SEARCH

SEARCH BY CITATION

Keywords:

  • Acute rejection;
  • kidney transplantation;
  • monocyte;
  • T-cell proliferation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

It is still disputed in which anatomical compartments of allograft recipients T-cells proliferate. After experimental renal transplantation, host monocytes and lymphocytes accumulate in the lumina of graft blood vessels. In this study, we test the hypothesis that T lymphocytes proliferate in the vascular bed of the graft. Kidneys were transplanted in the Dark Agouti to Lewis rat strain combination, an established experimental model for acute rejection. Isogeneic transplantation was performed as a control. Cells in the S-phase of mitosis were detected in situ three days posttransplantation by pulse-labeling with BrdU and by immunohistochemical detection of the proliferating cell nuclear antigen (PCNA). More than 20% of all T-cells in the lumina of allograft blood vessels incorporated BrdU and approximately 30% of them expressed PCNA. In the blood vessels of isografts as well as in other organs of allograft recipients, only few BrdU+ cells were detected. A majority of the BrdU+ cells in graft blood vessels expressed CD8. In conclusion, we demonstrate that CD8+ T lymphocytes proliferate in the lumina of the blood vessels of renal allografts during the onset of acute rejection.


Abbreviations: 
APAAP

alkaline phosphatase anti-alkaline phosphatase

DAB

3,3′-diaminobenzidine

NRS

normal rat serum

PCNA

proliferating cell nuclear antigen

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

According to the prevailing doctrine, naïve T-cells proliferate in secondary lymphoid organs before migrating to allografts (1–3), whereas activated and memory T cells can also proliferate in allograft tissues (4–6). This concept has been recently challenged by an elegant study on mouse lung allografts, which are rejected by naïve recipients lacking secondary lymphoid organs, suggesting proliferation of naïve T cells within the graft (7). Conclusions from this study, however, should be drawn with some care because the animal model used is highly artificial.

Several lines of experimental evidence suggest the provocative idea that proliferation of alloreactive T cells may in addition take place in the lumina of allograft blood vessels (8–14): Conclusive data on T-cell proliferation in intact allograft blood vessels in vivo, however, have not been published yet. Previously, we reported that renal and pulmonary allograft blood vessels become a crowded meeting place of host leukocytes and allogeneic endothelial cells. Monocytes and T cells are most abundant in allograft blood vessels and preliminary data suggest T-cell proliferation in blood vessels of lung allografts (15,16).

Herein, we test the hypothesis that T-cells proliferate inside renal graft blood vessels during the onset of acute rejection. We use an established model for acute rejection of renal allografts involving a fully allogeneic combination of wild-type rat strains. Of note, we exclusively use unmodified wild-type rats as donors and recipients and T-cell proliferation is investigated in vivo.

Material and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

Animal Experiments

Specified pathogen free male Lewis (RT1l) and Dark Agouti (RT1av1) rats were supplied by Harlan Winkelmann (Borchen, Germany) or Elevage Janvier (Le Genest-St-Isle, France). Rats were kept for up to two weeks in a conventional animal facility. Dark Agouti rats were used as donors for allogeneic and Lewis rats for isogeneic transplantation. Animal care and experiments were performed in accordance with German animal protection laws as well as the NIH principles of laboratory animal care. Kidneys were transplanted orthotopically to totally nephrectomized recipients weighing 260–300 g. Ischemic times remained below 30 min. After allogeneic transplantation, recipients die within 7.4 ± 0.7 days (mean ± SD, n = 10) after transplantation, whereas isograft recipients survive in good health (n = 10; Ref.17).

