Address reprint requests to Annette S. H. Gouw, MD, De- partment of Pathology ...... tion: The Cambridge/King's College Hospital Experi- ence. Edited by RY ...
American Journal of Pathology, Vol. 133, No. 1, October 1988 Copyright © American Association of Pathologists
Early Induction ofMHC Antigens in Human Liver Grafts An Immunohistologic Study
ANNETTE S. H. GOUW, MD,* SIPPIE HUITEMA,* JORIS GROND, MD,* MAARTENJ. H. SLOOFF, MD,f IDS J. KLOMPMAKER, MD,* CHRIS H. GIPS, MD,, and SIBRAND POPPEMA, MD*
From the Departments ofPathology,* Surgery, t and Hepatology,t University of Groningen, Groningen, The Netherlands
The present study documents major histocompatibility complex (MHC) Class I and II expression during early acute rejection of human liver grafts. Serial graft biopsies (pretransplant, time zero, and 1 week) were studied. Ten patients received azathioprine (AZA) and prednisone; the other six patients were treated with quadruple therapy (azathioprine, cyclosporine A, prednisone, and cyclophosphamide). To study the specificity of changes in MHC antigen expression, biopsies of six patients with minor or no morphologic abnormalities served as controls. In addition, phenotypes of inflammatory cells present during rejection were analyzed using a panel of monoclonal antibodies. The results show that during acute rejection expression of MHC Class I and II antigens increased significantly in the AZA-treated patients, in a pattern similar to that seen in the patients treated with quadruple therapy, showing enhanced MHC Class I expression on hepato-
cytes, bile duct epithelium, and sinusoidal endothelium, and Class II antigen on Kupffer cells and sinusoidal endothelium. Bile duct epithelium was consistently positive for Class II antigen; no significant difference with the nonrejection group was observed. T cells are the predominant inflammatory cells during rejection with equal quantities of CD4+ and CD8+ cells. A majority ofthe infiltrating T cells show expression of Class II antigen but do not react with anti-interleukin-2 receptor antibody. This may be the result of immunosuppressive therapy or a simple reflection of the temporary expression of interleukin-2 receptors during lymphocyte activation. The authors hypothesize that the induction of MHC antigens on bile duct epithelium leads to rejection whereas the expression on hepatocytes represents an epiphenomenon. (Am J Pathol 1988, 133:82-94)
SEVERAL RECENTLY PUBLISHED studies deal with the altered expression of major histocompatibility complex (MHC) antigens in rejected human liver grafts and primary liver diseases, such as primary biliary cirrhosis and chronic active hepatitis.'-' During acute rejection expression of MHC antigens is enhanced but the pathogenetic role of this increased expression of MHC antigens is unclear. MHC alloantigens appear to be major targets in the rejection reaction and expression and accessibility of MHC antigens are determining factors. " This study documents the alteration of MHC antigens expression in a selected group of liver transplant patients who experienced acute rejection in the first week after transplantation and were treated with two different types of immunosuppressive regimens.
Analysis of MHC expression in pretransplant, time zero, and rejection biopsies in parallel with assessment of phenotypes and activity states of the T cells in the graft enabled the question of whether increased MHC expression is an epiphenomenon or a "prelude" to acute rejection to be addressed.
Materials and Methods Patients As of October 1987, 77 orthotopic liver transplantations (OLT) were performed in the authors' center, Accepted for publication May 19, 1988. Address reprint requests to Annette S. H. Gouw, MD, Department of Pathology, University of Groningen, Oostersingel 63, 9713 EZ-Groningen, The Netherlands.
82
Vol. 133
*
MHC ANTIGENS IN LIVER GRAFTS
No. I
83
Table 1 -Data of the Patients with Rejection (AR group)
Patient OLT 3 OLT4 OLT 7 OLT 13 OLT 15 OLT 17 OLT 26 OLT 27 OLT 41
OLT43 OLT53 OLT 57 OLT 57R OLT 59 OLT 66 OLT69
Original disease HCC CAH CAH BuCh CC Al-CAH PBC PBC PSC CAH BDA PBC PBC PBC BuCh PSC
Age/sex at transplant (years)
Graft survival
42/F 55/F 44/M 27/M 34/M 17/F 46/F 48/F
19/M 40/F 7/F
46/F 46/F 51 /F 28/F 37/M
6y 1 mo* 6y 4y 4y 4y 4 mo* 3y 2 mo* 1y 1y 6 mo* 2 wk*
lOmo 6 wk* 1 mo
Therapy AZA AZA AZA AZA AZA AZA AZA AZA AZA AZA Q Q
Q Q Q
Day when HB 3 performed 8 14 8 8
11 7 7 7 7 7 8 4 7
6 7 7
Grade of AR 1 0 11 Il 11 11 11 11
11
Il 1 1 11
HCC, hepatocellular carcinoma; CAH, chronic active hepatitis; BuCh, Budd-Chiari disease; CC, cryptogenic cirrhosis; Al-CAH, auto immune chronic active hepatitis; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; BDA, bile duct atresia; R, retransplantation; AZA, azathioprine; 0, quadruple therapy (see text); 0, consistent with rejection; I, mild rejection; II, moderate rejection; III, severe rejection. *Died.
