Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Lungs are included among a variety of organs that undergo compensatory growth following partial resection or in response to severe functional demand that exceeds physiologic capacity (1). The lung in various mammalian species shows compensatory growth after removal of one or several pulmonary lobes. Although this compensatory lung growth in adult animals occurs in a much decelerated manner and often results in incomplete tissue restoration, as compared with that in immature animals, the potential for the adult lung to undergo this restorative growth has been confirmed (4, 5). Whether compensatory lung growth can occur in humans is controversial, whereas some physiologic studies indicated that such growth does occur to some extent in humans, primarily in infants and young children (6, 7).

Changes in the remaining lung following unilateral pneumonectomy have been studied for more than a century, and sources and the extent of compensatory lung growth have been examined (3). In rodents, this restorative process of the remaining lung mass consists not only of overinflation of existing airspaces or increased pulmonary blood flow, but also of cellular events, including proliferation of type II pneumocytes that contributes to formation of new alveoli, and increased interstitial cells such as fibroblasts and the endothelium (8, 9). Likewise, the increased mitotic index in the bronchiolar epithelium has been confirmed in cases of compensatory lung growth in rats (10). Although the stimuli and regulation of this growth process is not well understood, involvement of growth factors or cytokines has been implicated in the regulation of postpneumonectomy lung growth (11).

Hepatocyte growth factor (HGF), originally purified and cloned as a potent mitogen for mature hepatocytes (14, 15), has potent motogenic, mitogenic, and morphogenic activities on a wide variety of epithelial cells, including respiratory epithelia (16). Studies have shown that HGF, a multifunctional humoral mediator that is predominantly expressed in mesenchymal cells, acts on epithelial cells of various tissues through c-Met/HGF receptor tyrosine kinase (20). In the lung, HGF is produced by alveolar macrophages and vascular endothelial cells in the rat lung (24). HGF stimulates DNA synthesis of airway epithelial cells (18, 19) and alveolar type II cells that contribute to regenerating the alveolar structure as progenitor cells (16, 17, 25). During lung development, HGF is involved in other aspects of lung growth, i.e., epithelial morphogenesis, as a mediator in epithelial-mesenchymal interaction for pulmonary organogenesis (26). In addition to these findings in laboratory animals, serum HGF levels in patients with pneumonectomy or pulmonary lobectomy increased remarkably and were maintained at higher levels within several days of surgery (27). Although the biologic significance of this phenomenon remains to be addressed, the potential contribution of HGF to compensatory response after major lung resection warranted further study.

In the present study we focused on the role of endogenous HGF in the regulation of respiratory epithelial cell proliferation after unilateral pneumonectomy. We analyzed changes in HGF expression in a murine model of left pneumonectomy, the most commonly practiced surgical approach. Together with effects of biologic neutralization and supplement of recombinant HGF on DNA synthesis of cells in the remaining lung, our data strongly suggest that HGF plays a pulmotropic role in compensatory lung growth after unilateral pneumonectomy.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Surgical Procedures

Adult disease-free 8-wk-old female Institute of Cancer Research mice weighing ~ 25 g were anesthetized with 100 mg/kg ketamine and 10 mg/kg xylazine given subcutaneously. Following tracheostomy the lungs were ventilated with room air at a rate of 70 breaths/min with a tidal volume of 0.5 ml using an animal respirator (Model SN-480-7; Shinano Manufacturing Co. Ltd., Tokyo, Japan). The left lung was tied at the hilum and excised through a left thoracotomy at the fifth intercostal space. The thoracic incision was closed and the mouse was extubated when spontaneous breathing was restored. The sham-operated mice underwent simple left thoracotomy in which the left lung was not tied or removed. All animal experiments were done in accordance with NIH guidelines, as dictated by the Animal Care Facility at Osaka University Graduate School of Medicine.

Measurement of HGF in Plasma and Tissues

Blood samples were obtained from six mice at each time point in the pneumonectomy group and four mice at each time point in the sham-operated group. Plasma HGF concentrations were measured using enzyme-linked immunosorbent assay (ELISA) kits for rodent HGF (Institute of Immunology, Tokyo, Japan) (28). The lung, liver, and kidney were excised and instantly placed in liquid nitrogen. Preparation of tissue extracts and measurement of tissue HGF concentrations were done as described previously (28). Briefly, tissues were homogenized in 10 vols of buffer composed of 20 mM Tris-HCl (pH 7.5), 2 M NaCl, 0.01% Tween 80, 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.

