INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hepatocyte growth factor (HGF) was initially detected in the plasma of partially hepatectomized rats as a potent mitogen for mature hepatocytes in primary culture, and was purified to homogeneity from rat platelets (1). HGF is a heterodimer molecule composed of 69-kD -subunit and 34-kD -subunit, and contains four kringle domains in the -chain (4, 5). In addition to mitogenic activity for hepatocytes, recent studies have shown that HGF is a pleiotropic factor that is produced by mesenchymal cells and acts on a wide variety of epithelial cells, including renal tubular cells, melanocytes, and keratinocytes (6, 7). HGF strongly enhances cell motility of various epithelial cells as a motogen, and induces branching tubule formation as a morphogen (7). HGF is now recognized as a humoral mediator of epithelial-mesenchymal interaction in tissue regeneration (7). HGF messenger RNA (mRNA) and HGF activity markedly increase in the liver of rats after various liver insults (10), and intravenously injected recombinant HGF markedly enhances liver regeneration in mice (12). HGF mRNA and HGF activity also increase in the kidney after renal injury (13). Thus, HGF has been recently reported to function as a hepatotrophic or renotropic factor for hepatic or renal regeneration, respectively.

More recently, HGF has been noted to function as a pulmotrophic factor for regeneration of an injured lung. In the injured lung, the alveolar type II cells contribute to regenerating the alveolar structure as progenitor cells in the repair process (14). In vitro, HGF stimulates DNA synthesis of the alveolar type II cells (15, 16). HGF mRNA expression was rapidly induced and DNA synthesis of the alveolar type II cells was increased in the lung after acute lung injury caused by intratracheal administration of hydrochloride (HCl) solution (17). Furthermore, intravenous and intratracheal administration of recombinant HGF stimulates DNA synthesis of alveolar type II cells in the rat lung after acute lung injury (18, 19). Recent studies have reported that HGF levels are elevated in the serum and bronchoalveolar lavage fluid of patients with inflammatory lung diseases, and in the pulmonary edema fluid of patients with acute lung injury (17, 20). These findings prompted us to investigate the role of HGF in lung injury and subsequent regeneration after thoracic surgery.

We have recently demonstrated that serum HGF levels are markedly elevated during the early period after transthoracic esophagectomy, which is accompanied by lung compression and collapse under unilateral ventilation (23). From the view of HGF involvement in tissue regeneration, it is tempting to speculate that pulmonary ischemia-reperfusion (IR) caused lung injury, which in turn enhanced the expression of HGF for lung regeneration. In spite of recent increasing information about the activities of HGF, no study has yet been conducted to examine the relationship between pulmonary IR injury and HGF expression. In this study, we investigated the expression of HGF at the peptide and mRNA levels using a rat model of pulmonary IR injury, which was established by occlusion of the left lung. Furthermore, we administered an anti-HGF antibody to rats of pulmonary IR, and examined the influence of HGF neutralization on DNA synthesis of alveolar epithelial cells to assess whether HGF plays an important role in regeneration of an injured lung after pulmonary IR.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Male Wistar rats (Charles River Japan, Inc., Yokohama, Japan) weighing 250 to 300 g were used in this study. Animals were housed for at least 7 d under controlled light/dark conditions in which the light period was from 8:00 A.M. to 8:00 P.M. They were allowed free access to ordinary pellet diet and tap water ad libitum. All the experimental procedures were reviewed and approved by the Animal Care and Use Committee of the Nara Medical University.

Pulmonary IR Model and Experimental Groups

Rats were anesthetized by intraperitoneal administration of 50 mg/kg sodium pentobarbital and tracheostomized with a 14-gauge angiocatheter. The lungs were ventilated with 100% O₂ at a rate of 60 breaths/min with a tidal volume of 6 ml/kg body weight using a Harvard-type animal respirator (Harvard rodent ventilator Model 683; Harvard Apparatus Co., Millis, MA). A left thoracotomy was performed at the fifth intercostal space, and then the left pulmonary hilus was exposed and clamped at its base using a vascular clip followed by placement of the lung and clip in the pleural cavity for 1 h. The tidal volume was reduced to 3 ml/kg only during clamping of the left hilus. After surgically implemented ischemia, the left lung was reinflated and reperfused by simultaneous releasing of clamping of the left hilus and increasing the tidal volume to 6 ml/kg. After closing the thorax in two layers, the rats were set free to breathe room air. The sham-operated rats underwent simple left thoracotomy in which the left lung was not clamped, and the control rats were only ventilated without thoracotomy. The sham-operated and control lungs were ventilated with a tidal volume of 6 ml/kg for 1 h.

