Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Bronchial and pulmonary innervation are interrupted at lung transplantation. Post-transplantation degeneration of nerve fibers with cell bodies above the suture, such as the sensory C fibers, is expected, which may affect the response to nerve stimulation. Indeed, we have previously reported bronchial hyperresponsiveness to nerve stimulation after lung transplantation in the rat (1). Among the possible causes, an increased acetylcholine release may be due either to hyperexcitability of the postganglionic parasympathetic nerves or to loss of an inhibitory mechanism when neural connections with the central nervous system are interrupted. We have therefore studied the sensory nerves after transplantation through immunoreactivity (IR) of the sensory neuropeptide calcitonin gene-related peptide (CGRP) (2).

There is evidence of sympathetic reinnervation after human heart transplants (3) and rat intestinal transplants (4). The sensory neuropeptides CGRP and substance P have been hypothesized to participate in the trophic, regulatory, and reparative processes after motoneurone injury (5, 6). However, cardiac vagal afferent nerves do not appear to regenerate (7). The sensory C fibers are closely associated with mast cells in various tissues (8, 9), including the lung (10, 11). We have previously observed an increased number of mast cells in the lung after transplantation (12). Because mast cells synthesize, store, and release nerve growth factor (NGF) (13), which is involved in the growth and differentiation of the sensory neuropeptidergic fibers, we hypothesized this might help sensory nerve regeneration to occur in the lung after transplantation. We have therefore followed the occurrence of CGRP-IR in the lung at 1 and 5 mo after transplantation.

CGRP is also found, together with other neuropeptides, in neuroepithelial cells (14), either isolated or grouped, in the bronchial epithelium of mammalian lungs (15). The number and distribution of the neuroepithelial cells may be altered during denervation and during the post-transplantation remodeling processes. Therefore, we also studied the number and location of these cells after transplantation.

Our study of the occurrence of CGRP-IR was performed in the lung of Lewis rats 1 and 5 mo after syngeneic transplantation. We measured the length of immunoreactive C fibers and the number of CGRP-IR neuroepithelial cells. We compared these findings with those in control lungs at the same post-transplantation periods to assess possible reinnervation. These results are discussed in light of our previously published results about increased numbers of mast cells after transplantation (12).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Thirty male syngeneic Lewis rats weighing 300 g (Janvier, Le Jennest, France) were used. The left lungs from 10 donor rats were transplanted into ten recipient rats, as previously described. Briefly, the rats were anesthetized, intubated, and ventilated; a left thoracotomy was performed in the fifth intercostal space; the left hilus was dissected, the left lung excised, and the donor left lung implanted in the recipient rat by anastomosis of the pulmonary vein, bronchus, and artery (1, 12, 16). Lungs were isolated less than 40 min. Ten control rats did not undergo surgery.

The control and transplant rats underwent the following procedures, five in each group at 1 mo after transplantation and the rest at 5 mo after transplantation: they were killed by a blow to the head, their hearts and lungs were removed en bloc, and the left lungs were dissected out and prepared for immunohistochemistry.

Immunohistochemistry

Samples were immersed in a solution containing 4% paraformaldehyde (Merck, Darmstadt, Germany) and 0.3% picric acid (Merck) in phosphate-buffered saline (PBS) (0.1 M, pH 7.4) at 4°C for 3 h. They were then rinsed in a solution, changed daily, of 0.1 M PBS, pH 7.4, for 3 d at 4°C.

Samples were embedded in Tissue Tek (Miles, Epernon, France), frozen in deep-cooled isopentane, and maintained at 80°C. Serial 8-µm cryostat sections were performed on three different levels of 12 sections each, separated by five sections.

One section per level (three sections per rat) was pretreated in PBS containing 2% bovine serum albumin (wt/ vol; BSA fraction V; Boehringer-Mannheim, Bagnolet, France) and 0.3% Triton X100 (wt/vol; Sigma, L'Isle d'Abeau Chesnes, France) (PBS-BSA-Triton) for 30 min at room temperature. Sections were incubated overnight at 4°C with a rabbit polyclonal antiserum to rat CGRP (anti- CGRP; Amersham, Les Ulis, France) diluted 1:1,000 in PBS-BSA-Triton. They were washed twice with PBS-BSA-Triton and incubated in fluorescein-conjugated swine antibody to rabbit immunoglobulin G (Dako, Glostrup, Denmark), diluted 1:100 for 1 h at 4°C. After two washings in PBS-BSA, slides were immediately mounted with Mowiol 4-88 (Calbiochem, Meudon, France).

