| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print October 19, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Laboratory of Experimental Surgery, Department of General and Thoracic Surgery, Justus-Liebig-University Giessen, Giessen, Germany; and
Department of Clinical Chemistry and Molecular Diagnostics, Hospital of the Philipps-University, Marburg, Germany
2Correspondence: Laboratory of Experimental Surgery, Department of General and Thoracic Surgery, Justus-Liebig-University Giessen, Rudolf-Buchheim-Str. 7, D-35385 Giessen, Germany. E-mail: veronika.grau{at}chiru.med.uni-giessen.de
ABSTRACT
Alveolar macrophages (AMs) and peribronchial/perivascular macrophages are probably involved in lung allograft damage. We investigate leukocyte infiltration into graft tissue and address the question whether proliferation in situ contributes to macrophage homeostasis and accumulation. Lung transplantation was performed in the Lewis (LEW)-to-LEW and in the Dark Agouti-to-LEW rat strain combination. Graft infiltration by ED1+ and ED2+ (CD163) macrophages was analyzed by immunohistochemistry (IHC) and compared with infiltration by lymphocytes. Cells in the S-phase of the cell cycle were pulse-labeled with BrdU and detected immunohistochemically. Finally, the donor or recipient origin of AMs was determined by IHC and in situ hybridization. ED1+ AMs in allogeneic transplants increased by more than 25-fold from Days 1 to 5. In addition, large, peribronchial/perivascular infiltrates developed containing numerous ED1+ cells. Although AMs in normal rat lungs are CD163–, AMs up-regulated CD163 between Days 4 and 5, reaching maximum values on Day 6. Lymphocytes were less numerous than macrophages. About 16% of the AMs and 10% of the peribronchial/perivascular macrophages were in the S-phase of the cell cycle on Day 2 post-transplantation. No differences in the frequency of BrdU+ macrophages were obvious between isografts and allografts. AMs of donor origin increased in number considerably during allograft rejection. In conclusion, the cellular infiltrate in lung allografts is dominated by macrophages, which exhibit an unusual phenotype and a strong capacity for mitotic self-renewal.
Key Words: lung transplantation • alveolar macrophage • acute rejection • CD163
INTRODUCTION
The clinical outcome of lung transplantation is poor compared with other vascularized organs. Five years after lung transplantation, 
45% of the graft recipients are alive [1
], whereas 70% of the kidney transplant recipients are living with functioning grafts. Primary organ failure, infections, acute rejection, and the early onset of chronic rejection are the most important causes of mortality after lung transplantation [2
].
At least two populations of macrophages can be differentiated in the lung, alveolar macrophages (AMs) and peribronchial/perivascular macrophages, which reside in the connective tissue. In the lung of healthy rats, AMs are ED1+/ED2–, whereas peribronchial/perivascular lung macrophages are ED1–/ED2+ [3 ]. The mAb ED1 binds to a CD68-like lysosomal membrane antigen. ED2 recognizes the hemoglobin scavenger receptor CD163 [4 ]. As AMs can be easily isolated by bronchoalveolar lavage (BAL), this group of cells has been investigated thoroughly [5 ]. AMs are highly specialized cells restricted to the lung, potentially involved in mechanisms leading to the extraordinary fragility of lung transplants. They are indispensable effector cells in the innate defense against airborne pathogens. Although large amounts of nonpathogenic and pathogenic particles are inhaled and cleared by AMs continuously, AMs are poor APC and seem to down-regulate acquired immune responses in the healthy lung [6 ]. Experimental data underscore the relevance of AMs in lung transplants: AMs are involved in reperfusion injury [7 ], which is unavoidable in transplants. During acute rejection of lung allografts, AMs are activated; they strongly increase in number and produce proinflammatory mediators [8 9 10 11 12 13 ]. When allogeneic AMs are instilled into mouse lungs, together with dendritic cells (DC), histopathological changes resembling acute rejection develop; this does not happen when DC are applied alone [14 ]. AMs are involved in the pathogenesis of acute lung injury after allogeneic stem cell transplantation (for review, see ref. [15 ]). Compared with AMs, peribronchial/perivascular macrophages are more effective APC [5 ]. Their role in graft rejection, however, is largely unknown.