Measurement of HGF in Plasma

Blood samples were serially obtained from the tail vein of five rats in each group at five timed intervals: before surgery, and on the first, third, fifth, and the seventh postoperative days (PODs). Blood samples were collected in pyrogen-free tubes containing ethylenediaminetetraacetic acid (EDTA), centrifuged at 1,000 × g for 10 min, and the resultant plasma was stored at 80° C until the plasma HGF assay. HGF concentrations in plasma were measured using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Institute of Immunology, Tokyo, Japan) according to the protocol supplied by the manufacturer. The minimum detectable level of HGF with the ELISA kits was found to be 0.2 ng/ml. All the samples were assayed at least in duplicate.

Measurement of HGF in Tissue Extracts

The liver and lungs of five rats at each stated time point were excised after gentle perfusion with saline through the right ventricle to remove the blood thoroughly from these organs. During the perfusion, we confirmed that the color of the liver tissue changed from dark red to light red as the blood was successfully removed from the systemic circulation as well as from the pulmonary circulation. The liver and lung tissues were frozen instantly in liquid nitrogen. Tissue extracts were prepared as previously reported (24). Briefly, the liver and lung tissues were homogenized with Polytron (Kinematica AG, Littau, Switzerland) in 4 volumes of buffer composed of 20 mM Tris-HCl (pH 7.5), 2 M NaCl, and 0.01% Tween 80 containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride and 1 mM EDTA). The homogenate was centrifuged at 15,000 rpm for 30 min and the resultant supernatant was used as tissue extract. HGF concentrations in tissue extracts were measured using commercially available ELISA kits (Institute of Immunology) according to the protocol supplied by the manufacturer. All the samples were assayed at least in duplicate.

RNA Extraction and Northern Blot Analysis

Total RNA was extracted from the liver and lung tissues, using the acid guanidium thiocyanate-phenol-chloroform method (25). Ten micrograms of total RNA was subjected to electrophoresis on a 1% agarose- formaldehyde denaturing gel with 1× MOPS (4-morpholinepropanesulfonic acid) buffer and transferred to a Hybond-N nylon membrane filter (Amersham). Hybridization was performed at 42° C for 20 h in a solution composed of 50% (vol/vol) formamide, 5× SSPE (0.75 M NaCl, 50 mM sodium phosphate buffer, and 5 mM EDTA), 2× Denhardt's solution (1× Denhardt's solution consists of 0.02% Ficoll, 0.02% polyvinylpyrrolidone, and 0.02% bovine serum albumin), 0.5% sodium dodecyl sulfate (SDS), and 100 µg/ml salmon sperm DNA. The XbaI-BamHI fragment (2.6 kilobase pairs) of pBS rat was used as the hybridization probe and was labeled with [³²P]dCTP using a Random Primer Labeling Kit (Amersham), according to the manufacturer's instructions. The XbaI-BamHI fragment of pBS rat included the entire open reading frame of rat HGF cDNA (26). After hybridization, the filter was washed twice with 2× SSPE/0.5% SDS for 15 min at room temperature followed by twice washing with 0.2× SSPE/0.1% SDS for 30 min at 65° C, and then dried and autoradiographed. The same filter was later hybridized to a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe as an internal reference for the quantification of total mRNA. Densitometric analysis was performed with a Bio-Image Analyzer (Fuji Photo Film, Tokyo, Japan).

Quantitative Assessment of Lung Injury

Lung histology. Lung tissues obtained from five rats at each time point were fixed with 70% ethanol at 4° C for 12 h, and then embedded in paraffin. Each section was cut to a thickness of 4 µm and was stained with hematoxylin and eosin. For the quantitative histopathological analysis of lung injury, a scoring system was used on the basis of the following four criteria: (1) destruction of the alveolar architecture; (2) neutrophil infiltration; (3) capillary congestion; and (4) pulmonary edema. Each criterion was graded on a scale of 0 to 3: 0, absent; 1, mild; 2, moderate; and 3, severe, and a total score was calculated for each specimen. To prevent observer's bias, all specimens were randomly numbered and evaluated in a blinded fashion by a pathologist. The mean of the scores was calculated for each time point and considered as the lung injury score.

