Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The integrin v6 is a receptor for the extracellular matrix proteins fibronectin (1, 2), tenascin (3, 4), and vitronectin (X. Huang and D. Sheppard, unpublished observation). Expression of this integrin appears to be restricted to a subset of epithelial cells (5). In most normal adult tissues, v6 is expressed at low or undetectable levels, but expression is dramatically increased during development, following injury, and in a variety of epithelial neoplasms. Inactivation of the 6 subunit gene in mice produced a surprising phenotype characterized by focal accumulation of macrophages in the skin and persistent accumulation of activated lymphocytes in the lungs (6). These anatomic abnormalities were associated with baldness and exaggerated airway responsiveness to acetylcholine, suggesting that these inflammatory effects of the gene inactivation were functionally significant. These data suggested a previously unexpected role for this epithelial integrin in downregulating local inflammation, and raised the possibility that interventions targeting this integrin could affect the course of inflammatory disorders of the lungs and skin.

Identification of an unexpected phenotype by targeted gene inactivation raises the possibility that the targeting strategy used might have resulted in the inadvertent inactivation of adjacent genes. Furthermore, our previous results did not permit identification of the specific subset(s) of pulmonary epithelial cells responsible for the observed phenotype. To address each of these questions and to determine whether constitutive expression of v6 would itself produce any lung phenotype, we generated a line of mice constitutively expressing a human 6 transgene under the control of the human surfactant protein C (SPC) promoter, which previously has been shown to produce gene expression limited to alveolar type II cells and bronchiolar epithelial cells (7, 8). We then crossbred these transgenic mice with 6^/ mice and examined the phenotypes of littermates expressing or not expressing the transgene in mice homozygous for the wild type (6^+/+) or inactivated (6^/) allele. The transgene itself had no effect in wild-type mice but largely reversed the increases in bronchoalveolar lavage (BAL) cell number, BAL lymphocytosis, and lymphocyte and macrophage activation in 6^/ mice. These data definitively demonstrate that inactivation of the 6 subunit gene itself is responsible for the exaggerated airway inflammation seen in these mice and suggest that limited expression of v6 on alveolar type II cells and/or bronchiolar epithelial cells is sufficient to reverse most of the pulmonary abnormalities caused by inactivation of this gene.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

DNA Constructs and Generation of Transgenic Mice

To construct a human 6 expression vector, full-length human 6 cDNA was excised from the previously described plasmid pBluescript6 (2) by digestion with XhoI and XbaI and transferred to pMAMneoblue (Clontech, Palo Alto, CA) to obtain flanking SalI sites. The 6 cDNA was then removed by digestion with SalI and subcloned into the SPC expression plasmid pUC18SPC3.7 (9) (a gift from Jeffrey Whitsett, University of Cincinnati, Cincinnati, OH). The resultant plasmid, SPCh6, consisted of 3.7 kb of the human SPC promoter followed by human 6, the small t intron from SV40, and the SV40 polyadenylation sequence (Figure 1A). A fragment consisting of this expression cassette was digested away from the plasmid backbone with NdeI and NotI and purified by agarose gel electrophoresis.

View larger version (20K): [in this window] [in a new window]   Figure 1.   (A) SPCh6 transgene construct. Human 6 gene was fused to the 3' end of the SPC promoter. The expression fragment consisted of 3.7 kb of the human SPC promoter followed by human 6, the small t intron from SV40, and the SV40 polyadenylation sequence. (B) Southern blot analysis of mouse tail DNA digested with BamHI. The blot was probed with a fragment of 32P-labeled human 6 cDNA, and the expected 3.6-kb band was detected from transgenic mice.

Transgenic mice were prepared in F1 offspring of C57BL/6 and SJL strain mice by pronuclear injection. Transgene positive animals were identified by Southern blot analysis of BamHI-digested genomic DNA using a ³²P-labeled 2,000 nt human 6 cDNA fragment as a probe. Because this line was originally generated in advance of the 6^/ mice for studies that were intended to be performed in A/J strain mice, mice carrying the 6 transgene were backcrossed for one generation with A/J strain mice. However, when initial evaluation of these mice revealed no phenotypic effects of the transgene itself, the line was maintained by continuous inbreeding. For the studies described in this report, offspring of this inbred line expressing a single copy of the transgene were initially crossed with 6^/ mice. 6^+/ offspring that also expressed the SPCh6 transgene were then intercrossed to produce a series of littermates that were homozygous for 6 wild-type or null alleles in the presence or absence of the human transgene.

