Portal de Periódicos da CAPES

Enrichment of Regulatory T Cells in Acutely Rejected Human Liver Allografts

2012; Elsevier BV; Volume: 12; Issue: 12 Linguagem: Inglês

10.1111/j.1600-6143.2012.04264.x

1600-6143

Richard Taubert, Sven Pischke, Jérôme Schlué, Heiner Wedemeyer, Fatih Noyan, Albert Heim, Frank Lehner, Hannelore Barg‐Hock, Jürgen Klempnauer, Sven Olek, Michael P. Manns, Matthias Hardtke‐Wolenski, Elmar Jaeckel,

Immune Cell Function and Interaction

American Journal of TransplantationVolume 12, Issue 12 p. 3425-3436 Free Access Enrichment of Regulatory T Cells in Acutely Rejected Human Liver Allografts R. Taubert, R. Taubert Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorS. Pischke, S. Pischke Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorJ. Schlue, J. Schlue Department of Pathology, Hannover Medical School, GermanySearch for more papers by this authorH. Wedemeyer, H. Wedemeyer Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorF. Noyan, F. Noyan Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorA. Heim, A. Heim Department of Virology, Hannover Medical School, GermanySearch for more papers by this authorF. Lehner, F. Lehner Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorH. Barg-Hock, H. Barg-Hock Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorJ. Klempnauer, J. Klempnauer Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorS. Olek, S. Olek Ivana Türbachová Laboratory for Epigenetics, Epiontis GmbH, Berlin, GermanySearch for more papers by this authorM. P. Manns, M. P. Manns Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorM. Hardtke-Wolenski, M. Hardtke-Wolenski Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Germany Equal senior authors.Search for more papers by this authorE. Jaeckel, Corresponding Author E. Jaeckel Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Germany Equal senior authors.Elmar Jaeckel, jaeckel.elmar@mh-hannover.deSearch for more papers by this author R. Taubert, R. Taubert Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorS. Pischke, S. Pischke Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorJ. Schlue, J. Schlue Department of Pathology, Hannover Medical School, GermanySearch for more papers by this authorH. Wedemeyer, H. Wedemeyer Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorF. Noyan, F. Noyan Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorA. Heim, A. Heim Department of Virology, Hannover Medical School, GermanySearch for more papers by this authorF. Lehner, F. Lehner Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorH. Barg-Hock, H. Barg-Hock Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorJ. Klempnauer, J. Klempnauer Department of Visceral and Transplant Surgery, Hannover Medical School, GermanySearch for more papers by this authorS. Olek, S. Olek Ivana Türbachová Laboratory for Epigenetics, Epiontis GmbH, Berlin, GermanySearch for more papers by this authorM. P. Manns, M. P. Manns Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, GermanySearch for more papers by this authorM. Hardtke-Wolenski, M. Hardtke-Wolenski Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Germany Equal senior authors.Search for more papers by this authorE. Jaeckel, Corresponding Author E. Jaeckel Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Germany Equal senior authors.Elmar Jaeckel, jaeckel.elmar@mh-hannover.deSearch for more papers by this author First published: 20 September 2012 https://doi.org/10.1111/j.1600-6143.2012.04264.xCitations: 30 AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Acute cellular rejection (ACR) occurs frequently after liver transplantation and can usually be controlled. Triggering of allospecific immune responses and lack of immunoregulation are currently suggested as a cause of ACR, but there are no investigations of intrahepatic immune responses during ACR. Therefore we prospectively analyzed the intrahepatic T cell infiltration pattern in correlation to the severity of ACR in a cohort of patients with graft hepatitis (n = 