When aiming at a therapy affecting graft macrophages, the mechanisms involved in their renewal and accumulation should be considered in more detail. There is convincing evidence that under steady-state conditions as well as during inflammation, blood monocytes migrate into the alveoli [5 ]. Increased recruitment of blood monocytes into the lung and prolonged survival or impaired clearance from the respiratory tract are generally thought to be the most important mechanisms to increase AM numbers under pathological situations [16 ]. However, in addition, there are some careful experimental studies, indicating that the renewal of AMs may be accounted for by local proliferation in the healthy lung [17 , 18 ]. In these studies, Bitterman et al. [19 ] have shown that under normal conditions, 0.5% of the AMs obtained by BAL from humans incorporate 3H-thymidine. During inflammation, 3H-thymidine incorporation was seen in up to 6% of the AMs [19 20 21 ]. Nakata et al. [22 ], who analyzed proliferating cells by detection of the proliferating cell nuclear antigen after allogeneic bone marrow transplantation in humans, also obtained comparable results. Taken together, these studies indicate that recruitment of blood monocytes to the alveolar space and proliferation of AMs in situ are not mutually exclusive and probably occur concomitantly in an intact organism. Up to now, proliferation of peribronchial/perivascular macrophages and AMs in lung transplant has not been investigated.
The purpose of this study was to give a quantitative estimation of graft infiltration by macrophages in comparison with graft lymphocytes during the time course of acute rejection. We used the fully allogeneic Dark Agouti (DA)-to-Lewis (LEW) rat strain combination without immunosuppression to study unmodified fulminant rejection of lung transplants. Our immunohistochemical approach excluded that only a subpopulation of intra-alveolar cells harvested by BAL is analyzed. Furthermore, macrophage proliferation is investigated in situ in double-staining experiments after pulse-labeling with BrdU, and the donor or recipient origin of AMs was determined.
MATERIALS AND METHODS
Animals
LEW(RT1l) and DA(RT1av1) rats were purchased from Harlan Winkelmann (Borchen, Germany) or Elevage Janvier (Le Genest St. Isle, France) and kept under conventional conditions. Transplantation was performed in rats weighing 220–280 g. Animal care and animal experiments were performed following the current version of the German Law on the Protection of Animals as well as the National Institutes of Health "Principles of Laboratory Animal Care".
Lung transplantation
Animals were anaesthetized by short-time inhalation of isofluran (Forene, Abbot, Wiesbaden, Germany) followed by i.p. application of 90 mg/kg ketamine hydrochloride (Ketavet, Pfizer, Karlsruhe, Germany) and 0.1 mg/kg medetomidine hydrochloride (Domitor, Pfizer). In addition, they received 1000 IU heparin (Ratiopharm, Ulm, Germany) per kg body weight i.v. and 0.25 mg/kg atropin sulfate (Braun Melsungen, Melsungen, Germany) i.m. After orotracheal intubation, the lungs were ventilated mechanically with room air at a tidal volume of 2.5 ml, 100 breaths per min, and a positive end expiratory pressure of 5 cm H2O (Harvard rodent ventilator, South Natick, MA).
Lung transplantations using the cuff technique for the vascular anastomosis and sutures for the bronchial anastomosis were essentially performed as described [23 ]. The cuffs were made from 16 G Abbocath catheters (Abbott, Sligo, Ireland). For the bronchial sutures, we used Ethilon 9-0 (Ethicon, Norderstedt, Germany). After transplantation, anesthesia was antagonized with 0.5 mg/kg atipamezol hydrochloride (Antisedan, Pfizer), and the rats received one prophylactic dose of 150 mg ampicillin (Ratiopharm) i.p. No immunosuppression was applied.
For most experiments, male rats were used. To determine the origin of AMs by in situ hybridization (ISH), lungs from female donors were transplanted to male recipients.
