Abstract
Lung surfactant protein-D (SP-D), a collectin mainly produced by alveolar type II cells, initiates the effector mechanisms of innate immunity on binding to microbial carbohydrates. A panel of mRNAs from human tissues was screened for SP-D mRNA by RT-PCR. The lung was the main site of synthesis, but transcripts were readily amplified from trachea, brain, testis, salivary gland, heart, prostate gland, kidney, and pancreas. Minor sites of synthesis were uterus, small intestine, placenta, mammary gland, and stomach. The sequence of SP-D derived from parotid gland mRNA was identical with that of pulmonary SP-D. mAbs were raised against SP-D, and one was used to locate SP-D in cells and tissues by immunohistochemistry. SP-D immunoreactivity was found in alveolar type II cells, Clara cells, on and within alveolar macrophages, in epithelial cells of large and small ducts of the parotid gland, sweat glands, and lachrymal glands, in epithelial cells of the gall bladder and intrahepatic bile ducts, and in exocrine pancreatic ducts. SP-D was also present in epithelial cells of the skin, esophagus, small intestine, and urinary tract, as well as in the collecting ducts of the kidney. SP-D is generally present on mucosal surfaces and not restricted to a subset of cells in the lung. The localization and functions of SP-D indicate that this collectin is the counterpart in the innate immune system of IgA in the adaptive immune system.
Lung surfactant protein D (SP-D)3 is one of the collectins, a family of oligomeric proteins whose individual chains consist of a collagen region linked to a C-type lectin domain via an α helical neck region (1, 2). The structural subunits of SP-D are homotrimers of these chains, and the subunits are themselves oligomerized into cross-like tetramers and higher oligomers (3, 4, 5, 6, 7, 8).
SP-D binds to oligosaccharides on the surface of a variety of pathogenic microorganisms. This binding initiates several effector mechanisms, including the recruitment of inflammatory cells to destroy the pathogens. SP-D has been shown to bind to carbohydrate residues on influenza A virus, thereby inhibiting its hemagglutination activity and causing viral aggregation (9, 10). SP-D also enhances the binding of influenza A virus to neutrophil granulocytes and promotes the neutrophil respiratory burst in response to the virus (11). SP-D binds to and induces aggregation of other microorganisms, such as Gram-negative bacteria (12) and the fungi Cryptococcus neoformans and Aspergillus fumigatus (13, 14). SP-D binds directly to alveolar macrophages in the absence of microbial ligands, thereby mediating the generation of oxygen radicals (15), and also acts as a potent chemotactic agents for phagocytes (16, 17). SP-D has been localized in endocytic vesicles and lysosomal granules of alveolar macrophages (18, 19), suggesting a receptor-mediated uptake by these cells; a putative receptor for SP-D, gp-340, has been characterized (20). SP-D also seems to have an immunomodulatory function, inhibiting T lymphocyte proliferation and IL-2 production (21) as well as inhibiting specific IgE binding to allergens and blocking allergen-induced histamine release from human basophils (13, 22).
In addition to its role in antimicrobial defense, SP-D may be involved in pulmonary surfactant homeostasis. It interacts with phospho- and glycolipids in vitro (23, 24, 25), and mice made SP-D deficient by gene targeting accumulate surfactant lipids and alveolar macrophages in the alveolar space (26, 27). In this situation, the macrophages may appear as multinucleated foam cells, whereas the alveolar type II cells are hyperplastic and contain giant lamellar bodies.
SP-D is generally recognized as a molecule expressed in alveolar type II cells and Clara cells, but it has also been demonstrated in mucus cells of the rat gastric mucosa (28), and extrapulmonary expression has been detected in murine tissues (29). In this report, the extrapulmonary expression of human SP-D was investigated by RT-PCR and immunohistochemistry. SP-D was found to be expressed in relation to mucosal surfaces in numerous glands and organs.
Materials and Methods
SDS-PAGE and Western blotting
Electrophoresis was performed on 4–14% (w/v) polyacrylamide gradient gels with discontinuous buffers using the FAST system (Pharmacia, Piscataway, NJ). Samples were reduced by heating to 100°C for 3 min with 40 mM DTT in 0.1 M Tris-HCl buffer, pH 8.0, containing 1.5% (w/v) SDS and 5% (v/v) glycerol, and carboxamidated by the addition of iodoacetamide to 90 mM. Unreduced samples were heated in sample buffer with 90 mM iodoacetamide. After electrophoresis, proteins were electroblotted onto polyvinylidene difluoride membranes (Immobilon P, Millipore, Bedford, MA).