The specificity of the staining was controlled by: (1) omission of anti-CGRP antiserum and (2) incubation with antiserum pretreated with CGRP antigen (1:1,000 anti-CGRP and 1 µM CGRP).

Quantitative Analysis

Working with a fluorescent microscope equipped with a filter for green fluorescent light and fitted with a camera, we took five color photomicrographs from each of five areas selected at random from each section: one of large bronchi (diameter 500 to 800 µm), one of small bronchioles (250 to 350 µm), one of peripheral bronchioles (50 to 100 µm), one of the large pulmonary artery (500 to 700 µm), and the last of pulmonary arterioles (50 to 150 µm). Each slide was projected on a screen; fluorescent fibers (CGRP-IR fibers) and cells (CGRP-IR cells), together with the contiguous histologic structures, were drawn onto a transparency. A computerized morphometer (Reichert-Jung, Kontron, München, Germany) measured the length of the fibers and of the adjacent structures (basal membrane for bronchus; tunica intima for vessel). To standardize the measurements, the fiber length was always expressed by its ratio to the length of the adjacent structure.

CGRP-IR cells, isolated or grouped, were counted on three sections each of five rats. Two independent observers, one of whom was unaware of the study protocol, counted cells.

Expression of Results and Statistical Analysis

Quantitative distribution of CGRP-IR fibers was expressed by the ratio of fiber length to structure length, and results were expressed as means ± standard error of the mean. The total number of CGRP-IR cells, isolated and grouped, is reported. Statistical analysis, using a Wilcoxon sum-rank test (Statview II; Abacus Concepts, Berkeley, CA), compared observations for the transplanted lung with those for the lungs of control rats.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Control Groups

CGRP-immunoreactive fibers. Thin CGRP-IR fibers with bead varicosities were localized beneath the basal membrane of the bronchial epithelium and in the submucosa and the bronchial smooth muscle of large and small airways (Figure 1). They were also observed around the pulmonary vessels and within the pulmonary vascular smooth muscle (Figure 2). The number of CGRP-IR fibers gradually decreased as the airways and vessels became smaller (Figure 3). No immunofluorescent fibers were observed in the periphery (that is, airways with a diameter less than 30 µm), alveolar septa, or pleura.

View larger version (74K): [in this window] [in a new window]   Figure 1.   CGRP-IR fibers (arrows) within and around large bronchi (a, d, and g), bronchioles (b, e, and h), and peripheral bronchioles (c, f, and i) of lung from control rats (a, b, and c) and from rats 1 (d, e, and f  ) and 5 (g, h, and i) mo after lung transplantation. Scale bar = 50 µm.

View larger version (49K): [in this window] [in a new window]   Figure 2.   CGRP-IR fibers (arrows) around pulmonary artery and within smooth muscle (a, c, and e), and around arterioles (b, d, and f  ) of lung from control rats (a and b) and from rats 1 (c and d) and 5 (e and f  ) mo after lung transplantation. Scale bar = 50 µm.

View larger version (41K): [in this window] [in a new window]   Figure 3.   Length ratio of the CGRP-IR fibers per structure in five different areas in lungs from control animals (hatched columns) and from animals 1 (a, open columns) and 5 (b, open columns) mo after lung transplantation. The length ratio decreased after lung transplantation, particularly as the size of airways and vessels decreased (Br, Ar, Pbr). Results are means (columns) and SEM (bars) of five experiments each. **P < 0.02, ***P < 0.01 with control. LB, large bronchus (500 to 800 µm diameter); PA, pulmonary artery (500 to 700 µm diameter); Br, bronchiole (250 to 350 µm); Ar, around small vessel (50 to 150 µm); Pbr, peripheral bronchus (50 to 100 µm).

Immunoreactive cells. CGRP-IR cells were found only within the bronchial epithelium (Figure 4), where they were found either isolated or in small groups of three to five cells. They were sparse throughout the bronchial tree from the large bronchi to the peripheral bronchioles (Table 1). CGRP-IR was never found in cells from other locations, such as the parasympathetic ganglion cells.