Antibodies
mAb 10/78, ED1, ED2, OX6, OX33, OX76, and R73 were purchased from Serotec (Düsseldorf, Germany). The anti-BrdU mAb Bu20a was obtained from Dako (Hamburg, Germany), and mAb RP-1 was provided by BD PharMingen (San Diego, CA). DAKO supplied the peroxidase-conjugated rabbit antimouse Igs, the mouse EnVision peroxidase system, and the antimouse/rabbit EnVision alkaline phosphatase. Antimouse/rabbit EnVision is a reagent that binds to mouse and rabbit antibodies.
Paraffin histology
For the analysis of transplant infiltration by macrophages and lymphocytes, graft recipients were killed at intervals of 24 h from Days 1 to 6 after transplantation. The right native lung and the graft were cut in slices of 5 mm and fixed by immersion in 2% paraformaldehyde, 10 mM NaJO4, and 75 mM L-lysine [24
].
To study BrdU incorporation, the lungs were fixed by instillation of 4% paraformaldehyde, solubilized in 0.1 M phosphate buffer, pH 7.2, via the trachea with a hydrostatic pressure of 20 cm water column for 20 min, followed by 24 h immersion in the same fixative. Thereafter, the lungs were cut in slices of 5 mm and embedded for paraffin histology.
Histological sections of 6–8 µm, mounted on silane-coated glass slides, were dewaxed, rehydrated, and stained with hemalum eosin or used for immunohistology.
Immunohistochemistry (IHC)
Depending on the primary antibody, different pretreatment procedures were necessary. Detection of B cells with mAb OX33 was done without prior antigen retrieval. For staining with mAb ED2 and R73, slides were treated with 0.5 mg/ml Protease Type XIV (Sigma-Aldrich, Taufkirchen, Germany) in 50 mM Tris-HCl buffer, pH 7.6, 0.9% NaCl for 15 min at room temperature. The ED1 antigen was visualized after antigen retrieval in 0.01 M sodium citrate buffer, pH 6.0, for 15 min at 120°C and 1.1 bar. For mAb 10/78 and RP-1, all pretreatment protocols were tested, but in accordance with the data sheets of the suppliers, these antibodies did not stain paraffin sections.
Endogenous peroxidase activity was blocked with 1% H2O2 in PBS for 30 min. After washing in PBS, pH 7.2, the sections were incubated for 30 min with PBS, pH 7.2, 1% BSA (Serva, Heidelberg, Germany), 0.1% NaN3 (p.a., Merck, Darmstadt, Germany), followed by overnight incubation with an appropriate dilution of primary antibody diluted in the same solution at 4°C. On control sections, the primary antibody was omitted. Bound, primary antibodies were detected using the mouse EnVision peroxidase system containing 5% heat-inactivated normal rat serum and 3,3'-diaminobenzidine (DAB; Sigma-Aldrich). Selected sections were counterstained slightly with hemalum.
The origin of allograft AMs was determined on Days 3 and 4 post-transplantation. In a first step, protease-treated sections of lung allografts were incubated with control solution, mAb OX6, or mAb OX76. Bound antibodies were detected in blue using the antimouse/rabbit EnVision alkaline phosphatase (DAKO), containing 5% normal rat serum and the chromogen Fast Blue. In a second step, antigen retrieval with citrate buffer was performed at 120°C and 1.1 bar. This treatment completely removed the antibodies to MHC Class II as well as the detection system. Macrophages were detected in red with mAb ED1 and antimouse/rabbit EnVision alkaline phosphatase and Fast Red. Again, controls omitting mAb ED1 were included as negative controls. The sections were coverslipped in glycergel (DAKO).
Detection of proliferating macrophages
Cell proliferation was studied in the right and the left lung of isograft (n=4) and allograft (n=4) recipients as well as in untreated weight- and sex-matched LEW rats (n=4). To label cell nuclei in the S-phase of the cell cycle, 25 mg BrdU in 1 ml 0.9% saline was injected i.v. on Day 2 post-transplantation, 30 min before sacrificing the animals.