including 69 primary OLTs and eight re-OLTs. Graft biopsies were taken according to a protocol including a first biopsy before transplantation (HB 1), a second biopsy at the end of the surgical procedure (HB 2), a third one at the end of the first week (HB 3) provided that there were no clinical contraindications, eg, clotting disturbances, a fourth biopsy at discharge, and subsequent follow-up biopsies at yearly intervals. In addition, other biopsies were taken at clinical signs of graft dysfunction. This study includes the first three regular biopsies (HB 1, 2, 3). The patients include a group with a histologic diagnosis of acute rejection in the HB 3 (AR group, N = 16) and a group without rejection (non-AR group, N = 6). In the AR group HB 3 was taken at a mean of 7.7 ± 2.2 days after transplantation and in the non-AR group at 12.3 ± 4.8 days. The AR group was selected according to strict criteria that reasonably excluded causal factors of graft dysfunction other than acute rejection, such as viral infection, and biliary or vascular complications. The immunosuppressive regimen included azathioprine (AZA) and prednisone for ten patients, whereas the other six patients were treated with quadruple therapy consisting of AZA, prednisone, cyclophosphamide (100 mg/day for 1 week) and cyclosporin A (CSA) starting at day 2 after transplantation. In 12 of the 16 patients HB 3 was taken before anti-rejection treatment: OLT 4, 7, 17 and 26 received steroids before HB 3. Data of the AR group are shown in Tables 1 and 2. The results of a comparative morphologic study of the AZA-treated patients of this AR group and a similar group of patients from the University of
Minnesota maintained on CSA and low dose steroids (with or without AZA) are presented elsewhere. 12 The non-AR group consisted of six patients who were free of clinical and morphologic signs of graft dysfunction at the end of the first week after transplantation. These biopsies served as a control group for the study of MHC antigen expression. Subset phenotyping was not performed because these specimens were virtually free of inflammatory cells. The patients of this non-AR group were all treated with quadruple therapy and their data are shown in Tables 3 and 4. Morphologically normal HB 3s were not observed in patients treated with AZA and prednisone only. Histology and Immunocytochemistry The graft biopsy was performed with a 1.6 mm Menghini needle. One part was frozen in Freon (-90 C) and stored at -80 C for immunohistochemical stainings. Another part was fixed by immersion in 4% formaldehyde with 6% HgCl2, 5% glacial acetic acid'3 or in B5 and embedded in paraffin for routine histologic examination and for immunohistologic studies using monoclonal antibodies applicable on paraffin sections. Table 5 shows the antibodies used, their specificities, and sources. A two-step immunoperoxidase technique was applied for the immunohistochemical stainings on cryostat sections as described in detail previously.'0 A three-step immunoperoxidase technique was applied on the paraffin sections: 2-4 , sections were deparaffinized, rehydrated, desublimated, and pretreated for 30 minutes with 0.3% meth-
84
GOUW ET AL
AJP o October 1988
Table 2-Liver Tests in Patients with Rejection at the Time of HB 3 Bilirubin Patient
SGOT (30 U/L)
SGPT (30 U/L)
OLT 3 OLT 4 OLT 7 OLT 13 OLT 15 OLT17 OLT 26 OLT 27 OLT 41 OLT 43 OLT 53 OLT 57 OLT 57R OLT 59 OLT 66 OLT 69
57 189 246 83 171 299 44 168 42 81 72 95 1780 212 70 189
275 331 635 417 393 1245 152 525 192 266 201 158 1140 424 120 665
GGPT (65 U/L)
AP
(100 U/L) 490 93 369 209 88 153 136 377 235 305 148 95 146
Direct Total (6 to 17 ,umol/L)
305 54 790 250 61 53 226 155 195 334 146 NA 64 75 157
114
85 150
133 380 166 345 182 605 109 355 540 345 35 125 300 345 290 255
23 305 150 215 125 549 76 315 525 305 19 82 280 295 235 205
253
Maximal normal values appear in parenthesis. R, retransplantation; SGOT, serum glutamic-oxaloacetic transaminase; SGPT, serum glutamic-pyruvic transaminase; AP, alkaline phosphatase; GGPT, gamma glutamyl transpeptidase; NA, not available.
anol H202. In the first step the sections were incubated with undiluted LN2 and MB2 for 30 minutes. In the second step the sections were incubated with biotinylated rabbit anti-mouse Ig antibodies, diluted 1:400 (Dakopatts, Copenhagen, Denmark) for 15 minutes. This was followed by a third step using avidin-biotin complex for 30 minutes. The sections were stained with 3.3'-diaminobenzidine tetrahydrochloride (Sigma, St. Louis, MO) and H202 for 10 minutes, counterstained with Mayer's hemalum, dehydrated, and mounted in xylene soluble mounting medium. Between the steps the sections were rinsed thoroughly with PBS. Assessment ofthe staining results was done in a semiquantitative way in which - equals no positive cells, ± equaled minimal, + equaled mild, ++ equaled moderate quantities of positive cells, and +++ equaled abundant positive cells.