Myeloperoxidase (MPO) assay. The extent of neutrophil sequestration was quantified by measuring MPO activity in the lung tissue according to the method described by Mullane and colleagues (27). Briefly, frozen lung tissue was homogenized in hexadecyltrimethyl ammonium bromide phosphate buffer (0.5%, pH 6.0), and the homogenate was centrifuged at 40,000 × g for 15 min. Fifty microliters of supernatant was added to 1.4 ml of 50 mM phosphate buffer (pH 6.0) containing 0.001% hydrogen peroxide, and then combined with 50 µl of 30 mM o-dianisidine hydrochloride (Sigma). MPO activity was assayed by measuring the change in spectrophotometric absorbance (optical density, O.D.) at 460 nm for 1 min and expressed as O.D./g of lung tissue.

Wet-to-dry weight ratio. Each harvested lung sample was weighed for the determination of wet weight and then dried in an oven at 70° C for 48 h. The dry tissue weight was determined and the lung wet-to-dry (W/D) ratio was calculated to assess pulmonary edema.

Immunohistochemistry

Lung tissues were fixed with 70% ethanol at 4° C for 12 h, dehydrated, and embedded in paraffin. Each section was cut to a thickness of 4 µm and stained for endogenous HGF, with an avidin-biotin-coupling (ABC) immunoperoxidase technique using a commercial kit (Vectastain Elite ABC; Vector Laboratories, Burlingame, CA), according to the instructions of the manufacturer. After deparaffinization, the sections were incubated at 4° C overnight with a rabbit immunoglobulin G (IgG) against rat HGF (1:1,000) (Institute of Immunology) for the primary reaction. After three washes with phosphate-buffered saline (PBS), the sections were further reacted with biotinylated goat anti-rabbit IgG at room temperature for 2 h. Immunostaining for HGF was visualized with 3,3'-diaminobenzidine tetrahydrochloride containing 0.01% hydrogen peroxide. To identify alveolar macrophages, mouse IgG against rat macrophage (ED-1) (1:500) (Serotec, Raleigh, NC) was applied on the dewaxed sections for the primary reactions. Following washing, the sections were incubated with a secondary rabbit anti-mouse IgG at room temperature for 30 min. After incubation with the alkaline phosphatase-conjugated anti-alkaline phosphatase (APAAP) complex (Dako Japan, Kyoto, Japan), alkaline phosphatase (AP) activity was developed in red with the New Fuchsin AP substrate system (Dako Japan) according to the instructions of the manufacturer.

Measurement of DNA Synthesis in the Lung

To determine the regenerative cell proliferation after pulmonary IR injury, cells undergoing DNA synthesis were identified by immunohistochemical staining for proliferative cellular nuclear antigen (PCNA). PCNA staining was performed on the dewaxed sections, using a commercial kit with a monoclonal antibody against PCNA (clone: PC-10) (EPOS kit; Dako Japan). The PCNA labeling index was determined by counting more than 1,000 nuclei of alveolar epithelial cells in randomly selected microscopic fields.

Anti-HGF Antibody Treatment

To assess the mitogenic effect of endogenous HGF on alveolar epithelial cells of an injured lung, another 12 rats with pulmonary IR were randomly divided into two groups and injected intraperitoneally with the neutralizing anti-rat HGF rabbit IgG (n = 6) or normal rabbit IgG (n = 6) at 1, 12, 24, and 36 h after declamping of the left hilus. A dose of an anti-HGF antibody was 400 µg/rat at 1 h, and 200 µg/rat at 12, 24, and 36 h, respectively, after declamping. An anti-rat HGF IgG was prepared as previously described (28). These rats were sacrificed at 48 h after declamping of the left hilus, and the PCNA labeling index was determined as described above.

Statistical Analysis

The results are expressed as the mean ± standard deviation. The means of different groups were compared using a one-way analysis of variance. Statistical analysis was performed with the unpaired Student's t test. A p value < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histological Changes after Pulmonary IR