151). While CD4+ cells dominated the portal infiltrates in mild–moderate ACR, CD8+ cells prevailed in severe ACR. Furthermore portal CD8+ and not CD4+ infiltration correlated with serum transaminases and with the likelihood of subsequent ACRs. Surprisingly, the rise of portal effector T cells density during ACR was surpassed by the increase in portal infiltration of regulatory T cells by a factor of two. Thus ACRs rather showed an increase and not a lack of regulation, as was suggested by analysis of peripheral blood mononuclear cells. Despite the pattern of enhanced immunoregulation, patients with severe ACR had a higher risk for subsequent rejections and showed a trend to a reduced survival. Thus, patients with severe rejections might need a modification of their immunosuppression to improve prognosis. Abbreviations: ACR acute cellular rejection CNI calcineurin inhibitors CSA cyclosporine FFPE formalin-fixed paraffin-embedded IS immunosuppression LFT liver function test LT liver transplantation MMF mycophenolate mofetil Pred. prednisolone PB protocol biopsy RAI rejection activity index SR Spearman rho rank correlation coefficient TAC tacrolimus Treg regulatory T cells Teff effector T cells Introduction Orthotopic liver transplantation (LT) is the ultimate therapy of end stage liver disease and of early stages of hepatocellular carcinoma. Although the liver is immunological privileged with good long-term survival and low incidence of chronic rejection (∼2%) compared to other transplanted solid organs, acute cellular rejection (ACR) still occurs frequently (∼30%; 1). ACR of liver grafts is characterized by a trias of mixed mononuclear portal infiltrate, containing lymphocytes, neutrophils and frequently eosinophils, bile duct damage and sub endothelial inflammation of veins or terminal hepatic venules. The severity of ACR in liver grafts is classified by the international accepted rejection activity index (RAI) or BANFF score that includes the grades of all three criteria (2). Regulatory T cells (Treg) are one of the key regulators of immune homeostasis. Their substantial quantitative and/or functional deficiency, e.g. in hereditary immune disorders like IPEX, can cause lethal autoimmunity (3-5). Animal models support a role of Tregs in prevention of graft rejection as well, but many of those studies were performed with tolerance inducing protocols (6) and not under classical chronic immunosuppression (IS). Despite that, adoptive transfer of Tregs could just control rejection under lymphopenic conditions, whereas under nonlymphopenic conditions, adoptive transfer of polyspecific Tregs was largely unsuccessful to control transplant rejection (7). The lack of exclusive markers hampers an unambiguous characterization and functional analysis of human Tregs. Especially, CD25 and FOXP3 are expressed by activated effector T cells (Teff) as well (8). CD127 can be used to distinguish between activated Teff (CD127+) and Treg (CD127low; 9, 10). But only the demethylation of a Treg-specific demethylated region (TSDR) in the FOXP3 gene indicates stable FOXP3 expression and suppressive capacity without contamination of T cells producing proinflammatory cytokines like IL2, IL17 or IFNγ (11-14). Therefore, we consider CD4+CD25highCD127lowFOXP3+/high with a demethylated TSDR as the current gold standard definition for stable and functional Tregs in humans. After LT in humans, a lack of peripheral Tregs was associated with ACRs supporting the hypothesis that a lack of regulation by Tregs is causative for the development of rejection (15-18). However, the analysis of peripheral Tregs might not mirror the immunoregulation within the graft and animal models highlighted the importance of intragraft Tregs for prevention of rejection (19). Studies of intrahepatic Tregs after LT were limited to quantitative FOXP3 mRNA analysis or semiquantitative assessment of Treg numbers in single color immunohistochemistry (20-23). These methodological limitations might also explain the contradictory