Generation of Rabbit Monoclonal Antibodies

Rabbits were immunized subcutaneously with recombinant secreted human v6 (2) in Freund's adjuvant. Rabbit splenocytes were harvested and fused with 240E rabbit plasmacytoma cells (10). Supernatants generated by individual hybridomas were screened by flow cytometry with 6- and mock-transfected SW480 cells (2). Antibodies found to recognize 6-transfected but not mock-transfected SW480 cells were further characterized for reactivity with human and murine 6 by Western blotting of human cell lines and murine tissues and by flow cytometry of murine keratinocytes.

Histology and Immunohistochemistry

Freshly isolated organs were embedded in ornithine carbamyl transferase and quick frozen in liquid nitrogen. Serial 5-µm sections were prepared and fixed in 2% paraformaldehyde (Fisher Scientific) for hematoxylin and eosin staining. For immunohistochemistry, frozen sections were fixed in cold acetone for 5-10 min and air dried. Sections were blocked for endogenous peroxidase and biotin activities with Peroxoblock solution (Zymed Labs, South San Francisco, CA) and Avidin/Biotin Blocking Kit (Vector, Burlingame, CA) at room temperature. After rinsing, sections were blocked with 0.25% casein/0.025% thimerosal in phosphate-buffered saline (PBS) for 15 min and then incubated overnight at 4°C in rabbit antihuman 6 monoclonal antibody. After washing, sections were incubated in biotin- labeled goat antirabbit antibody followed by ABC avidin/ peroxidase reagent (Vector) for 1 h at room temperature. Chromagen was developed using the DAB Plus Kit (Zymed). Finally, sections were dehydrated and mounted with permount onto clean slides.

Western Blot Analysis

Freshly isolated mouse lungs were minced and then homogenized in lysis buffer (200 mM n-octylglucoside in 100 mM Tris buffer). The homogenates were centrifuged and supernatants were saved. SW480 cells were lysed in Laemmli sample buffer. Aliquots containing 30 µg of total protein were separated on a 7.5% polyacrylamide gel and transferred onto Immobilon membrane (Millipore, Bedford, MA) using a Hoefer transfer apparatus. The membrane was blocked with 5% powdered milk for 2-4 h at room temperature and incubated with rabbit anti-6 antibody followed by HRP-conjugated goat antirabbit secondary antibody. The membrane was exposed to film after a brief incubation in Luminol (Amersham, Arlington Heights, IL).

Bronchoalveolar Lavage and Flow Cytometry

Mice were killed by cervical dislocation and a blunt needle was inserted into the upper trachea. Lavage was performed by introducing five sequential 0.8-ml aliquots of PBS into the lungs and carefully withdrawing the fluid. The BAL fluid was centrifuged at 130 × g for 5 min and the cell pellets were resuspended in 1-ml red blood cell lysis buffer (Sigma Chemical Co., St. Louis, MO). Total cells were counted in a hemocytometer and a Cytospin slide preparation was made from an aliquot of each sample. Slides were stained with Diff-Quik stain set (Dade Diagnostics of Puerto Rico Inc., Aguada, PR), and a differential count of 300 cells was made.

The remaining BAL cells were blocked with normal goat serum (Vector) at 4°C for 10 min and then labeled with monoclonal antibodies against murine CD4 (fluorescein isothiocyanate [FITC], conjugated), CD25 (phycoerythrin conjugated), or integrin M (FITC conjugated) (Caltag, South San Francisco, CA) for 20 min at 4°C. After washing twice with PBS, stained cells were resuspended in 100 ml of PBS and analyzed by flow cytometry on a Becton Dickinson FACSort (San Jose, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of Rabbit Monoclonal Antibodies to v6

