Tissue Distribution of Surfactant Proteins A and D in the Mouse

Abstract

Surfactant proteins A and D, collagen-like lectins (collectins), were first isolated from the lung. In the lung, SP-A and SP-D have roles in surfactant homeostasis and innate immunity. In this study we show that SP-A and SP-D mRNA can be detected in a significant number of non-pulmonary tissues but the proteins have a more limited distribution. SP-D protein was detected in lung, uterus, ovary, and lacrimal gland, whereas SP-A protein was detected only in the lung. The results suggest that SP-D participates in mucosal immunity throughout the body.

SP-A and SP-D are members of the collectin protein family, defined by linked collagen-like and carbohydrate recognition domains (Hoppe and Reid 1994). The collectins contribute to host defense by binding to a wide range of pathogens and allergens and enhancing phagocytosis. Different members of the collectins also influence the immune response by activation of complement and regulation of macrophage and lymphocyte activity (Wright 1997; Crouch and Wright 2001). Initially, both SP-A and SP-D were believed to be restricted to the lung and to have roles in the activity, structure, and metabolism of pulmonary surfactant. More recently, the host defense functions of SP-A and SP-D have been emphasized (Wright 1997; Crouch and Wright 2001), and non-pulmonary expression of both SP-A and SP-D has been reported in rats, pigs, rabbits, fish, and humans (Bourbon and Chailley-Heu 2001). The specific tissue expression of these two proteins is different and also varies significantly among species.

Mice overexpressing or lacking SP-A and SP-D are useful models in which to explore the roles of these proteins in host defense. Mice lacking SP-D develop spontaneous pulmonary inflammation (Botas et al. 1998; Wert et al. 2000) and clear respiratory viruses poorly (LeVine et al. 2001). Mice lacking SP-A clear bacterial pathogens slowly and have exaggerated inflammatory responses to microbial challenge of the lung (LeVine et al. 2000). The tissue distribution of SP-A and SP-D in the mouse has not been reported.

To define the tissue distribution of SP-A and SP-D in the mouse, we harvested tissues from both male and female C57/BL6 mice 8–12 weeks old. Tissues from gene-targeted C57/BL6 mice deficient in SP-A or SP-D gene expression, generated in our laboratory, were harvested as controls for antibody specificity. RNA was isolated by the guanidinium isothiocyanate method using RNA STAT-60 reagents (Tel-Test; Friendswood, TX). Two μg of total RNA from each tissue was reverse-transcribed using a RETROscript kit (Ambion, Austin, TX). The SP-A primers were located in exon V (5′) and exon VI to yield a 276-bp product. The primers of SP-D were located in exon VII and exon VIII to yield a 469-bp product. RT-PCR was performed for 35 cycles using 500 ng of template cDNA and an annealing temperature of 60C. SP-A and SP-D PCR products were probed by Southern blotting with ³²P-labeled nested oligonucleotides. Controls without reverse transcriptase or template were run with each reaction and were always negative. For Northern analysis, total RNA (20 μg for all samples) isolated from mouse tissues was electrophoresed, blotted, and probed sequentially for SP-A, SP-D, and 18S RNA. The SP-A cDNA probe was a 345-bp fragment from exon VI. The SP-D probe was a 270-bp Bgl II fragment of exon VIII. The specificity of the primers and probes was confirmed by negative amplification or hybridization to mRNA from the lungs of SP-A–/– and SP-D–/– mice using SP-A and SP-D primers and probes respectively (not shown).

For Western blotting analysis, freshly harvested mouse tissues (100 mg) from wild-type and gene-targeted mice were sonicated in 500 μl 50 mM NaHCO₃ (pH 9.0) containing 20 μl protease inhibitor cocktail (Sigma P8340; St Louis, MO). The samples were centrifuged to clear and used immediately without any freeze-thaw cycles. Tears were also collected from anesthetized mice for Western analysis. The antibody to mouse SP-D was raised in rabbits against mouse SP-D expressed in Chinese hamster ovary cells. The antibody against mouse SP-A was raised in rabbits against the CRD fragment of mouse SP-A expressed in E. coli using the pET23 vector system (Novagen; Madison, WI). Antibody specificity was confirmed by Western analysis using recombinant SP-A and SP-D and by Western analysis and immunocytochemistry on tissues from SP-A–/– and SP-D–/– mice (not shown). Based on the strength of the mRNA and protein signals, the lung, uterus, and lacrimal gland were selected for immunocytochemistry using standard protocols that included antigen retrieval. Controls without the primary antibody and reactions run on –/– mice gave no reaction product.