Three days after isogeneic (n = 5) or allogeneic (n = 4) kidney transplantation, the recipients received an injection of 25 mg BrdU (Sigma-Aldrich, Taufkirchen, Germany) in 1 ml 0.9% saline i.v. and were sacrificed 30 min later. Grafts, spleens, hearts, livers, pancreata and small intestines were removed immediately and cut into small pieces. The specimens were sliced, snap-frozen and stored in liquid nitrogen. One slice of each graft was fixed in 4% buffered paraformaldehyde for conventional paraffin histology and staining of histological sections with hemalum eosin. Additional transplantations were performed to isolate graft blood leukocytes for flow cytometry and immunocytochemistry (see later). Kidneys from untreated Lewis rats were used as additional controls.

Immunohistochemistry

Antibodies used in this study are listed in Table S1. Cryostat sections (5 μm) were collected on silane coated glass, fixed in isopropanol for 10 min at 4°C, air-dried for 1 h at room temperature and stored at –20°C until use. Sections were incubated for 30 min in 1% H2O2 in PBS at room temperature and rinsed in PBS pH 7.2. To stain vascular endothelial cells, mAb HIS52, diluted 1:3000 in PBS, pH 7.2, 1% BSA (Serva, Heidelberg, Germany), 0.1% NaN3 (p.a. Merck, Darmstadt, Germany; PBS/BSA/NaN3) was applied to the sections for 1 h at room temperature. Bound antibodies were detected with rabbit anti-mouse immunoglobulins conjugated to horseradish peroxidase (DAKO, Hamburg, Germany) containing 5% heat inactivated normal rat serum (NRS, Harlan Winkelmann) and the chromogen 3,3′-diaminobenzidine (DAB, Sigma-Aldrich). In a second staining step, mAbs to the α/βTCR (R73, 1:500), to CD4 (W3/25, 1:500), to the CD8 α-chain (Ox8, 1:500), to the CD8 β-chain (341, 1:500), to CD45RC (Ox22, 1:500), or to CD161 (10/78, 1:500) diluted in PBS/BSA/NaN3 were applied. Rabbit anti-mouse immunoglobulins and alkaline phosphatase anti-alkaline phosphatase (APAAP; both DAKO) in combination with the chromogen Fast Blue were used to detect the second set of primary antibodies. In the third staining step, cell nuclei labeled with BrdU were detected. DNA was denatured in 0.1 M HCl and 0.9% NaCl at 60°C for 10 min. This procedure also removes antibodies and enzymes from the previous staining steps but neither the DAB nor the Fast Blue product. Slides were washed in PBS and incubated with mAb Bu20a (1:400) followed by rabbit anti-mouse immunoglobulins conjugated to peroxidase in the presence of 5% NRS. Finally, peroxidase was visualized with DAB. To detect PCNA, cryostat sections were postfixed with 4% buffered paraformaldehyde for 8 min at 4°C and washed with PBS. After blocking endogenous peroxidase activity, mAb PC10, 1:1000 was applied and detected with anti-mouse immunoglobulins conjugated to peroxidase and DAB. Thereafter, RECA-1 and finally R73 were applied and detected as described. As a control, the staining procedure was performed in the absence of mAbs to leukocyte surface antigens and to BrdU or PCNA. In addition, each primary antibody was used alone. The small intestine was included as a positive control for cell proliferation. Sections were cover-slipped in glycergel (DAKO), evaluated with an Olympus BX51 (Olympus Optical Co., Hamburg, Germany) microscope and documented using the analysis software.

To measure the proportion of proliferating (BrdU+ or PCNA+) T cells in renal blood vessels, at least 50 intravascular cells stained with either mAb R73 or Ox22 were counted per experiment. Furthermore, the proportions of α/βTCR/BrdU+ or α/βTCR/PCNA+ cells were determined as well as the proportions of intravascular BrdU+ cells expressing CD4, CD8 α-chain, CD8 β-chain or CD161.

Flow cytometry and immunocytochemistry

Leukocytes, including the population interacting with vascular endothelial cells were harvested from the blood vessels of renal transplants by intensive single organ perfusion as described previously (12). The methods for flow cytometry and immunocytochemistry are indicated in the data supplement of the online version of this paper.