Dewaxed sections of paraffin-embedded lung tissue were treated with protease Type XIV, followed by 1% H2O2, and incubated with mAb ED1 or ED2. For antibody detection, antimouse/rabbit EnVision alkaline phosphatase (DAKO) containing 5% normal rat serum was combined with the chromogen Fast Blue. The DNA was subsequently denatured in 0.1 M HCl and 0.9% NaCl at 60°C for 10 min. This procedure also removes bound mAb and EnVision but not the Fast Blue staining. The slides were washed in PBS and incubated with the anti-BrdU mAb followed by rabbit antimouse Ig conjugated to peroxidase in the presence of 5% normal rat serum. Finally, bound peroxidase was visualized with DAB. Sections were coverslipped in glycergel (DAKO). As a control, the staining procedure was performed in the absence of mAb to macrophages and to BrdU. In addition, each primary antibody was used alone.
ISH
Female LEW and DA rats were used as donors for lung transplantation to male LEW rats. Graft-infiltrating AMs were detected on Day 4 post-transplantation by ISH using a probe for a repetitive rat Y chromosome-specific sequence und consecutive IHC with mAb ED1. The right native lungs of graft recipients were included in each experiment as a positive control and sections of an untreated female lung as a negative control.
The plasmid "9.IES 8, y-probe" containing a rat Y chromosome-specific DNA sequence was kindly provided by Hoebee and co-workers [25 ]. Competent One Shot TOP 10F' Escherichia coli cells (Gibco-Invitrogen, Karlsruhe, Germany) were transfected at 109 cells/50 µl with 170 ng plasmid DNA using the heat-shock transfection protocol provided by the manufacturer. Cells were grown overnight on ampicillin (Roche, Heidelberg, Germany)-containing agar plates, and plasmid-positive clones were further grown in Luria-Bertani medium (Gibco-Invitrogen) with 50 ng/ml ampicillin. Plasmid DNA was prepared using the Qiagen plasmid maxi kit (Qiagen, Hilden, Germany), and the insert size was checked by EcoRI (Roche) restriction digestion with subsequent agarose gel analysis. The probe was labeled with digoxigenin using the DIG-Nick Translation kit (Roche).
Lung sections were dewaxed and pretreated at 37°C with 100 µg/ml RNase (Sigma No. R5503) in 2x SSC for 1 h, 0.2 n HCl for 1 min, and 2 mg/ml pepsin (Sigma No. P7012) in 0.01 n HCl for 20 min. Thereafter, the sections were rinsed with PBS and fixed for 10 min with 1% freshly dissolved parafomaldehyde in PBS. After equilibration in 2x SSC, prehybridization was performed for 1.5 h at 37°C using the hybridization mixture without probe. The hybridization mixture consisted of 50% deionized formamide (Sigma No. F9037), 10% dextran sulfate (Sigma No. D8906), 0.33 mg/ml salmon testes DNA (Sigma No. D1626), and 0.11 mM EDTA in 2x SSC. After prehybridization, the sections were rinsed with 2x SSC, transferred to water, dehydrated in 70%, 90%, and 100% ethanol, and air-dried. Hybridization mixture containing 1 ng/µl digoxigenin-labeled probe was applied to the sections, denatured at 90°C for 10 min, and hybridized at 37°C for 3 days. Sections were washed for 5 min in 2x SSC at 40°C, 15 min in 50% formamide in 2x SSC at 44°C, 15 min in 0.1x SSC at 40°C, and 3 min in 4x SSC containing 0.1% Tween 20 at ambient temperature. Thereafter, the slides were transferred to PBS, and digoxigenin was detected by IHC using sheep antidigoxigenin FAB fragments coupled to alkaline phosphatase and NBT/5-bromo-4-chloro-3-indolyl-phosphate as chromogen. In a last step, AMs were stained by IHC with mAb ED1 using the EnVision alkaline phosphatase system and Fast Red as described.
Quantification and statistics
Sections were evaluated with an Olympus BX51 microscope and the analySIS software. The percentage of the immunopositive area on sections stained with mAb ED1, ED2, OX33, and R73 was determined in three randomly selected fields of 0.58 mm2 each. Only fields of the lung parenchyma were evaluated, which include leukocytes infiltrating the alveolar space and the interstitium of alveolar septa as well as leukocytes localized in capillaries. Regions containing bronchi, bronchioles, arteries, or veins were excluded. To determine the origin of AMs on sections stained by IHC or ISH, at least 50 ED1+ with the morphology and localization typical for AMs were counted per specimen. All data are expressed as mean ± SD.