The 4-µm-thick sections of lung tissues were stained with hematoxylin and eosin, and the visual time course of the histopathological findings are shown in Figure 1. In normal lung, the intima of thin alveolar septa consist of simple squamous epithelia (Figure 1A). In the control left lung on POD 1, minimal morphological changes were detected (Figure 1B). In the sham-operated left lung on POD 1, moderate infiltration of neutrophils and macrophages into the alveoli with focal fibrinous exudate were observed (Figure 1C). In the nonischemic opposite right lung of rats with left pulmonary IR on POD 1, moderate infiltration of macrophages into the alveoli and mild hyperemia of the alveolar septa were observed. Extensive edema and fibrinous exudates in alveolar lumina were also obvious (Figure 1D). In contrast, in the ischemic reperfused left lung of rats with pulmonary IR on POD 1, neutrophils markedly infiltrated into the alveolar septa and alveoli, and moderate infiltration of macrophages into the alveoli with focal fibrinous exudate were observed as well as severe destruction of the alveolar architecture (Figure 1E). Thereafter, severe infiltration of macrophages and increased alveolar epithelial cell proliferation were observed on POD 3 (Figure 1F), and mild infiltration of macrophages was observed on POD 5 (Figure 1G). In the reperfused left lung on POD 7, alveolar epithelial cell proliferation had decreased, and few macrophages into the alveoli were observed (Figure 1H).

View larger version (116K): [in this window] [in a new window]   Figure 1.   Histological changes in lung tissues after pulmonary IR. (A) Normal lung; (B) control left lung on POD 1; (C ) sham-operated left lung on POD 1; (D) nonischemic opposite right lung of rats with left pulmonary IR on POD 1; (E-H ) ischemic-reperfused left lung of rats with pulmonary IR. (A) In normal lung, the intima of the thin alveolar septa consist of simple squamous epithelia. (B) In the control left lung, minimal morphological changes were detected. (C ) In the sham-operated left lung, moderate infiltration of neutrophils and macrophages into the alveoli with focal fibrinous exudate were observed. (D) In the nonischemic opposite right lung of rats with left pulmonary IR, moderate infiltration of macrophages into the alveoli and mild hyperemia of the alveolar septa were observed. Extensive edema and fibrinous exudates in alveolar lumina were also obvious. (E ) In the ischemic reperfused left lung of rats with pulmonary IR on POD 1, neutrophils markedly infiltrated into the alveolar septa and alveoli, and moderate infiltration of macrophages into the alveoli with focal fibrinous exudate was observed as well as severe destruction of the alveolar architecture. (F  ) In the reperfused left lung on POD 3, severe infiltration of macrophages and increased alveolar epithelial cell proliferation were observed. (G) In the reperfused left lung on POD 5, mild infiltration of macrophages was observed. (H ) In the reperfused left lung on POD 7, alveolar epithelial cell proliferation had decreased, and few macrophages into the alveoli were observed (hematoxylin- eosin stain; original magnification: ×120).

Quantitative Evaluation of Lung Injury

For the quantitative analysis of lung injury, the lung injury score evaluated by histopathological examination, lung tissue MPO activity, and W/D weight ratio were examined, and the time courses of the three groups are shown in Table 1. In the sham-operated left lung, the lung injury score was slightly increased on POD 1, being about 8-fold higher than in the control left lung (p < 0.01). In contrast, in the ischemia-reperfused left lung of rats with pulmonary IR, the lung injury score was markedly increased and reached the peak level on POD 1, being about 20-fold higher than in the control left lung, and about 2.5-fold higher than in the sham-operated left lung (p < 0.01). A significant increase in lung injury score was also observed in the nonischemic opposite right lung of rats with left pulmonary IR on POD 1 compared with the sham-operated right lung (3.4 ± 0.7 versus 0.5 ± 0.3, p < 0.01) (data not shown). Consistent with the histopathological findings, MPO activity, a marker of tissue neutrophil infiltration, was significantly higher in the ischemic-reperfused left lung on POD 1 compared with the sham-operated left lung (p < 0.01). MPO activity in the sham-operated left lung was also significantly increased on POD 1 compared with the control left lung (p < 0.05). Lung tissue W/D ratio was significantly increased in the ischemic-reperfused left lung compared with the sham-operated left lung on POD 1 (p < 0.01). A significant increase in the W/D ratio was observed in the nonischemic opposite right lung of rats with left pulmonary IR on POD 1 compared with the sham-operated right lung (data not shown).