Statistical Evaluation The analysis of variance was applied for the assessment of MHC antigen expression and the Mann-
Whitney U test for the liver tests' values with P < 0.05 the level of significance.
as
Results The pretransplant and time zero biopsies of both groups showed no morphologic abnormalities except for one case. The portal tracts in HB 2 of OLT 3 contained some inflammatory cells and minimal cholestasis in the centrizonal areas. Acute rejection was characterized by portal inflammation, venous endothelialitis of portal and terminal hepatic venules, bile duct damage, and variable degrees of parenchymal hepatocellular damage with acidophilic degeneration of hepatocytes in periportal and centrilobular areas and centrilobular ballooning and atrophy of liver cells. Table 1 lists the severity of the rejection episode ofeach patient, graded according to the criteria proposed by Snover.'4 The HB 3 of the non-AR group showed no histologic abnormalities except for the mild to moderate
Table 3-Data of the Patients Without Rejection (non-AR Group)
Patient
disease
Age/sex at transplant (years)
OLT 45R
BC PBC BuCh PBC CC PSC
18/M 48/F 52/F 54/F 27/F 19/F
Original OLT52 OLT56 OLT 58 OLT 62 OLT 65
Graft survival 1, 5 y 1y 1y
lOmo 7 mo 5 mo
Therapy
Day when HB 3 performed
Q Q Q Q Q Q
21 14 7 12 10 10
Cholestasis grade + ++
+ + ++
BC, secondary biliary cirrhosis; PBC, primary biliary cirrhosis; BuCh, Budd-Chiari disease; CC, cryptogenic cirrhosis; PSC, primary sclerosing cholangitis; AZA, azathioprine; 0, quadruple therapy (see text); -, no cholestasis; ±, minimal cholestasis; +, mild cholestasis; ++, moderate cholestasis; +++, severe cholestasis.
MHC ANTIGENS IN LIVER GRAFTS
Vol. 133 * No. I
85
Table 4-Liver Tests in Patients Without Rejection at the Time of HB 3
Bilirubin
SGOT
SGPT
Patient
(30 U/L)
(30 U/L)
OLT 45R OLT 52 OLT 56 OLT 58 OLT 62 OLT 65
21 46
30 105 215 295 168 185
26 73 66 37
AP (100 U/L) 526 114
220 214 144 119
GGPT
Direct Total (6 to 17 Amol/L)
(65 U/L) 118 107 NA 162 181 143
52 163 26 34 72 123
74 239 41 57 110 166
Maximal normal values appear in parenthesis. R, retransplantation; SGOT, serum glutamic oxaloacetic transaminase; SGPT, serum glutamic pyruvic transaminase; AP, alkaline phosphatase; GGPT, gamma glutamyl transpeptidase; NA, not available.
cholestasis that was present in all but one biopsy without signs of bile obstruction or cholangitis, eg, portal edema, marginal bile duct proliferation, and infiltration by neutrophilic granulocytes (Table 3). The liver tests' values of the same day on which the HB 3 was taken showed a significant difference in the elevation of serum glutamic-oxaloacetic transaminase (SGOT) (P = 0.005), serum glutamic-pyruvic transaminase (SGPT) (P = 0.03), direct bilirubin (P = 0.03), and total bilirubin (P = 0.02) ofthe AR group compared with the non-AR group. Alkaline phosphatase (AP) and gamma glutamyl transpeptidase (GGTP), although elevated, did not differ significantly from the non-AR group (Tables 2 and 4). The results of subset phenotyping are presented in Table 6. T cells were the predominant mononuclear cells in acute rejection. B cells were scant. Among the T cells, CD8+ (OKT8+) cells were predominantly present in moderate (++) quantities (8 of 13). CD4+ cells (Leu 3+) were present in moderate quantities (++) in 7 of 13 biopsies and in abundant quantities (+++) in 2 patients. The CD4+ cells were located diffusely and in the central parts of the portal tracts. CD8+ cells were located in the portal periphery. A minority of the CD4+ cells represented a group of Kupffer cells or histiocytes that are known to be CD4+.'5 Bile duct epithelium was infiltrated by both CD4+ and CD8+ cells (Figure lA,B). Both cell types were found in the sinusoids also but in much lesser quantities than in the portal tracts (0-1+). Kupffer cells were OKM 1-. In 8 out of 12 cases portal tracts regularly contained Leu 3a+ and OKM + cells other than granulocytes in mild (1 +) quantities. Five monoclonal antibodies recognizing different T cell activation antigens, were used on the biopsies showing acute rejection. The results are shown in Table 7. In 13 of 14 biopsies, Class II positive (LN2+) mononuclear cells were present in moderate (++) to abundant (+++) quantities without showing discrete locations in the portal tracts either central or periph-