Statistical analyses

Results are reported as mean ± SD. Results obtained for allografts were compared to isografts and statistical significance was analyzed by Mann–Whitney rank sum test (SPSS software, Munich, Germany) with p ≤ 0.05 set as level for significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

Most T-cells proliferate in the blood vessels of renal allografts

Dark Agouti kidneys were transplanted to Lewis rats, which is a fully allogeneic combination of immuno-competent wild-type inbred rats. Three days after transplantation, cell nuclei in the S phase of mitosis were pulse-labeled in vivo for 30 min with BrdU immediately before sacrificing the animals. To identify T cells synthesizing DNA in the lumina of graft blood vessels, triple-staining was performed to detect the endothelial cell marker RECA-1, the α/β T cell receptor (α/βTCR) and BrdU. A remarkably high proportion of intravascular T-cells with BrdU+ nuclei (22.1 ± 6.3%) was detected in renal allografts (n = 4), whereas in isografts (n = 5) only 4.3 ± 0.9% blood T cells were BrdU+ (p = 0.014; Figures 1A and B). In allografts, most BrdU+ leukocytes expressed the α/βTCR (84.7 ± 10.2%), whereas their proportion was lower in isografts (42.0 ± 11.1%, α/βTCR+/BrdU+ cells). Because of their low absolute number, we did not identify BrdU+/α/βTCR− cells.

image

Figure 1. Proliferation and immune phenotype of T-cells in blood vessels of rat renal grafts on day three after transplantation. (A, B) Cells were pulse-labeled in vivo with 5-bromo-2’-deoxyuridine (BrdU) and analyzed by immunohistochemical triple-staining with antibodies to BrdU, to the α/β TCR and RECA-1 (vascular endothelial cells). (C, D) The proliferating cell nuclear antigen (PCNA) was detected together with the α/β T-cell receptor and RECA-1. (A, C) T-cells are visualized in blue, BrdU+ or PCNA+ cell nuclei and endothelial cells are stained in brown. The arrows point to BrdU+ or PCNA+ intravascular T-cells. (B) Quantification of intravascular BrdU+ T-cells on immunohistologic sections of renal isografts (iso, n = 5) and allografts (allo, n = 4). (C) Quantification of intravascular PCNA+ T-cells in healthy control kidneys (con), isografts and allografts (n = 5 each). Box plots indicate median and percentiles 0, 25, 75 and 100. The open circle represents a value beyond ± 3 x standard deviation. (E, F) Cytospin preparations of mononuclear leukocytes originating from a renal allograft on day three posttransplantation. (E) The α/β T-cell receptor is stained in brown, MHC class II antigens are stained in blue. The arrow is pointing to a putative immunological synapse. (F) The α/β T-cell receptor is stained in red and cell nuclei are counter-stained with hemalum to visualize mitotic figures.

Download figure to PowerPoint

To verify this result independently, we detected PCNA. As expected, PCNA+ nuclei were more frequent than BrdU+ nuclei in α/βTCR+ allograft T-cells (approximately 30%) and only a minor proportion of α/βTCR+ T-cells from isografts and normal control kidneys were PCNA+ (Figures 1C and D). In addition, mitotic figures were seen in T-cells on cytospin preparations of leukocytes isolated from allograft blood vessels (Figure 1F).

Proliferation of blood T-cells is restricted to the allograft

To test if there is a general induction of T-cell proliferation in the blood, we analyzed BrdU incorporation by blood α/βTCR+ cells in other organs of the same allograft recipients (n = 4 each). Only 2.1 ± 1.5%α/βTCR/BrdU double-positive T-cells were detected in the heart, 1.7 ± 0.6% in the liver and 0.9 ± 1.0% in the pancreas, indicating that BrdU-incorporation into T-cell nuclei predominantly occurs in the blood vessels of the allograft.