SP-D mRNA was detected by RT-PCR in a wide range of tissues, including the salivary gland, lacrimal gland, ovary, uterus, esophagus, stomach, testes, thyroid, and heart (Figure 1). No SP-D mRNA was detected in the bladder, brain, colon, duodenum, jejunum, kidney, liver, mammary gland, muscle, pancreas, prostate, seminal vesicle, spleen or tongue. A more restricted pattern of SP-D expression (lung, uterus, and lacrimal gland with weaker signals in the ovary, esophagus, and trachea) was detected by Northern blotting analysis (Figure 2). SP-D protein was detected by Western analysis in an even more restricted set of tissues: lung, trachea, uterus, ovary, and lacrimal gland (Figure 3). Clearly, the sensitivity of the methods used for mRNA and protein detection is markedly different and low-level protein expression in some tissues could have been missed. In the lung, SP-D was localized to alveolar epithelial type II cells and non-ciliated bronchiolar cells (Figure 4) as previously described in the mouse (Wong et al. 1996) and other species. SP-D was also localized to ducts of the lacrimal gland (Figure 5) and epithelial cells lining the uterine lumen and the ducts of uterine myometrial glands (Figure 6). Material in the lumen of the uterine myometrial glands stained intensely for SP-D, and we detected SP-D in mouse tears by Western analysis (Figure 3). In situ hybridization will be required to determine the cells expressing SP-D in these tissues. No reaction product was detected in any tissue from SP-D –/– mice (not shown).

SP-D mRNA and protein were recently detected in a large number of human tissues (Madsen et al. 2000). Our results suggest there are distinct differences in the tissue distribution and relative abundance of SP-D mRNA in mouse and human. Madsen et al. (2000) reported abundant mRNA in human kidney, brain, and pancreas, with a much lower level of expression in the uterus. In the C57/BL6 mouse we found no SP-D mRNA in the kidney, brain, and pancreas but high-level expression in the uterus. We found little evidence for SP-D expression in the mouse gastrointestinal tract. Even by RT-PCR the mRNA was restricted to the stomach and jejunum, and no SP-D protein was detected in the gastrointestinal tract. In contrast, SP-D is expressed quite strongly in a subset of cells in the antrum of the rat stomach (Fisher and Mason 1995), and SP-D mRNA is easily detected by Northern blotting throughout the pig small intestine (van Eijk et al. 2000). Several factors might account for the apparent species differences in SP-D message abundance in different tissues. These include simple sampling artifacts caused by taking small samples from larger tissues that might have included more or less of a specific cell type, e.g., duct epithelium, or the potential effects of genetic background, age, estrous cycle, inflammation, and infection on SP-D expression. In other species SP-D has also been detected in tissues we did not study. In the human SP-D is expressed by the biliary epithelium, epithelium of the urinary tract, sweat glands, placenta, and basal cells of the epidermis (Madsen et al. 2000). SP-D has been found in the Eustachian tube and middle ear of pigs, the middle ear and nasal sinuses of rabbits, and in peritoneal mesothelial cells in rat and human (Bourbon and Chailley-Heu 2001).

SP-A mRNA was readily detected by RT-PCR only in lung, trachea, and esophagus of the mouse, a very different pattern of expression from that of SP-D (Figure 1). A weak signal for SP-A mRNA was also detected in liver, stomach, jejunum, pancreas, spleen, heart, testes, and ovary but Northern blotting analysis only detected SP-A mRNA in the lung (Figure 2). SP-A protein was also detected only in the lung by Western analysis (Figure 7), suggesting a much more restricted distribution of protein expression relative to SP-D. In the lung, SP-A and SP-D have similar cellular distributions (Figure 4). We are uncertain as to the significance of the patchy detection of SP-A in the lumen of uterine gland ducts (Figure 6B) because we were unable to confirm the presence of SP-A in the uterus by Western blotting analysis. We did not detect any reaction product in the uterus of SP-A–/– mice (not shown), suggesting that the finding was specific, and SP-A mRNA was detected in the uterus with 40 cycles of amplification (not shown). The more restricted distribution of SP-A compared to SP-D is consistent with most other reports, but there may be significant interspecies variation. In the rat, SP-A mRNA has been detected in mesentery and in selected sites throughout the gastrointestinal tract, and in the intestine and swim bladder of the carp (Bourbon and Chailley-Heu 2001).