eral (Figure 2). OKT 10 positive cells were less conspicuous. Interleukin-2 receptor (Tii-69), Ki-67, and OKT9 positive cells were scarce. The distribution of MHC antigens in the various liver cell populations is shown in Tables 8 and 9. A comparable distribution of Class I and II antigen expression was observed in the first and second biopsy. Hepatocytes failed to express both classes of antigens apart from a mild expression in one HB 2. Bile duct epithelium and Kupffer cells were Class I positive. Expression of Class II antigens on bile duct epithelium was absent except for one biopsy taken before transplantation without histologic abnormalities. Kupffer cells were Class II positive in all cases. Arterial, ve-
Table 5-Monoclonal Antibodies, Their Specificities and Sources
Antibody Leu 1 Leu 3
OKT 8 OKM 1 1 B9
B9-12-1 B8-12-2 OKT 10
Tu-69 Ki-67 LN 2 OKT 9 MB2
Specificities
Source
CD 5. Mature peripheral T cells. CD 4. T helper/inducer cells, histiocytes. CD 8. T cytotoxic/suppressor cells. CD 11 b. Monocytes, granulocytes, NK cells CD 20. B cells. MHC-class I antigens. MHC-class II antigens. CD 38. Activated T and B cells; plasma cells, histiocytes. CD 25. lnterleukin-2 receptor. Proliferating (non GO) cells. MHC-class II antigens. Transferrin receptor. B cells, endothelial- and some epithelial cells.
a a
b b c
d d b e
f e
b
g
a, Beckton and Dickinson, Mountain View, CA. b, Ortho Pharnaceutical Corp., Raritan, NJ. c, Prof. Dr. S. Poppema, Cross Cancer Inst. Edmonton, Canada. d, Drs. M. Giphart and F. Koning, Dept. of Immunohernatology, Univ. of Leiden, The Netherlands. e, Biotest, Dreieich, West Germany.
f, Dakopatts, Copenhagen, Denmark. g, Poppema et al.29
86
GOUW ET AL
AJP o October 1988
Table 6-Phenotypes of Mononuclear Cells in Acute Rejection Patient
CD5
OLT3 OLT4 OLT7 OLT 13 OLT15 OLT 17 OLT 26 OLT 27 OLT 41 OLT 43 OLT 53 OLT 57 OLT 57R OLT 59 OLT 66 OLT 69
N N + +
N +++ ++ + ++ +++ ++ +++ +++ + ++ ++
CD4
CD8
CD11 b
CD20
MB2
N N N + N +
N N + + N +
N
N N
++ ++ ++ +++ + +++ ++
N N + ++ N + ++ ++ + ++ + ++ ++
_ -
+ N
+
+
+ ++
++ ++
N N ++ + N ++
+ ++
+ ++
+ + + + ++ +
+ N N N
N N N
N N
+ + ++
+ N + +
-, no positive cells; ±, minimal; +, mild; ++, moderate; +++, abundant; N, no (frozen) sections available.
nous, and sinusoidal endothelium showed mild to moderate expression of Class I and low expression of Class II antigens. Class I and II antigen expression did not differ significantly between the pretransplant biopsies (HB 1) of the AR and non-AR group. In both groups expression of Class I and II antigens increased significantly after 1 week (HB 3) (Figures 3 and 4). In the AR group, this increase concerned all cell types for Class I antigen but it was limited to bile duct epithelium, sinusoidal, and vascular endothelium for Class II antigen. In the non-AR group, a significant increase of Class I antigen was present on bile duct epithelium and vascular endothelium, whereas enhancement of Class II antigen was observed on sinusoidal and vascular endothelium. The 1-week biopsies (HB 3) with acute rejection showed a striking change of MHC antigens expression. In these specimens hepatocytes were almost diffusely positive for Class I antigens (Figure 5A-C).
No hepatocytic Class II antigen expression was observed. HLA Class II antigens did appear on bile duct epithelium in addition to the Class I antigens already present before the rejection episode (Figure 6A,B). Kupffer cells were positive again for both groups of antigens. Sinusoidal, arterial, and venous endothelium showed increased Class I and II antigen expression. Mean values and standard deviations of Class I and II expression in HB 1, 2, and 3 in the AR and the nonAR groups are presented in Figures 3 and 4, respectively. Figure 7 presents these data for HB 3 in the AR and non-AR groups. The results of statistical analysis ofMHC expression data are summarized in Table 10.