Real-Time Quantitative Reverse Transcriptase/Polymerase Chain Reaction

Total RNA was prepared from the right lung, liver, and kidney, using ISOGEN (Nippon Gene, Tokyo, Japan). One microgram of total RNA was reverse-transcribed into first strand cDNA with random hexaprimer using Superscript II reverse transcriptase (RT) (Life Technologies Inc., Rockville, MD). Quantitative polymerase chain reaction (PCR) was performed, using the ABI PRISM 7,700 Sequence Detector System (Perkin-Elmer Biosystems, Foster City, CA) as described elsewhere (29, 30). Briefly, sequences for primers and TaqMan fluorogenic probes (Perkin-Elmer Biosystems) were as follows: HGF, forward primer, 5'-AAG AGT GGC ATC AAG TGC CAG-3', reverse primer, 5'-CTG GAT TGC TTG TGA AAC ACC-3', probe, 5'(FAM)-TGA TCC CCC ATG AAC ACA GCT TTT TG-(TAMRA)3'; c-Met, forward primer, 5'-GTA CGG TGT CTC CAG CAT TTT T-3', reverse primer, 5'-AGA GCA CCA CCT GCA TGA AG-3', probe, 5'(FAM)-ACC ACG AGC ACT GTT TCA ATA GGA CCC-(TAMRA)3'; glyceraldehyde 3-phosphate dehydrogenase (GAPDH), forward primer, 5'-CCA TCA CTG CCA CTC AGA AGA C-3', reverse primer, 5'-TCA TAC TTG GCA GGT TTC TCC A-3', probe, 5'(FAM)-CGT GTT CCT ACC CCC AAT GTA TCC GT-(TAMRA)3'. Experimental samples were matched to a standard curve generated by amplifying serially diluted products, using the same PCR protocol. To correct for variability in RNA recovery and efficiency of reverse transcription, GAPDH cDNA was amplified and quantitated in each cDNA preparation.

Immunohistochemical Detection of the c-Met/HGF Receptor

The lung was fixed in 70% ethanol at 4°C for 12 h, dehydrated, and embedded in paraffin. The antibody used for the primary reaction on the deparaffinized sections was rabbit polyclonal anti-mouse c-Met (1:200) (SP 260; Santa Cruz Biotechnology Inc., Santa Cruz, CA). The negative control was also prepared by preabsorption of the anti-c-Met antibody with the antigenic synthetic peptide. After three washes with phosphate-buffered saline (PBS), the sections were further reacted with a secondary biotinylated anti-rabbit IgG (1:200) for 2 h. An avidin-biotin coupling (ABC) immunoperoxidase technique was used for sections, at room temperature, and using a commercial kit (Vectastain Elite ABC; Vector Laboratories, Burlingame, CA), according to the manufacturer's instructions. Immunostaining for c-Met/HGF receptor was visualized in PBS containing 0.02% 3,3'-diaminobenzidine tetrahydrochloride (DAB) and 0.01% H₂O₂.

Evaluation of DNA Synthesis in the Lung

DNA synthesis of cells in tissues was measured by incorporation of 5-bromo-2'-deoxyuridine (BrdU) into nuclei and subsequent immunohistochemical staining, using an anti-BrdU monoclonal antibody, as described previously (24). Briefly, BrdU (Sigma, St. Louis, MO) dissolved in saline was intraperitoneally injected at 100 mg/kg at 1 h before the animals were killed. Tissues were embedded in paraffin as described above. After deparaffinization, the sections were subjected to inactivation of endogenous peroxidase in 0.3% H₂O₂ for 30 min and washed with PBS. DNA was denatured in 2 M HCl for 1 h, and sections were neutralized in 0.1 M borate buffer (pH 8.4). The nonspecific binding of antibodies were blocked with PBS containing 10% horse serum. After washing with PBS, the sections were incubated for 1 h with anti-BrdU monoclonal antibody (Takara, Kyoto, Japan) diluted to 1:2,000 with PBS containing 5% horse serum. After washing, the sections were successively incubated for 30 min with biotinylated horse anti-mouse IgG (1:1,000) and peroxidase-conjugated avidin-biotin complex (Vector Laboratories) for 5 min. Immune complexes were visualized in the substrate solution composed of PBS containing 0.02% DAB and 0.01% H₂O_2. The BrdU labeling index that represents a ratio of labeled cells to total cells counted was determined by counting more than 1,000 nuclei in randomly selected microscopic fields.