The percentage of proliferating macrophages was determined by counting at least 100 macrophages on at least two different fields per lung. Data were analyzed with nonparametric Mann-Whitney rank sum test using SPSS software (SPSS Software, Munich, Germany), and P < 0.05 was set as the level for significance.
RESULTS
Phases of acute lung allograft rejection
Orthotopic left lung transplantation was performed in the fully allogeneic DA-to-LEW (n=3 per day) rat strain combination and in the isogeneic LEW-to-LEW (n=1 per day) combination. Fixation was performed by immersion of lung slices in paraformaldehyde-lysine-periodate. This procedure led to good preservation of leukocyte cell surface antigens but to poor histomorphology from the esthetic point of view. We chose this technique, as in contrast to fixation by instillation or perfusion, AMs and intravascular leukocytes are not displaced. The time course of the histopathological changes during rejection of DA-to-LEW lung allografts has not been reported before but resembled the Brown Norway-to-LEW rat strain combination described previously [26
]. However, lung allografts were destroyed more quickly in the DA-to-LEW combination.
We subdivided the process of allograft rejection into three consecutive phases. During Phase I from Day 1 to Day 2 post-transplantation, similar changes were seen in isografts and allografts (Fig. 1A and 1B ). The morphological changes were moderate and probably a result of reimplantation responses. Perivascular edema was obvious, together with mild intra-alveolar edema. Minimal perivascular, peribronchial, and alveolar leukocyte accumulations were seen. Phase II allografts on Days 3 and 4 were characterized by an increase in mononuclear infiltrates in perivascular and peribronchial regions (Fig. 1C) . The alveolar septa were thickened and contained leukocytes. Numerous intra-alveolar leukocytes were seen. In the lumina of blood vessels, mononuclear leukocytes accumulated. The respiratory epithelium displayed goblet cell hyperplasia. During Phases I and II, only small numbers of granulocytes were seen. Phase III allografts on Days 5 and 6 exhibited severe, irreversible damages (Fig. 1E) : The lungs became atelectatic on Day 5, and the alveoli were filled with inflammatory cells and fibrillar material. Alveolar septa were thickened, and dense perivascular and peribonchial infiltrates, mainly consisting of mononuclear leukocytes, were obvious. Although the infiltrate was predominantly composed of mononuclear cells, granulocytes were detected more frequently during Phase III of allograft rejection compared with Phases I and II. Day 6 allografts exhibited patches of necrotic lesions and extensive hemorrhage. In contrast, isograft-infiltrating leukocytes only increased moderately in number until Day 3, and the moderate changes nearly completely resolved in Days 5 and 6 isografts (Fig. 1D and 1F) . No substantial, pathological findings were seen in native right lungs of isograft and allograft recipients (data not shown).
|
|
|
16% of the area was stained with mAb ED1 (Fig. 3A)
. Only a mild increase in the number of AMs was seen in isografts on Days 2 and 3 post-transplantation, which resolved thereafter (Figs. 2
and 3A)
. As expected, no substantial changes were seen in the number of ED1+ macrophages in the right native lung of isograft and allograft recipients (Fig. 3A)
. In the perivascular and peribronchial regions of Phase I allografts, of all isografts, and of right native lungs, almost no cells were ED1+. However, peribronchial/perivascular macrophages in Phase II allografts faintly expressed the ED1 antigen, and a strong expression was obvious in Phase III allografts (Fig. 2) .
ED2+ macrophages in lung transplants
Rat CD163 is recognized by mAb ED2 and is normally expressed by peribronchial/perivascular lung macrophages but in striking contrast to ED1, not by AMs. In a healthy lung, some ED2+ cells are located in the connective tissue surrounding blood vessels and airways. Only few ED2+ cells are present in the alveolar walls [3
, 4
, 27
].
During Days 1–3 (Phase I and early Phase II) of allograft rejection, ED2+ cells were increased moderately in number compared with healthy lungs. They were mainly located in perivascular and peribronchial regions, whereas AMs were still ED2– (Figs. 2 and 3B) . However, on Day 4 (late Phase II), AMs started to become ED2+ (Fig. 2) . Phase III (Days 5 and 6) of allograft rejection was characterized by a strong and pronounced expression of CD163, not only by peribronchial/perivascular macrophages but also by interstitial macrophages of the alveolar wall as well as by AMs (Figs. 2 and 3B) . CD163 expression in isografts and right native lungs closely resembled healthy lungs (Fig. 2) .