The membrane was incubated with primary Ab (monoclonal SP-D Ab, 50 ng/ml) and secondary alkaline phosphatase-coupled rabbit anti-mouse Ig in 10 mM Tris-HCl, pH 7.4, containing 0.5 M NaCl, 15 mM NaN3, and 0.05% (v/v) Tween 20 (polyoxyethylene sorbitan monolaurate, Merck-Schuchardt, Hohenbrunn, Germany). The membranes were washed and developed with nitroblue tetrazolium and potassium 5-bromo-4-chloro-3-indolylphosphate.
Reverse-transcriptase PCR
Total RNA was obtained from whole organs from various human tissues (Clontech, Palo Alto, CA). Two micrograms of total RNA were used for first-strand synthesis using the antisense SP-D primer, SP-DR1, 5′-TCAGAACTCGCAGACCA-3′. This sequence corresponds to nucleotides 1282–1298 of the cDNA sequence of SP-D (2). The RNA was incubated in a volume of 20 μl with 20 U Moloney murine leukemia virus reverse transcriptase (Stratagene, La Jolla, CA) at 37°C for 1 h. One-fourth of the resulting cDNA was subjected to 40 cycles of PCR amplification in a volume of 30 μl with a denaturation temperature of 94°C for 30 s, annealing at 62°C for 1 min, extension at 72°C for 2 min, and finally an extension at 72°C for 7 min, using Taq polymerase (Stratagene) with two SP-D primers spanning the neck and carbohydrate recognition domain region of human SP-D: a sense primer, SP-DF1, 5′-ATGTTGCTTCTCTGAGG-3′, corresponding to nucleotides 839–855; and an antisense primer, SP-DR3, 5′-TCAGAACTCGCAGACCACAAG-3′ corresponding to nucleotides 1278–1298. As a control, RT-PCR was performed on β-actin under the same conditions as for SP-D, except that the annealing temperature was 58°C and the primers used were: 5′-actin (sense), 5′-GGCATGGCTTTATTTGTTTT-3′; and 3′-actin (antisense), 5′-GTAAGCCCTGGCTGCCTC-3′. The reaction mixtures were run on a 1.2% agarose gel, blotted onto a nylon membrane (Biotrace HP, Gelman Science, Ann Arbor, MI) by alkaline Southern blotting, and hybridized with the respective SP-D and β-actin probes labeled with [α-32P]dATP.
The SP-D probe was amplified by PCR using a human cDNA clone of SP-D (kindly provided by Dr. Jinhua Lu, Singapore University, Singapore) as template and the same conditions and primers as used for RT-PCR. The β-actin probe was generated from a template of total RNA from the uterus, under the same conditions and with the same primers as those used for β-actin RT-PCR. Each reaction mixture was run on an 1.2% agarose gel; the bands were extracted from the gel by means of the QIAquick Gel Extraction Kit (Qiagen, Chatsworth, CA) and labeled with [α-32P]dATP using “large fragments of DNA polymerase I” (Life Technologies) and random hexamers as primers.
The PCR products obtained from lung and salivary gland were cut out from the agarose gel, subcloned with the TA cloning kit (Invitrogen) and sequenced.
Production of mAbs against SP-D
SP-D was purified from amniotic fluid by the method described in Ref. 30 .
Mice were immunized with 10 μg purified human SP-D adsorbed onto aluminum hydroxide gel and emulsified with 0.25 ml Freund’s incomplete adjuvant. Fusions were performed with the cell line X63Ag8.6.5.3, and positive clones were identified by ELISA using purified SP-D coated onto the wells. Cells from wells with Abs against SP-D were recloned three times by the limiting dilution method. The specificity of the mAbs for SP-D was checked by Western blotting of partially purified SP-D mixed with Triton X-100 solubilized alveolar macrophages separated on SDS-PAGE in the reduced and unreduced state. A panel of nine Abs was obtained, and Hyb 245-1 was selected for immunohistochemistry.