Neutralization of HGF by an Anti-HGF Antibody

Anti-rat HGF neutralizing antibody was prepared as described previously (26), and the IgG fraction was purified on a protein A-Sepharose column (Pharmacia Biotech AB, Uppsala, Sweden). In cell scattering assay with Madin-Darby canine kidney cells (an authentic bioassay for HGF), 5 µg/ml of anti-rat HGF IgG completely inhibited the cell scattering induced by 5 ng/ml rat HGF. Specific binding of this antibody to HGF but not other growth factors has previously been confirmed both in vitro and in vivo (26, 30). Pneumonectomized mice were randomly divided into two groups and injected intraperitoneally with the neutralizing anti-rat HGF rabbit IgG or normal rabbit IgG at 24-h intervals after surgery with the first administration being given at the time of surgery. The animals were killed on Days 3 and 5 after pneumonectomy to examine changes in cell proliferation and weight of the remaining lung (including upper, middle, lower, and cardiac lobes). Six mice were included in each experimental group.

Recombinant HGF Treatment

Human recombinant HGF was purified from culture media of Chinese hamster ovary cells transfected with an expression vector containing human HGF cDNA as described previously (15, 19, 30). The purity of HGF was > 98%, as determined by sodium dodecylsulfate polyacrylamide gel electrophoresis and the following protein staining. Pneumonectomized mice were randomly divided into two groups and mice were injected subcutaneously with 500 µg/kg body weight of recombinant HGF dissolved in saline or an identical volume of saline alone. The first administration was given at the time of surgery, followed by administrations at 12-h intervals. The animals were killed on Days 1, 2, 3, and 5 after pneumonectomy to examine changes in cell proliferation and weight of the remaining lung. Six mice were used in each experimental group.

Statistical Analysis

Data are expressed as the mean ± SEM. The means of different groups were compared using a one-way analysis of variance. For statistical analysis we used unpaired Student's t test and a P value of < 0.05 was considered to be statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Proliferation in the Remaining Lung after Pneumonectomy

It was reported that unilateral pneumonectomy resulted in proliferation of various types of cells in the contralateral remaining lung (8, 9). To examine DNA synthesis of cells in the remaining right lung, cells undergoing DNA synthesis were labeled by incorporation of BrdU followed by immunochemical staining. In the intact lung, there were few cells undergoing DNA synthesis (Figure 1A), and the labeling indexes (% of cells undergoing DNA synthesis) of both alveolar and airway epithelial cells were less than 0.1%. On Day 5 after the surgery, several epithelial cells in both alveolar and airway regions underwent DNA synthesis (Figures 1B, 1C, 1D, and 1E). Most of the labeled cells localized in the alveolar regions showed typical characteristics of alveolar type II epithelial cells, with a relatively large cytoplasm at the alveolar surface and usually located in corners (Figures 1B and 1C), whereas mostly labeled airway epithelial cells localized in bronchioles (Figures 1D and 1E). The time course of change in the labeling index in both alveolar and airway epithelial cells indicates that the indexes increased from Day 1 in both alveolar and airway epithelial cells, and reached respective maximum values on Day 5: 2.24 ± 0.16% in alveoli; 2.33 ± 0.15% in airways (Figures 1F and 1G). On Day 10, the labeling index in airway epithelial cells reverted to normal levels, and the labeling index in alveolar epithelial cells was higher than normal but was much less than the peak value on Day 5. In contrast, significant changes in DNA synthesis in the distant organs, including the liver and kidney, were not seen during the same period as above (data not shown). The time course of change in DNA synthesis of interstitial cells (mostly fibroblasts, endothelial cells, and macrophages) was similar to that of epithelial cells with the peak of ~ 1% on Day 5 (data not shown).

View larger version (101K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Figure 1.   Distribution of cells undergoing DNA synthesis and change in the number of alveolar and airway epithelial cells undergoing DNA synthesis in the remaining lung following left pneumonectomy. (A-E) Distribution of cells undergoing DNA synthesis in the right lung. Cells undergoing DNA synthesis were detected by BrdU incorporation into the nuclei and subsequent immunohistochemical staining. A, Normal lung; B and C, alveolar region at Day 5 after left pneumonectomy; D and E, around bronchiole at Day 5 after pneumonectomy. Original magnifications: A ×100; B and D, ×200; C and E, ×400. (F and G) Change in the number of cells undergoing DNA synthesis (BrdU labeling index) in alveolar epithelial cells (F  ) and airway epithelial cells (G) in the right lung. Mice were being either pneumonectomized (Pneum, closed circles) or simply thoracotomized (Sham, open circles) and killed on Day 0 (before surgery; control), 1, 3, 5, and 10 after surgery. Labeling indexes represent the ratios of cells labeled with BrdU to total cells counted for more than 1,000 nuclei. Each value represents the mean ± SEM of values obtained using 4-6 mice at each time point.