Plasma HGF Level

Blood samples were obtained from rats divided into three groups as follows: control, rats ventilated for 1 h without thoracotomy; sham, rats underwent simple left thoracotomy for 1 h; IR, rats underwent thoracotomy with left pulmonary ischemia for 1 h. The HGF concentration in plasma was measured by ELISA (Figure 2). The plasma HGF levels in the sham- operated rats gradually increased to a maximum on POD 3, and returned to the base value on POD 7. The maximum level was about 3-fold higher than that of the control rats (p < 0.01). In contrast, the plasma HGF levels in rats with pulmonary IR were markedly elevated on POD 1, reaching the peak level on POD 3, and then remained at high level for a week after surgery. The maximum level was about 5-fold higher than that of the sham-operated rats, and about 15-fold higher than that of the control rats (p < 0.01).

View larger version (13K): [in this window] [in a new window]   Figure 2.   Time course of the plasma HGF concentration. Blood samples were obtained from rats divided into three groups as follows: control, rats ventilated for 1 h without thoracotomy (open circles); sham, rats underwent simple left thoracotomy for 1 h (closed circles); IR, rats underwent thoracotomy with left pulmonary ischemia for 1 h (closed squares). Values are mean ± SD for five rats in each group. *p < 0.05, **p < 0.01 versus control. p < 0.01 versus sham (Student's t test).

Tissue HGF Level in Liver and Lung

To assess the distribution of HGF, the tissue HGF concentration in the liver and lung was measured by ELISA (Figure 3). The tissue HGF levels in the liver of rats with pulmonary IR and the sham-operated rats were significantly higher than those of the control rats on POD 3 (p < 0.05). In the right lung, the tissue HGF levels of rats with pulmonary IR and the sham-operated rats were increased and reached the peak value on POD 3, being about 1.5-fold higher than that of the control rats (p < 0.05). In the left lung of the sham-operated rats, the tissue HGF level was slightly increased to 512 ± 107 ng/g tissue on POD 3. In contrast, in the left lung of rats with pulmonary IR, the tissue HGF level was markedly increased to 643 ± 161 ng/g tissue on POD 1 (p < 0.05) and reached the peak level of 1026 ± 185 ng/g tissue on POD 3, being about 2-fold higher than that of the sham-operated rats and about 3-fold higher than that of the control rats (p < 0.01). A significant increase was also observed in the ischemic-reperfused left lung on POD 5 compared with the control left lung (p < 0.01).

View larger version (23K): [in this window] [in a new window]   Figure 3.   Time course of the tissue HGF concentration in the liver, right and left lungs. Liver and lung tissue extracts were obtained from rats divided into three groups as follows: control, rats ventilated for 1 h without thoracotomy (open bars); sham, rats underwent simple left thoracotomy for 1 h (hatched bars); IR, rats underwent thoracotomy with left pulmonary ischemia for 1 h (closed bars). POD = postoperative day; POD 0 = before surgery. Values are mean ± SD for five rats at each time point. *p < 0.05, **p < 0.01 versus control.  p < 0.01 versus sham (Student's t test).

HGF mRNA Expression in Liver and Lung

To determine whether the upregulation of HGF is selective for the ischemic-reperfused lung, we examined its expression at the mRNA level by Northern blot analysis in the liver and the nonischemic right lung as well as in the ischemic-reperfused left lung of rats with pulmonary IR. The densitometric analysis of HGF normalized to the GAPDH transcripts is shown in Figure 4A. Figure 4B shows the representative results of rat HGF mRNA expression in the liver and lungs of three groups on POD 1. The level of rat HGF mRNA was significantly increased on POD 1 in the ischemic-reperfused left lung (56 ± 14%) compared with the sham-operated left lung (24 ± 7%) and the control left lung (15 ± 4%) (p < 0.01). Furthermore, HGF mRNA expression was also increased in the sham-operated left lung on POD 1 compared with the control left lung (p < 0.05), despite a similar level of HGF mRNA in the right lung between the two groups. Interestingly, a significant increase in HGF mRNA expression was also found in the nonischemic right lung of rats with pulmonary IR on POD 1 (29 ± 9%; p < 0.05 versus the control right lung). There were no significant differences in the HGF mRNA expression in the liver among the three groups.