Supernatants from hybridomas generated from rabbits immunized with secreted v6 were initially screened for differential staining of mock- and 6-transfected SW480 cells (2). Flow cytometry demonstrating specificity of three clones, B1, 4B5, and D5, is shown in Figure 2. One of these, clone B1, also recognized murine v6, as shown by staining of keratinocytes from wild-type mice (Figure 2). Antibodies B1 and 4B5 were further characterized for their ability to recognize the 6 subunit itself by Western blotting of cell lysates of murine lung and 6-transfected SW480 cells. Both antibodies recognized bands of the appropriate molecular mass to be mature and immature human 6 in lysates of 6- but not mock-transfected SW480 cells (Figure 3A). B1 also recognized a band of the appropriate molecular mass to be 6 in lysates of lung tissue from wild-type, but not 6^/ mice (Figure 3B). Mature and immature forms were not resolved on this gel.

View larger version (31K): [in this window] [in a new window]   Figure 2.   Flow cytometry of rabbit monoclonal antibodies. SW480 cells or murine keratinocytes were stained with no antibody (white peaks) or with antibodies B1, 4B5, or D5 (black peaks) followed by PE-labeled goat antirabbit IgG, and then subjected to flow cytometry.

View larger version (15K): [in this window] [in a new window]   Figure 3.   (A) Western blot analysis of B1 and 4B5 on SW480 cells. Lysates from mock- and 6-transfected SW480 cells were separated by 7.5% sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) under nonreducing conditions, transferred to Immobilon membranes and probed with antibodies B1 or 4B5. (B) Western blot analysis of mouse tissue. Lysates from lungs of 6+/+ and 6/ mice were separated by 7.5% SDS-PAGE under nonreducing conditions, and transferred and probed with antibody B1 as before. The expected apparent molecular mass of mature 6 is denoted to the left of each blot. The lower band seen in A corresponds to the expected apparent molecular mass of incompletely glycosylated 6.

The Human 6 Transgene Is Expressed in Mouse Lung

Mice generated by injection of a plasmid containing the human 6 transgene under the control of the human SPC promoter were initially screened by Southern blot analysis of tail DNA (Figure 1B). Offspring of transgene positive founders were screened for protein expression by immunohistochemistry of lung tissue with antibodies that recognized human, but not murine, v6. The founder line demonstrating the highest level of protein expression was maintained for subsequent experiments. The lungs of these SPCh6 transgenic mice demonstrated the expected pattern of human 6 expression in alveolar type II cells and bronchiolar epithelial cells (Figure 4A), and the expression was detected in the lungs but not in the kidneys (Figure 4B). The SPCh6 transgenic mice were able to reproduce and grow normally, and there were no gross or histologic abnormalities of the lungs, heart, liver, kidney, spleen, or intestine of any of the SPCh6+ mice analyzed.

View larger version (50K): [in this window] [in a new window]   Figure 4.   (A) SPCh6 transgene expression in murine alveolar type II cells and bronchiolar epithelial cells. Acetone-fixed frozen sections from 6/ mice with or without expression of SPCh6 were stained with antibody 4B5 that recognizes human 6 subunit. Cells in the appropriate anatomic locations to be alveolar type II cells (arrowhead) and bronchiolar epithelial cells (arrow) were stained by this antibody in lungs from SPCh6+ but not in SPCh6 mice. (B) Western blotting of human 6. Lysates from kidney (K) and lungs (L) of 6/ mice with or without expression of SPCh6 were separated by 7.5% SDS-PAGE under nonreducing conditions and then blotted with antibody 4B5.

6^/ Mice of Mixed Genetic Background Demonstrate Airway Eosinophilia in Addition to the Previously Described Features of Airway Inflammation

