Stimuli
Basal Saline Systemic stimuli LPS (25 mg) day 3 allogeneic day 7 rIFN-γ (100 000 U) day 3 oxazolone day 6 Local injury ATN day 7 gentamicin day 7
Mice WTb
GKO
GRKO
IRF-1KO
1
1x
1x
1x
5x–7x 8x–16x 8x 6.5x
1x 1x 8x 1x
1x 1x 1x 1x
1x 2x 1x 1x
4–6x 3x
2x 2–3x
n/d 2–3x
2x n/d
a Based on standard curves, the change in specific counts per minute (cpm) bound in the radiolabelled antibody binding assay underestimates the degree of change in antigen expression: each twofold change in cpm corresponds to threefold increases in antigen levels, i.e., (cpmWT/cpmKO) x 3/2= fold increase. b WT mice are either BALB/c or 129/J.
sentinels surveying the environment for infection, damage, or stress. Basal class II expression in kidney (i.e., expression on renal IDCs) is normal in germ-free mice but is reduced by lethal irradiation, in keeping with its dependency on marrow-derived cells [10]. T-cell-deficient nude and SCID mice have normal or increased expression, which is readily increased by environmental stimuli. There is no reduction in basal class II expression in kidneys of GKO mice [18], GRKO mice (Takei et al., in press), or IRF-1 KO mice [31, 32] compared to their wild-type controls. Thus IFN-γ, its receptor, and its signalling pathway through IRF-1 are not required for basal class II molecule expression on IDC in the mouse. 2.2. Inflammatory stimuli
Class II expression is induced strongly in kidney by intense inflammation anywhere in the host in response to graft versus host disease, graft rejection, skin sensitization to oxazalone, or bacterial LPS. This response to inflammation is highly dependent on IFN-γ, but may at times reflect synergy with other cytokines, particularly TNF-α (see below). Thus intense class II expression in a rejecting transplant is a systemic process; class II is intensely induced in the host kidneys and other organs as well as in the transplanted organ (Halloran P.F., Miller L.W. and Solez, K., unpublished results). Likewise, potent local inflammation such as oxazalone skin painting induces MHC expression systemically. Administration of LPS also induces systemic class I and class II expression. IFN-γ (20 000 to 100 000 units i.p.) is the only recombinant cytokine that strongly induces class II in renal epithelial cells and endothelial cells in vivo. Class II on the renal epithelium is expressed on the basolateral membrane, presumably for surveillance by T cells in the interstitium. During intense stimulation, class II is also induced on the endothelium of arteries. TNF-α synergizes with IFN-γ to induce class II in renal epithelium, but does little without IFN-γ [9, 24]. No other cytokine that we tested induces class II in the parenchymal and endothelial cells of the kidney in vivo. In WT mice, class II is systemically increased in many epithelial and endothelial cells in response to the inflam904
matory stimuli mentioned above [18, 27, 32, 47]. These responses are essentially prevented by administration of anti-IFN-γ antibodies [18, 47] and are severely reduced in GKO, GRKO, and IRF-1 knockout mice [18] (Takei et al., in press). Thus the upregulation of class II molecule expression in response to systemic inflammatory stimuli is highly IFN-γ-dependent, but synergy with other factors such as TNF-α may be involved. In GKO mice, rIFN-γ induces renal class II expression as strongly as in WT mice [18], but GRKO mice have no response to rIFN-γ (Takei et al., in press). We previously showed that IFN-γ could induce its own mRNA in kidney [25], presenting the possibility of self amplification. Thus mice with intact IFN-γ genes might have stronger responses to rIFN-γ than do GKO mice. However, we have shown that this does not occur, at least with single doses of rIFN-γ [18]. IRF-1 is transcriptionally induced by IFN-γ and activates the transcription of target genes through sites in their promoters (ISRE, interferon-stimulated response element) [15]. When mice are challenged with LPS, oxazalone, tumor allograft, or injected rIFN-γ, the renal tubules of IRF-1KO mice show markedly less class II molecule expression than WT mice as assessed by immunoperoxidase staining of tissue sections (figure 1). The lack of class II induction in the IRF-1KO mice has been verified by RABA and mRNA assessments [31, 32]. This action of IRF-1 is probably through CIITA expression. In addition to epithelial and endothelial induction of class II, inflammatory stimuli change the IDC population. When mice are injected with rIFN-γ, LPS, or P815, their kidneys show a decrease in the number of class-II-positive IDCs (measured by counting class II stained interstitial cells in tissue sections) despite the massive induction of class II in epithelia and endothelia [18, 31]. While cessation of class II expression could explain the data, it is more likely that this change reflects migration of IDCs [1]. Thus IFN-γ may have multiple effects on renal MHC expression including upregulation of class I and class II in the parenchymal cells and some endothelial cells, as well as mobilization of the IDCs. Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
Forum in Immunology
Figure 1. IFN-γ induction of MHC class II in renal tubules is IRF-1-dependent. 129/J mice were injected i.p. with 100,000 U of rIFN-γ and their kidneys harvested at day 3. Kidneys were sectioned and MHC class II visualized by immunoperoxidase staining with magnification at 25X. Small arrows indicate interstitial cells and big arrows indicate renal tubules. A. 129/J mouse sham(PBS)-injected. B. 129/J mouse injected with rIFN-γ. C. IRF-1KO mice sham-injected. D. IRF-1KO mouse injected with rIFN-γ. Not shown: control slides showing no staining without primary antibody. Microbes and Infection 1999, 903-912
905
Forum in Immunology
2.3. The local injury response
Beginning several days after renal injury, likely during the regeneration phase, class I and class II expression is transiently induced in epithelium of the injured kidney [52]. This induction is weaker than observed with systemic stimuli (about a 2-3-fold increase) but is accompanied by an increase in class-II-positive IDCs [18, 20, 47, 55]. We observed the same response in both ischemic and toxic acute tubular necrosis (ATN) (figure 2). In unilateral ischemic ATN, the contralateral kidney, serving as a control, shows neither injury nor enhanced class II production. Ischemic or toxic injury in GKO, GRKO, and IRF1KO mice induces class II expression but at a reduced level compared to WT mice [18, 55]. Thus, class II induction by renal injury is partially dependent on IFN-γ and IRF-1 but is also partly IFN-γ-independent. The response to injury is stereotyped and independent of the injurious agent (ischemia versus toxins). The mechanisms of class II induction in epithelium are not known but may be similar to those inducing class I in the same injured epithelium. Such changes may be relevant to the ability of injury to alter the immunogenicity of the tissue in autoimmune disease and transplantation [53]. 2.4. Kidney and heart transplants
During graft rejection, class I and II molecules are massively induced both in the rejecting organ and in the host organs. Class-II-positive IDCs are reduced in the host organs while the transplanted organ becomes massively infiltrated with mononuclear cells that are class-I and -II-positive. Using vascularized heart and kidney transplants, we investigated the role of IFN-γ in the rejection of allografts in GKO hosts and by rejection in normal hosts of transplants from GRKO donors. WT mice have increases in class II induction both in the graft and in the host during allograft rejection. WT grafts into GKO hosts show very little MHC induction, yet GKO hosts reject allografts in an accelerated fashion with increased infarction and hemorrhage [3, 49]. GKO and WT mice mobilize similar cytotoxic T-cell gene expression and circulating alloantibody levels. Nevertheless, it is difficult to determine whether the protective effect of IFN-γ was on the host or the graft. Accordingly, we studied the rejection of GRKO kidney allografts in normal mice. The grafts cannot receive IFN-γ, and showed increased vascular injury and infarction at day 7. Thus IFN-γ transiently protects grafts against vascular damage and infarction around day 7 [29, 38], by a direct effect on the graft. Whether protection against injury is due to the MHC induction or to other effects of IFN-γ (i.e., induction of endothelial protective genes) is not known. The GRKO allografts at day 7 showed greatly reduced MHC class I and II induction in the graft, and had reduced infiltration by cells bearing CD3, CD8, and CD45 markers and a particularly marked reduction in CD4 cells. This observation suggests that donor class II induction in the graft promotes infiltration by host CD4 T cells, and that MHC induction in the transplant is biologically significant for the T-cell response. Our working hypothesis is that IFN-γ actually protects against alloantibody injury, but whether this is due to MHC induction or another action of IFN-γ is uncertain. Indeed, the apparent paradox may be 906
Sims and Halloran
due to the multiplicity of potential effector mechanisms operating on the graft – alloantibody versus T cells. Thus IFN-γ (and potentially MHC induction) could protect against alloantibody while promoting the somewhat slower T-cell-mediated rejection.
