Background
Lung morphogenesis requires coordinated input from a multitude of diverse molecules ranging from transcription factors to growth factors to cytokines and extracellular matrix. Basement membranes are specialized extracellular matrices that have vital roles in cell adhesion, migration, differentiation, as well as in tissue organization and development [
1]. Laminins, type IV collagen, entactin/nidogen, and sulfated proteoglycans are the main components of basement membranes. Laminins are heterotrimers composed of one α, one β, and one γ chain. To date, 5α chains, 4β chains, and 3γ chains are present in humans and mice and these laminin chains self-assemble to form at least 15 laminins [
2].
The lung is rich in laminin chains and all but the laminin γ3 chain (which is not present in lung basement membrane [
3]) are detected in the lung at some point during development and in adult lungs. The laminin α1-α5, β1-β3, and γ1-γ2 chains are present in embryonic lung; laminin α2-α5, β1-β3, and γ1-γ2 chains are present in the adult lung [
4‐
7]. Studies of lung development from our laboratory and others have shown that laminin and its interactions are crucial for lung morphogenesis. Epithelial-derived laminin chains are important for lung development since addition of either laminin-111 (formerly laminin-1) antibodies or proteolytic fragments to lung bud cultures perturbs branching morphogenesis [
8]. Interference with entactin/nidogen binding to laminin through ablation of the nidogen-binding site on laminin γ1 in vivo affects sacculation [
9]. Mesenchymal cell-derived laminin α2 is required for bronchial smooth muscle cell differentiation in vitro [
10]. Targeted deletion of laminin α5 in the mouse results in abnormal lobar septation, absence of visceral pleura basement membrane, and ectopic deposition of laminin α4 in lungs through embryonic day 16.5 at which time these mice die [
11]. Ablation of laminin α5 expression by lung epithelial cells alone via the SP-C promoter and the Cre/LoxP system enabled examination of lungs up to post-natal day 1. The lungs had grossly enlarged distal airspaces and markedly impaired distal epithelial cell differentiation [
11]. Thus, multiple in vitro and in vivo studies have shown that laminins are important for lung development at different stages.
Laminin γ2 which is unique to laminin-332 (formerly laminin-5), localizes to airway epithelial basement membranes during lung development leading to speculation that it is required for lung development [
12‐
14]. Laminin γ2 null
(Lamc2-/-) mice exhibit blistering and erosions of the skin and die a few days after birth presumably due to malnutrition as a result of blistering and erosions in the oral cavity [
15], but the lungs have not been described. In this investigation, we report findings regarding lung development in
Lamc2-/- mice.
Methods
Production, breeding and genotyping of mice
Production of
Lamc2-/- mice has been described previously in detail [
15].
Lamc2-/- mice were maintained on a mixed 129/C57BL/6J background. Timed matings of mice were established to produce
Lamc2-/- and littermate control offspring at various ages. Noon on the day of detection of a vaginal plug was designated as E0.5. Genomic DNA was obtained from tails using the Qiagen DNA kit (Qiagen, Carlsbad CA). Genotyping was completed by PCR with laminin γ2 specific primers, wild type forward 5'-CCG CTT GCT GAC TTG TAT CC-3',
Lamc2-/- forward 5'-AGC TAA TAC GGG TTC AGC C-3', reverse 5'-TGT AAC CAG AAG CAC ATT CC-3'. The Washington University Animal Studies Committee approved all experiments.
Antibodies
Rat monoclonal antibody to murine laminin α1 was from Dale Abrahamson (University of Kansas Medical Center). Rabbit polyclonal antibodies to laminin α3A and α3B were from Takako Sasaki (Max Planck Institute of Biochemistry, Martinsreid, Germany) [
16]. Rabbit polyclonal antibodies to entactin/nidogen [
17], laminin α5 [
4], laminin α4 [
18], and laminin γ2 [
19] were produced as described. Antibody against integrin α3 was from C. Michael DiPersio (Albany Medical College). Antibody against BP180 (also known as bullous pemphigoid BP antigen 2 and type XVII collagen, a major component of the epidermal anchoring complex) was from Zhi Liu (University of North Carolina, Chapel Hill, NC). Antibodies purchased from commercial suppliers were anti laminin α2 (4H8-2, Alexis Biochemicals, San Diego, CA), PECAM, integrin β4 (Pharmingen, San Diego, CA), laminin γ1, laminin β1, integrin α6, aquaporin-5, CC26, pro-surfactant protein C (SP-C) (Chemicon, Temecula, CA), β-tubulin IV (Biogenex, San Ramon, CA), and α-smooth muscle actin (Sigma, St. Louis, MO). FITC- and TRITC-conjugated anti-rabbit and anti-rat secondary antibodies were from Jackson ImmunoResearch Laboratories (West Grove, PA). FITC-anti-mouse IgG2b secondary antibody was from ICN Biomedicals, Inc. (Costa Mesa, CA).