Changes of HGF mRNA in the Lung, Liver, and Kidney

To investigate a potential involvement of HGF in compensatory lung regeneration, we analyzed changes of HGF mRNA in the remaining right lung, liver, and kidney following pneumonectomy. Total RNA was prepared from these tissues and changes of HGF mRNA levels were determined by real-time quantitative RT-PCR (Figure 2). In the remaining lung, HGF mRNA expression increased as early as Day 1 after pneumonectomy (274 ± 34% of the level in the normal lung; P < 0.01) (Figure 2A). HGF mRNA level in the remaining lung reached the peak on Day 5 (360 ± 28% of the normal level; P < 0.01), then decreased, yet remained at higher levels than normal until Day 10. HGF mRNA in the right lung of sham-operated mice also increased on Day 1, reaching a level similar to that seen in the remaining lung after pneumonectomy. Thereafter it decreased toward normal levels. Interestingly, the expression of HGF mRNA was also upregulated in the liver and kidneys after pneumonectomy (Figures 2B and 2C). HGF mRNA levels in these distant organs increased during the first 3 d after pneumonectomy (up to 212 ± 17% of control in the liver, and 225 ± 26% of control in the kidney, P < 0.01).

View larger version (21K): [in this window] [in a new window]   Figure 2.   Changes in HGF mRNA expression in the right lung (A), liver (B), and kidney (C) following left pneumonectomy. Mice were killed on Day 0 (before surgery; control) 1, 3, 5, and 10. Total RNAs were obtained from mice with either left pneumonectomy (Pneum, closed circles) or sham left thoracotomy (Sham, open circles) at each time point and HGF mRNA levels were determined by real-time quantitative RT-PCR. Each value (% of control) represents the mean ± SEM of values obtained using 4-6 mice at each time point. *P < 0.05, **P < 0.01 versus control. P < 0.05. P < 0.01 versus sham (Student's t test).

Changes of Tissue HGF Levels in Lung, Liver, and Kidney

To measure changes in the protein level of HGF in the right lung, liver, and kidney, tissue HGF levels were measured using ELISA (Figure 3). In the right lung of the sham-operated mice, tissue HGF levels were transiently elevated on Day 1, then decreased to almost a normal level on Day 10 (Figure 3A). HGF level of the right lung of pneumonectomized mice also increased on Day 1, to a level similar to that seen in sham-operated mice, although it thereafter increased to higher levels: lung HGF level in pneumonectomized mice was ~ 1.5-fold higher than that of the sham-operated mice on Day 10 (140 ± 9 ng/g tissue versus 94 ± 8 ng/g tissue, P < 0.01). HGF levels in the liver and kidney of the sham-operated mice remained constant for 10 d after the surgery (Figures 3B and 3C). The hepatic HGF level in pneumonectomized mice initially decreased on Day 1, then increased, being significantly higher than findings in the sham-operated mice on Day 3 (114 ± 6% versus 96 ± 5% of normal level, P < 0.05) and Day 10 (121 ± 6% versus 101 ± 4% of normal level, P < 0.01). The renal HGF level of the pneumonectomized mice was initially decreased on Days 1 and 3 (80 ± 4%; P < 0.01, and 83 ± 6%; P < 0.05, respectively, of normal level in the kidney) then increased to almost a normal level on Days 5-10. Therefore, the protein levels of hepatic and renal HGF decreased transiently during 3 d after surgery, whereas mRNA levels in these organs increased during this period (Figures 2B and 2C). A potential explanation for this observation is that HGF sequestered in the liver and kidney might be mobilized into the blood circulation in response to pneumonectomy, to be followed by de novo synthesis of hepatic and renal HGF to compensate for early decreases and/or in response to inductive stimuli for HGF expression following pneumonectomy. Likewise, there is a discrepancy between changes in mRNA expression and protein levels of HGF: although HGF mRNA levels in the remaining lung and liver on Day 10 after left pneumonectomy were lower than their maximal levels, HGF protein levels in these tissues increased during 5-10 d after surgery. Although we have no clear explanation for this discrepancy, HGF in these tissues may be stored intracellularly and/or HGF secreted from cells may be sequestered extracellularly, because HGF has a relatively high affinity for proteoglycans and other extracellular matrix proteins (31).