3. CIITA regulation CIITA is essential to most class II expression [14, 41, 48, 63] and a CIITA deficiency causes a severe immunodeficiency due to failure of CD4+ T-cell development and activation [7, 57]. We cloned and sequenced mouse CIITA mRNA to document the functional domains in the predicted protein [54]. Comparing the human and mouse CIITA predicted protein sequence [57] revealed regions of high and low homologies, indicating that previously described potential protein-binding domains are highly conserved but are separated by nonconserved regions [54, 62]. The highly conserved regions are the minimal transcriptional activation domain, the amino terminus and two regions in the carboxyl region. There are collagen-like sequences that may allow for flexibility during transactivation. In vivo changes in CIITA expression closely correlate with changes in class II expression [8, 63]. The basal expression of CIITA and class II mRNA in mice with disrupted IFN-γ genes (GKO mice) is similar to that in WT mice. Injection of IFN-γ strongly induces CIITA and class II mRNA: CIITA mRNA increased at 2 h and declined to baseline by 48 h, whereas class II mRNA increased at 24 h and returned to baseline at 7 days. Proinflammatory stimuli that induce IFN-c production (allogeneic cells and LPS) induce CIITA and class II expression in WT mice, but not in GKO mice. Ischemic and toxic injury induces CIITA and class II expression in the kidney (see figure 2), and this induction is present but reduced in mice lacking IFN-γ. We are in the process of determining the effect of CIITA gene disruption on the expression of class II in the basal, inflammation-induced, and injury states in mouse kidney. Preliminary experiments suggest that at least some of the basal expression in IDCs may be independent of CIITA. At the end of these studies we will be able to indicate the extent to which individual genes in the class II antigen presentation system (i.e., the heavy chains, Ii, H2-M, etc.) are CIITA-dependent in each of the three states of MHC expression. CIITA induction by IFN-γ is reduced by cycloheximide, suggesting that another protein is required for CIITA induction. The principal candidate is IRF-1. The time-course of the response to a single rIFN-γ injection shows that steady state mRNA levels of IRF-1 increase earlier than CIITA [32]. IRF-1KO mice have greatly reduced CIITA induction by systemic inflammatory stimuli and by rIFN-γ, and moderately reduced CIITA induction by injury. 3.1. The role of individual CIITA promoters
CIITA expression is controlled by three promoters, PI, PIII, and PIV. Each integrates messages from separate stimuli and is differentially inducible depending on the cell type (figure 3) [39, 44, 60]. In vitro, CIITA expression Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
Forum in Immunology
Figure 2. MHC class II induction in response to toxic injury by gentamicin. Mice were injected s.c. with gentamicin at 250 mg/kg for 3 days. Their kidneys were harvested at day 7, 14, or 21. A. The kidneys were homogenized and used in radiolabelled antibody (anti-class II) binding assays or mRNA was used for northern blots (class II) or in an RT-PCR system (CIITA). B. Kidneys were snap-frozen, sectioned, and stained for class II. Arrows show MHC class II expression in the kidney tubules. Panel a: sham-injected; panel b: day 7; panel c: day 14; panel d: day 21. Reprinted with permission [55]. Microbes and Infection 1999, 903-912
907
Forum in Immunology
Sims and Halloran
Figure 3. Depiction of the 5-prime specific mouse CIITA mRNAs. This drawing is representative of the utilization of the individual promoters (on left) as templates for RT-PCR. Probes specific for each unique exon 1 or a common exon 2 probe detected the amplified mRNAs. S, sense; AS, antisense. is regulated by PI in dendritic cell lines and by PIII in B cell lines, whereas PIV displays GAS and IRF-1 sites and regulates IFN-γ-inducible CIITA expression [44]. We are investigating the correlation between the steady state mRNA from each promoter-associated 5-prime mRNA and the states of class II expression in tissues. By reverse transcription-polymerase chain reaction (RT-PCR), CIITA mRNA levels of either the unique first exon (indicative of individual promoter usage) or the common exons (indicative of total full-length CIITA transcripts) are amplified. In the basal state, the spleen has high expression of mRNA from all three promoters, likely due to its mixed population of immune cells: dendritic cells, B cells, and cytokine-induced cells. To investigate the dependency of CIITA on IFN-γ in the basal state, CIITA expression was assessed in WT, GKO and IRF-1KO mice. Each promoter and the total CIITA was unchanged in the GKO, GRKO, and IRF-1KO mice. Thus basal CIITA promoter usage is independent of IFN-γ or IRF-1 (table II).
A single rIFN-c injection induces CIITA mRNA in kidney by 2 h to 4 h and corresponds to PIV mRNA levels. rIFN-γ fails to activate PIII at any time but, surprisingly, weakly increases the steady state levels of PI mRNA (the dendritic cell promoter) in kidney between 12 h and 96 h after injection. The mechanism and significance of this effect is unknown but may be related to mobilization of IDC traffic. The rIFN-γ significantly increases PIV mRNA levels in the kidney from both WT and IRF-1KO mice but the induction is reduced in the IRF-1KO mice. This induction in the IRF-1KO mice is compatible with activation via a STAT-1α site within PIV and the reduced induction is compatible with the loss of activation through the IRF-1 site. Thus the activation of PIV of CIITA at 4 h is partially IRF-1-dependent. In kidney, our data indicate that PIV probably mediates most of the CIITA mRNA induction after inflammatory stimulation because the pattern of total CIITA expression closely follows that of PIV. Further, PIV and total CIITA
Table II. Summary of CIITA promoter induction in kidney compared to WT micea,b Approximate fold inductions of CIITA mRNA d
GKO
WT
IRF-1KO
mRNA Stimulic Basal saline Systemic stimuli LPS at 12 h LPS at 96 h rIFN-γ at 4 h allogeneic Local injury ATN
PI
PIII
PIV
Total CIITA
PI
PIII
PIV
Total CIITA
PI
PIII
PIV
Total CIITA
1
1
1
1
1x
1x
1x
1x
1x
2x
1x
1x
1x 2x 1x 3x
1x 1x 2x 1x
6x 5x 8x 9x
7x 5x 8x 9x
1x 2x n/d 1x
1x 3x n/d 4x
1x 1x n/d 1x
1x 1x n/d 1x
n/d n/d 1x 3x
n/d n/d 1x 2x
n/d n/d 4x 4x
n/d n/d 3x 4x
2x
4x
5x
5x
2x
2x
2x
3x
n/d
n/d
n/d
n/d
a Based on phosphorimaging counts of the specific probe hybridized to the amplified mRNA promoter regions. bComparisons are made only within the column or between the same columns but comparing different mouse strains cStimuli include: single injections of LPS (S. minnesota at 25 mg), rIFN-γ (100,000 U), or P815 (20 million cells) and ATN (acute tubular necrosis). dWT mice are either BALB/c or 129/J. n/d, not done.
908
Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
induction is reduced or lost in mice lacking IFN-γ or IRF-1. However, some of these responses may be of an indirect or complex nature. For example, lipopolysaccharide (LPS) induces PIV and total CIITA mRNA levels in two peaks: the expected peak at 12 h and a later peak at 96 h, both of which are lost in GKO mice. Local injury in the kidney induced by ischemia or gentamicin induces the mRNA steady state levels from all three promoters in the kidneys from both WT and GKO mice [60]. ATN in GKO mice induces increases in PIV mRNA levels, but to a lesser extent than in the WT counterparts. This indicates that even the levels of the IFN-γinducible CIITA promoter, PIV, are increased by both IFN-γ-dependent and IFN-γ-independent mechanisms in response to local injury. This demonstrates that PIV is inducible by stimuli other than IFN-γ. Thus these correlations suggest (but do not prove) that changes in class II expression in kidney are mediated principally by changes in CIITA expression, governed by PIV expression in various inflammatory states. Basal CIITA expression is not dependent on either IFN-γ or IRF-1. Injury induces a complex pattern that suggests induction of all three promoters.
4. Regulation of H2-M in mouse kidney Since the antigen processing machinery is also regulated and is responsive to IFN-γ [6], we asked if the machinery is regulated in the same manner as class II in response to inflammatory stimuli and injury. We studied the regulation of mRNA for the class II peptide loader, H2-M, in the kidney in vivo after systemic stimulation or renal injury. As a marker of the response of the heterodimer, we assess H2-Ma chain (H2-Ma) steady-state mRNA levels by RT-PCR in WT, IRF-1KO, and GKO mice [56]. In WT (BALB/c and 129/J) mice, rIFN-γ injection increases class II expression at the mRNA and protein levels, to a maximum at day 3. The H2-Ma mRNA levels are increased by rIFN-γ at an earlier time point (12 h). GKO mice have similar levels of H2-Ma mRNA induction after rIFN-γ injection compared to WT mice, as expected, while IRF-1KO mice have reduced induction. In response to inflammatory stimuli, WT mice increase H2-Ma and class II. This response is reduced in IRF1-KO mice and lost in GKO mice. After injury, H2-Ma mRNA levels are induced in WT and GKO mice, but the levels of H2-Ma mRNA are reduced in GKO mice. Thus induction of H2-Ma is dependent on IFN-γ and is at least partially dependent on IRF-1, likely through CIITA. The levels of H2-Ma in response to systemic stimuli and ischemic injury parallel changes in class II expression.
Forum in Immunology
demonstrated in vitro. Renal class II expression is a prototype of class II expression in nonlymphoid tissues, and reflects a composite pattern of changes in gene expression and changes in cell populations (namely IDCs). The picture emerges of a dynamic pattern of MHC changes not just in disease states but in day-to-day encounters with tissue injury and infections. The events in the nonlymphoid organs include altered gene expression in parenchymal and endothelial cells and altered populations of interstitial cells. Although MHC changes correlate with many disease processes, the causal role of regulated class II changes during immune events in the tissues is difficult to resolve. Class II induction by IFN-γ in tubular epithelial cells confers some APC-like properties, as shown by their ability to take up and present foreign and self antigen in vitro and to stimulate hybridomas [59]. For example, SLEderived nephritis in mouse models is promoted by IFN-γ but whether MHC induction is the mechanism of this effect is not clear [30, 61]. The lack of costimulator expression (i.e., B7) by renal epithelial cells may limit their ability to trigger primary responses and may render class II on kidney epithelium tolerogenic, although this area is controversial [23, 33, 35, 51, 59]. It remains likely that the immune reactions in nonlymphoid organs are triggered primarily by IDCs (and possibly endothelium) but that MHC in parenchymal cells is used for surveillance of infected cells at the effector level. In transplantation this is further complicated by the powerful alloantibody response against MHC antigens, which can destroy endothelium and perhaps parenchymal cells that express MHC. Thus in transplants, MHC can be the priming antigen (by direct or indirect routes), the target of T effector cells, the target of alloantibody, as well as the tolerizing antigen.