Immunofluorescence microscopy
Newborn pups were sacrificed by decapitation and immediately immersed in Tissue Tek OCT embedding medium (Sakura Finetek, Torrance, CA), frozen in liquid nitrogen-cooled 2-methylbutane, and sectioned at 6 μm on a cryostat. Sections were blocked with 10% normal goat serum in 1% BSA/PBS, then incubated with primary antibody diluted in 1% BSA/PBS. Slides were washed with PBS, and incubated with secondary antibody diluted in 1% BSA/PBS. Slides were again washed with PBS and mounted in Vectashield (Vector Laboratories, Temecula, CA). For laminin α4 staining, sections were fixed in 4% paraformaldehyde for 10 minutes, washed in PBS, treated with 0.1 M glycine, pH 3.5, for 10 minutes, washed in PBS, treated with 0.1% SDS at 55°C for 1 hour, washed in PBS, blocked, and stained as with other antibodies. For PECAM staining, sections were fixed in 100% ethanol prior to addition of the PECAM antibody. Antibody dilutions were 1:500 for laminins α1-α4 , 1:600 for laminin α5, 1:200 for laminin γ1, 1:800 for laminin γ2, and 1:50 for aquaporin-5.
Electron microscopy
For transmission electron microscopy, lungs from newborn pups were fixed in 3% glutaraldehyde in 0.1 M sodium cacodylate buffer, postfixed with aqueous 1.25% osmium tetroxide, stained with 4% aqueous uranyl acetate, dehydrated through an ethanol series, embedded in Polybed, sectioned on a Reichert-Jung Ultra Cut, post-stained in 4% uranyl acetate and lead citrate, and viewed on a Zeiss 902 electron microscope. All reagents for electron microscopy were purchased from Electron Microscopy Sciences (Ft. Washington, PA) except Polybed (Polysciences, Warrington, PA).
Histology and immunohistochemistry
Thoraces of newborn pups were isolated by decapitation and transection at the level of the liver. Thoraces were fixed in 4% paraformaldehyde in PBS, dehydrated in graded ethanols, embedded in paraffin, sectioned, and stained with hematoxylin and eosin for light microscopy.
For staining with β-tubulin, CC26, and SP-C antibodies 5 μm paraffin sections were immunostained using Vectastain Elite ABC and MOM staining kits (Vector Laboratories). Antigen unmasking prior to immunostaining with the SP-C antibody was performed using citrate buffer together with a pressure cooker (Biocare Medical, Carlsbad, CA) according to the manufacturer's instructions. Antibodies were developed with DAB (Vector Laboratories) and counterstained with hematoxylin. Antibody dilutions were 1:2500 for β-tubulin, 1:200 for CC26, and 1:5000 for SP-C.
Lung bud culture
Lung rudiments dissected from E12.5 embryos were maintained for 72 h at 37°C on Transwell filters (Corning Costar, Cambridge, MA) in DMEM with 10% fetal bovine serum, 2 mM glutamine and 1% antibiotic/antimycotic [
18]. Branching morphogenesis was quantified by counting terminal peripheral end buds at 0, 24, 48, and 72 h in culture.
Morphometry
Chord length measurements for saccules were performed as described [
20]. Briefly, ten serial slides at least 50 μm apart, which contained 5 μm sections of paraffin-embedded newborn
Lamc2-/- and littermate control lungs (at least 3 of each) were stained with H&E. Random images were acquired with the 40× objective on a Nikon Optiphot II microscope and a Zeiss Axiocam digital camera. Chord length was determined using the NIH image program. Fields containing large airways and vessels were excluded. Statistical analysis was determined by two-tailed Student's t-Test.