View larger version (66K): [in this window] [in a new window]   Figure 4.   Grouped CGRP-IR cells (large arrows) in the epithelium of large bronchi (800 µm diameter; a and c) and bronchioles (250 to 350 µm diameter; b and d) in lungs from control rats (a and b) and from rats 1 mo after lung transplantation (c and d). CGRP-IR fibers are indicated by small arrows. Scale bar = 50 µm.

One Month after Left Lung Transplantation

CGRP-immunoreactive fibers. One month after transplantation, only a few CGRP-IR fibers were seen in the large bronchi and pulmonary vessels (Figures 1 and 2). Sparse immunolabeling was observed in the bronchioles, with no immunofluorescence detectable around the peripheral bronchioles or arterioles. When immunoreactive fibers were present, their localization was the same as in the control rat lungs; that is, beneath the epithelium, in the submucosa and smooth muscle of the airways, around the pulmonary vessels, and in the vascular smooth muscle (Figures 1 and 2).

Immunoreactive cells. One month after left lung transplantation, isolated and grouped CGRP-IR cells were observed in the bronchial epithelium (Figure 4). In the epithelium of large bronchi, substantially more of these cells were observed in transplanted than in control rats (P < 0.001) (Table 1); whereas in the bronchioles, the two groups of rats did not differ as to the number of CGRP-IR. CGRP-IR was not found in any cells from other locations.

Five Months after Left Lung Transplantation

CGRP-immunoreactive fibers. Five months after transplantation, CGRP-IR fibers were observed in the same locations as in the controls, even in the peripheral bronchioles (Figures 1 and 2).

Immunoreactive cells. Five months after left lung transplantation, the number, location, and distribution of isolated and grouped CGRP-IR cells were similar to those in the control lungs (Table 1). The epithelium of the large bronchi contained significantly fewer such cells than it had at 1 mo after transplantation (Table 1).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our study shows nerve fibers and cells immunoreactive for CGRP in the adult control rat lung and at 1 and 5 mo after lung transplantation. Cells were sparse in control lungs and their numbers increased significantly in large bronchi 1 mo after transplantation, to normalize after 5 mo. Fibers were present in control lung from the central to the peripheral airways, decreasing in number as the airways became smaller. One month after the transplantation, the length of CGRP-IR fibers had decreased markedly around the large bronchi and vessels and had totally disappeared from peripheral airways. Five months after transplantation, these fibers were significantly longer than at 1 mo, and immunoreactive nerves were found even in the periphery. These findings suggest some degree of reinnervation, even though significantly fewer nerve fibers were found in the transplanted than in the control lungs.

The nerves containing CGRP were located as expected from the studies in the upper and lower respiratory tracts of various mammals, including humans (17, 18). At lung transplantation, that is at interruption of the bronchial innervation, the CGRP-IR sensory C fibers should, in theory, degenerate because their cell bodies are located above the suture. Indeed, 1 mo after the transplantation here, CGRP-IR nerve fibers had totally disappeared from the peripheral bronchioles. In addition, from small to large airways, the length and number of fibers increased slightly but remained markedly lower than in controls. This reappearance of CGRP fibers suggests two alternative hypotheses: (1) regeneration of the C fibers in the transplanted lung through the suture, or (2) phenotypic change in nerves normally not expressing CGRP, such as parasympathetic fibers, the nerve bodies of which remain within the lung at transplantation (1). Further analysis of CGRP-IR fibers 5 mo after transplantation showed that nerve fibers were present up to the peripheral bronchioles, in greater length and number than at 1 mo after transplantation. Moreover, reinnervation of the original structures had occurred. These results strongly support the hypothesis of nerve growth, i.e., of lung reinnervation by C fibers, as opposed to the hypothesis of a phenotypic change of persistent nerves. This was proposed following the report by Springall and colleagues (19) that after lung transplantation, human intramural parasympathetic ganglion cells express tyrosine hydroxylase and are immunoreactive for neuropeptide Y, both of which are normally expressed in sympathetic neurons. In our experiments, however, CGRP-IR was never observed within the parasympathetic ganglion cells of the transplanted lung. Moreover, the location of CGRP-IR fibers was similar in the transplanted and control lungs, even though there were fewer fibers in the transplanted lungs. These findings suggest that CGRP is present in sensory rather than in parasympathetic cholinergic nerves and, again, strongly support the hypothesis of a slow sensory reinnervation of the rat lung after transplantation. The possibility of a role for the persistent parasympathetic innervation in the observed sensory reinnervation process must be retained, however, in light of the recent demonstration that vasoactive intestinal peptide can induce an activity-dependent neurotrophic factor (ADNF-14) (20, 21).