View larger version (21K): [in this window] [in a new window]   View larger version (29K): [in this window] [in a new window]   Figure 4.   (A) Time course of the densitometric relative quantification of HGF mRNA in the liver, right and left lungs. The relative amount of HGF message was determined by scanning the bands with a Bio-Image Analyzer. For relative quantification, the densitometer value of HGF was normalized to that of GAPDH. Total RNAs were obtained from rats divided into three groups as follows: control, rats ventilated for 1 h without thoracotomy (open bars); sham, rats underwent simple left thoracotomy for 1 h (hatched bars); IR, rats underwent thoracotomy with left pulmonary ischemia for 1 h (closed bars). POD = postoperative day; POD 0 = before surgery. Values (%) are mean ± SD for five rats at each time point. *p < 0.05, **p < 0.01 versus control.  p < 0.05, p < 0.01 versus sham (Student's t test). (B) HGF mRNA expression in the liver, right and left lungs. The representative results on the first postoperative day are shown. HGF mRNA levels were evaluated by Northern blot analysis, as described in METHODS. Arrows indicate 3 kbp and 6 kbp of rat HGF mRNA. GAPDH transcripts are shown as an internal reference for the quantification of total mRNA. C = control; S = sham; IR = pulmonary ischemia-reperfusion.

Lung Immunohistochemistry

To determine the localization of HGF protein in the lung, we performed immunohistochemical analysis. The representative results of the ischemic reperfused left lung obtained from rats with pulmonary IR on POD 3 are shown in Figure 5. The cytoplasm of interstitial cells was intensely stained for HGF, but the alveolar epithelial cells were negative for HGF (Figure 5A). Most of these interstitial cells were also positive for macrophage marker (ED-1) (Figure 5B).

View larger version (80K): [in this window] [in a new window]   Figure 5.   Immunohistochemical staining for HGF (A) and macrophage marker (ED-1) (B). The representative results of the reperfused lung obtained from rats with pulmonary IR on the third postoperative day are shown. Most of the interstitial cells that are stained for HGF are also positive for ED-1, but the alveolar epithelial cells are negative for HGF. Original magnification: ×200.

DNA Synthesis in the Lung after Pulmonary IR Injury

To determine the regenerative cell proliferation after pulmonary IR injury, cells undergoing DNA synthesis were identified by immunohistochemical staining using anti-PCNA monoclonal antibody. In the ischemic-reperfused left lung of rats with pulmonary IR, PCNA-positive cells were mainly alveolar epithelial cells (Figure 6). Several endothelial cells and macrophages in the alveolar space were also positive for PCNA, although the number of these cells undergoing DNA synthesis was much fewer than in alveolar epithelial cells. The PCNA labeling index for the alveolar epithelial cells was increased and reached the peak level on POD 3 in the ischemic-reperfused left lung, being about 3-fold higher than that in the sham-operated left lung (p < 0.01) (Table 2).

View larger version (134K): [in this window] [in a new window]   Figure 6.   Distribution of cells undergoing DNA synthesis in the ischemic-reperfused left lung. DNA synthesis was identified by immunohistochemical staining using anti-PCNA monoclonal antibody, and the representative result obtained from rats with pulmonary IR on the third postoperative day is shown. Dark brown nuclei represent positive staining. PCNA-positive cells were mainly alveolar epithelial cells. Several endothelial cells and macrophages in the alveolar space were also positive for PCNA, although the number of these cells undergoing DNA synthesis was much fewer than in alveolar epithelial cells. Original magnification: ×200.

Influence of HGF Neutralization on Lung Injury and DNA Synthesis

To assess the mitogenic effect of endogenous HGF on alveolar epithelial cells of an injured lung, we injected neutralizing anti-HGF rabbit IgG (or normal rabbit IgG as a placebo) into rats with pulmonary IR and examined the influence of HGF neutralization on DNA synthesis of alveolar epithelial cells. Furthermore, we also examined its influence on lung morphology. Lung histology revealed that administration of anti-HGF IgG aggravated lung injury characterized by pulmonary edema and destruction of the alveolar architecture with few alveolar epithelial proliferation, which were more severe in rats given anti-HGF IgG than in the normal IgG-injected rats (Figure 7A). The lung injury score was significantly higher in the anti-HGF IgG-treated rats than in the normal IgG-injected rats (7.6 ± 1.4 versus 4.1 ± 1.3, p < 0.01) (Figure 7B). Lung immunohistochemistry revealed that HGF neutralization reduced the number of PCNA-positive alveolar epithelial cells (Figure 7C), and the PCNA labeling index for alveolar epithelial cells in the anti-HGF IgG-treated rats was much lower than that in the normal IgG-injected rats (3.6 ± 1.8 versus 15.3 ± 4.2, p < 0.01) (Figure 7D).