View larger version (21K): [in this window] [in a new window]   Figure 3.   Changes in tissue HGF levels in the right lung (A), liver (B), and kidney (C) following left pneumonectomy. Mice were killed on Day 0 (before surgery; control), 1, 3, 5, and 10. Tissue extracts were obtained from mice with either left pneumonectomy (Pneum, closed circles) or sham left thracotomy (Sham, open circles) at each time point and HGF levels in tissue extracts were measured using ELISA. Each value represents the mean ± SEM of values obtained using 4-6 mice at each time point. *P < 0.05, **P < 0.01 versus control. P < 0.05, P < 0.01 versus sham (Student's t test).

Plasma HGF Level

Plasma HGF levels in the sham-operated mice rapidly increased to a maximum on Day 1, then decreased and were maintained at slightly higher levels than normal (Figure 4). This transient increase in plasma HGF level seems to coincide with an early transient induction of HGF in the lung of the sham-operated mice (Figures 2A, 3A, and 4). Incision of muscles in the chest and neck following thoracotomy and tracheostomy might be involved in the initial increase in the plasma HGF level (32, 33), as well as in the transient increase in the remaining lung (Figures 2A and 3A). In contrast, plasma HGF levels of pneumonectomized mice were markedly elevated as early as Day 1, and reached a peak level on Day 3. Thereafter, there was a decrease in plasma HGF levels, but they remained higher than those seen in sham-operated mice for 10 d after the surgery. The maximum level on Day 3 was ~ 3-fold higher than that of normal mice, and plasma HGF levels in pneumonectomized mice were significantly higher than those of sham-operated mice on Day 3 (1.12 ± 0.06 ng/ml versus 0.62 ± 0.07 ng/ml, P < 0.01) and Day 5 (0.96 ± 0.08 ng/ml versus 0.63 ± 0.06 ng/ml, P < 0.01). Increased plasma HGF levels may be attributed partly to hepatic and renal HGF released into the circulating blood, as suggested by early decreases in these organs.

View larger version (15K): [in this window] [in a new window]   Figure 4.   Changes in plasma HGF levels following left pneumonectomy. Plasma was obtained from mice with either left pneumonectomy (Pneum, closed circles) or sham left thoracotomy (Sham, open circles), and plasma HGF levels were measured using ELISA. Mice were killed on Day 0 (before surgery; control), 1, 3, 5, and 10. Each value represents the mean ± SEM of values obtained using 4-6 mice at each time point. *P < 0.05, **P < 0.01 versus control. P < 0.05, P < 0.01 versus sham (Student's t test).

Expression of c-Met/HGF Receptor mRNA

To determine if c-Met/HGF receptor expression is regulated in the remaining lung and distant organs, we analyzed changes in c-Met/HGF receptor mRNA expression in these organs after surgery, using real-time quantitative RT-PCR (Figures 5A, 5B, and 5C). In the remaining lung following pneumonectomy, the c-Met/HGF receptor mRNA expression transiently increased on Day 3, being 2-fold higher than that of controls (194 ± 12% of normal level, P < 0.001); however, c-Met/HGF receptor mRNA induction in the lung was not evident in sham-operated animals. c-Met/HGF receptor mRNA levels in the liver and kidney did not change. Thus the lung-specific enhancement of c-Met/HGF receptor mRNA expression after pneumonectomy correlates with the increased DNA synthesis of the epithelia in the remaining lung. To specify localization of cells which express the c-Met/HGF receptor in the remaining lung 5 d after pneumonectomy, tissue sections were examined immunohistochemically (Figures 5D, 5E, 5F, and 5G). c-Met/HGF receptor expression was localized in airway epithelial cells and alveolar cells, tissues likely to be type II pneumocytes and mostly located in corners of alveoli (Figures 5F and 5G).