Discussion Sequential liver biopsies from the graft (before, after, and 1 week following OLT) of all patients transplanted in the authors' center, taken independently of
Figure IA-CD4+ cells in a portal tract during acute rejection infiltrating bile duct epithelium (arrow). (x432) A infiltrating the same bile duct (arrow). (X432)
B-CD8+ cells in the same portal tract as in
MHC ANTIGENS IN LIVER GRAVTS
Vol. 133 * No.lI
87
Table 7-T Cell Activation Dunng Acute Rejection
Patient
CD38
Tu-69
Ki-67
LN2
OKT9
OLT3 OLT4 OLT 7 OLT 13 OLT 15 OLT17 OLT 26 OLT 27 OLT41 OLT 43 OLT 53 OLT 57 OLT 57R OLT 59 OLT 66 OLT 69
N N N ++ N + + + + ++ + + + + + +
N N N N
N N N N
-
-
-
-
-
-
++ ++ +++ +++ N ++ +++ +++ + ++ ++ ++ ++ N
N N N N N N N N N N _
-
-
-
-
+
-
+ -
+ -
-
+_ -
-, no positive cells; +, minimal; +, mild; ++, moderate; +++, abundant; N, no (frozen) sections available.
their clinical course, offered the opportunity to study the alteration of MHC expression during acute rejection separately from that due to other factors, such as liver preservation or operation procedures. The observations that MHC antigen expression increased during acute rejection confirm data from other centers.'" The specific stimulus for an altered MHC expression during hepatic graft rejection is unknown. The preservation and transplantation procedure per se, with a subsequent inflammatory response, may provoke release of lymphokines leading to induction of MHC antigens expression. Interferons, in particular gamma-interferon, have been reported to induce MHC product expression both in vitro and in vivo. 16-18 The posttransplant increase of MHC antigen expression as observed in the non-AR group may be caused by such unspecific stimuli. Conceivably, the
Figure 2-Class II positive cells in a portal tract
during acute rejection. (LN2, x1330)
rejection-related increase of MHC antigens is then limited only to hepatocytes, Kupffer cells, and sinusoidal endothelium for Class I and to Kupffer cells and bile duct epithelium for Class II, because these were not increased in the non-AR group. There was still a significant difference between the AR and non-AR group 1 week after transplant concerning Class I expression on bile duct epithelium, however, indicating that only part of the increase on this cell type is due to unspecific stimuli. The increased expression of Class II antigen on bile duct epithelium during AR (HB 3AR) did not differ significantly from the 1 week biopsies of the non-AR group (P = 0.07). A previous study hypothesized on the possible role of donor-specific, Class II-rich Kupffer cells in the induction ofearly alloresponse by acting as antigen-presenting cells (APC) parallel to the role of passenger
88
GOUW ET AL
AJP . October 1988
Table 8-Expression of MHC Antigens in HB 1, 2, and 3 in the AR Group
MHC Class I Patient OLT 3
Cell type
HB 1
he
N
HB 2
MHC Class II HB 3
HB 1
N
N
HB 2
HB 3 N
bd Ku Si
av
OLT 4
he bd Ku Si
av
OLT 7
he bd Ku
N
N
N ++
+ + +
N
N +++ +++ ++
Si
av
OLT 13
he bd Ku
++ ++ ++
N
he
N
av
he bd Ku
N
++
av
+++ ++ +
he bd Ku Si
N +
N ++
N
++ ++
N
++ ++
+
he bd Ku
+++
av
OLT 57
he bd Ku Si av
N
N
+
++ +++ ++
++
++
++ +
N +
++
++ ++
N
++ ++ ++ +
Si
Si av
++
++ +
he bd Ku
++ ++ ++
++
av
he bd Ku
N
++ ++ +++
Si
OLT 53
N
++
+ N
av
OLT 43
++ +
av
OLT 41
N
Si
he bd Ku
N
+
+++
Si
OLT 27
N ++ +
Si
OLT 26
+ +
N+
bd Ku
OLT 17
N
++
Si
av
OLT 15
N
+
+++ ++
++ ++ ++ +
++
N ++ ++ + +
Vol.
133 * No.
89
MHC ANTIGENS IN LIVER GRAFTS
I
Table 8-Continued MHC Class I Patient
Cell type
HB 1
HB2
OLT 57R
he bd Ku
-
N
Si
+
MHC Class 11 HB 3
+ ++
HB 1
HB 2
+
-
N
++ ++ ++
++ +
av
OLT 59
+ ++
N
Si
+
av
OLT 66
OLT 69
he bd Ku Si av
he bd Ku
++ ++ ++
+
+
he bd Ku
+
-
++ ++ ++
+ ++ +
N ++ ++ ++
+
-
N
+ ++
+
++
+
++ ++
++
N + ++ ++
+
+
-
N
++
+
Si
N +-++
+ ++ ++
+ +
+ -
++ +
+
+
av
HB 3
No portal tract assessable. he, hepatocytes; bd, bile duct epithelium; Ku, Kupffer cells; si, sinusoidal endothelium; av, arterial/venous endothelium; R, retransplantation; N, no frozen section available; -, no positive cells; ±, minimal; +, mild; ++, moderate; +++, abundant.