View larger version (69K): [in this window] [in a new window]   Figure 7.   Influence of HGF-neutralizing treatment with an anti-HGF antibody on lung injury and DNA synthesis of alveolar epithelial cells. (A) Histological findings of the ischemic-reperfused lung in rats given normal or anti-HGF IgG (hematoxylin-eosin stain; original magnification: ×100), and (B) the quantitative analysis of lung injury, as determined by lung injury score. (C ) Immunohistochemical staining for PCNA in the ischemic-reperfused lung of rats given normal or anti-HGF IgG (original magnification: ×100), and (D) the quantitative analysis of DNA synthesis, using PCNA labeling index for alveolar epithelial cells. Values are mean ± SD for six rats in each group. *p < 0.01 compared with the value in normal IgG-injected group (Student's t test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we have shown that thoracotomy with 1 h of pulmonary ischemia resulted in significantly higher levels of plasma HGF compared with simple thoracotomy alone. Histopathologically, massive neutrophil infiltration in interstitium and alveoli was observed as well as destruction of the alveolar architecture in the ischemic-reperfused lung. Furthermore, lung tissue MPO activity, a marker of tissue neutrophil infiltration, was also augmented in the reperfused lung. Sekido and colleagues (29) have provided evidence that ablation of neutrophils attenuated the degree of tissue injury in rabbit models of pulmonary IR injury, indicating a pivotal role of neutrophils in the pathogenesis of IR injury. Therefore, our results suggest that HGF production may be associated with lung injury caused by pulmonary IR in our rat model, which was established by occlusion of the left lung. In our pilot study, thoracotomy with 1 h of pulmonary compression, in which the left lung was compressed using a balloon filled with water, resulted in significantly higher levels of plasma HGF compared with simple thoracotomy, but significantly lower HGF levels compared with thoracotomy with pulmonary ischemia. We also measured the volumetric lung tissue blood flow in the compressed lung using a noncontact laser doppler blood flowmeter, and those values rapidly decreased to 30% of the normal values during simple thoracotomy (data not shown). These findings suggest that the plasma HGF level reflects the severity of acute lung injury caused by hypoxia and subsequent reoxygenation.

Northern blot analysis revealed that HGF mRNA was dramatically upregulated in the reperfused lung after pulmonary ischemia. However, HGF mRNA was upregulated even in the nonischemic left lung of rats with simple left thoracotomy, suggesting that HGF expression may be induced by surgical trauma such as thoracotomy alone, unrelated to pulmonary IR injury. Nevertheless, the HGF mRNA level was significantly higher in the ischemic-reperfused left lung compared with the nonischemic left lung of rats with simple left thoracotomy, indicating that enhanced expression of HGF mRNA is associated with pulmonary IR injury. Two sequence elements, an interleukin-6 (IL-6) response element and a potential binding site for nuclear factor IL-6, are located near the transcription initiation site of the human HGF gene, and they might be involved in the regulation of HGF gene expression (30). It has been also reported that inflammatory cytokines such as IL-1 and tumor necrosis factor- stimulate HGF production (31). Recently, we reported that serum IL-6 levels are markedly increased after thoracic surgery (23), and IL-1 gene expression is upregulated in pulmonary IR injury (32). These findings suggest that pulmonary IR injury and surgical trauma may induce the production of these inflammatory cytokines, which in turn enhance the upregulation of HGF gene expression in an injured lung.

In this study, HGF mRNA was also upregulated even in the nonischemic opposite right lung after left pulmonary ischemia, although the mRNA level was much higher in the injured left lung compared with the intact right lung. Moreover, no significant increase in HGF mRNA level was found in the right lung of the sham-operated rats. These results suggest that HGF mRNA in the intact right lung after left pulmonary ischemia may be regulated by factors other than pure pulmonary IR injury. The relative increase in blood flow to the opposite right lung during occlusion of the left lung may be in part responsible for the upregulation of HGF mRNA in the intact right lung. Indeed, the histological examination revealed mild hyperemia of the alveolar septa and severe intraalveolar edema in the opposite right lung of rats with left pulmonary IR. However, further investigations are needed to explain the expression of HGF in the nonischemic right lung.