View larger version (19K): [in this window] [in a new window]   View larger version (145K): [in this window] [in a new window]   Figure 5.   Changes in c-Met/HGF receptor mRNA expression in the right lung (A), liver (B), and kidney (C), and expression of the c-Met/HGF receptor in the right lung (D-G) following left pneumonectomy. (A-C) Changes in the c-Met/HGF receptor mRNA expression. Total RNAs were obtained from each tissue of mice with either left pneumonectomy (Pneum, closed circles) or sham left thoracotomy (Sham, open circles) and c-Met/HGF receptor mRNA levels were determined using real-time quantitive RT-PCR. Mice were killed on Day 0 (before surgery; control), 1, 3, 5, and 10. Each value (% of control) represents the mean ± SEM of values obtained using 4-6 mice at each time point. P < 0.001 versus control and sham (Student's t test). (D-G) Immunohistochemical staining for the c-Met receptor in the right lung 5 d after left pneumonectomy. D and E: negative control; F and G: tissue sections were subjected to immunohistochemical staining for the c-Met/HGF receptor using anti-c-Met/HGF receptor antibody. Most airway and alveolar type II epithelial cells stained for the c-Met/HGF receptor, whereas the endothelial cells were negative for such staining. Original magnifications: D-G, ×200.

Change in Compensatory Cell Proliferation by Neutralization of HGF

Changes in the expression of HGF and the c-Met/HGF receptor after left pneumonectomy suggested that HGF might play a role in compensatory lung regeneration. To determine the involvement of endogenous HGF on compensatory proliferation of alveolar and airway epithelial cells in a remaining lung, we administered neutralizing anti-HGF rabbit IgG or normal rabbit IgG to pneumonectomized mice and examined effects on DNA synthesis in these epithelial cells. Pneumonectomized mice were treated with 200 and 600 µg/head of an anti-HGF antibody or identical doses of normal IgG, and subjected to analysis of DNA synthesis, using BrdU incorporation and subsequent immunohistochemistry (Figure 6). In alveolar epithelial cells, the BrdU labeling index in the anti-HGF IgG-treated mice was significantly lower than in the normal IgG-injected mice on Day 3 (0.56 ± 0.04% at 200 µg/head and 0.60 ± 0.04% at 600 µg/head versus 0.86 ± 0.08%, P < 0.05) and Day 5 (1.34 ± 0.10% at 200 µg/head and 1.39 ± 0.12% at 600 µg/ head versus 2.43 ± 0.09% in control, P < 0.01) (Figures 6A, 6B, and 6C). In airway epithelial cells, the labeling index in anti-HGF IgG-treated mice showed no significant difference from that in mice treated with normal IgG on Day 3; however, neutralization of endogenous HGF reduced the labeling index on Day 5; 1.69 ± 0.15% at 200 µg/head and 1.63 ± 0.25% at 600 µg/head versus 2.23 ± 0.16% in control (P < 0.05) (Figure 6D). Thus, neutralization of endogenous HGF suppressed proliferation of lung epithelial cells after pneumonectomy, although it was less effective in airway epithelial cells than in alveolar epithelial cells. To determine if neutralization of HGF would result in anomalous lung morphology, proliferative change in the interstitial region was also evaluated. There was no significant difference in proliferation of interstitial cells between the HGF-neutralized group and the placebo control group (data not shown).

View larger version (49K): [in this window] [in a new window]   Figure 6.   Effect of neutralizing antibody against HGF on DNA synthesis of epithelial cells in the remaining right lung following left pneumonectomy. (A and B) Distribution of cells undergoing DNA synthesis on Day 5 in alveolar regions of the remaining lung of mice administered normal IgG (A) or anti-HGF IgG (B). Original magnifications: ×200. (C and D) Changes in the number of cells undergoing DNA synthesis (BrdU labeling index) in alveolar (C) and airway (D) epithelial cells. Mice were placed into three groups as follows: normal IgG, mice given 200 µg/head of normal IgG (open bars); anti-HGF/200, mice given 200 µg/head of anti-HGF IgG (closed bars); anti-HGF/600, mice given 600 µg/ head of anti-HGF IgG (hatched bars). Each value represents the mean ± SEM of values obtained using 4-6 mice at each time point. *P < 0.05, **P < 0.01 compared with the value in normal IgG-injected group (Student's t test).