5. Discussion
In vivo, the expression of class II, CIITA, and H2-Ma in the kidney exists in three different states including in the basal state, in response to systemic cytokine release during an infection, and in response to local injury. Our data suggest that if CIITA is involved in basal class II expression, it primarily reflects CIITA expression in IDCs, independent of IFN-γ and IRF-1. Induction by potent inflammatory stimuli is highly dependent on IFN-γ and reflects induction in epithelial and endothelial cells, mediated by the PIV of CIITA and largely dependent on IRF-1. This state is accompanied by poorly understood changes in PI expression and by changes in the IDC class II expression. Local tissue injury induces class II in the injured and/or recovering epithelial cells and an accumulation of a class-IIpositive interstitial infiltrate. The injury-associated changes are not as great in magnitude as the systemic IFN-γ response to inflammation but are of great complexity reflecting changes in at least PI and PIV but also in PIII, and reflecting both population changes and transcriptional changes.
It may be surprising but reassuring to those who study class II expression in vitro that class II expression in vivo in kidney follows many of the same rules as first revealed in cell lines. What is remarkable is that some of these changes are both as rapid and as intense as changes
In vivo regulation of class II expression is a readily demonstrable, consistent, and quantitatively impressive accompaniment to inflammation and tissue injury. As mechanisms of regulation are unravelled in vitro and in vivo, we need to focus more attention on experimental
Microbes and Infection 1999, 903-912
909
Forum in Immunology
situations where the significance of class II regulation in immune responses and disease states in vivo can be established.
Acknowledgments We thank Ms. Lina Kung and Dr. Lisa Helms for critical reading of the manuscript and Drs. Yutaka Takei, Nelson Goes and Mike Hobart, Ms. Joan Urmson and Mr. Vido Ramassar for their contributions to these experiments. We also thank Ms. Pam Publicover for her assistance in preparation of the manuscript. This work is supported by grants from The Medical Research Council of Canada, The Kidney Foundation of Canada, The Muttart Foundation, The Royal Canadian Legion, Fujisawa Canada, Novartis Pharmaceuticals Canada, and Roche Canada. T.N.S. is supported by a Studentship from the Alberta Heritage Foundation for Medical Research. P.F.H. is the Muttart Chair in Clinical Immunology.
References [1] Austyn J.M., New insights into the mobilization and phagocytic activity of dendritic cells, J. Exp. Med. 183 (1996) 1287. [2] Barrett M., Milton A.D., Barrett J. et al Needle biopsy evaluation of class II major histocompatibility complex antigen expression for the differential diagnosis of cyclosporine nephrotoxicity from kidney graft rejection, Transplantation 44 (1987) 223. [3] Belitsky P., Miller S.M., Gupta R., Lee S., Ghose T., Induction of MHC class II expression in recipient tissues caused by allograft rejection, Transplantation 49 (1990) 472. [4] Benoist C., Mathis D., Selection for survival? Science 276 (1997) 2000. [5] Campos L., Naji A., Deli B.C. et al Survival of MHCdeficient mouse heterotopic cardiac allografts, Transplantation 59 (1995) 187. [6] Chang C.H., Flavell R.A., Class II transactivator regulates the expression of multiple genes involved in antigen presentation, J Exp Med. 181 (1995) 765. [7] Chang C.H., Guerder S., Hong S.C., VanEwijk W., Flavell R.A., Mice lacking the MHC class II transactivator (CIITA) show tissue-specific impairment of MHC class II expression, Immunity 4 (1996) 167. [8] Chang C.H., Fontes J.D., Peterlin M., Flavell R.A., Class II transactivator (CIITA) is sufficient for the inducible expression of major histocompatibility complex class II genes, J. Exp. Med. 180 (1994) 1367. [9] Cockfield S.M., Ramassar V., Halloran P.F., Regulation of IFN-γ and tumor necrosis factor-a expression in vivo, J. Immunol. 150 (1993) 342. [10] Cockfield S.M., Urmson J., Pleasants J.R., Halloran P.F., The regulation of expression of MHC products in mice factors determining the level of expression in kidneys of normal mice, J. Immunol. 144 (1990) 2967. 910
Sims and Halloran
[11] Daar A.S., Fuggle S.V., Fabre J.W., Ting A., Morris P.J., The detailed distribution of MHC class II antigens in normal human organs, Transplantation 38 (1984) 293. [12] Denzin L.K., Cresswell P., HLA-DM induces CLIP dissociation from MHC class II alpha/beta dimers and facilitates peptide loading, Cell 82 (1995) 155. [13] Denzin L.K., Hammond C., Cresswell P., HLA-DM interactions with intermediates in HLA-DR maturation and a role for HLA-DM in stabilizing empty HLA-DR molecules, J. Exp. Med. 184 (1996) 2153. [14] Douhan III J., Lieberson R., Knoll J.H.M., Zhou H., Glimcher L.H., An isotype-specific activator of major histocompatibility complex (MHC) class II genes that is independent of class II transactivator, J. Exp. Med. 185 (1997) 1885. [15] Escalante C.R., Yie J., Thanos D., Aggarwal A.K., Structure of IRF-1 with bound DNA reveals determinants of interferon regulation, Nature 391 (1998) 103. [16] Fellstrom B., Klareskog L., Larsson E. et al Tissue distribution of macrophages, class II transplantation antigens, and receptors for platelet-derived growth factor in normal and rejected human kidneys, Transplant. Proc. 19 (1987) 3625. [17] Finn P.W., Kara C.J., Van T.T., Douhan III J., Boothby M.R., Glimcher L.H., The presence of a DNA binding complex correlates with E beta class II MHC gene expression, EMBO J. 9 (1990) 1543. [18] Goes N., Sims T., Urmson J., Vincent D., Ramassar V., Halloran P.F., Disturbed MHC regulation in the interferon-γ knockout mouse, J. Immunol. 155 (1995) 4559. [19] Goes N., Urmson J., Hobart M., Halloran P.F., The unique role of interferon-γ in the regulation of MHC expression on arterial endothelium, Transplantation 62 (1996) 1889. [20] Goes N., Urmson J., Ramassar V., Halloran P.F., Ischemic acute tubular necrosis induces an extensive local cytokine response evidence for induction of interferon-γ, transforming growth factorβ-1, granulocyte-macrophage colonystimulating factor, interleukin-2 and interleukin-10, Transplantation 59 (1995) 565. [21] Goes N., Urmson J., Vincent D., Ramassar V., Halloran P.F., Induction of major histocompatibility complex and inflammatory cytokines after ischemic injury to the kidney Lessons from the interferon-γ gene knockout mice, Transplant. Proc. 27 (1995) 771. [22] Guardiola J., Maffei A., Control of MHC class II gene expression in autoimmune, infectious, and neoplastic diseases, Crit. Rev. Immunol. 13 (1993) 247. [23] Hagerty D.T., Intercellular adhesion molecule-1 is necessary but not sufficient to activate CD4+ T cells. Discovery of a novel costimulator on kidney tubule cells, J. Immunol. 156 (1996) 3652. [24] Halloran P.F., Interferon-γ, prototype of the proinflammatory cytokines - importance in activation, suppression, and maintenance of the immune response, Transplant. Proc. 25 (1993) 10. [25] Halloran P.F., Autenried P., Ramassar V., Urmson J., Cockfield S., Local T cell responses induce widespread MHC expression. Evidence that IFN-γ induces its own expression in remote sites, J. Immunol. 148 (1992) 3837. Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