Table 9-Expression of MHC Antigens in the non-AR Group
MHC Class I Patient
Cell types
OLT 45R
he bd Ku Si av he
OLT 52
bd Ku Si
HB 1
HB 2
HB 3
HB 1
HB 2
-
N
+ +++
-
N
+
++ +
++ +
-
+ +
OLT 58
OLT 62
OLT 65
HB 3 + +++ +
-
+ +
+
++
++
++
++
++
++
+++
+++
+
+
+
_
_
-
-
+ ++ -
++ -
++ + +
av
OLT 56
MHC Class 11
+
he bd Ku Si av he bd Ku Si av he bd
+
Ku
++
Si av he bd Ku Si av
+
++ +
N
*
+
++
++ + +
++ + +
N
-
N
N
++ ++ +
++ ++
+
+
-
N
+ ++ + +
+
+ ++ + ++
-
++ +
++ ++ + +
-
N ++ ++
++
+
++ _
++ + +
++ + +
No portal tract assessable. he, hepatocytes; bd, bile duct epithelium, Ku, Kupffer cells; si, sinusoidal endothelium; av, arterial/venous endothelium; R, retransplantation; N, no frozen section available; -, no positive cells; ±, minimal; +, mild; ++, moderate; +++, abundant.
90
AJP o October 1988
GOUW ET AL
4
4
1
T
3-
3
T
2
T J
I
0
HE
go:
HSI
KU
BD
so:
He2
Si
HE
AV
BD
KU
Si
AV
M:HN3
Figure 3-Mean values and SD of Class (left) and Class II (right) expression in the AR group in HB 1 (pretransplant), HB 2 (time zero), and HB 3 (1 week). Xaxis: semiquantitative grading of MHC expression 0-3+. HE, hepatocytes; BD, bile duct epithelium; KU, Kupffer cells; Si, sinusoidal endothelium; AV, arterial/ venous endothelium.
leukocytes, as suggested by Lechler and Batchelor.)0" 9 It might be expected that this induction mechanism takes place in each graft-host interaction; it has even been described in tolerant rat-strain combinations.20'2' The question arises as to why this early graft-host interaction does not provoke an inflammatory response in all individuals, eg, those of the non-AR group. There is a higher response in the AZAtreated patients than in those treated with quadruple therapy: the non-AR group was represented by patients treated with quadruple therapy only. No morphologically normal HB 3 was present in the AZAtreated patients. This difference between AZA-treated patients and those receiving quadruple therapy that included CSA has been reflected in the lower incidence of rejection in CSA-treated patients in the first
weeks after transplantation.22 CSA has been found to be effective in inhibiting human monocyte antigen presentation, leading to inhibition of antigen-dependent T cell proliferation without affecting Class II expression.23 CSA also inhibits production of lymphokines by blocking the antigen-mediated signal to the nucleus of the T cell.24 Both of these mechanisms can be responsible for a less successful induction ofalloresponse by APC in the host's immune system of the non-AR group compared with the AR-group. According to Hall, the migration of activated effector cells into tissues is a random process. Vascular endothelium detains relevant circulating activated effector cells and accounts for the homing of effector cells into grafts. Local proliferation follows contact with relevant antigens in the graft.25 Cells with a high
4.-
4
3
3
TT 2
T1
-
T
2 T
0
0.-
HE Eza: HS1I
KU
BO S
:
H2
_
Si
AV
HE
BD
KU
Si
AV
HD 3
Figure 4-Mean values and SD of Class I (left) and Class II (right) expression in the non-AR group in HB 1 (pretransplant), HB 2 (time zero), and HB 3 (1 week). X-axis: semiquantitative grading of MHC expression 0-3+. HE, hepatocytes; BD, bile duct epithelium; KU, Kupffer cells; Si, sinusoidal endothelium; AV, arterial/venous endothelium.
Vol. 133
*
91
MHC ANTIGENS IN LIVER GRAFTS
No.lI
..-b
.. .'
h
jjo
.
_
~...
I;
w
.. :
?,
IT.Io.
* .
r
k. ..:.
+
iL,,z,
:I, I
...,
Figure 5A-Diffuse expression of Class antigen on hepatocytes during acute rejection. Two portal tracts are shown in the left lower corner and right upper corner. (B9-12-1, x532) B-Focal expression of Class antigen on hepatocytes around a terminal hepatic venule (left). A portal tract is visible with class I positive bile duct epithelium (right). (B9-12-1, x532) C-Absence of Class antigen expression on hepatocytes in a biopsy of the nonrejection group while bile duct epithelium shows expression of Class I antigen (arrow). (B9-12-1,
I,
/:
X
.
i1-,u
-s
i
-w;
x532)
1~~~~~~~~~
-
A.
if
.4 4i
.;
If " Is.
i
~~~~~4%"
....
.
s
,,ti< ,-
,6
-.:
I. `.".....
44. ..
.. .....
D
...
* '..
N.
Y. : .:-.
~~~~~~~~~~~~~7%
Ct
92
GOUW ET AL
AJP o October 1988 .