In a previous report conducted in intercrosses of 129 and C57Bl/6 mice, we described the development of lymphocytic airway inflammation in 6^/ mice (6). We have subsequently observed the presence of foamy macrophages expressing the murine activation marker integrin M in these mice, as well as in purebred 129 or C57Bl/6 animals. However, because the SPCh6 transgenic animals described in this study were generated in advance of the 6^/ mice and in a different genetic background, it was important to determine whether the phenotype we originally described would also be expressed in the complicated genetic background that resulted from crossing 6^/ mice with the SPCh6 transgenics. To address this issue, we generated 6^+/ mice that were heterozygous for the SPCh6 transgene, and then intercrossed these animals to generate 6^// SPCh6+, 6^+/+/SPCh6+, 6^//SPCh6, and 6^+/+/ SPCh6 mice for use in the present study. In this fashion, any phenotypic differences due to differences in genetic background would be randomly distributed in all four groups. In comparison with 6^+/+/SPCh6 mice, 6^//SPCh6 mice had marked increases in the total number of cells obtained by BAL, in the numbers of CD4+ lymphocytes, in the number of lymphocytes expressing the activation marker CD25 (Figure 5), and in the numbers of M expressing alveolar macrophages (Figure 6), all features shared by 6^/ mice in the 129, C57Bl/6, and 129 by C57Bl/6 backgrounds. Surprisingly, in this genetic background 6^/ mice also demonstrated a marked increase in eosinophils in BAL fluid (Figure 7).

View larger version (18K): [in this window] [in a new window]   Figure 5.   Effects of 6 knockout and SPCh6 expression on total cell number, CD4+ and CD4/CD25 double positive cells in BAL fluid. Mean ± SEM of total cells obtained from BAL of each of six mice in the four experimental groups was counted and then stained with antibodies for murine CD4 and the T-cell activation marker, CD25. N.D. denotes none detected.

View larger version (20K): [in this window] [in a new window]   Figure 6.   Effects of 6 knockout and SPCh6 on expression of the murine macrophage activation marker integrin M. Cells obtained from BAL of animals from each of the four experimental groups were stained with no antibody (white peaks) or antibody specific for mouse-activated macrophage marker-anti-integrin M (black peaks). A demonstrates the flow cytometric histograms obtained from a representative animal from each group. B demonstrates the difference of mean fluorescence intensity on stained and unstained BAL macrophages (net MFI ± SEM) calculated for cells obtained from six animals in each of the four experimental groups.

View larger version (20K): [in this window] [in a new window]   Figure 7.   Effects of 6 knockout and SPCh6 on differential cell counts of BAL cells. Cells obtained by BAL from six mice in each experimental group were centrifuged onto glass slides and stained with Diff-Quik. Differential cell counts were performed on 300 cells from each slide. Data are expressed as means ± SEM. N.D. denotes none detected.

The Expression of Human 6 Largely Rescues the BAL Abnormalities in 6^/ Mice

Expression of the human 6 transgene in 6^+/+ mice had no effect on the number of cells obtained by BAL. As in transgene negative 6^+/+ mice, nearly all of the BAL cells were macrophages, and these macrophages did not express the activation marker integrin M. However, expression of the 6 transgene in the distal lung epithelium of 6^/ mice largely corrected most of the BAL abnormalities seen in these animals, reducing CD4+ lymphocytes and CD25-expressing lymphocytes by more than 80% (Figure 5). The increased M expression on BAL macrophages of 6^/ mice was also inhibited by expression of the human 6 transgene (Figure 6). Similar effects were seen for morphologic evidence of macrophage activation. Most of the BAL macrophages from 6^//SPCh6 mice demonstrated varying amounts of vacuolated cytoplasm, whereas BAL macrophages from 6^//SPCh6+ mice were much more uniform, similar to the macrophages from wild-type mice (data not shown). The increase in BAL eosinophils seen in these mice was also reduced by expression of the human 6 transgene (Figure 7).

We have previously described airway lesions in 6^/ mice composed of accumulations of lymphocytes around conducting airways and adjacent veins (6). As we previously described in 129 mice by C57Bl/6 intercrosses, these lesions were not seen in any of the 10 6^+/+ mice evaluated in the present study. Among the 6^/ mice, 14 of 14 6^// SPCh6 mice had such lesions, whereas lesions were only apparent in 6 of 14 6^//SPCh6+ mice.