Most BrdU+ allograft blood leukocytes are CD8+ T-cells

BrdU+ graft blood cells were further characterized by immunohistochemistry: 72 ± 5.3% of them expressed the α-chain of CD8, 59.8 ± 3.1% expressed the CD8 β-chain and 9.8 ± 1.7% expressed CD4, which can also be expressed by rat monocytes (15,16). CD161 (NKR-P1), a cell surface marker for NK cells and natural killer T-cells was expressed by 39 ± 6.0% of the BrdU+ blood cells. However, no coexpression of the constant Vβ8.2 T-cell receptor chain typical for rat natural killer T-cells (18) was detected.

In contrast to the T-cells inside blood vessels, only 2.0 ± 0.8% (n = 3) of the CD8+ cells were BrdU+ in the periarteriolar lymphatic sheath (T-cell area) of spleens of allograft recipients (Figure S1). The proportion of proliferating CD8+ spleen cells was similar in isograft recipients (1.6 ± 0.3%, n = 3).

Most CD8+ T-cells from allograft blood vessels are naïve or recently activated

Next, we isolated intravascular isograft and allograft leukocytes by intensive perfusion of graft blood vessels on day 3 posttransplantation and analyzed their cellular composition by flow cytometry. The amount of leukocytes harvested from allografts was more than threefold higher compared to isografts. Monocytes were most frequent in isografts and allografts, followed by lymphocytes, NK cells and granulocytes (Table 1). The percentage of CD8+ T-cells among all CD5+ T-cells in allograft blood vessels was approximately 55% and slightly increased in comparison to isografts (Table S2).

Table 1.  Percentage of leukocytes obtained from the vasculature of isotransplanted (iso) and allotransplanted (allo) kidneys on day 3 after transplantation. The first value indicates the composition of the perfusate. The second value indicates the cellular composition of mononuclear leukocytes purified by Percoll density gradient centrifugation
Total cell number (x 106)Monocytes (%) ED9 (low granularity)Granulocytes (%) ED9 (high granularity)T-cells (%) Ox19B cells (%) Ox33NK cells (%) 10/78 (strong expression)
  1. Note: The data are expressed as mean ± SD; p Values refer to differences between isografts and allografts.

iso (n = 4) 8.8 ± 0.9 6.6 ± 1.163.9 ± 5.3 66.6 ± 4.85.5 ± 1.7 0.4 ± 0.212.9 ± 1.9 15.2 ± 2.37.4 ± 2.6 8.3 ± 3.28.6 ± 2.4 9.5 ± 2.2
allo (n = 4) 32.7 ± 6.2 (p = 0.021) 24.7 ± 4.0 (p = 0.021)76.2 ± 4.1 (p = 0.021) 79.1 ± 1.6 (p = 0.021)2.5 ± 1.1 (p = 0.021) 0.2 ± 0.1 (p = 0.083)7.6 ± 1.7 (p = 0.021) 8.5 ± 1.5 (p = 0.021)8.4 ± 2.0 (p = 0.773) 7.9 ± 1.3 (p = 0.773)3.2 ± 1.7 (p = 0.043) 3.6 ± 2.1 (p = 0.043)

We investigated the state of activation of T-cells from day three allograft blood vessels. CD25 (Figure 2A), CD71 and CD62L (Figure 2B) are early T-cell activation markers, whereas CD43 is down-regulated upon activation (15,16). CD45RC is a marker expressed by naïve or recently activated rat T-cells but absent from effector and central memory cells (19). Between the total T-cell and the CD8+ T-cell population, a small proportion of isograft T-cells expressed CD25 and CD71, approximately half of them expressed CD62L and most T-cells were CD43+ (Table 2, Table S2). The proportion of CD25+ and CD71+ T-cells slightly increased in allografts, whereas no changes were seen in the proportion of CD62L+, CD45RC and CD43+ T-cells (Table 2, Table S2). Notably, CD45RC was expressed by 70% of the CD8+ T-cells, indicating that only about 30% of them were memory T-cells.

image

Figure 2. Three color flow cytometry of mononuclear leukocytes isolated from the blood vessels of renal grafts. A gate was set on CD5+ leukocytes and the differential expression of CD8 versus CD25 (A) and CD62L (B) is depicted. Values are indicated as mean ± SD.