.0
I
JA~ ~
~
~
~
~
~~
A
Figure 6A-Class II expression on bile duct epithelium (arrow), Kupffer cells and mild expression on sinusoidal endothelium (left upper corner) during acute B-Class II expression on bile duct epithelium during acute rejection. (18-12-2, X432) rejection. (B8-12-2, x532)
for the induction of Class II enhancement on bile duct epithelium, maintaining the local activation of T cells and the rejection process. Class II antigen is reported to be the primary stimulating determinant for helper T cells that subsequently produce lymphokines necessary for cytotoxic T cells to mature and proliferate.27 The absence of the T lymphocyte activation marker IL2R on the AR biopsies may be related to the early appearance and temporary character ofthis receptor,28 especially because one ofthe late activation antigens, Class II (LN2), clearly was present.
constitutive expression of MHC will be the first target of effector T cells. 16'25 Following this interaction lymphokines are released to induce MHC expression on the surrounding cells to make them vulnerable to lysis.25 In the liver graft, T cells entering portal tracts are confronted with bile duct epithelium, which, in contrast with hepatocytes, is a naturally Class I-rich structure. Bile duct epithelium is therefore a suitable target for cytolysis, illustrated by the consequent finding of cellular infiltration of bile duct epithelium during rejection.'2"'4 26 The subsequent release of lymphokines might be responsible 4
3
I
H
2
1
I
Figure 7-Mean values and SD of Class and 11 expression on the 1 week biopsies (HB 3) of the AR (acute rejection) and the non-AR groups. X-axis: semiquantitative grading of MHC expression 0-3+. HE, hepatocytes; BD, bile duct epithelium; KU, Kupffer cells; Si, sinusoidal endothelium; AV, arterial/venous endothelium.
0 HE
BD
class I D: AR class if: E :AR
KU
U : non-AR E : non- AR
Si
AV
MHC ANTIGENS IN LIVER GRAVF1S
Vol. 133 * No. I
Table 1 0-Summary of the Statistical Analysis of MHC Expression HB3 HB 1-HB 2 HB 2-HB 3 HB 1-HB 3 AR vs. non-AR Class I AR Non-AR
Class II AR
NS*
P < 0.001
P < 0.001*
AV P = 0.02
NS*
BD P < 0.001 AV P= 0.003
BD/KU/SI
BD/SI/AV P < 0.01
BD P = 0.03 AV P < 0.001
P < 0.001
HE: P < 0.001 BD: P = 0.02 SI: P < 0.005
KU: P = 0.03 SI: P = 0.003
SI/AV Non-AR
NS*
NS*
P < 0.02
* All structures. NS, not significant; HE, hepatocytes; BD, bile duct epithelium; KU, Kupffer cells; SI, sinusoidal endothelium; AV, arterial/venous endothelium.
The finding of equal quantities of CD4+ and CD8+ cells during AR is compatible with the thesis that both subsets are necessary to induce rejection.25 The lymphokines released by the interaction of inflammatory cells with bile duct epithelium may induce Class I expression on hepatocytes but this is apparently not sufficient for induction of cytolysis of the parenchymal cells. This is in concordance with the rare occurrence of lobular infiltration or piecemeal necrosis during AR.12"'426 The relation between enhanced class I expression on hepatocytes and the significant increase of the transaminases in the AR group that indicates hepatocellular damage is not readily apparent. This is also true for the relation between enhanced expression of MHC antigens on bile duct epithelium and elevated bilirubin levels. In conclusion, these data demonstrate that MHC induction on liver graft cell populations such as bile duct epithelium accelerates an alloresponse. Because enhanced MHC expression on hepatocytes does not induce inflammation and cytolysis, this may be regarded as an epiphenomenon. Further studies are necessary to establish the role of bile duct epithelium in maintaining the alloresponse in the portal tracts and to clarify the relation between enhanced MHC antigens expression and functional deterioration of liver cell constituents.