As expected, all of the 6^//SPCh6+ and the 6^// SPCh6 mice developed the same pattern of inflammatory baldness we have previously reported in 6^/ mice generated in the 129 mice by C57Bl/6 background. No baldness was seen in any of the mice homozygous for the wild-type 6 allele or in any of the 6 heterozygotes.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of the present study demonstrate that limited expression of a human 6 transgene in alveolar type II cells and bronchiolar epithelial cells results in dramatic reversal of many of the features of airway inflammation seen in 6^/ mice. These data strongly suggest that the airway inflammation caused by inactivation of the 6 subunit gene is a direct consequence of the loss of 6 expression and not due to inadvertent inactivation of other nearby genes. These data confirm a role for the product of this gene, the v6 integrin, in the control of tissue inflammation. The fact that this rescue was accomplished by expression of a transgene in only a small minority of respiratory epithelial cells present in the most distal regions of the lung suggests that control of local inflammation does not require v6 expression in all airway epithelial cells. These results also raise the possibility that induction or transfer of the 6 gene in a fraction of respiratory epithelial cells (as might be accomplished by gene therapy, for example) could produce a meaningful impact on lung inflammation. Whether similar effects on inflammation could be accomplished by expressing the 6 subunit in epithelial cells in other regions of the airway cannot be determined from the current study.

An interesting sidelight of the current report is the observation that inactivation of the 6 subunit gene resulted in the accumulation of eosinophils in the airways, in addition to the activated lymphocytes and macrophages we have previously observed. This finding is probably a fortuitous consequence of the complex genetic background of the mice used in these studies, which included contributions from SJL and A/J strains in addition to 129 and C57Bl/6 types. This finding suggests that loss of 6 interacts with other genes that may be differentially expressed among these strains of mice to affect the cellular composition of lung inflammation. Identification of which strain or strains are responsible for this effect must await studies of the effects of 6 inactivation in purebred SJL and A/J mice.

Expression of the SPC-driven human transgene used in these studies largely inhibited the effects of inactivation of the 6 subunit on macrophage and lymphocyte accumulation and activation in BAL fluid. However, the airway eosinophilia and the peribronchial accumulations of lymphocytes that resulted from inactivation of this gene were not completely prevented by expression of the human transgene. These results, especially the failure to prevent completely the peribronchial and perivascular accumulation of lymphocytes, should not be surprising given the significant physical distance between some of these lesions and the closest bronchiolar epithelial cells and alveolar type II cells. The results suggest either that 6 expression in more proximal airway cells would be necessary to rescue these aspects of the knockout phenotype completely, or that the level of transgene expression achieved in the mice studied was too low to reverse these effects completely.

It would clearly be of interest to know whether distal lung expression of the 6 subunit would be sufficient to prevent the airway hyperresponsiveness to acetylcholine that we previously described as a consequence of inactivation of the 6 subunit gene. Unfortunately, this question could not be addressed in the current study because of the complicated genetic background of the mice studied. Marked genetically determined differences in acetylcholine sensitivity have been described previously between A/J and C57Bl/6 mice. Therefore, it is not surprising that when we attempted to examine airway responsiveness in these groups of animals, there was a broad range of responses among 6^+/+ animals, a range in excess of the effects of 6 inactivation we reported previously. Analysis of the effects of SPC-driven expression of human 6 on this aspect of the phenotype must thus await the development of SPCh6 transgenics on a pure genetic background.

The dramatic effect of expressing the 6 subunit in a small minority of airway epithelial cells seen in this study, including protection from inflammatory lesions that develop at some distance from the alveolar lumen, suggests that effects of epithelial v6 in inhibiting lung inflammation are likely due to induction of one or more secreted anti-inflammatory factors from lung epithelial cells. The respiratory epithelium has the capacity to secrete a number of potentially anti-inflammatory cytokines and growth factors, including transforming growth factor- (TGF-) (11, 12), IL-6 (13), IL-10 (16), and IL-11 (17). Thus far, we have not been able to identify any systematic differences in expression of these known anti-inflammatory factors that would explain the phenotypic differences between wild-type and 6^/ mice. Until such factor or factors are identified, hypotheses about the molecular mechanisms by which epithelial v6 regulates tissue inflammation will remain speculative. Nonetheless, the dramatic reduction in inflammatory cells that resulted from limited expression of v6 in this study suggests that interventions aimed at augmenting 6 expression or enhancing its effects could be useful for the treatment and/or prevention of pulmonary inflammation.