Lymphocytes in lung transplants
T cells expressing the 
/ß-TCR were found in small numbers in all compartments of a healthy rat lung. The time course and the localization of allograft infiltration by T lymphocytes resembled that of ED1+ macrophages (Figs. 2
and 3)
. However, three differences were obvious in comparison with ED1+ macrophages (Fig. 3)
: Allograft infiltration by T cells was delayed; whereas T cell infiltration started on Day 3, increased amounts of ED1+ macrophages were detectable on Day 2 after transplantation. T cells were less numerous; on Day 5 after surgery, a maximum of 3% of a histological section of the lung parenchyma was positive for the /ß-TCR. On Day 6, when allografts were irreversibly destroyed, T cell numbers decreased in comparison with Day 5.
In isografts, no conspicuous T cell infiltrates were seen. However, in the right native lung of allograft recipients, the number of T cells increased on Day 5 after transplantation but decreased to levels comparable with healthy lungs on Day 6 (Fig. 3C) . In contrast, T cells did not infiltrate the native lung of isograft recipients.
In comparison with macrophages and T lymphocytes, only a small number of B lymphocytes entered allogeneic lung transplants. Allograft infiltration by B cells peaked on Day 3 post-transplantation when 0.3% of the lung parenchyma was stained with mAb OX33 (data not shown). In isografts and right native lungs of graft recipients, no B cell infiltrates were obvious.
Proliferation of macrophages
Macrophage proliferation was investigated on Day 2 post-transplantation, as the population of AMs increased by about threefold between Days 2 and 3 (Fig. 3A)
. We studied proliferation of ED1+ and ED2+ macrophages in isogeneic (n=4) and allogeneic (n=4) grafts as well as in untreated LEW rats (n=4) after incorporation of BrdU in vivo for 30 min. Immunohistochemical double-staining experiments were performed using ED1 or ED2 in combination with mAb to BrdU. Only ED1+ cells with typical morphology and intra-alveolar localization compatible with AMs were counted. On sections stained with mAb ED2, only macrophages in the perivascular/peribronchial tissue were analyzed.
In transplanted lungs, 16% of the ED1+ AMs incorporated BrdU, indicating that they were in the process of DNA synthesis (Figs. 4A
and 5A
). No differences in the percentage of BrdU+ AMs were obvious between isografts (16.6±2.6%) and allografts (16.0±6.4%). The number of proliferating AMs in isografts and allografts was significantly (P<0.05) higher in comparison with left lungs from untreated controls (5.9±0.9%). The percentage of proliferating cells in the right native lung of isograft recipients (8.2±4.4%) and allograft recipients (7.7±4.2%) tended to be increased in comparison with control lungs (Fig. 5A)
. However, this difference was not statistically significant.
|
|
Origin of AMs in lung transplants
First, we determined the proportion of AMs in pulmonary allografts expressing MHC Class II antigens in double-staining experiments with mAb OX6. This antibody binds to monomorphic regions present on MHC Class II antigens of donor and recipient origin. On Day 3 (n=3), 20 ± 5.3% of the AMs were MHC Class II+ and 27 ± 2.9% on Day 4 (n=3). Subsequently, double-staining experiments using mAb ED1 and mAb OX76 were performed (Fig. 6A
). OX76 is directed to a polymorphic region of the MHC Class II antigen present in the donor (DA) but not in the recipient (LEW) rat strain. The proportion of donor cells among the MHC Class II+ AMs was calculated: 39 ± 8.7% and 24 ± 4.0% of these cells originated from the donor on Days 3 and 4, respectively. In the right native lungs of transplant recipients, which were included as negative controls, no OX76+ cells were detected.
|
50% of the sections of AMs. Assuming that 100% of the AMs in the right lung were male cells, the proportion of male recipient AMs in the pulmonary grafts of female origin was calculated: 30 ± 16% male AMs and 44 ± 12% male AMs were present in isografts and allografts, respectively. DISCUSSION
Our study led to the following main results: During all phases of acute lung transplant rejection, macrophages are more numerous in the lung parenchyma compared with lymphocytes; the expression of ED2/CD163 is induced on AMs during graft rejection; graft AMs and peribronchial/perivascular macrophages strongly proliferate in situ; and AMs of donor and recipient origin increase in number during allograft rejection.