[26] Halloran P.F., Goes N., Urmson J. et al MHC expression in organ transplants lessons from the knock-out mice, Transplant. Proc. 29 (1997) 1041. [27] Halloran P.F., Jephthah-Ochola J., Urmson J., Farkas S., Systemic immunologic stimuli increase class I and II antigen expression in mouse kidney, J. Immunol. 135 (1985) 1053. [28] Halloran P.F., Urmson J., Ramassar V., Laskin C., Autenried P., Increased class I and class II MHC products and mRNA in kidneys of MRL-lpr/lpr mice during autoimmune nephritis and inhibition by cyclosporine, J. Immunol. 141 (1988) 2303. [29] Halloran P.F., Urmson J., Solez K., IFN-γ-induced MHC expression in epithelium of rejecting kidneys may protect the graft against early graft destruction by alloantibody (Abstract), J. Am. Soc. Nephrol. 9 (1998) 650A. [30] Haverty T.P., Watanabe M., Neilson E.G., Kelly C.J., Protective modulation of class II MHC gene expression in tubular epithelium by target antigen-specific antibodies. Cell-surface directed down-regulation of transcription can influence susceptibility to murine tubulointerstitial nephritis, J. Immunol. 143 (1989) 1133. [31] Hobart M., Ramassar V., Goes N., Urmson J., Halloran P.F., The induction of class I and II major histocompatibility complex by allogeneic stimulation is dependent on the transcription factor interferon regulatory factor 1 (IRF-1). Observations in IRF-1 knockout mice, Transplantation 62 (1996) 1895. [32] Hobart M., Ramassar V., Goes N., Urmson J., Halloran P.F., IFN regulatory factor 1 (IRF-1) plays a central role in the regulation of the expression of class I and II MHC genes in vivo, J. Immunol. 158 (1997) 4260. [33] Inaba K., Witmer-Pack M., Inaba M. et al The tissue distribution of the B7-2 costimulator in mice abundant expression on dendritic cells in situ and during maturation in vitro, J. Exp. Med. 180 (1994) 1849. [34] JanewayJr C.A., Bottomly K., Babich J. et al Quantitative variation in Ia antigen expression plays a central role in immune regulation, Immunol. Today 5 (1984) 99. [35] Jevnikar A.M., Singer G.G., Coffman T., Glimcher L.H., Rubin-Kelley V.E., Transgenictubular cell expression of class IIIs sufficient to initiate immune renal injury,J. Am. Soc. Nephrol. 3 (1993) 1972. [36] Katz J.F., Stebbins C., Appella E., Sant A.J., Invariant chain and DM edit self-peptide presentation by major histocompatibility complex (MHC) class II molecules, J. Exp. Med. 184 (1996) 1747. [37] Kirberg J., Berns A., VonBoehmer H., Peripheral T cell survival requires continual ligation of the T cell receptor to major histocompatibility complex-encoded molecules, J. Exp. Med. 186 (1997) 1269. [38] Konieczny B.T., Dai Z., Elwood E.T. et al IFN-γ is critical for long-term allograft survival induced by blocking the CD28 and CD40 ligand T cell costimulation pathways, J. Immunol. 160 (1998) 2059. [39] Lennon A.M., Ottone C., Rigaud G. et al Isolation of a B-cell-specific promoter for the human class II transactivator, Immunogenetics 45 (1997) 266. Microbes and Infection 1999, 903-912
Forum in Immunology
[40] Letterio J.J., Geiser A.G., Kulkarni A.B. et al Autoimmunity associated with TGF-β 1-deficiency in mice is dependent on MHC class II antigen expression, J. Clin. Invest. 98 (1996) 2109. [41] Mach B., Steimle V., Martinez-Soria E., Reith W., Regulation of MHC class II genes lessons from a disease, Ann. Rev. Immunol. 14 (1996) 301. [42] Milton A.D., Fabre J.W., Massive induction of donor type class I and class II major histocompatibility complex antigens in rejecting cardiac allografts in the rat, J. Exp. Med. 161 (1985) 98. [43] Morris P., Shaman J., Attaya M. et al An essential role for HLA-DM in antigen presentation by class II major histocompatibility molecules, Nature 368 (1994) 551. [44] Muhlethaler-Mottet A., Otten L.A., Steimle V., Mach B., Expression of MHC class II molecules in different cellular and functional compartments is controlled by differential usage of multiple promoters of the transactivator CIITA, EMBO J. 16 (1997) 2851. [45] Natali P.G., Quaranta V., Nicotra M.R., Apollonj C., Pellegrino M.A., Ferrone S., Tissue distribution of Ia-like antigens in different species analysis with monoclonal antibodies, Immunobiology 169 (1985) 228. [46] Neuberger M.S., Antigen receptor signaling gives lymphocytes a long life, Cell 90 (1997) 971. [47] Ramassar V., Goes N., Hobart M., Halloran P.F., Evidence for the in vivo role of class II transactivator in basal and IFN-γ induced class II expression in mouse tissue, Transplantation 62 (1996) 1901. [48] Riley J.L., Westerheide S.D., Price J.A., Brown J.A., Boss J.M., Activation of class II MHC genes requires both the X box region and the class II transactivator (CIITA), Immunity 2 (1995) 533. [49] Saleem S., Konieczny B.T., Lowry R.P., Baddoura F.K., Lakkis F.G., Acute rejection of vascularized heart allografts in the absence of IFN-γ, Transplantation 62 (1996) 1908. [50] Sanderson F., Thomas C., Neefjes J., Trowsdale J., Association between HLA-DM and HLA-DR in vivo. Immunity 4 (1996) 87. [51] Schwartz R.H., Models of T-cell anergy is there a common molecular mechanism? J. Exp. Med. 184 (1996) 1. [52] Shoskes D., Parfrey N.A., Halloran P.F., Increased major histocompatibility complex antigen expression in unilateral ischemic acute tubular necrosis in the mouse, Transplantation 49 (1990) 201. [53] Shoskes D.A., Halloran P.F., Delayed graft function in renal transplantation etiology, management and long-term significance, J. Urol. 155 (1996) 1831. [54] Sims T.N., Elliott J.F., Ramassar V., DenneyJr D.W., Halloran P.F., Mouse CIITA cDNA sequence and amino acid comparison with human CIITA, Immunogenetics 45 (1997) 220. [55] Sims T.N., Goes N.B., Ramassar V., Urmson J., Halloran P.F., In vivo class II transactivator expression in mouse is induced by a non-IFN-γ mechanism in response to local injury, Transplantation 64 (1997) 1657. [56] Sims T.N., Pahl A., Ramassar V., Halloran P.F., Regulation of LMP7, TAP1 and H2-MA RNA levels in response to IFN-γ and injury in the kidney (Abstract), FASEB J. 12 (1998) A590. 911
Forum in Immunology
[57] Steimle V., Otten L.A., Zufferey M., Mach B., Complementation cloning of an MHC class II transactivator mutated in hereditary MHC class II deficiency (or Bare Lymphocyte Syndrome), Cell 75 (1993) 135. [58] Steimle V., Siegrist C.A., Mottet A., Lisowska-Grospierre B., Mach B., Regulation of MHC class II expression by interferon-γ mediated by the transactivator gene CIITA, Science 265 (1994) 106. [59] Strutz F., Neilson E.G., The role of lymphocytes in the progression of interstitial disease, Kidney Int. 45 (S45) (1994) 106. [60] Takei Y., Sims T.N., Halloran P.F., Differential usage of the class II transactivator (CIITA) promoters in vivo (Abstract), Transplantation 65 (1998) S30.
912
Sims and Halloran
[61] Wuthrich R.P., Yui M.A., Mazoujian G., Nabavi N., Glimcher L.H., Kelley V.E., Enhanced MHC class II expression in renal proximal tubules precedes loss of renal function in MRL/lpr mice with lupus nephritis, Am. J. Pathol. 134 (1989) 45. [62] Zhou H., Glimcher L.H., Human MHC class II gene transcription directed by the carboxyl terminus of CIITA, one of the defective genes in type II MHC combined immune deficiency, Immunity2 (1995) 545. [63] Zhou H., Su H.S., Zhang X., DouhanIii J., Glimcher L.H., CIITA-dependent and -independent class II MHC expression revealed by a dominant negative mutant, J. Immunol. 158 (1997) 4741.
Microbes and Infection 1999, 903-912
MHC class II regulation in vivo in the mouse kidney Tasha N. Simsa, Philip F. Hallorana,b* b
a Department of Medical Microbiology and Immunology, University of Alberta, #303, 8249–114 Street, Edmonton, Alberta T6G 2R8, Canada Department of Medicine, Division of Nephrology and Immunology, University of Alberta, #303, 8249-114 Street, Edmonton, Alberta T6G 2R8, Canada
1. Introduction The expression of MHC class II products is relevant to T-cell development and to T-cell activation in host defence against infection, in autoimmune disease, and in graft rejection. In addition, peripheral T cells require continuous ‘tonic’ stimulation through their antigen receptors to survive [46], indicating a role for MHC expression in vivo in maintaining the peripheral T-cell population [4, 37]. MHC genes are ‘housekeeping’ genes, expressed or inducible in many tissues. MHC molecules are expressed at a basal level in the healthy host and are increased during infections, tissue injury, autoimmune disease, and graft rejection [2, 5, 22, 28, 34, 40, 42]. The basal level of MHC expression in each cell reflects a variety of influences, including constitutive expression and induction by exogenous signals such as cytokines and extracellular matrix, effects of tissue injury and repair, and changing cell populations. For example, class I expression is often considered constitutive but, in mouse, class I in the basal state is partly dependent on IFN-γ [18, 19]. Thus basal IFN-γ production can contribute to basal MHC expression in nonlymphoid tissue. Inducible class II expression is regulated at the level of transcription by the class II transactivator (CIITA) [7, 47, 55, 57, 58], which is itself likely regulated transcriptionally. Short-term changes in class II expression in vivo are preceded by changes in CIITA. Other transcription factors bound to the class II promoter presumably determine the degree of transcription of individual class II genes, making CIITA more an on-off switch for the whole class II antigen presentation system than a fine tuner of expression of each class II gene. Cell-surface expression of class II molecules requires sequential interaction with other molecules in the antigen presentation pathway, namely the invariant chain (Ii), and the peptide loader HLA-DM (H2-M), the availability of which are also regulated at least in part by CIITA [6, 12, 13, 36, 43, 50]. Thus, class II expression in nonmarrow-derived tissues is potentially regulated at multiple levels.