Immunohistochemistry
Normal human tissues were from the tissue bank at the Department of Pathology, Odense University Hospital (Odense, Denmark). The tissues were fixed in 4% formalin in 10 mM sodium phosphate buffer, pH 7.4, containing 140 mM NaCl (PBS) for 24 h and then conventionally dehydrated and embedded in paraffin.
A biotin-streptavidin immunoperoxidase technique was used on paraffin sections. Briefly, the paraffin sections were pretreated in 10 mM sodium citrate buffer, pH 6.0, in a microwave oven for three 5-min periods at 650 W. The sections were left in citrate buffer for 15 min; washed in 10 mM Tris-HCl, pH 7.2, containing 140 mM NaCl and 7.5 mM NaN3 (TBS); preincubated with 2% (w/v) BSA in TBS for 10 min; incubated for 30 min with Hyb 245-1, 5 μg/ml in TBS containing 1% (w/v) BSA; washed with TBS; incubated for 30 min with biotin-labeled goat anti-mouse Ig (Dako, Carpinteria, CA) diluted 1:200 in TBS; washed with TBS; incubated with HRP-coupled streptavidin (Dako) diluted 1:300 in TBS without NaN3; washed with TBS and water; incubated for 20 min with 0.04% 3-amino-9-ethylcarbazole (Sigma, St. Louis, MO) and 0.015% H2O2 in 50 mM sodium acetate buffer, pH 5.0; washed; counterstained with Mayer’s hematoxylin for 2 min; and mounted in Aquatex (Sigma). The specificity of immunostaining was verified by replacing the primary Ab with an unrelated mAb of the same subclass as the SP-D Ab, as well as other conventional staining controls.
Results
Expression of SP-D in human tissues analyzed by RT-PCR
A fragment of 461 bp corresponding to the neck and carbohydrate recognition domain region of human SP-D was amplified by RT-PCR applied to total RNA from 19 different human tissues (Fig. 1⇓). Expression of SP-D mRNA was most pronounced in the lung but was also found in the kidney, trachea, brain, testis, pancreas, salivary gland, heart, prostate, small intestine, and placenta. Low expression was found in the uterus, stomach, mammary gland, spleen, adrenal gland, and liver. No expression was observed in skeletal muscle or thymus. All samples were normalized with respect to β-actin. The PCR products obtained from lung and salivary gland were isolated and sequenced. In both tissues the sequences obtained were identical with the corresponding sequence published for human SP-D, confirming that the correct transcript was being amplified in the PCR.
RT-PCR analysis of SP-D expression in human tissues. Total mRNA (2 μg) from 19 different tissues were amplified, separated in 1.2% (w/v) agarose gel, blotted onto nylon membrane, and hybridized with either a human SP-D or a β-actin oligonucleotide probe as described in Materials and Methods.
Characterization of Abs
The specificity of the Ab used for immunohistochemistry was verified by Western blots of partially purified SP-D that has been mixed with Triton X-100-solubilized alveolar macrophages and separated on SDS-PAGE in the reduced state (Fig. 2⇓).
Specificity of the monoclonal SP-D Ab analyzed by Western blotting. Lane 1, reduced partially purified SP-D mixed with Triton X-100 solubilized alveolar macrophages. The blots were incubated with 50 ng/ml Hyb 245-1. The bound Ab was visualized by means of alkaline phosphatase-labeled goat anti-mouse IgG and substrate. Lane 2, Gold staining of reduced partially purified SP-D mixed with Triton X-100 solubilized alveolar macrophages.
Immunohistochemical analysis of SP-D
In the lung, strong immunoreactivity was observed in alveolar type II cells (Fig. 3⇓, a and b), in unciliated bronchial cells (Fig. 3⇓, b and c) and in a subset of alveolar macrophages (Fig. 3⇓, d–f). Some alveolar macrophages were stained intracellularly (Fig. 3⇓f), whereas others showed a distinct staining of the cell membrane (Fig. 3⇓, d and e).
Immunohistochemical localization of SP-D in normal human lung. SP-D was found in alveolar type II cells (a) and Clara cells (b and c). SP-D was also found on and within alveolar macrophages (d–f). The tissues were stained by an indirect immunoperoxidase technique and counterstained with Mayer’s hematoxylin as described in Materials and Methods. Original magnification, ×400.