References 1. Nagafuchi Y, Thomas HC, Hobbs KEF, Scheuer PJ: Expression of beta-2-microglobulin on hepatocytes after liver transplantation. Lancet 1985, i:551-554 2. Takacs L, Szende B, Monostori E, Rot A, Lapis K, Szecseny A, Ando I: Expression of HLA-DR antigens on bile ducts of rejected liver transplant. Lancet 1983, i: 1500
93
3. Demetris JA, Lasky S, VanThiel DH, Starzl TE, Whiteside T: Induction of DR/Ia antigens in human liver allografts: An immunocytochemical and clinicopathologic analysis of twenty failed grafts. Transplantation 1985, 40:504-509 4. SteinhoffG, Wonigeit K, Ringe B, Pichlmayr R: Induction of MHC antigens in human liver after transplantation and in liver disease. J Hepatol 1986, 3(suppl 1):S95 5. Ballardini G, Bianchi FB, Doniachi D, Mirakian R, Pisi E, Botazzo GF: Aberrant expression of HLA-DR antigens on bile duct epithelium in primary biliary cirrosis: Relevance to pathogenesis. Lancet 1984, i: 1009-1113 6. Fukusato T, Gerber MA, Thung SN, Ferrone S, Schaffner F: Expression of HLA class I antigens on hepatocytes in liver disease. Am J Pathol 1986, 123:264270 7. Montano L, Miescher GC, Goodal AH, Janossy G, Thomas HC: Hepatitis B virus and HLA antigen display in chronic hepatitis B virus infection. Hepatology 1982, 2:557-561 8. Barbatis C, Woods J, Morton JA, Fleming KA, McMichael A, McGee JOD: Immunocytochemical analysis of HLA (A, B, C) antigens in liver disease using a monoclonal antibody. Gut 1981, 22:985-991 9. So SKS, Platt JL, Ascher NL, Snover DC: Increased expression of class I MHC antigens on hepatocytes in rejecting human liver allografts. Transplantation 1987, 43:79-85 10. Gouw ASH, HouthoffHJ, Huitema S, Beelen JM, Gips CH, Poppema S: Expression of major histocompatibility antigens and replacement of donor cells by recipient ones in human liver grafts. Transplantation 1987, 43: 291-296 11. Milton AD, Fabre JW: Massive induction of donor type class I and class II major histocompatibility complex antigens in rejecting cardiac allografts in the rat. J Exp Med 1985, 161:98-112 12. Gouw ASH, Snover DC, Grond J, Huitema S, Gips CH, SlooffMJH, Poppema S: Acute rejection in human liver grafts: A comparative histologic study of cases maintained on azathioprine and prednisone versus cyclosporine A and low dose steroids. Human Pathol 1988, in press 13. Bosman FT, Lindeman J, Kuiper G, van der Wal A, Kreunig J: The influence offixation on immunoperoxidase staining of plasma cells in paraffin sections ofintestinal biopsy specimens. Histochemistry 1977, 53:57-62 14. Snover DC: The pathology of acute rejection. Transplant Proc 1986, 19:50-56 15. Poppema S, de Leij L: Unexpected specificities of monoclonal antibodies, Protides of the Biological Fluids: Proc 30th Colloquium. Edited by H Peeters. Oxford, Pergamon Press, 1982, pp 437-440 16. Halloran PF, Wadgymar A, Autenried P: The regulation of expression of major histocompatibility complex products. Transplantation 1986,41:413-420 17. Koene RAP, de Waal RMW, Bogman MJT: Variable expression of major histocompatibility antigens: Role in transplantation immunology. Kidney Int 1986, 30: 1-8 18. Unanue ER, Allen PM: Comment on the finding of Ia expression in nonlymphoid cells. Lab Invest 1986, 55, 2: 123-125
94
GOUW ET AL
19. Lechler RI, Batchelor JR: Restoration of immunogenicity to passenger cell depleted kidney allografts by the addition of donor strain dendritic cells. J Exp Med 1982, 155:31-41 20. Baudot P, Capron-Laudereau M, Emond J, Gane P, Rouger P, Houssin D: Early disappearance of donorspecific immunofluorescence in spontaneously tolerated liver allografts in inbred rats. Transplant Proc 1987, 19, 1:200-204 21. Gassel HJ, Engemann R, Thiede A, Hamelmann H: Replacement of donor Kupffer cells by recipient cells after orthotopic rat liver transplantation. Transplant Proc 1987, 19:351-353 22. Krom RAF, Wiesner RH, Haagsma EB, Krom RAF, Wiesner RH, Haagsma EB, Ludwig J, Gips CH, Grond AJK, Houthoff HJ: A comparison of azathioprine and cyclosporine in liver transplantation: A study of two personal series. Transplant Proc 1987, 19:2440-2442 23. Snyder D, Wright CL, Ting C: Inhibition of human monocyte antigen presentation, but not HLA-DR expression of cyclosporine. Transplantation 1987, 44: 407-411 24. Hodgkin PD, Hapel AJ, Johnson RM, Young IG, Lafferty KJ: Blocking of delivery of the antigen medi-
AJP * October 1988
25. 26.
27.
28.
29.
ated signal to the nucleus of T cells by cyclosporine. Transplantation 1987, 43:685-692 Hall BM: Cellular infiltrates in allografts. Transplant Proc 1987, 19:50-56 Wight DGD: Pathology ofrejection, Liver Transplantation: The Cambridge/King's College Hospital Experience. Edited by RY Calne, New York, Grune and Stratton, 1987, pp 385-435 Bach FH, Sachs DH: Transplantation immunology. N Engl J Med 1987, 317:489-492 Waldmann TA. The structure, function and expression of interleukin-2 receptors on normal and malignant lymphocytes. Science 1986,232:727-732 Poppema S, Hollema H, Visser L, Vos H: Monoclonal antibodies (MT 1, MT2, MB 1, MB2, MB3) reactive with leucocyte subsets in paraffin embedded tissue sections. Am J Pathol 1987, 127:418-429
Acknowledgment The authors thank Hilbrand Wierenga for skillful photographic help and Jose Kop for typing the manuscript.