* Correspondence and reprints Microbes and Infection 1999, 903-912
2. MHC class II regulation in mouse kidney Our laboratory studies class II regulation in vivo in nonlymphoid tissues as a possible regulator of immune recognition during organ-specific disease states and graft rejection. We use the mouse kidney as a model because of its relevance to human kidney transplantation and kidney disease, and because the kidney has little contamination from other tissues (i.e., lymphoid tissue). We use a variety of gene-disrupted (‘knockout’) mice [including mice lacking IFN-γ (GKO), IFN-γ receptor a-chains (GRKO), or the transcription factor IRF-1 (IRF-1KO)] to examine the roles of IFN-γ and its signalling pathways in various states. We have developed a battery of standard stimuli that alter renal MHC expression: rejection of a kidney allograft, rejection of a P815 ascites tumour allograft, oxazaloneinduced skin inflammation, and bacterial lipopolysaccharide (LPS) of Salmonella minnesota (25 µg i.p.) [31, 32, 47]. In addition, we induce renal tissue injury by unilateral ischemia (produced by cross-clamping the vascular pedicle of the right kidney for 60 min to produce reversible acute tubular necrosis) and by administering nephrotoxins [18, 20, 21, 52, 55]. Using these stimuli, we have identified in mouse kidneys three states of class II expression: basal, systemically induced, and local injury induced (table I). Changes in class II expression can be mediated by altered gene expression or by population changes, underscoring the complexity of studying class II regulation in vivo. 2.1. The basal state
In the basal state in humans, class II is expressed in the lymphoid organs but is also found in nonlymphoid organs on some endothelial and epithelial cells and on interstitial dendritic-like cells (IDCs) [11, 17]. (We use the term IDC to describe the interstitial class-II-positive population, including macrophages and true dendritic cells). Human kidney expresses class II on the IDCs with low but variable expression on the tubules, vascular endothelium, and glomeruli [11, 16]. In the basal state, mice do not express class II on parenchymal or endothelial cells [45]; only IDCs express class II [26]. These IDCs are presumably 903
Forum in Immunology
Sims and Halloran
Table 1. Summary of approximate MHC class II molecule fold induction compared to WT saline-injected micea. Stimuli
Basal Saline Systemic stimuli LPS (25 mg) day 3 allogeneic day 7 rIFN-γ (100 000 U) day 3 oxazolone day 6 Local injury ATN day 7 gentamicin day 7
Mice WTb
GKO
GRKO
IRF-1KO
1
1x
1x
1x
5x–7x 8x–16x 8x 6.5x
1x 1x 8x 1x
1x 1x 1x 1x
1x 2x 1x 1x
4–6x 3x
2x 2–3x
n/d 2–3x
2x n/d
a Based on standard curves, the change in specific counts per minute (cpm) bound in the radiolabelled antibody binding assay underestimates the degree of change in antigen expression: each twofold change in cpm corresponds to threefold increases in antigen levels, i.e., (cpmWT/cpmKO) x 3/2= fold increase. b WT mice are either BALB/c or 129/J.
sentinels surveying the environment for infection, damage, or stress. Basal class II expression in kidney (i.e., expression on renal IDCs) is normal in germ-free mice but is reduced by lethal irradiation, in keeping with its dependency on marrow-derived cells [10]. T-cell-deficient nude and SCID mice have normal or increased expression, which is readily increased by environmental stimuli. There is no reduction in basal class II expression in kidneys of GKO mice [18], GRKO mice (Takei et al., in press), or IRF-1 KO mice [31, 32] compared to their wild-type controls. Thus IFN-γ, its receptor, and its signalling pathway through IRF-1 are not required for basal class II molecule expression on IDC in the mouse. 2.2. Inflammatory stimuli
Class II expression is induced strongly in kidney by intense inflammation anywhere in the host in response to graft versus host disease, graft rejection, skin sensitization to oxazalone, or bacterial LPS. This response to inflammation is highly dependent on IFN-γ, but may at times reflect synergy with other cytokines, particularly TNF-α (see below). Thus intense class II expression in a rejecting transplant is a systemic process; class II is intensely induced in the host kidneys and other organs as well as in the transplanted organ (Halloran P.F., Miller L.W. and Solez, K., unpublished results). Likewise, potent local inflammation such as oxazalone skin painting induces MHC expression systemically. Administration of LPS also induces systemic class I and class II expression. IFN-γ (20 000 to 100 000 units i.p.) is the only recombinant cytokine that strongly induces class II in renal epithelial cells and endothelial cells in vivo. Class II on the renal epithelium is expressed on the basolateral membrane, presumably for surveillance by T cells in the interstitium. During intense stimulation, class II is also induced on the endothelium of arteries. TNF-α synergizes with IFN-γ to induce class II in renal epithelium, but does little without IFN-γ [9, 24]. No other cytokine that we tested induces class II in the parenchymal and endothelial cells of the kidney in vivo. In WT mice, class II is systemically increased in many epithelial and endothelial cells in response to the inflam904
matory stimuli mentioned above [18, 27, 32, 47]. These responses are essentially prevented by administration of anti-IFN-γ antibodies [18, 47] and are severely reduced in GKO, GRKO, and IRF-1 knockout mice [18] (Takei et al., in press). Thus the upregulation of class II molecule expression in response to systemic inflammatory stimuli is highly IFN-γ-dependent, but synergy with other factors such as TNF-α may be involved. In GKO mice, rIFN-γ induces renal class II expression as strongly as in WT mice [18], but GRKO mice have no response to rIFN-γ (Takei et al., in press). We previously showed that IFN-γ could induce its own mRNA in kidney [25], presenting the possibility of self amplification. Thus mice with intact IFN-γ genes might have stronger responses to rIFN-γ than do GKO mice. However, we have shown that this does not occur, at least with single doses of rIFN-γ [18]. IRF-1 is transcriptionally induced by IFN-γ and activates the transcription of target genes through sites in their promoters (ISRE, interferon-stimulated response element) [15]. When mice are challenged with LPS, oxazalone, tumor allograft, or injected rIFN-γ, the renal tubules of IRF-1KO mice show markedly less class II molecule expression than WT mice as assessed by immunoperoxidase staining of tissue sections (figure 1). The lack of class II induction in the IRF-1KO mice has been verified by RABA and mRNA assessments [31, 32]. This action of IRF-1 is probably through CIITA expression. In addition to epithelial and endothelial induction of class II, inflammatory stimuli change the IDC population. When mice are injected with rIFN-γ, LPS, or P815, their kidneys show a decrease in the number of class-II-positive IDCs (measured by counting class II stained interstitial cells in tissue sections) despite the massive induction of class II in epithelia and endothelia [18, 31]. While cessation of class II expression could explain the data, it is more likely that this change reflects migration of IDCs [1]. Thus IFN-γ may have multiple effects on renal MHC expression including upregulation of class I and class II in the parenchymal cells and some endothelial cells, as well as mobilization of the IDCs. Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
Forum in Immunology
Figure 1. IFN-γ induction of MHC class II in renal tubules is IRF-1-dependent. 129/J mice were injected i.p. with 100,000 U of rIFN-γ and their kidneys harvested at day 3. Kidneys were sectioned and MHC class II visualized by immunoperoxidase staining with magnification at 25X. Small arrows indicate interstitial cells and big arrows indicate renal tubules. A. 129/J mouse sham(PBS)-injected. B. 129/J mouse injected with rIFN-γ. C. IRF-1KO mice sham-injected. D. IRF-1KO mouse injected with rIFN-γ. Not shown: control slides showing no staining without primary antibody. Microbes and Infection 1999, 903-912
905
Forum in Immunology
2.3. The local injury response
Beginning several days after renal injury, likely during the regeneration phase, class I and class II expression is transiently induced in epithelium of the injured kidney [52]. This induction is weaker than observed with systemic stimuli (about a 2-3-fold increase) but is accompanied by an increase in class-II-positive IDCs [18, 20, 47, 55]. We observed the same response in both ischemic and toxic acute tubular necrosis (ATN) (figure 2). In unilateral ischemic ATN, the contralateral kidney, serving as a control, shows neither injury nor enhanced class II production. Ischemic or toxic injury in GKO, GRKO, and IRF1KO mice induces class II expression but at a reduced level compared to WT mice [18, 55]. Thus, class II induction by renal injury is partially dependent on IFN-γ and IRF-1 but is also partly IFN-γ-independent. The response to injury is stereotyped and independent of the injurious agent (ischemia versus toxins). The mechanisms of class II induction in epithelium are not known but may be similar to those inducing class I in the same injured epithelium. Such changes may be relevant to the ability of injury to alter the immunogenicity of the tissue in autoimmune disease and transplantation [53]. 2.4. Kidney and heart transplants
During graft rejection, class I and II molecules are massively induced both in the rejecting organ and in the host organs. Class-II-positive IDCs are reduced in the host organs while the transplanted organ becomes massively infiltrated with mononuclear cells that are class-I and -II-positive. Using vascularized heart and kidney transplants, we investigated the role of IFN-γ in the rejection of allografts in GKO hosts and by rejection in normal hosts of transplants from GRKO donors. WT mice have increases in class II induction both in the graft and in the host during allograft rejection. WT grafts into GKO hosts show very little MHC induction, yet GKO hosts reject allografts in an accelerated fashion with increased infarction and hemorrhage [3, 49]. GKO and WT mice mobilize similar cytotoxic T-cell gene expression and circulating alloantibody levels. Nevertheless, it is difficult to determine whether the protective effect of IFN-γ was on the host or the graft. Accordingly, we studied the rejection of GRKO kidney allografts in normal mice. The grafts cannot receive IFN-γ, and showed increased vascular injury and infarction at day 7. Thus IFN-γ transiently protects grafts against vascular damage and infarction around day 7 [29, 38], by a direct effect on the graft. Whether protection against injury is due to the MHC induction or to other effects of IFN-γ (i.e., induction of endothelial protective genes) is not known. The GRKO allografts at day 7 showed greatly reduced MHC class I and II induction in the graft, and had reduced infiltration by cells bearing CD3, CD8, and CD45 markers and a particularly marked reduction in CD4 cells. This observation suggests that donor class II induction in the graft promotes infiltration by host CD4 T cells, and that MHC induction in the transplant is biologically significant for the T-cell response. Our working hypothesis is that IFN-γ actually protects against alloantibody injury, but whether this is due to MHC induction or another action of IFN-γ is uncertain. Indeed, the apparent paradox may be 906
Sims and Halloran
due to the multiplicity of potential effector mechanisms operating on the graft – alloantibody versus T cells. Thus IFN-γ (and potentially MHC induction) could protect against alloantibody while promoting the somewhat slower T-cell-mediated rejection.