results: Although Li et al. reported an increase of FOXP3+ cells in operational tolerant patients off IS, they could not detect Tregs in chronic rejection supporting the above mentioned hypothesis on the role of Tregs in ACR (22). In contrast, Stenard et al. reported an increase of FOXP3+ cells in ACR (23), whereas Eguchi et al. did not find a correlation between rejection index and numbers of FOXP3+ cells (21). As Treg numbers are usually increasing with severity of inflammation, the mere report of Treg numbers might not be sufficient. Instead concentrations of Tregs in relation to effector cells should be provided. We, therefore, developed a four-color immunofluorescence of formalin-fixed paraffin-embedded tissues (FFPE) and analyzed the concentration of Tregs in relation to Teff quantitatively. If Tregs are important in immunoregulation their contribution should vary with more severe rejection. Therefore, we intentionally included mild and severe ACRs to correlate our findings with clinical and histomorphological data. Using this methodology we could demonstrate that ACRs were associated with a predominant infiltration of CD8+ T cells. More importantly, intrahepatic immune responses during ACR were not associated with a lack but rather an increase of Treg infiltration, which was independent of the IS used. Thus our results suggest that a lack of immunoregulation is not responsible for development of graft rejection but rather support an active role of Tregs in controlling rejection. Materials and Methods Subjects Percutaneous needle biopsies of 151 patients after LT, characterized by at least two times elevation of liver function tests (LFT; ALT, AST, AP, gGT, GLDH, bilirubin), were prospectively taken and examined in our transplantation center between 2004 and 2008. Patients with proven virus replication of HBV, HCV or CMV pp65 antigen in peripheral blood were excluded. The clinical data are listed in Table 1. Patient's age at biopsy ranged from 17 to 72 years. The liver grafts were transplanted between 1975 and 2008. Patients and graft survival were observed till March 2009. Table 1. Demographic and clinical data of the patient collectives 151 patients 54 biopsies 10 protocol biopsies Age (years) 45.9 ± 13.3 43.6 ± 11.9 47.2 ± 9.5 Sex (male/female) 94/57 30/24 7/3 Rejection activity index (RAI) 2.7 ± 2.5 3.8 ± 2.5 0.1 ± 0.3 Time Tx to Biopsy (month) 32.8 ± 55.6 27.1 ± 51.2 13.7 ± 10.3 Graft survival after biopsy (month) 23.7 ± 15.2 25.1 ± 13.7 Death 32 7 ReTx 6 0 Cause of liver transplantation PSC 21% 24% 20% HBV 9% 6% 10% Alcoholic cirrhosis 18% 24% 10% Liver failure 8% 17% 10% M. Wilson 5% 4% 10% AIH 4% 0% 10% PBC 7% 4% 0% HCV 0% 0% 10% Others 20% 22% 20% Immunosuppression Cyclosporine A 54% 63% 50% Tacrolimus 42% 35% 50% No CNI 4% 2% 0% Prednisolone 69% 74% 90% Mycophenolate mofetil 50% 44% 90% Liver function test ALT (U/L) 198 ± 214 235 ± 196 24 ± 6 AST (U/L) 146 ± 160 172 ± 183 22 ± 8 GLDH (U/L) 33 ± 63 39 ± 62 2 ± 1 gGT (U/L) 360 ± 332 355 ± 290 30 ± 12 AP (U/L) 306 ± 258 305 ± 207 84 ± 35 Bilirubin (μmol/L) 68 ± 110 61 ± 87 10 ± 3 Demographic data and clinical characteristics of the graft hepatitis group (left column) and subgroup with further immunohistological analysis (central column) and the control group of protocol biopsies (right column). The distribution of RAI of both groups with graft hepatitis is shown in Figure 1. For analysis of portal T cell infiltration, additional immunohistological stainings of 54 biopsies out of this collective have been performed. For correlation analysis over all grades of ACR, we randomly chose patients according to their RAI (Figure 1). Figure 1Open in figure viewerPowerPoint Distribution of ACR grades during episodes of graft hepatitis. The actual distribution of RAI in our collective of 151 patients with graft hepatitis is shown in (A). To study T cell pattern during different grades of ACR a more equal distribution of RAI was intended in our subgroup of 54 biopsies (B). Therefore, we randomly chose biopsies according