Mitogenic Effect of Recombinant HGF on Epithelial Cells in the Remaining Lung

Based on the above results, we administered human recombinant HGF or saline alone as a placebo to pneumonectomized mice and examined the effect of exogenous HGF on DNA synthesis of cells in the remaining lung. In alveolar regions, HGF significantly stimulated DNA synthesis of cells on Day 3 (Figure 7C). Appearance and distribution of cells undergoing DNA synthesis in alveolar regions on Day 3 indicated that these cells seemed to be mostly alveolar type II epithelial cells (Figures 7A and 7B). On Day 3, the labeling index in alveolar cells was stimulated by HGF to levels 2-fold higher than those seen in control mice (1.66 ± 0.13% versus 0.83 ± 0.06% in placebo group, P < 0.01). In airway regions, HGF stimulated DNA synthesis of airway epithelial cells on Days 1-3, whereas statistically significant stimulatory effects were seen on Day 3 (1.73 ± 0.16% versus 1.31 ± 0.10% in placebo group, P < 0.05). The stimulatory effect of HGF on DNA synthesis was not seen on Day 5 in either alveolar or airway epithelial cells. In addition, the stimulatory effect of HGF on DNA synthesis of interstitial cells was not seen, compared with findings in the saline-injected control group (data not shown). Likewise, it is noteworthy that enhanced cell proliferation by HGF did not lead to abnormal morphology, including interstitial expansion that might impair gas exchange capacity.

View larger version (55K): [in this window] [in a new window]   Figure 7.   Effect of recombinant HGF on DNA synthesis of epithelial cells in the remaining lung following left pneumonectomy. (A and B) Distribution of cells undergoing DNA synthesis on Day 3 in alveolar regions of the remaining lung of mice administered saline alone (A) or recombinant HGF (B). Cells undergoing DNA synthesis were detected by BrdU incorporation and subsequent immunohistochemical staining using and anti-BrdU monoclonal antibody. Original magnification: ×200. (C and D) Changes in the number of cells undergoing DNA synthesis (BrdU labeling index) in alveolar (C) and airway (D) epithelial cells in the remaining lung in the HGF-treated group (HGF, mice given 500 µg/kg of human recombinant HGF dissolved in saline, closed bars) and control group (Saline, mice given identical volume of saline, open bars). Each value represents the mean ± SEM of values obtained using 4-6 mice at each time point. *P < 0.05, **P < 0.01 compared with the value in normal saline-injected group (Student's t test).

Effects of Neutralization or Administration of HGF on Lung Weight

To determine whether neutralization or administration of HGF alters increase in weight of the remaining right lung, we determined lung weight (expressed as the ratio of wet lung weight to body weight) during the compensatory lung growth (Table 1). Neutralization of endogenous HGF with daily administration of antibody (200 µg/head) significantly attenuated the increase in lung weight on Day 5 (P < 0.05), whereas administering recombinant human HGF enhanced increase in lung weight on Day 3 (P < 0.05). These results strongly suggest that HGF has stimulatory effects on actual compensatory growth of the lung mass. Differences in lung weight between the HGF-neutralized or HGF-administered groups and their respective control groups were less compared with those in the BrdU labeling index in epithelial cells. This might be because HGF is involved in the mitogenic response predominantly in the epithelial cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

HGF plays a role as a mitogen for both alveolar and bronchial epithelial cells (17). Using distinct approaches with a neutralizing anti-HGF antibody and recombinant HGF, we obtained data for attenuation and enhancement of compensatory proliferation of both alveolar type II and airway epithelial cells, as well as lung weight, respectively, by neutralization and supplementation of HGF. Together with expression of the c-Met/HGF receptor in these epithelial cells in the remaining lung and changes in expression of HGF in response to unilateral pneumonectomy, we propose that HGF has a role as a pulmotrophic factor for compensatory regeneration of the mature lung after unilateral pneumonectomy.

Changes of HGF mRNA in the remaining lung, liver, and kidney during compensatory lung growth were evident. HGF mRNA was increased not only in the remaining lung, but also in the liver and kidney after pneumonectomy. Interestingly, previous studies noted that HGF mRNA was induced in intact distant organs following partial hepatectomy and unilateral nephrectomy, as well as in the remaining liver and kidney, respectively (24, 34). Although the inducers and mediators in this response between distal organs remain to be determined, the involvement of humoral mediators has been implicated in this type of upregulation of HGF (35). Therefore, our results suggest the potential involvement of an endocrine-like action of HGF mediated by humoral inductive signals for compensatory regeneration of organs following partial resection, including unilateral pneumonectomy. On the other hand, protein levels of HGF in the liver and kidney transiently decreased following pneumonectomy, whereas pulmonary HGF levels increased in a time-dependent manner. One potential explanation is that hepatic and renal HGF might be mobilized into the blood circulation and be involved in increased plasma HGF levels and compensatory cell proliferation in the remaining lung in response to pneumonectomy. Other investigators reported that mobilization of HGF in the remaining liver is involved in compensatory regeneration after partial hepatectomy, and that the mobilization of HGF is mediated by activation of urokinase-type plasminogen activator, presumably because this event stimulated the release of HGF from the hepatic extracellular matrix (38).