We analyzed the time course of graft infiltration by ED1+ macrophages, ED2+ macrophages, and lymphocytes by densitometry. Macrophages were the dominating cell type during all phases of rejection. Similar data were obtained in DA-to-LEW kidney transplants [28
]. In contrast, after lung transplantation in the Brown Norway-to-LEW rat strain combination, BAL fluid contained more than 60% T lymphocytes and only 20% AMs [9
]. This discrepancy may be a result of differences between rat strain combinations. In addition, technical reasons might be responsible: Only the subpopulation of intra-alveolar leukocytes could be investigated when analyzing BAL fluids. Furthermore, it cannot be taken for granted, especially under pathologic conditions, that BAL fluids contain a representative sample of the entire intra-alveolar compartment. Certain leukocyte populations might be enriched during BAL, whereas other populations remain inside the alveoli. To exclude these limitations, we investigated infiltration of the lung parenchyma in situ by IHC, followed by densitometrical analysis of the immunopositive area in histological sections. By this approach, interstitial, intravascular, as well as intra-alveolar leukocytes were included in our analysis. As alveolar macrophages are larger than lymphocytes, they might be overestimated when measuring the immunopositve surface area. Therefore, we also counted lymphocytes and ED1+ macrophages in randomly selected areas. Based on the cell number per area, the macrophage/T cell ratio was at least 2.5 at all points in time (data not shown).
AMs did not only increase in number in allografts, they additionally started expressing ED2/CD163 on Day 4 post-transplantation, which is uncommon in rat AMs. CD163, a regulated hemoglobin scavenger receptor, has been shown to be involved in the down-modulation of inflammatory responses [29 ]. Up-regulation of CD163 expression by AMs can be interpreted as a part of a negative feedback loop, limiting immune responses. However, in our experimental model of vigorous acute rejection, this anti-inflammatory response may not influence graft outcome, as maximal ED2 staining was seen on Days 5 and 6 when grafts were already irreversibly destroyed.
Macrophage proliferation was investigated on Day 2 post-transplantation. At this point in time, 16% of the AMs in lung transplants incorporated BrdU during pulse-labeling, clearly indicating that they were in the S-phase of the cell cycle. Pulse-labeling with BrdU is the standard method for the assessment of cell proliferation [30
]. Similar labeling indices of AMs were found in isografts and allografts, indicating that AM proliferation is triggered by transplantation and not by rejection. The proliferation of AMs in situ together with additional mechanisms, such as increased recruitment of blood monocytes and reduced elimination from the lung, are likely to be responsible for the observed accumulation of AMs in lung allografts. Much to our surprise, 5% of the AMs in untreated control lungs in healthy rats also incorporated BrdU. Histopathological evaluation of these lungs did not reveal any kind of inflammation.
In contrast to AMs, peribronchial/perivascular ED2+ macrophages neither proliferated in control lungs or in the right native lung of graft recipients. However, similar to AMs in isografts and allografts, a large proportion of proliferating ED2+ perivascular/peribronchial macrophages was also seen. This indicates that similar to renal allografts [28 ], proliferation in situ is an important mechanism for the development of macrophage infiltrates.