Discussion
Laminin-332, the only laminin containing the laminin β3 and γ2 chains, was first detected in human basement membranes and underneath hemidesmosomes decades ago [
26]. The primary structures of the individual chains of laminin-332 were determined in the early 1990s and laminin-332 and its components have been studied extensively in subsequent years. Laminin-332 containing the laminin γ2 chain is produced by epithelial cells and is widely distributed in basement membranes of most epithelia, including skin, lung, gastrointestinal tract, kidney, prostate, ovary, and blood vessels of spleen and thymus [
12,
27,
28]. Knockout and transgenic mice technologies have enabled exploration of specific functions of individual laminin-332 components in mouse development. Targeted deletion of the laminin α3 chain leads to perinatal death with a severe blistering disease similar to human junctional epidermolysis bullosa, and defective late stage differentiation of ameloblasts in developing incisors [
29]. A null mutation in the
Lamb3 gene from spontaneous insertion of an intracisternal-A particle at an exon/intron junction also results in blisters and death hours after birth [
30].
Lamc2-/- mice suffer perinatal death and exhibit skin lesions that recapitulate human junctional epidermolysis bullosa with induced apoptosis in the basal cells of the abnormal skin [
15]. Although laminin-332 is seen in many organs, most of the focus has been on skin during the characterization of mice with mutations in any of the laminin-332 constituents. In this report, we focus on lung development in
Lamc2-/- mice.
Lamc2, present in laminin-332, is found in the epithelial airway and alveolar basement membranes of adult lungs and epithelial basement membranes of lung from the pseudoglandular to the alveolar stage of lung development [
14]. In human lung during the pseudoglandular stage, immunodetection of laminin γ2 and laminin-332 revealed higher intensity of fluorescence in the clefts of the ramifications of the growing respiratory tubules leading the authors to hypothesize a role in branching morphogenesis [
14]. Because laminin-332 co-localizes with laminin-111, which is a known effector of lung branching morphogenesis in vitro, speculation of a role for laminin-332 in branching morphogenesis was plausible. In our study, we were able to directly examine branching morphogenesis in the absence of laminin γ2 and laminin-332 through use of the
Lamc2-/- mouse. We found that deficiency of laminin γ2 did not affect lung branching morphogenesis of in vitro lung bud cultures. This result is reminiscent of our finding normal lung branching morphogenesis in the
Lama5-/- mouse, even though laminin α5 co-localized with laminin α1 [
18]. Of note, the null mutation of integrin α3, a major ligand for both laminin α3 and laminin α5 containing laminins, led to abnormal branching morphogenesis [
31] thus normal branching morphogenesis in laminin 332 deficient and laminin α5 null mice is rather unexpected. In the case of the
Lamc2-/- mouse, only laminin-332 is absent so that other laminin α3 chain containing laminins (laminins-311 or -321) are still present and can contribute to the process of branching morphogenesis. An alternate, and perhaps more attractive, conclusion is that within the epithelial-derived laminin chains (those with laminin α1, α3, and α5 chains), no redundancy of function exists for laminin chains and only laminins containing the laminin α1 chain exert effects on lung branching morphogenesis. To resolve this, one needs to examine lung branching morphogenesis in a
Lama1-/- mouse or a double
Lama3/Lama5 knockout mouse.
Again, based on localization of laminin-332 and γ2 during development, roles in lung epithelial differentiation and alveolization were also suggested [
12,
14]. This idea fits with studies showing that laminin-332 stabilizes the phenotype of primary alveolar epithelial cells in culture [
32‐
34]. However, we found normal expression and localization of markers of airway and alveolar epithelial cells in
Lamc2-/- lungs. This finding contrasts with the lungs of mice lacking laminin α5 in which there is a marked impairment in differentiation of distal epithelial cells [
11]. With respect to lung alveolization, we did note a mild increase in saccule size in the
Lamc2-/- compared with littermate controls. Whether laminin γ2 or laminin-332 is important for later stages of lung development, specifically alveolization remains to be determined since
Lamc2-/- mice died before alveolization occurs. That lack of laminin γ2 did not significantly affect lung epithelial cell differentiation while perturbation of laminin α5 had a dramatic effect again indicates that laminins have specific, non-overlapping functions during lung development.