Although HGF mRNA expression in the liver and kidney increased with concomitant elevation of plasma HGF levels, cells in the remaining lung (Figures 1B, 1C, 1D, and 1E), but not in the liver and kidney (not shown), specifically underwent compensatory DNA synthesis. Likewise, previous studies demonstrated that internalization of the c-Met/HGF receptor, a general process that occurs following ligand-dependent activation of receptor tyrosine kinases (20), specifically occurred in the remaining tissue after hepatectomy or nephrectomy, even though plasma HGF levels increased following resection of these organs (39). Because the specific internalization of the c-Met/ HGF receptor may be the track of ligand-dependent activation of the c-Met/HGF receptor, these results do suggest that HGF specifically acts on the remaining organs, under these conditions. Although mechanisms by which HGF selectively activates the c-Met/HGF receptor in the remaining tissues after partial resection of organs are unknown, some possibilities can be given consideration. In mature hepatocytes in primary culture, HGF exerts mitogenic actions in a cell density-dependent manner. HGF potently enhances DNA synthesis of hepatocytes cultured at a low cell density but does not do so in case of a high cell density (40). Cell-cell adhesion and communication may regulate the function of the c-Met/HGF receptor, such that HGF exerts biologic activities in an injured tissue-specific manner. In addition, in contrast to increase in HGF mRNA expression, c-Met/HGF receptor mRNA expression specifically increased in the remaining lung but not in the liver and kidney. Therefore, local upregulation of the c-Met/ HGF receptor may possibly make HGF susceptible to this receptor in the remaining lung. A similar upregulation of the c-Met/HGF receptor expression in the kidney was noted in cases of unilateral nephrectomy and nephrotoxin-induced acute renal injury (41).

Neutralization of HGF inhibited DNA synthesis of alveolar and airway epithelial cells and the increase in the lung weight in compensatory lung growth. However, this inhibitory effect was partial and incomplete, which suggests that other pulmotrophic factors may also participate in proliferation of these cells. There are data on the stimulatory effects of growth factors on compensatory lung growth, including insulin-like growth factor-I (IGF-I) (11) and epidermal growth factor (EGF) (13): however, direct evidence for effects of these factors on the proliferation of pneumocytes or airway epithelial cells has not been documented in cases of postpneumonectomy lung growth. Changes observed in lung IGF-I expression following pneumonectomy did not represent major contributions to the regulation of compensatory lung growth (12). Although the volume and weight of the remaining lung following pneumonectomy were elevated by administration of EGF, histologic evaluation that indicates the effect of EGF on proliferation of lung cells has not been reported (13). The potential involvement of other mitogenic growth factors for lung cells such as acidic fibroblast growth factor (42), keratinocyte growth factor (43), and heparin-binding EGF-like growth factor (44) in compensatory lung regeneration remains to be addressed.

In addition to compensatory lung regeneration following pneumonectomy, involvement of HGF on regeneration and protection of the lung after acute lung injuries has been noted. HGF mRNA expression was rapidly induced in the injured lung following transtracheal hydrochloride challenge (45, 46), and administration of HGF stimulated DNA synthesis of alveolar and bronchiolar epithelial cells after acute lung injury (19, 46). Expression of HGF mRNA markedly increased in the injured lung after pulmonary ischemia-reperfusion, and neutralization of HGF greatly enhanced the pulmonary injury and thus retarded tissue regeneration (47). Likewise, increased serum HGF levels were noted in patients with lung diseases (45) and pneumonectomy or pulmonary lobectomy (27). Moreover, HGF ameliorated the onset of lung fibrosis induced by bleomycin (48, 49). Taken together with the present study, HGF plays an important role as an intrinsic pulmotrophic factor for lung regeneration following lung injuries or pneumonectomy. HGF as a therapeutic factor deserves attention as it may prevent pulmonary fibrosis, mitigate ischemia-reperfusion injury in the setting of lung transplantation, and accelerate compensatory growth of the remaining lung after pulmonary resection.