3. CIITA regulation CIITA is essential to most class II expression [14, 41, 48, 63] and a CIITA deficiency causes a severe immunodeficiency due to failure of CD4+ T-cell development and activation [7, 57]. We cloned and sequenced mouse CIITA mRNA to document the functional domains in the predicted protein [54]. Comparing the human and mouse CIITA predicted protein sequence [57] revealed regions of high and low homologies, indicating that previously described potential protein-binding domains are highly conserved but are separated by nonconserved regions [54, 62]. The highly conserved regions are the minimal transcriptional activation domain, the amino terminus and two regions in the carboxyl region. There are collagen-like sequences that may allow for flexibility during transactivation. In vivo changes in CIITA expression closely correlate with changes in class II expression [8, 63]. The basal expression of CIITA and class II mRNA in mice with disrupted IFN-γ genes (GKO mice) is similar to that in WT mice. Injection of IFN-γ strongly induces CIITA and class II mRNA: CIITA mRNA increased at 2 h and declined to baseline by 48 h, whereas class II mRNA increased at 24 h and returned to baseline at 7 days. Proinflammatory stimuli that induce IFN-c production (allogeneic cells and LPS) induce CIITA and class II expression in WT mice, but not in GKO mice. Ischemic and toxic injury induces CIITA and class II expression in the kidney (see figure 2), and this induction is present but reduced in mice lacking IFN-γ. We are in the process of determining the effect of CIITA gene disruption on the expression of class II in the basal, inflammation-induced, and injury states in mouse kidney. Preliminary experiments suggest that at least some of the basal expression in IDCs may be independent of CIITA. At the end of these studies we will be able to indicate the extent to which individual genes in the class II antigen presentation system (i.e., the heavy chains, Ii, H2-M, etc.) are CIITA-dependent in each of the three states of MHC expression. CIITA induction by IFN-γ is reduced by cycloheximide, suggesting that another protein is required for CIITA induction. The principal candidate is IRF-1. The time-course of the response to a single rIFN-γ injection shows that steady state mRNA levels of IRF-1 increase earlier than CIITA [32]. IRF-1KO mice have greatly reduced CIITA induction by systemic inflammatory stimuli and by rIFN-γ, and moderately reduced CIITA induction by injury. 3.1. The role of individual CIITA promoters
CIITA expression is controlled by three promoters, PI, PIII, and PIV. Each integrates messages from separate stimuli and is differentially inducible depending on the cell type (figure 3) [39, 44, 60]. In vitro, CIITA expression Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
Forum in Immunology
Figure 2. MHC class II induction in response to toxic injury by gentamicin. Mice were injected s.c. with gentamicin at 250 mg/kg for 3 days. Their kidneys were harvested at day 7, 14, or 21. A. The kidneys were homogenized and used in radiolabelled antibody (anti-class II) binding assays or mRNA was used for northern blots (class II) or in an RT-PCR system (CIITA). B. Kidneys were snap-frozen, sectioned, and stained for class II. Arrows show MHC class II expression in the kidney tubules. Panel a: sham-injected; panel b: day 7; panel c: day 14; panel d: day 21. Reprinted with permission [55]. Microbes and Infection 1999, 903-912
907
Forum in Immunology
Sims and Halloran
Figure 3. Depiction of the 5-prime specific mouse CIITA mRNAs. This drawing is representative of the utilization of the individual promoters (on left) as templates for RT-PCR. Probes specific for each unique exon 1 or a common exon 2 probe detected the amplified mRNAs. S, sense; AS, antisense. is regulated by PI in dendritic cell lines and by PIII in B cell lines, whereas PIV displays GAS and IRF-1 sites and regulates IFN-γ-inducible CIITA expression [44]. We are investigating the correlation between the steady state mRNA from each promoter-associated 5-prime mRNA and the states of class II expression in tissues. By reverse transcription-polymerase chain reaction (RT-PCR), CIITA mRNA levels of either the unique first exon (indicative of individual promoter usage) or the common exons (indicative of total full-length CIITA transcripts) are amplified. In the basal state, the spleen has high expression of mRNA from all three promoters, likely due to its mixed population of immune cells: dendritic cells, B cells, and cytokine-induced cells. To investigate the dependency of CIITA on IFN-γ in the basal state, CIITA expression was assessed in WT, GKO and IRF-1KO mice. Each promoter and the total CIITA was unchanged in the GKO, GRKO, and IRF-1KO mice. Thus basal CIITA promoter usage is independent of IFN-γ or IRF-1 (table II).
A single rIFN-c injection induces CIITA mRNA in kidney by 2 h to 4 h and corresponds to PIV mRNA levels. rIFN-γ fails to activate PIII at any time but, surprisingly, weakly increases the steady state levels of PI mRNA (the dendritic cell promoter) in kidney between 12 h and 96 h after injection. The mechanism and significance of this effect is unknown but may be related to mobilization of IDC traffic. The rIFN-γ significantly increases PIV mRNA levels in the kidney from both WT and IRF-1KO mice but the induction is reduced in the IRF-1KO mice. This induction in the IRF-1KO mice is compatible with activation via a STAT-1α site within PIV and the reduced induction is compatible with the loss of activation through the IRF-1 site. Thus the activation of PIV of CIITA at 4 h is partially IRF-1-dependent. In kidney, our data indicate that PIV probably mediates most of the CIITA mRNA induction after inflammatory stimulation because the pattern of total CIITA expression closely follows that of PIV. Further, PIV and total CIITA
Table II. Summary of CIITA promoter induction in kidney compared to WT micea,b Approximate fold inductions of CIITA mRNA d
GKO
WT
IRF-1KO
mRNA Stimulic Basal saline Systemic stimuli LPS at 12 h LPS at 96 h rIFN-γ at 4 h allogeneic Local injury ATN
PI
PIII
PIV
Total CIITA
PI
PIII
PIV
Total CIITA
PI
PIII
PIV
Total CIITA
1
1
1
1
1x
1x
1x
1x
1x
2x
1x
1x
1x 2x 1x 3x
1x 1x 2x 1x
6x 5x 8x 9x
7x 5x 8x 9x
1x 2x n/d 1x
1x 3x n/d 4x
1x 1x n/d 1x
1x 1x n/d 1x
n/d n/d 1x 3x
n/d n/d 1x 2x
n/d n/d 4x 4x
n/d n/d 3x 4x
2x
4x
5x
5x
2x
2x
2x
3x
n/d
n/d
n/d
n/d
a Based on phosphorimaging counts of the specific probe hybridized to the amplified mRNA promoter regions. bComparisons are made only within the column or between the same columns but comparing different mouse strains cStimuli include: single injections of LPS (S. minnesota at 25 mg), rIFN-γ (100,000 U), or P815 (20 million cells) and ATN (acute tubular necrosis). dWT mice are either BALB/c or 129/J. n/d, not done.
908
Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
induction is reduced or lost in mice lacking IFN-γ or IRF-1. However, some of these responses may be of an indirect or complex nature. For example, lipopolysaccharide (LPS) induces PIV and total CIITA mRNA levels in two peaks: the expected peak at 12 h and a later peak at 96 h, both of which are lost in GKO mice. Local injury in the kidney induced by ischemia or gentamicin induces the mRNA steady state levels from all three promoters in the kidneys from both WT and GKO mice [60]. ATN in GKO mice induces increases in PIV mRNA levels, but to a lesser extent than in the WT counterparts. This indicates that even the levels of the IFN-γinducible CIITA promoter, PIV, are increased by both IFN-γ-dependent and IFN-γ-independent mechanisms in response to local injury. This demonstrates that PIV is inducible by stimuli other than IFN-γ. Thus these correlations suggest (but do not prove) that changes in class II expression in kidney are mediated principally by changes in CIITA expression, governed by PIV expression in various inflammatory states. Basal CIITA expression is not dependent on either IFN-γ or IRF-1. Injury induces a complex pattern that suggests induction of all three promoters.