Download figure to PowerPoint

Table 2.  Expression of cell surface antigens by CD8α+/CD5+ T-cells isolated from the blood vessels of renal isografts and allografts on day 3 posttransplantation. Cell surface antigen expression was determined by flow cytometry and the data are expressed as mean ± SD
 α/βTCRCD25 (%)CD43 (%)CD45RC (%)CD62L (%)CD71 (%)CD134 (%)CD161 (%)
  1. Note: p-Values refer to differences between isografts and allografts.

  2. * n = 3 instead of n = 5, due to technical problems.

Monoclonal antibodyR73Ox39W3 of 13Ox22Ox85Ox26Ox4010/78
Isograft (n = 4)86.1 ± 3.96.5 ± 3.586.7 ± 6.163.1 ± 9.355.6 ± 7.99.9 ± 4.06.1 ± 1.950.5 ± 14.6
Allograft (n = 5)89.4 ± 3.9 p = 0.24817.6 ± 6.1 p = 0.04385.5 ± 11.2 p = 170.1 ± 13.2 p = 0.38651.8 ± 10.8 p = 0.56423.5 ± 5.7 p = 0.0438.8 ± 3.7* p = 0.28950.7 ± 5.8 p = 0.564

To directly demonstrate that CD45RC+ cells incorporated BrdU, immunohistochemistry was performed (Figure S2). Although the staining intensity was weak on tissue sections including spleens from healthy untreated rats, 36.4 ± 8.6% of all BrdU+ blood cells were labeled by antibodies to CD45RC.

T-cells interact with monocytes in allograft blood vessels

T-cells frequently interacted with monocytes in allograft blood vessels, which was seen on cytospin preparations (Figure 1E). Approximately 20% of the cell surface was involved in this cell contact and both, MHC class II antigens and T-cell receptors seemed to concentrate in this region.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

In this study, we demonstrate that the majority of the T-cells present in blood vessels of experimental renal allografts proliferate, as early as three days posttransplantation. Three independent experimental methods were applied to demonstrate T-cell proliferation. (i) BrdU pulse labeling was used, which is still a standard method for the detection of cell nuclei in the S phase of mitosis (20). As the S phase takes less than one third of the time needed to complete a mitotic cell cycle and approximately 20% of the T-cells in the blood vessels of the graft were BrdU+, most allograft blood T-cells seemed to proliferate. (ii) This assumption was corroborated by immunohistochemical detection of PCNA, another reliable method for the detection of proliferating cells (20). PCNA is the auxiliary protein of DNA polymerase-δ, which is directly involved in DNA replication, expressed during the S phase and shortly thereafter (20). Accordingly, the percentage of PCNA+ T-cells was approximately 30%. (iii) Finally, staining of cell nuclei of leukocytes obtained by vascular perfusion of renal allografts directly evidenced mitotic figures in T-cells.

Intravascular T-cell proliferation was essentially restricted to allografts as only a small proportion of T-cells proliferated in the blood vessels of other organs of allograft recipients. Similarly, T-cell proliferation was a rare event in isografts. From our data, however, we cannot conclude where T-cell activation takes place. Monocytes, which are abundant in allograft blood vessels, have the potential to present alloantigen to T-cells in a semi-direct fashion (13,14) and endothelial cells might directly present alloantigen (8–13). Both mechanisms would enable a stimulation of alloreactive CD8+ and CD4+ T-cells via MHC class I and class II antigens, respectively. We indeed observed interactions of monocytes with T-cells in the blood vessels of the graft resembling immunological synapses. The conclusion that graft monocytes function as antigen presenting cells and induce T-cell proliferation, however, is too ambitious as functional evidence is missing. Alternatively, recently activated T-cells might emigrate from secondary lymphoid organs or from the graft tissue itself, accumulate in the blood vessels of the allograft and accomplish at least a part of mitosis at this place.