In the absence of a laminin chain, compensation by ectopic expression of another laminin of the same chain group can occur.
Lamb1 compensates for lack of
Lamb2 in the kidney, upregulation of
Lama4 is seen with loss of
Lama2 in muscle, deletion of
Lama5 leads to ectopic
Lama2 and
Lama4 in ectoderm and intestines [
21,
22,
35,
36]. However, a compensatory response was not detected with deletion of laminin γ2 in the lung or in the skin. The reason for this is unknown but it may relate to the uniqueness of laminin-332 in that it is the only laminin known to contain the β3 and γ2 chains and it is the only laminin present in hemidesmosomes.
By transmission electron microscopy, tracheal hemidesmosomes in
Lamc2-/- mice were different from those of the littermate control. This finding is consistent with cutaneous hemidesmosomes in the
Lamc2-/- and the
Lama3-/- mice. However, even though the hemidesmosomes appeared abnormal at the ultrastructural level, immunofluorescence staining for other components of hemidesmosomes was similar between
Lamc2-/- and littermate control lungs. In contrast, immunostaining for cutaneous basement membrane zone proteins in
Lamc2-/- and
Lama3-/- both showed abnormal distribution of these proteins compared with controls [
15,
29]. In addition, skin epithelial and oral and bladder mucosa of
Lamc2-/- and
Lama3-/- had areas of blister formation while no areas of blistered epithelium were found in
Lamc2-/- tracheas. This suggests that abnormal hemidesmosomes in
Lamc2-/- tracheas did not produce a functional defect or that the tracheas are not mechanically stressed enough to blister. Alternatively, tracheal hemidesmosomes may have different function compared to hemidesmosomes in other tissues. In people with epidermolysis bullosa, the main pathologic feature is skin blistering with abnormal hemidesmosomes. Rare cases of laryngotracheal involvement have been reported but airway obstruction has not been implicated as a significant cause of mortality in these patients [
37‐
39]. Thus, our finding of normal appearance, integrity, and presumably function, of tracheal epithelium despite abnormal hemidesmosomes in the
Lamc2-/- mice is consistent with infrequent abnormalities in humans.
While we did not observe a significant role for laminin γ2 and laminin-332 in lung development, physiologic roles of this laminin must exist. Laminin-332 may facilitate alveolar epithelial repair via effects on cell migration. By in situ hybridization, immunohistochemistry, and immunoelectron microscopy, regenerating epithelial cells in cryptogenic organizing pneumonia and in idiopathic pulmonary fibrosis both express laminin γ2 in response to injury [
40]. In addition, a recent report shows that laminin γ2 is not only present in the basement membrane but also in the cytoplasm of injured epithelial cells and in columnar epithelium of allergic asthmatics [
41]. Moreover, this laminin may influence tumor cell behavior [
42]. Tumor cell lines often express laminin-332 and the expression is enhanced by epidermal growth factor [
43]. In lung tumors, expression of the laminin γ2 chain was strong in squamous cell carcinomas, adenocarcinomas, and large cell carcinomas, with immunoreactive cells localizing to the epithelial-stromal interface of tumor clusters [
44]. Inactivation of laminin-322 genes by aberrant methylation in prostate cancer and bladder cancer samples correlated with poor prognosis [
45‐
47]. Laminin γ2 can be found in the cytoplasm of carcinoma cells invading into interstitial stroma while laminin α3 and β3 chains are only found in the basement membrane [
48].
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
NN participated in the design of the study, carried out the lung bud cultures, immunofluorescence staining, statistical analysis, supervised the optimization of immunostaining, evaluated the data, and drafted the manuscript. JS carried out the immunohistochemical staining, morphometry, evaluated the data, and helped to draft the manuscript. LP and JU produced the Lamc2-/- mouse and helped to draft the manuscript. GM made the laminin γ2 antibody and helped to draft the manuscript. RS conceived the study, participated in its design and coordination, evaluated the data, and helped to draft the manuscript. All authors read and approved the final manuscript.