4. Regulation of H2-M in mouse kidney Since the antigen processing machinery is also regulated and is responsive to IFN-γ [6], we asked if the machinery is regulated in the same manner as class II in response to inflammatory stimuli and injury. We studied the regulation of mRNA for the class II peptide loader, H2-M, in the kidney in vivo after systemic stimulation or renal injury. As a marker of the response of the heterodimer, we assess H2-Ma chain (H2-Ma) steady-state mRNA levels by RT-PCR in WT, IRF-1KO, and GKO mice [56]. In WT (BALB/c and 129/J) mice, rIFN-γ injection increases class II expression at the mRNA and protein levels, to a maximum at day 3. The H2-Ma mRNA levels are increased by rIFN-γ at an earlier time point (12 h). GKO mice have similar levels of H2-Ma mRNA induction after rIFN-γ injection compared to WT mice, as expected, while IRF-1KO mice have reduced induction. In response to inflammatory stimuli, WT mice increase H2-Ma and class II. This response is reduced in IRF1-KO mice and lost in GKO mice. After injury, H2-Ma mRNA levels are induced in WT and GKO mice, but the levels of H2-Ma mRNA are reduced in GKO mice. Thus induction of H2-Ma is dependent on IFN-γ and is at least partially dependent on IRF-1, likely through CIITA. The levels of H2-Ma in response to systemic stimuli and ischemic injury parallel changes in class II expression.
Forum in Immunology
demonstrated in vitro. Renal class II expression is a prototype of class II expression in nonlymphoid tissues, and reflects a composite pattern of changes in gene expression and changes in cell populations (namely IDCs). The picture emerges of a dynamic pattern of MHC changes not just in disease states but in day-to-day encounters with tissue injury and infections. The events in the nonlymphoid organs include altered gene expression in parenchymal and endothelial cells and altered populations of interstitial cells. Although MHC changes correlate with many disease processes, the causal role of regulated class II changes during immune events in the tissues is difficult to resolve. Class II induction by IFN-γ in tubular epithelial cells confers some APC-like properties, as shown by their ability to take up and present foreign and self antigen in vitro and to stimulate hybridomas [59]. For example, SLEderived nephritis in mouse models is promoted by IFN-γ but whether MHC induction is the mechanism of this effect is not clear [30, 61]. The lack of costimulator expression (i.e., B7) by renal epithelial cells may limit their ability to trigger primary responses and may render class II on kidney epithelium tolerogenic, although this area is controversial [23, 33, 35, 51, 59]. It remains likely that the immune reactions in nonlymphoid organs are triggered primarily by IDCs (and possibly endothelium) but that MHC in parenchymal cells is used for surveillance of infected cells at the effector level. In transplantation this is further complicated by the powerful alloantibody response against MHC antigens, which can destroy endothelium and perhaps parenchymal cells that express MHC. Thus in transplants, MHC can be the priming antigen (by direct or indirect routes), the target of T effector cells, the target of alloantibody, as well as the tolerizing antigen.
5. Discussion
In vivo, the expression of class II, CIITA, and H2-Ma in the kidney exists in three different states including in the basal state, in response to systemic cytokine release during an infection, and in response to local injury. Our data suggest that if CIITA is involved in basal class II expression, it primarily reflects CIITA expression in IDCs, independent of IFN-γ and IRF-1. Induction by potent inflammatory stimuli is highly dependent on IFN-γ and reflects induction in epithelial and endothelial cells, mediated by the PIV of CIITA and largely dependent on IRF-1. This state is accompanied by poorly understood changes in PI expression and by changes in the IDC class II expression. Local tissue injury induces class II in the injured and/or recovering epithelial cells and an accumulation of a class-IIpositive interstitial infiltrate. The injury-associated changes are not as great in magnitude as the systemic IFN-γ response to inflammation but are of great complexity reflecting changes in at least PI and PIV but also in PIII, and reflecting both population changes and transcriptional changes.
It may be surprising but reassuring to those who study class II expression in vitro that class II expression in vivo in kidney follows many of the same rules as first revealed in cell lines. What is remarkable is that some of these changes are both as rapid and as intense as changes
In vivo regulation of class II expression is a readily demonstrable, consistent, and quantitatively impressive accompaniment to inflammation and tissue injury. As mechanisms of regulation are unravelled in vitro and in vivo, we need to focus more attention on experimental
Microbes and Infection 1999, 903-912
909
Forum in Immunology
situations where the significance of class II regulation in immune responses and disease states in vivo can be established.
Acknowledgments We thank Ms. Lina Kung and Dr. Lisa Helms for critical reading of the manuscript and Drs. Yutaka Takei, Nelson Goes and Mike Hobart, Ms. Joan Urmson and Mr. Vido Ramassar for their contributions to these experiments. We also thank Ms. Pam Publicover for her assistance in preparation of the manuscript. This work is supported by grants from The Medical Research Council of Canada, The Kidney Foundation of Canada, The Muttart Foundation, The Royal Canadian Legion, Fujisawa Canada, Novartis Pharmaceuticals Canada, and Roche Canada. T.N.S. is supported by a Studentship from the Alberta Heritage Foundation for Medical Research. P.F.H. is the Muttart Chair in Clinical Immunology.
References [1] Austyn J.M., New insights into the mobilization and phagocytic activity of dendritic cells, J. Exp. Med. 183 (1996) 1287. [2] Barrett M., Milton A.D., Barrett J. et al Needle biopsy evaluation of class II major histocompatibility complex antigen expression for the differential diagnosis of cyclosporine nephrotoxicity from kidney graft rejection, Transplantation 44 (1987) 223. [3] Belitsky P., Miller S.M., Gupta R., Lee S., Ghose T., Induction of MHC class II expression in recipient tissues caused by allograft rejection, Transplantation 49 (1990) 472. [4] Benoist C., Mathis D., Selection for survival? Science 276 (1997) 2000. [5] Campos L., Naji A., Deli B.C. et al Survival of MHCdeficient mouse heterotopic cardiac allografts, Transplantation 59 (1995) 187. [6] Chang C.H., Flavell R.A., Class II transactivator regulates the expression of multiple genes involved in antigen presentation, J Exp Med. 181 (1995) 765. [7] Chang C.H., Guerder S., Hong S.C., VanEwijk W., Flavell R.A., Mice lacking the MHC class II transactivator (CIITA) show tissue-specific impairment of MHC class II expression, Immunity 4 (1996) 167. [8] Chang C.H., Fontes J.D., Peterlin M., Flavell R.A., Class II transactivator (CIITA) is sufficient for the inducible expression of major histocompatibility complex class II genes, J. Exp. Med. 180 (1994) 1367. [9] Cockfield S.M., Ramassar V., Halloran P.F., Regulation of IFN-γ and tumor necrosis factor-a expression in vivo, J. Immunol. 150 (1993) 342. [10] Cockfield S.M., Urmson J., Pleasants J.R., Halloran P.F., The regulation of expression of MHC products in mice factors determining the level of expression in kidneys of normal mice, J. Immunol. 144 (1990) 2967. 910
Sims and Halloran
[11] Daar A.S., Fuggle S.V., Fabre J.W., Ting A., Morris P.J., The detailed distribution of MHC class II antigens in normal human organs, Transplantation 38 (1984) 293. [12] Denzin L.K., Cresswell P., HLA-DM induces CLIP dissociation from MHC class II alpha/beta dimers and facilitates peptide loading, Cell 82 (1995) 155. [13] Denzin L.K., Hammond C., Cresswell P., HLA-DM interactions with intermediates in HLA-DR maturation and a role for HLA-DM in stabilizing empty HLA-DR molecules, J. Exp. Med. 184 (1996) 2153. [14] Douhan III J., Lieberson R., Knoll J.H.M., Zhou H., Glimcher L.H., An isotype-specific activator of major histocompatibility complex (MHC) class II genes that is independent of class II transactivator, J. Exp. Med. 185 (1997) 1885. [15] Escalante C.R., Yie J., Thanos D., Aggarwal A.K., Structure of IRF-1 with bound DNA reveals determinants of interferon regulation, Nature 391 (1998) 103. [16] Fellstrom B., Klareskog L., Larsson E. et al Tissue distribution of macrophages, class II transplantation antigens, and receptors for platelet-derived growth factor in normal and rejected human kidneys, Transplant. Proc. 19 (1987) 3625. [17] Finn P.W., Kara C.J., Van T.T., Douhan III J., Boothby M.R., Glimcher L.H., The presence of a DNA binding complex correlates with E beta class II MHC gene expression, EMBO J. 9 (1990) 1543. [18] Goes N., Sims T., Urmson J., Vincent D., Ramassar V., Halloran P.F., Disturbed MHC regulation in the interferon-γ knockout mouse, J. Immunol. 