Various exocrine glands were stained. In the parotid gland both large and small ducts were stained (Fig. 4⇓, a and b), as were ducts of oral mucous glands (Fig. 4⇓c). In the pancreas, the intercalated ducts showed a strong granular staining (Fig. 4⇓d). Small bile ducts were stained in the liver, but no staining was observed in hepatocytes (Fig. 4⇓e). Staining was seen in the ducts of sweat glands (Fig. 4⇓f), and a similar staining pattern was observed in mammary and lachrymal gland (not shown).
Immunohistochemical localization of SP-D in exocrine ducts in normal human tissues. SP-D was found in duct epithelial cells within the salivary gland (a–-c), in the intercalated ducts of pancreas (d), in the small bile ductules of the liver (c), and in the sweat glands (f). The tissues were stained by an indirect immunoperoxidase technique and counterstained with Mayer’s hematoxylin as described in Materials and Methods. Original magnifications: a and f, ×200; b–e, ×400.
In the epidermis, only the basal cells were stained (Fig. 5⇓a), whereas staining was seen throughout the unkeratinized stratified squamous epithelium of the esophagus (Fig. 5⇓b). Little or no staining was observed in the stomach and large intestine, whereas a slight but significant staining was seen in the epithelial cells of the small intestine (Fig. 5⇓c). The gallbladder epithelium was stained (Fig. 5⇓d).
Immunohistochemical localization of SP-D in epithelial cells in normal human tissues. SP-D was found in the skin (a), esophagus (b), small intestine (c), gallbladder epithelium (d), collecting ducts of the kidney (e–g), and the epithelia of the ureter (h) and urinary bladder (i). The tissues were stained by an indirect immunoperoxidase technique and counterstained with Mayer’s hematoxylin as described in Materials and Methods. Original magnifications: e and h, ×100; b–d, ×200; a, f, and g, ×400.
In the kidney, the collecting ducts were stained (Fig. 5⇑, e–g), and faint staining was also seen in the loops of Henle, while the glomeruli were unstained. The epithelium of both ureter and bladder were stained (Fig. 5⇑, h and i). Some of the prostate gland epithelium was clearly stained, whereas gland epithelium from other parts of the prostate showed only faint staining (not shown). Staining and substitution controls were negative.
Discussion
The present report describes the tissue expression of SP-D in the lung and 18 extrapulmonary tissues as determined by RT-PCR and immunohistochemistry. SP-D was demonstrated in epithelial cells in a variety of tissues throughout the body.
It has long been established that alveolar type II cells and unciliated bronchial epithelial cells (Clara cells) in the lung are the major sites of synthesis of SP-D (18, 31, 32). Northern blotting and RT-PCR have previously been used to determine SP-D expression by selected extrapulmonary tissues of different species. In humans, SP-D transcripts have been detected in heart, pancreas, stomach, small, and large intestine (29, 33), whereas they have been found in the stomach, kidney, and heart of the mouse (29). In the rat, SP-D has been found in the gastric mucosa (28) and mesentery cells (34) by RT-PCR, immunohistochemistry, and Western blotting, whereas it could not be demonstrated in the small or large intestine. Other cells expressing SP-D outside the lung have not been characterized in the rat.
In the present study, SP-D was found to be strongly expressed in the human lung and trachea, but extrapulmonary organs such as the kidney, brain, testis, pancreas, salivary gland, heart, prostate, small intestine, and placenta also produced clear signals, whereas the uterus, stomach, mammary gland, spleen, adrenal gland, and liver showed weak expression. Sequencing of the RT-PCR product from lung and parotid gland confirmed that correct transcript had been amplified.
Of the panel of 9 mAbs raised against purified human SP-D, two could be used for immunohistochemical staining on paraffin-embedded tissue, and the staining pattern of these two Abs was identical in lung and salivary glands. Hyb 245-1 was selected for its good immunohistochemical staining properties. In Western blotting, this Ab reacted with a 43-kDa band from the reduced SP-D preparation, corresponding to single chain of SP-D chains. In the unreduced preparation, the Ab reacted with a 180-kDa band corresponding to SP-D chain oligomers (not shown).