Intra-alveolar proliferation of AMs in healthy and damaged lungs is discussed controversially in the literature. Some authors claim that AMs have almost totally lost their proliferative potential [31 ], whereas other publications demonstrate that the renewal of AMs may only be accounted for by local proliferation [17 , 32 , 33 ]. Our data are in line with the assumption that local proliferation contributes considerably to the homeostasis of AMs in healthy and inflamed lungs [19 , 20 , 22 ]. However, the proliferation rates published so far for normal and pathological lungs are clearly below the amount of proliferation found in our experimental model. It cannot be excluded that lung transplantation produces a unique situation leading to extraordinarily high AM proliferation rates. It is also possible that for different reasons, our technique is highly sensitive when compared with other techniques: High concentrations of BrdU are applied i.v. for pulse-labeling; in contrast, i.p. application might result in low local doses of 3H-thymidine in the lung and therefore, limit the detection of proliferating AMs [20 ]. Cell nuclei in the S-phase of the cell cycle are labeled in the living animal, whereas in other studies, BAL cells were collected and analyzed in an in vitro culture system [21 , 22 , 31 ]. In our experimental model, S-phase AMs are identified by double-staining with mAb ED1 and anti-BrdU, allowing the unequivocal identification of AMs.
After renal transplantation in the same rat strain combination, we were also able to detect proliferating ED1+ and ED2+ tissue macrophages [28 ]. In contrast to our results in the lung transplant model, proliferation rates are lower in renal isografts compared with allografts. Therefore, graft rejection seems to provide the main stimulus, which drives macrophage proliferation in the kidney. However, proliferation in renal grafts was studied on Day 3. At this point in time, severe mononuclear infiltrates are present in renal allografts but not in isografts, whereas Day 2 lung isografts and allografts are only mildly infiltrated.
Our hypothesis that proliferation in situ significantly contributes to the population of AMs in pulmonary grafts is corroborated by the high proportion of AMs of donor origin as determined by IHC and ISH on Day 4 post-transplantation. Furthermore, it can be concluded that resident alveolar macrophages, which entered the healthy donor lung before transplantation, are able to proliferate after transplantation. IHC, using strain-specific antibodies to recipient MHC Class II, revealed 24% AMs of donor origin in Day 4 allografts, whereas ISH for recipient Y chromosomes revealed 66% donor macrophages. Taking into account that the ED1+ area increases by 20-fold from Days 1 to 4, donor-derived macrophages must have proliferated in the graft. However, less donor AMs are detected by IHC compared with ISH. This is probably a result of the fact that only MHC Class II+ macrophages are detected by IHC, whereas ISH is not restricted to a subpopulation of donor AMs. In a previous study on kidneys transplanted in the same rat strain combination, we detected 50% MHC Class II+ macrophages of donor origin in the infiltrate [28
]. In contrast to AMs, most kidney macrophages were MHC Class II+ on Day 4 after allogeneic transplantation [28
].
In conclusion, macrophages in transplanted lungs display unexpected properties. AMs express CD163 during the effector phase of acute rejection, which is characteristic of macrophages involved in anti-inflammatory responses. Proliferation in situ contributes considerably to the renewal of AMs and peribronchial/perivascular macrophages. Strategies aiming at a limitation of AM accumulation in lung transplants should inhibit extravasation of blood monocytes as well as local AM replication.
ACKNOWLEDGEMENTS
This study was supported by the Deutsche Forschungsgemeinschaft (FE 287/6-1). The authors thank Claudia Cybon and Sandra Iffländer, who cared for the experimental animals, as well as Anja Seiler and Renate Plass for excellent technical assistance. We are grateful to Jürg Hamacher for patiently teaching the technique of lung transplantation and to Wolfgang Kummer for statistical advice. We thank Barbara Hoebee for kindly providing the Y chromosome-specific probe.
FOOTNOTES
1 These authors contributed equally to this work. 
Received June 7, 2006; revised September 4, 2006; accepted September 18, 2006.
REFERENCES
production, lymphocytic bronchitis, and vasculitis in recipient murine lungs J. Leukoc. Biol. 64,578-586[Abstract], T helper 1/T helper 2 cytokines, IgG subclasses, and pathology in a rat model of lung transplantation J. Immunol. 159,4084-4093[Abstract]This article has been cited by other articles:
|
|
 |
|
  J. Sadovski, T. Kuchenbuch, C. Ruppert, A. Fehrenbach, M. Hirschburger, W. Padberg, A. Gunther, J. M. Hohlfeld, H. Fehrenbach, and V. Grau Keratinocyte growth factor prevents intra-alveolar oedema in experimental lung isografts Eur. Respir. J., January 1, 2008; 31(1): 21 - 28. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