HGF is secreted as a single-chain precursor. In response to tissue injury, the single-chain precursor is converted to a biologically active heterodimer by a serine protease, the activity of which is initiated in the injured tissue (33). We found here that the tissue HGF levels of the sham-operated rats were significantly higher not only in the left lung but also in liver and right lung compared with those of the control rats, despite similar levels of HGF mRNA in the liver and right lung between the two groups. The discrepancy in the degree of increase in HGF levels between HGF mRNA and the tissue HGF may be due to the difference in measurement. HGF mRNA correlates to the full-length single-chain HGF, whereas the antibody used for measurement of tissue HGF level was raised against a peptide corresponding to the -chain of HGF, and thus recognizes the intracellular monomeric forms as well as the mature heterodimeric forms (24). Furthermore, HGF has a specific affinity for heparin, and there are the high- and low-affinity HGF receptors present on the target cell surface (34). The high-affinity receptor, which corresponds to a signal-transducing receptor, is the c-met/HGF receptor, whereas the low-affinity receptor is thought to be mainly heparin sulfate-like polysulfated glycosaminoglycan. Thus, the tissue HGF level may include not only the cytoplasmic HGF but also the receptor-bound HGF and the membrane-bound HGF captured and retained by proteoglycans present on the extracellular matrix and cell surface, although the blood was removed thoroughly from the organs by saline perfusion. These variant forms of HGF protein and the receptor-bound HGF could explain the disparity between the elevated protein values in sham liver and right lung where mRNA did not increase.

In this model, pulmonary IR caused neutrophil infiltration and subsequently increased cellularity with alveolar macrophages and alveolar epithelial cell proliferation. Furthermore, DNA synthesis of alveolar epithelial cells was significantly increased in the ischemic-reperfused lung. In the injured lung, the bronchial and alveolar type II epithelium replications are essential for tissue repair and regeneration (14). Recent studies have demonstrated that HGF stimulates proliferation of these epithelial cells through DNA synthesis in vitro and in vivo (15). Taking these results together, it is tempting to speculate that pulmonary IR caused lung injury, which in turn enhanced the expression of HGF for regeneration of an injured lung. To examine our hypothesis, we administered neutralizing anti-HGF antibody into rats with pulmonary IR. HGF neutralization dramatically inhibited DNA synthesis of alveolar epithelial cells in the ischemic-reperfused lung. Furthermore, anti-HGF treatment aggravated lung injury, whereas the control antibody did not. These results suggest that HGF may act as a pulmotrophic factor for regeneration of an injured lung after pulmonary IR. In the lung, the major sources of HGF are mesenchymal cells, such as alveolar macrophages, fibroblasts, and endothelial cells (15, 21, 35). We found here the alveolar macrophages were intensely stained for HGF. These findings suggest that HGF produced by the alveolar macrophages may induce proliferation of the alveolar epithelial cells in the injured lung.

This experimental study supports and extends our recent work, which has demonstrated that the serum HGF level is markedly elevated during the early period after transthoracic esophagectomy (23). We have clinically noticed that pulmonary infections are frequent after transthoracic esophagectomy and some cases progressed to severe respiratory failure. Taking our results together, severe surgical trauma such as pulmonary ischemia during thoracic surgery may cause lung injury, which facilitated pulmonary infections through destruction of the alveolar architecture that acts as a physiological barrier. In light of this information, prevention or early treatment of lung injury after thoracic surgery may reduce the frequency of postoperative pulmonary infections. Recent studies have shown a therapeutic potential of HGF in acute lung injury and bleomycin-induced lung fibrosis (36, 37). The present study suggests that HGF may act as a pulmotrophic factor for lung regeneration after acute lung injury caused by pulmonary IR. In this context, exogenous HGF may be a potential candidate as a preventive agent for pulmonary infections after major thoracic surgery. Furthermore, even in the established postoperative pulmonary infections, exogenous HGF may also have a therapeutic potential through regeneration of an injured lung.

In conclusion, we have shown that enhanced expression of HGF and DNA synthesis of alveolar epithelial cells were induced in the ischemic-reperfused lung. Furthermore, HGF-neutralizing treatment with an anti-HGF antibody dramatically reduced DNA synthesis of alveolar epithelial cells in the reperfused lung and aggravated lung injury. These findings suggest that HGF may play an important role in regeneration of an injured lung after pulmonary IR. This study may provide an important consideration in understanding the basic biological process of lung regeneration after pulmonary IR injury. In addition, this work may contribute to new preventive and therapeutic strategies for lung injury after major thoracic surgery, such as transthoracic esophagectomy and lung transplantation, accompanied by a prolonged pulmonary ischemia. Further clarification of the biological and physiological significance of HGF in pulmonary IR will likely have important clinical implications in light of "regenerating medicine" in the future.