Immunohistochemistry with this Ab confirmed the presence of SP-D immunoreactivity in alveolar type II cells, Clara cells, and the mesothelial cells lining the pleural cavity; those lining the peritoneal cavity were also stained for SP-D (not shown). The weak signal seen in the RT-PCR analysis of, e.g., the spleen could be derived from mesothelial cells. In most of the alveolar macrophages, the staining appeared to be present in the phagolysosome compartment, but a few macrophages showed a distinct granular staining of the cell membrane. Ultrastructural studies have previously shown the presence of SP-D in the endocytic compartment of alveolar macrophages but not in the biosynthetic organelles (18). This suggests that SP-D is not produced by these cells but is taken up by endocytosis. The localization of SP-D to the membrane is the first step in such an event, and a putative receptor for SP-D that may be responsible for this membrane binding has recently been characterized (20, 35). This glycoprotein, named gp-340, exists both in a soluble form and in a form associated with the membrane of the alveolar macrophages. It is a multidomain protein composed of 14 scavenger receptor cysteine-rich domains, four CUB domains, and a zona pellucida domain (35) and appears to be an alternative splicing product of the DMBT1 (deleted in malignant brain tumor 1) gene (36).
SP-D was identified in epithelial cells lining the ducts of a number of exocrine glands and the biliary and urinary tracts. SP-D may thus be secreted at the lining of these ducts, pointing to a major potential role in the mucosal defense system against invading microorganisms.
Surprisingly, SP-D was also located in the basal cells of epidermis and throughout the epithelial cell layer of the esophagus. SP-D could be released from these cells in inflammatory states and play a role in wound healing and combating cutaneous infection. There is evidence of some species variation in SP-D expression in the gastrointestinal tract. Whereas SP-D was clearly detected in the gastric mucosa but not in the small intestine of the rat (28), both the body and pyloric mucosa of human gastric mucosa showed faint immunostaining, and only a weak RT-PCR signal was found when RNA from whole gastric mucosa was used as template. Both these parameters were stronger in the small intestine.
In the urinary tract, SP-D was found in the renal collecting ducts and loops of Henle, as well as in the epithelium of the ureter and bladder. Hensin, the rabbit analogue of DMBT1, is found in the collecting ducts of the kidney and has been implicated in the terminal differentiation of the intercalated cells (37, 38). Because SP-D colocalizes with gp-340 in the collecting ducts, as well as in epithelial cells in the lung, salivary glands, pancreas, and small intestine, it is possible that the interaction between SP-D and gp-340, also a DMBT1 gene product, may be involved in processes other than immunoprotection. DMBT1 has been suggested as a candidate tumor suppressor for brain, gastrointestinal, and lung cancers (36, 38, 39, 40, 41), and it is well known that interruption of terminal differentiation pathways can lead to tumor formation.
We conclude that the localization and synthesis of SP-D are not restricted to the respiratory system. SP-D was found to be widely distributed in exocrine glands and epithelial cells throughout the body. In many tissues, SP-D colocalizes with gp-340, and it cannot be excluded that SP-D may have additional functions related to its interaction with gp-340. SP-D also colocalizes with IgA of the adaptive immune system. The main function of SP-D, like that of IgA, is to prevent microbial colonization of the epithelium and inhibit growth of pathogens once attached to the epithelium, attracting phagocytes and promoting phagocytosis by binding to specific receptors as part of this function. In contrast to IgA, however, SP-D is constitutively produced from birth and may be up-regulated in the course of infection (42).
Like mannan-binding lectin, SP-D may be considered as an “ante-Ab” with a broad binding profile, and the localization and function of SP-D indicates that this collectin is the counterpart in the innate immune system to IgA in the adaptive immune system.
Acknowledgments
We thank Mr. Ole Nielsen for skilled technical assistance.
Footnotes
-
↵1 This work was supported by the Danish Medical Research Council, Michaelsen Fonden, the Novo Nordisk Foundation, Fonden til Lægevidenskabens Fremme, Nationalforeningens Fond, and the Benzon Foundation.
-
↵2 Address correspondence and reprint requests to Dr. Uffe Holmskov, Immunology and Microbiology, Institute of Medical Biology, University of Southern Denmark, Odense University, DK-5000 Odense, Denmark. E-mail address: holmskov{at}imbmed.sdu.dk
-
↵3 Abbreviations used in this paper: SP-D, lung surfactant protein D; DMBT1, deleted in malignant brain tumors 1.
- Received July 15, 1999.
- Accepted March 21, 2000.
- Copyright © 2000 by The American Association of Immunologists
References
In this issue