to their RAI. Ten protocol biopsies (PB) without histological signs of rejection, without abnormal LFT and without proven viral replication (HCV, HBV, CMV) at the time point of biopsy were randomly selected from our PB program as a control group. This study was approved by the local research Ethics Committee. Written informed consent was obtained from all patients before performing liver biopsy. Immunohistological methods We used up to four serial but not adjacent sections of FFPE liver biopsies per patient. For dewaxing, tissue sections were immersed in xylol and for rehydration in alcohol bathes with diminishing ethanol concentration and were finally stored in deionized water. We performed heat-induced epitope recovery in 1 mM EDTA solution. Following antibodies and detergents were used: mouse anti-human CD4 (BC/1F6, Abcam, Cambridge, UK), rabbit anti-human CD8 (SP16, Abcam), goat CY5 conjugated anti-mouse (Jackson Immunoresearch, Suffolk, UK), goat AF488 conjugated anti-rabbit (Invitrogen, Carlsbad, CA, USA), rat biotin conjugated antihuman FOXP3 (PCH101, eBioscience, San Diego, CA, USA), Cy3 conjugated streptavidin (Jackson Immunoresearch), DAPI. Further staining steps were performed in Tris-buffered saline with 0.05% Tween 20 at room temperature. Sections were blocked before incubation with CD4 and CD8 antibodies, washed and incubated with secondary antimouse and antirabbit antibodies. Then further blocking steps with Avidin (0.01%), Biotin (0.005%) were performed. Followed by FOXP3 antibody and incubated with streptavidin-Cy3. Nuclei were stained with DAPI. Finally, sections were covered with Mowiol. ACR grading and quantification of portal T cell infiltration Biopsies were fixed in 4% neutral buffered formalin, embedded in paraffin and 2-μm-thick sections were stained with hematoxylin and eosin. Histological examination and grading of RAI (2) was performed by an experienced liver pathologist. Immunofluorescent staining and quantification of portal T cell infiltration were performed in a blinded fashion. Taking of photographs of representative portal fields (Axio Imager.M1 at a magnification of 200x, Carl Zeiss, Jena, Germany), counting of cells and quantification of infiltration area was performed with AxioVision 4.6 software (Carl Zeiss Microimaging). DNA methylation analysis DNA methylation analysis was performed by a quantitative RT-PCR analysis (13). All epigenetic analyses were performed by Epiontis (Berlin, Germany). Statistical analysis Statistical analysis was performed with SPSS 15.0 software. T-test was used for comparison between two groups. For correlation analysis the Spearman rank correlation coefficient was calculated. Survival of patient groups was analyzed according to Kaplan–Meier. Note that p values < 0.05 (two tailed) were considered statistically significant in all analyses. Results Low to moderate interportal variation of T cell infiltration pattern We analyzed the portal T cell infiltration pattern in 54 biopsies out of a collective of 151 patients during an episode of graft hepatitis characterized by an elevation of LFT twice above normal (Table 1). We excluded HBV and HCV reinfection or CMV reactivation as other causes of graft hepatitis to focus on immune-mediated mechanisms caused by allospecific rejection. For correlation analysis over all grades of rejection, the biopsies were chosen based on the RAI to enrich for more severe grades of rejection. We established a highly reproducible procedure of deparaffinizing and heat-induced epitope recovery (24) of the FFPE biopsies that enabled us to simultaneously stain surface (CD4, CD8) and nuclear antigens (FOXP3; Figure 2). Figure 2Open in figure viewerPowerPoint Multicolor immunofluorescence of human FFPE liver biopsies. Surface staining of CD4 (red) and CD8 (green) can be clearly distinguished from nuclear localization of FOXP3 (blue, white arrows and additionally in D with DAPI counterstaining in white) in liver sections (A and in higher magnification C and D). Strong autofluorescence