In summary, we have shown that SP-D is expressed in several non-pulmonary sites, specifically the uterus and lacrimal gland, in the C57/BL6 mouse. It is possible that SP-D contributes to the immune defense, similar to the purported role of SP-D in the lung, or has other unrecognized functions in these tissues. In contrast, SP-A protein appears to be predominantly expressed in the lung in the mouse, at least within the group of 25 tissues examined in this study. Given that mRNA for both collectins was detected in many other tissues, we cannot exclude the possibility that the distribution and abundance of the proteins might be more widespread in the context of infection or inflammation. The mouse should prove a useful model in which to further study the function of SP-D and SP-A at non-pulmonary sites.

Figure 1

Tissue distribution of SP-D and SP-A mRNA detected by RT-PCR. Top, SP-D; middle, SP-A; bottom, ethidium bromide-stained gels of PCR reactions using RETROscript primers to control for the reverse transcription reaction. Lane a, esophagus; Lane b, stomach; Lane c, testes; Lane d, thyroid; Lane e, lung; Lane f, trachea; Lane g, salivary gland; Lane h, lacrimal gland; Lane i, ovary; Lane j, uterus; Lane k, bladder; Lane l, brain; Lane m, colon; Lane n, duodenum; Lane o, heart; Lane p, ileum; Lane q, jejunum; Lane r, kidney; Lane s, liver; Lane t, mammary gland; Lane u, muscle; Lane v, pancreas; Lane w, prostate; Lane x, seminal vesicle; Lane y, spleen; Lane z, tongue. The SP-D product is the expected 469 bp and the SP-A product is 276 bp.

Figure 2

Tissue distribution of SP-D and SP-A mRNA detected by Northern analysis. Top, SP-D; middle, SP-A; bottom, 18S RNA. Lane a, bladder; Lane b, brain; Lane c, colon; Lane d, duodenum; Lane e, esophagus; Lane f, heart; Lane g, ileum; Lane h, jejunum; Lane i, kidney; Lane j, liver; Lane k, ovary; Lane l, pancreas; Lane m, urethral gland; Lane n, spleen; Lane o, stomach; Lane p, testes; Lane q, thyroid; Lane r, tongue; Lane s, uterus; Lane t, muscle; Lane u, seminal vesicle; Lane v, trachea; Lane w, lung; Lane x, salivary gland; Lane y, lacrimal gland. SP-D mRNA is 1.35 kb and SP-A mRNA is 1.45 kb.

Figure 3

Tissue distribution of SP-D protein detected by Western analysis. One hundred mg of each tissue was sonicated in 500 μl buffer, centrifuged and 40 μl loaded on the gel. Lane a, esophagus; Lane b, trachea; Lane c, stomach; Lane d, testes; Lane e, tears; Lane f, lung; Lane g, blank; Lane h, thryoid; Lane i, lacrimal gland; Lane j, salivary gland; Lane k, SP-D+/+ uterus; Lane l, SP-D–/– uterus; Lane m, SP-D+/+ ovaries; Lane n, SP-D–/– ovaries.

Figure 4

Localization of SP-D and SP-A in the lung. Immunofluorescence (A) and phase-contrast (C) of mouse lung stained with anti-SP-D. Immunofluorescence (B) and phase-contrast (D) of mouse lung stained with anti-SP-A. Bar = 50 μm.

Figure 5

Localization of SP-D and SP-A in the lacrimal gland. Immunofluorescence (A) and phase-contrast (C) of mouse lacrimal gland stained with anti-SP-D. Immunofluorescence (B) and phase-contrast (D) of mouse lacrimal gland stained with anti-SP-A. Duct epithelium (arrows) is positive for SP-D but not SP-A. Bar = 50 μm.

Figure 6

Localization of SP-D and SP-A in the uterus. Immunofluorescence (A) and phase-contrast (C) of mouse uterus stained with anti-SP-D. Immunofluorescence (B) and phase-contrast (D) of mouse uterus stained with anti-SP-A. The duct epithelium of the myometrial glands (black arrows) and the luminal epithelium (^∗) are positive for SP-D. Faint staining for SP-A is seen in the lumen of a few myometrial gland ducts (white arrows). Bar = 50 μm.

Figure 7

Tissue distribution of SP-A protein detected by Western analysis. One hundred mg of each tissue was sonicated in 500 μl buffer, centrifuged, and 40 μl loaded on the gel. Lane a, lung; Lane b, lacrimal gland; Lane c, uterus; Lane d, esophagus. Samples from +/+ and SP-A–/– mice were examined for each tissue.