155 (1995) 4559. [19] Goes N., Urmson J., Hobart M., Halloran P.F., The unique role of interferon-γ in the regulation of MHC expression on arterial endothelium, Transplantation 62 (1996) 1889. [20] Goes N., Urmson J., Ramassar V., Halloran P.F., Ischemic acute tubular necrosis induces an extensive local cytokine response evidence for induction of interferon-γ, transforming growth factorβ-1, granulocyte-macrophage colonystimulating factor, interleukin-2 and interleukin-10, Transplantation 59 (1995) 565. [21] Goes N., Urmson J., Vincent D., Ramassar V., Halloran P.F., Induction of major histocompatibility complex and inflammatory cytokines after ischemic injury to the kidney Lessons from the interferon-γ gene knockout mice, Transplant. Proc. 27 (1995) 771. [22] Guardiola J., Maffei A., Control of MHC class II gene expression in autoimmune, infectious, and neoplastic diseases, Crit. Rev. Immunol. 13 (1993) 247. [23] Hagerty D.T., Intercellular adhesion molecule-1 is necessary but not sufficient to activate CD4+ T cells. Discovery of a novel costimulator on kidney tubule cells, J. Immunol. 156 (1996) 3652. [24] Halloran P.F., Interferon-γ, prototype of the proinflammatory cytokines - importance in activation, suppression, and maintenance of the immune response, Transplant. Proc. 25 (1993) 10. [25] Halloran P.F., Autenried P., Ramassar V., Urmson J., Cockfield S., Local T cell responses induce widespread MHC expression. Evidence that IFN-γ induces its own expression in remote sites, J. Immunol. 148 (1992) 3837. Microbes and Infection 1999, 903-912
Regulation of MHC class II expression
[26] Halloran P.F., Goes N., Urmson J. et al MHC expression in organ transplants lessons from the knock-out mice, Transplant. Proc. 29 (1997) 1041. [27] Halloran P.F., Jephthah-Ochola J., Urmson J., Farkas S., Systemic immunologic stimuli increase class I and II antigen expression in mouse kidney, J. Immunol. 135 (1985) 1053. [28] Halloran P.F., Urmson J., Ramassar V., Laskin C., Autenried P., Increased class I and class II MHC products and mRNA in kidneys of MRL-lpr/lpr mice during autoimmune nephritis and inhibition by cyclosporine, J. Immunol. 141 (1988) 2303. [29] Halloran P.F., Urmson J., Solez K., IFN-γ-induced MHC expression in epithelium of rejecting kidneys may protect the graft against early graft destruction by alloantibody (Abstract), J. Am. Soc. Nephrol. 9 (1998) 650A. [30] Haverty T.P., Watanabe M., Neilson E.G., Kelly C.J., Protective modulation of class II MHC gene expression in tubular epithelium by target antigen-specific antibodies. Cell-surface directed down-regulation of transcription can influence susceptibility to murine tubulointerstitial nephritis, J. Immunol. 143 (1989) 1133. [31] Hobart M., Ramassar V., Goes N., Urmson J., Halloran P.F., The induction of class I and II major histocompatibility complex by allogeneic stimulation is dependent on the transcription factor interferon regulatory factor 1 (IRF-1). Observations in IRF-1 knockout mice, Transplantation 62 (1996) 1895. [32] Hobart M., Ramassar V., Goes N., Urmson J., Halloran P.F., IFN regulatory factor 1 (IRF-1) plays a central role in the regulation of the expression of class I and II MHC genes in vivo, J. Immunol. 158 (1997) 4260. [33] Inaba K., Witmer-Pack M., Inaba M. et al The tissue distribution of the B7-2 costimulator in mice abundant expression on dendritic cells in situ and during maturation in vitro, J. Exp. Med. 180 (1994) 1849. [34] JanewayJr C.A., Bottomly K., Babich J. et al Quantitative variation in Ia antigen expression plays a central role in immune regulation, Immunol. Today 5 (1984) 99. [35] Jevnikar A.M., Singer G.G., Coffman T., Glimcher L.H., Rubin-Kelley V.E., Transgenictubular cell expression of class IIIs sufficient to initiate immune renal injury,J. Am. Soc. Nephrol. 3 (1993) 1972. [36] Katz J.F., Stebbins C., Appella E., Sant A.J., Invariant chain and DM edit self-peptide presentation by major histocompatibility complex (MHC) class II molecules, J. Exp. Med. 184 (1996) 1747. [37] Kirberg J., Berns A., VonBoehmer H., Peripheral T cell survival requires continual ligation of the T cell receptor to major histocompatibility complex-encoded molecules, J. Exp. Med. 186 (1997) 1269. [38] Konieczny B.T., Dai Z., Elwood E.T. et al IFN-γ is critical for long-term allograft survival induced by blocking the CD28 and CD40 ligand T cell costimulation pathways, J. Immunol. 160 (1998) 2059. [39] Lennon A.M., Ottone C., Rigaud G. et al Isolation of a B-cell-specific promoter for the human class II transactivator, Immunogenetics 45 (1997) 266. Microbes and Infection 1999, 903-912
Forum in Immunology
[40] Letterio J.J., Geiser A.G., Kulkarni A.B. et al Autoimmunity associated with TGF-β 1-deficiency in mice is dependent on MHC class II antigen expression, J. Clin. Invest. 98 (1996) 2109. [41] Mach B., Steimle V., Martinez-Soria E., Reith W., Regulation of MHC class II genes lessons from a disease, Ann. Rev. Immunol. 14 (1996) 301. [42] Milton A.D., Fabre J.W., Massive induction of donor type class I and class II major histocompatibility complex antigens in rejecting cardiac allografts in the rat, J. Exp. Med. 161 (1985) 98. [43] Morris P., Shaman J., Attaya M. et al An essential role for HLA-DM in antigen presentation by class II major histocompatibility molecules, Nature 368 (1994) 551. [44] Muhlethaler-Mottet A., Otten L.A., Steimle V., Mach B., Expression of MHC class II molecules in different cellular and functional compartments is controlled by differential usage of multiple promoters of the transactivator CIITA, EMBO J. 16 (1997) 2851. [45] Natali P.G., Quaranta V., Nicotra M.R., Apollonj C., Pellegrino M.A., Ferrone S., Tissue distribution of Ia-like antigens in different species analysis with monoclonal antibodies, Immunobiology 169 (1985) 228. [46] Neuberger M.S., Antigen receptor signaling gives lymphocytes a long life, Cell 90 (1997) 971. [47] Ramassar V., Goes N., Hobart M., Halloran P.F., Evidence for the in vivo role of class II transactivator in basal and IFN-γ induced class II expression in mouse tissue, Transplantation 62 (1996) 1901. [48] Riley J.L., Westerheide S.D., Price J.A., Brown J.A., Boss J.M., Activation of class II MHC genes requires both the X box region and the class II transactivator (CIITA), Immunity 2 (1995) 533. [49] Saleem S., Konieczny B.T., Lowry R.P., Baddoura F.K., Lakkis F.G., Acute rejection of vascularized heart allografts in the absence of IFN-γ, Transplantation 62 (1996) 1908. [50] Sanderson F., Thomas C., Neefjes J., Trowsdale J., Association between HLA-DM and HLA-DR in vivo. Immunity 4 (1996) 87. [51] Schwartz R.H., Models of T-cell anergy is there a common molecular mechanism? J. Exp. Med. 184 (1996) 1. [52] Shoskes D., Parfrey N.A., Halloran P.F., Increased major histocompatibility complex antigen expression in unilateral ischemic acute tubular necrosis in the mouse, Transplantation 49 (1990) 201. [53] Shoskes D.A., Halloran P.F., Delayed graft function in renal transplantation etiology, management and long-term significance, J. Urol. 155 (1996) 1831. [54] Sims T.N., Elliott J.F., Ramassar V., DenneyJr D.W., Halloran P.F., Mouse CIITA cDNA sequence and amino acid comparison with human CIITA, Immunogenetics 45 (1997) 220. [55] Sims T.N., Goes N.B., Ramassar V., Urmson J., Halloran P.F., In vivo class II transactivator expression in mouse is induced by a non-IFN-γ mechanism in response to local injury, Transplantation 64 (1997) 1657. [56] Sims T.N., Pahl A., Ramassar V., Halloran P.F., Regulation of LMP7, TAP1 and H2-MA RNA levels in response to IFN-γ and injury in the kidney (Abstract), FASEB J. 12 (1998) A590. 911
Forum in Immunology
[57] Steimle V., Otten L.A., Zufferey M., Mach B., Complementation cloning of an MHC class II transactivator mutated in hereditary MHC class II deficiency (or Bare Lymphocyte Syndrome), Cell 75 (1993) 135. [58] Steimle V., Siegrist C.A., Mottet A., Lisowska-Grospierre B., Mach B., Regulation of MHC class II expression by interferon-γ mediated by the transactivator gene CIITA, Science 265 (1994) 106. [59] Strutz F., Neilson E.G., The role of lymphocytes in the progression of interstitial disease, Kidney Int. 45 (S45) (1994) 106. [60] Takei Y., Sims T.N., Halloran P.F., Differential usage of the class II transactivator (CIITA) promoters in vivo (Abstract), Transplantation 65 (1998) S30.
912
Sims and Halloran
[61] Wuthrich R.P., Yui M.A., Mazoujian G., Nabavi N., Glimcher L.H., Kelley V.E., Enhanced MHC class II expression in renal proximal tubules precedes loss of renal function in MRL/lpr mice with lupus nephritis, Am. J. Pathol. 134 (1989) 45. [62] Zhou H., Glimcher L.H., Human MHC class II gene transcription directed by the carboxyl terminus of CIITA, one of the defective genes in type II MHC combined immune deficiency, Immunity2 (1995) 545. [63] Zhou H., Su H.S., Zhang X., DouhanIii J., Glimcher L.H., CIITA-dependent and -independent class II MHC expression revealed by a dominant negative mutant, J. Immunol. 158 (1997) 4741.
Microbes and Infection 1999, 903-912