caused the varying colors of hepatocytes as seen in the secondary reagent control (B). Liver sinusoidal epithelial cells weakly express CD4 and can be distinguished from T cells by strength of CD4 expression, shape and localization of cells. Light extracellular dots are mostly erythrocytes. Portal T cell infiltration increased with severity of ACR (RAI 2: E, RAI 4: F, RAI 7: G). The red lines surround portal infiltrations and exclude lumen of veins, arteries and bile ducts. We counted portal lymphocytes (CD4+, CD8+, CD4+FOXP3+, CD8+FOXP3+) until we reached around 200–400 cells of each T cell lineage (CD4+, CD8+) or evaluated all portal fields of up to four sequential biopsy sections. Over all, up to 14 portal fields in a single section and up to 17 portal fields in serial but not adjacent sections were analyzed per patient. The area of evaluated portal infiltrations (excluding the lumen of all veins, bile ducts and arteries) was quantified (Figure 2). Subsequently, we were able to measure the portal infiltration densities and ratios of CD4+, CD8+, CD4+FOXP3+ and CD8+FOXP3+. The coefficients of variation between the portal fields of one biopsy were low for CD4+ and CD8+ and moderate for CD4+FOXP3+ infiltration density (Figure 3A). The higher variation of CD4+FOXP3+ was because of the 10–20 times lower portal cell frequency. In average, five–six portal fields were evaluated to compensate for higher variation of rarer cells. Thus, a comparison between biopsies with varying numbers of evaluated portal fields was justified. Figure 3Open in figure viewerPowerPoint Reliable detection of Treg in FFPE liver biopsies. (A) Because of varying size of portal infiltrates and varying density of portal T cell infiltration up to 17 portal fields per biopsy were evaluated to reach 200–400 cells of each T cell linage (CD4+, CD8+). The variation of T cell infiltration between the portal fields within a biopsy was low for frequent cells (CD8+, CD4+) and moderate for rare cells (CD4+FOXP3+) and thus allowed the comparison of liver biopsies with different numbers of evaluated portal fields. All 64 biopsies of this study (graft hepatitis and protocol biopsies (Tab 1) have been included in the variation analysis. (B) The histologically determined portal CD4+FOXP3+/CD4++CD8+ ratios were comparable to the FOXP3/CD3 ratios determined by a DNA methylation analysis of the whole FFPE liver biopsies (2x ACR with RAI 7 [black dots]; 2x protocol biopsy without ACR [black rings]) and in a FFPE inflamed tonsil (triangle) as a control. Lymphocyte infiltration was predominantly localized in the portal fields and with varying extend in the lobules. Lobular infiltration is biased by normal lymphocytic circulation through sinusoids and cannot be standardized to the infiltration density. The Banff rejection scoring system incorporates centrilobular changes ([peri]venulitis) as well (2). A relevant extension of the venulitis to the parenchyma (score 3 in the category venous endothelial inflammation) was only diagnosed in 3/54 biopsies. Thus, we focused our further analysis exclusively on the portal infiltration. Reliable detection of intrahepatic Treg and Teff Despite substantial testing, we failed to include CD127, LAP, GARP or CD137 staining into our setting of FFPE biopsies to distinguish activated CD4+FOXP3+ Teff from CD4+FOXP3+ Treg. To validate that our immunofluorescence technique just recognizes true Tregs, we evaluated the Treg content of sorted human peripheral blood mononuclear cell (PBMC) populations (CD3+CD25high, CD3+CD25−) with three independent methods (Table S1): (1) we quantified Tregs by flow cytometry according to their CD4+CD25highCD127lowFOXP3+ phenotype. (2) In parallel, we analyzed the percentage of Tregs by quantitative epigenetic analysis of the TSDR. (3) Finally, the sorted cell population was injected into spleen and liver of RAG1−/− mice lacking endogenous lymphocytes. The murine organs were handled in parallel with the liver biopsies of transplanted patients with our immunophenotyping (Figure S1). The number of CD8+ T cells in sorted CD3+CD25high cells was low. Most important, the frequency of Tregs detected by immunofluorescence of FFPE samples were related to Treg determined as CD4+CD25highCD127lowFOXP3+ in flow cytometry and beneath the frequency of demethylated TSDR (Table S1). Thus, our histological technique detected Treg comparable to FACS and trended to underestimate the epigenetically detected Treg frequency. In addition, sorted CD3+CD25− cells (Teff) did not contain relevant amounts of demethylated TSDR or FOXP3+ cells by immunofluorescence. These results were confirmed in liver biopsies of transplanted patients by showing a close correlation between FOXP3/CD3 ratios as determined by TSDR analysis and our immunofluorescence. Even in an inflamed tonsil, Treg frequencies by TSDR analysis correlated closely with the Treg number obtained by our staining method (Figure 3B). Thus, we consider CD4+FOXP3+ cells as Treg rather than FOXP3+ Teff in our immunofluorescence technique. Different portal infiltration pattern of T cells subset during ACR Portal CD4+ cells densities rose from no/mild to moderate rejection (RAI 0–4) and were highly correlated with the RAI (Spearman rho [SR] 0.73; p < 0.001), whereas they reached a plateau in moderate–high grades of ACR (RAI 5–8; SR =–0.002; p = 0.994; Figure 4A). In contrast, portal density of CD8+ cells continuously increased with RAI and dominated in severe ACR (RAI ≥ 7). CD8+ cell infiltration was correlated with RAI over all grades of ACRs (SR = 0.399; p = 0.003; Figure 4B). As a result, the portal CD4/CD8 ratio continuously decreased to RAI 8 and was inversely correlated with RAI (SR =–0.379; p = 0.005; Figure 4D). But overall portal Teff density (CD4++CD8+) rose continuously to RAI 8 and was correlated with RAI (SR = 0.341; p = 0.012). Figure 4Open in figure viewerPowerPoint Portal T cell infiltration pattern in ACR. Although mild to moderate grades of ACR were predominantly characterized by portal CD4+ T cell infiltration (A) density of CD8+ T cell infiltration rose continuously and dominated in severe ACR (RAI ≥ 7) (B) thus CD4/CD8 ratio decreased with severity of ACR (D). Surprisingly not only portal Treg (CD4+FOXP3+) infiltration density increased continuously with ACR (C) but also the fold increased of Treg surpassed Teff by factor of two (F; CD4+ dotted line; CD8+ dashed line; Treg black line) resulting in a continuous increase of Treg/Teff ratio with severity of ACR (E). Because of the ordinal scale of the RAI the spearman rank correlation test for correlation analysis of portal T cell infiltration and RAI was applied (SR: Spearman rho; A–E). Numbers of biopsies per rejection grade are depicted in Figure 1(B). In parallel with Teff portal Treg (CD4+FOXP3+) infiltration increased with severity of ACR (Figure 4C) and the fold increase of portal Treg infiltration surpassed the Teff increase which resulted in continuous increase of portal Treg/Teff ratio, that correlated with the RAI (SR = 0.36; p = 0.007; Figure 4E). Finally, portal Treg infiltration density was correlated with RAI too (SR = 0.402; p = 0.003). In summary, mild–moderate grades of ACR in liver grafts were dominated by CD4+ cells whereas portal infiltration of CD8+ cells increased continuously and dominated in severe ACR (RAI ≥ 7). Surprisingly, portal Treg infiltration was correlated with RAI and surpassed the fold increase of Teff leading to a continuous increase of the portal Treg/Teff ratio with severity of rejection. Sixteen of the 54 biopsies (29.6%) contained portal CD8+FOXP3+ cells. These biopsies had higher RAI (p = 0.005) and subsequently higher Treg/Teff ratios (p = 0.022), whereas the portal Treg density was not significantly increased (p = 0.06). Nonetheless, CD8+FOXP3+ were just detected infrequently and at very low frequency questioning their regulatory significance. Comparison of liver biopsies and PBMCs during ACR and in patients without rejection supports a role of active regulation in the liver Comparison