Immunohistochemical double-staining revealed that most of the BrdU+ cells expressed both, the α- and the β-chain of CD8. However, a slightly lower proportion of BrDU+ cells expressed the β-chain of CD8, which may be due to a reduced sensitivity of the immunohistochemical staining in comparison to the α-chain. Cytotoxic CD8+ T-cells are of foremost interest in the context of acute organ rejection because they can directly destroy allogeneic cells. We do not know, however, if a minority of CD4+ T lymphocytes also incorporates BrdU, as CD4 can also be expressed by a subpopulation of rat monocytes (15,16). Because of their low abundance, we did not further characterize the CD4+/BrdU+ cell population.

Although the panel of antibodies directed to rat T-cell activation markers is limited, flow cytometry clearly revealed that most intravascular allograft CD8+ T-cells displayed a naïve or recently activated phenotype. Less than 25% of the allograft CD8+ T-cells coexpressed the activation marker CD25 or CD71 and no reduction in the expression of CD43 was observed in comparison to T-cells isolated from isografts. About 70% of all CD8+ T-cells expressed CD45RC, a marker, which is absent on central and effector rat memory T-cells (19). Furthermore, we directly evidenced BrdU incorporation in cell nuclei of CD45RC+ cells. Together with the fact that proliferation is observed as early as three days posttransplantation, these data suggest that naïve or just recently activated T-cells are involved.

This study has several limitations. It describes proliferation of CD8+ T-cells in the blood vessels of the allograft, the localization as well as the cellular and molecular mechanisms of T-cell activation, however, are unknown. We do not know if the absolute number of CD8+ T-cells proliferating in the graft exceeds the number CD8+ T-cells proliferating in secondary lymphoid tissues of allograft recipients. Although CD8+ T-cells are probably important effector cells involved in acute rejection, the role of intravascular CD8+ T-cells and of their local proliferation remain to be studied. Furthermore, we ignore the specificity of the T-cell receptors and the exact mechanism of leukocyte accumulation in allograft blood vessels. Finally, only a single time point after transplantation was examined. The major advantages of our experimental approach, however, are the use of wild-type rats with a fully functional immune system and that T-cell proliferation is investigated in the living animal.

In conclusion, this is the first report on proliferating nontransformed blood T lymphocytes. Our results challenge the common notion that blood mainly transports T-cells from lymphoid organs to peripheral tissues. Considering the functional importance of CD8+ T-cells, our data may also be relevant for the development of new therapeutic strategies preventing allograft rejection.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

This study was supported by a research grant from the University Medical Center Giessen and Marburg. The authors thank Thomas Herrmann (Würzburg, Germany) and Diethard Gemsa (Marburg, Germany) for valuable suggestions, Hartmut Dietrich, Kathrin Petri and Sandra Iffländer (all Giessen, Germany) for expert technical support.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Reference

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. Reference
  10. Supporting Information

Supplemental Materials & Methods: Flow cytometry, Immunocytochemistry

Table S1: Primary antibodies used in this study

Table S2: Expression of cell surface antigens by CD5+ T cells isolated from renal isografts and allografts on day 3 posttransplantation followed by density gradient centrifugation

Figure S1: Proliferation of CD8+ cells in the spleen of isograft (A) and allograft (B) recipients three days after kidney transplantation

Figure S2: Proliferation and expression of CD45RC of T-cells in blood vessels of rat renal grafts on day 3 after transplantation

FilenameFormatSizeDescription
AJT_3642_sm_SuppMat.doc1413KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.