Trophic support is necessary for sensory neuron development after nerve injury (22). NGF is another likely candidate because it induces nerve growth (for review, see 25) and directly regulates the expression of neuropeptides in vitro and in vivo (26). An increase in CGRP-IR fibers after sympathetic denervation has been observed in rat lungs (29). This led to the suggestion that the sympathetic and neuropeptidergic fibers compete for the endogenous NGF needed for the growth of each (30). NGF may be synthesized by neurons and Schwann cells (22), and the latter help guide nerve outgrowth (31, 32), such as might help regeneration of CGRP-IR nerves in rat lung after transplantation.

More recently, Leon and associates (13) reported that the mast cell also may synthesize, store, and release NGF in the rat. Mast cells and sensory nerves are closely associated in various tissues, forming an anatomic and functional unit (9, 11). We have previously reported an increase in the number of mast cells in peripheral airways 1 mo after transplantation (12), when reinnervation has not yet taken place (the present study). In contrast, the number of mast cells did not increase in the larger airways (12), where, as we see here, sensory nerves were already present. Mast cells, therefore, might participate in the sensory reinnervation of rat lung after transplantation by providing NGF for nerve growth and for regulating CGRP expression. This hypothesis is supported by the results reported in animals treated with capsaicin: the number of mast cells increases in the lung (9), where sensory innervation is absent. Nonetheless, the contribution to sensory nerve regeneration of trophic factors besides NGF, including insulin-like growth factor-1 or neurotrophin-3, cannot be excluded (24, 33).

We also identified CGRP-IR cells, that is, the neuroendocrine cells described by Shimosegawa and Said (14), in the lung. They were located in the epithelium along the bronchial tree, as reported in studies of other mammals, including humans (15, 17, 18). The number of CGRP-IR cells was low in control subjects, and significantly increased 1 mo after lung transplantation in the large bronchi, where reinnervation was poor. Their number became similar to that in controls 5 mo after transplantation, when reinnervation tended to normalize (this study). This might support the hypothesis of an increase in CGRP production by neuroepithelial cells to counterbalance the lack of CGRP from the nerves. However, the number of neuroendocrine cells remained stable in the peripheral bronchi, where a total sensory denervation existed 1 mo after transplantation. The increase in CGRP-IR cells thus appears to be independent of the absence of CGRP-IR fibers, at least in the periphery.

The increased number of neuroepithelial cells in the large bronchi might also be related to post-transplantation tissue repair. The neuroendocrine cells participate in the vascular response to hypoxia (34). Because their number increased in the large bronchi and not in the small and peripheral bronchioles, ischemia accompanying transplantation is an unlikely explanation for the increase in neuroepithelial cells. Neuroendocrine cells and bodies also containing bombesin and related peptides are most numerous during fetal life, and may play a role in the growth and proliferation of lung cells (35). Hence, the increased number of neuroepithelial cells after transplantation might help post-transplantation tissue regeneration in the short term, then normalizing within 5 mo in the rat lung.

In conclusion, our study found more CGRP-IR fibers in rat lung at 5 mo than at 1 mo after transplantation, and thus suggests the growth of the sensory C fibers. This study, together with our previous finding of an increased number of mast cells in the lung 1 mo after transplantation, supports the hypothesis that the mast cell, which contains NGF (13), plays a role in nerve growth and tissue repair after transplantation (12). In addition, the number of neuroepithelial cells increased 1 mo after lung transplantation in bronchi where reinnervation was not fully achieved. This increase may help in post-transplantation tissue regeneration and remodeling. The sensory nerve regeneration occurring in the lung might play a role in the observed bronchial hyperresponsiveness to nerve stimulation of the transplanted bronchi (1).