of intrahepatic infiltrates during ACR with biopsies of patients receiving PB (with normal liver enzymes) further supported an active role of regulation during rejection: (1) The portal infiltration area, which did not correlate with RAI (p = 0.141), was enlarged in ACR (Figure 5A). (2) Teff density in patients with ACR was not significantly different compared to patients with PB. Thus, the overall increase in Teff numbers in ACR was rather because of an increase in portal infiltration area (Figure 5B). (3) In contrast, portal Treg density was increased in ACR as compared to PB leading to an increased Treg/Teff ratio (Figure 5C and D). Figure 5Open in figure viewerPowerPoint Comparison of intrahepatic T cell pattern during ACR and in protocol biopsies after liver transplantation. Compared to protocol biopsies (PB, with normal liver function tests and without histological signs of ACR) after LT the size of portal infiltrates were significantly enlarged during various grades of ACR (A, mild–moderate ACR: RAI 3–6; severe ACR: RAI ≥ 7). Although portal Teff density remained the same in these enlarged infiltrates (B), Treg filled up these portal infiltrates with a significantly higher density during ACR (C). Because portal Treg/Teff ratio was subsequently higher during ACR an active role for immunoregulation within the rejected allograft is suggested (D). Numbers of analyzed samples: PB: 10; mild–moderate ACR: 27; severe ACR: 9. (n.s. p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001). Although the Treg/Teff ratios were unchanged in peripheral blood during ACR they were increased within the liver (Figure S2B and C). In addition, the majority of peripheral blood Tregs expressed the effector molecule CD39 highlighting the functional properties of Tregs after transplantation (Figure S2D). Comparison of intrahepatic T cell pattern with the clinical course Although histological grading is the gold standard to diagnose rejection, therapeutic decisions are often based on LFTs as well. Therefore, we correlated LFTs with RAI and the portal T cell pattern. In our group of 54 biopsies, serum markers of hepatocyte damage (ALT, AST) and not cholestatic markers (gGT, AP) or bilirubin correlated with RAI (AST: SR = 0.475; p < 0.001; ALT: SR = 0.375; p = 0.005; Figure 6A). Elevation of serum transaminases correlated with portal CD8+ (AST: SR = 0.299; p = 0.028; ALT: SR = 0.269; p = 0.049) and not CD4+ infiltration (AST: p = 0.47; ALT: p = 0.266; data not shown). Figure 6Open in figure viewerPowerPoint Correlation of liver function test with histological characteristics of ACR. Liver enzymes as serum markers of graft hepatitis were correlated with RAI (A) and portal infiltration pattern (B). Because of ordinal scale of the RAI and to compensate outliers in LFTs we used the Spearman rank correlation coefficient (SR: Spearman rho). Only transaminases as markers of hepatocyte damage and not cholestatic markers or bilirubin correlated with RAI (A). Portal CD8+ and not CD4+ infiltration correlated with hepatocyte damage (B). Number of samples per RAI is depicted in Figure 1. The clinical diagnoses, treatment of ACR and epicrisis are summarized in Table 2. Table 2. Treatment response of ACR Clinical diagnosis Immunosuppression Treatment response Steroid bolus Increased Modified Unclear Remission Refractory Death No follow-up Rejection 34 26 12 9 2 27 5 1 1 No rejection 20 0 0 0 3 10 2 3 5 Seven patients (13%) experienced a second ACR in average 24.4 ± 13 months after the initial episode. These initial biopsies were characterized by a lower portal CD4/CD8 ratio (p = 0.005) and a higher RAI (p = 0.005; Figure 7A and B). In addition, patients with severe ACR (RAI ≥ 7) tend to have a lower graft survival compared to lower ACR grades (RAI 3–6) and other causes of graft hepatitis (RAI < 3; Figure 7C). This trend did not reach statistical significance because of the low number of patients with severe stages ACRs. Figure 7Open in figure viewerPowerPoint Clinical course after graft hepatitis. I