HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 22, 19929-19937, May 31, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/22/19929    most recent M201537200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wang, W.-J. Articles by Beers, M. F. Search for Related Content PubMed PubMed Citation Articles by Wang, W.-J. Articles by Beers, M. F. Social Bookmarking                      What's this?

Biosynthesis of Surfactant Protein C (SP-C)

SORTING OF SP-C PROPROTEIN INVOLVES HOMOMERIC ASSOCIATION VIA A SIGNAL ANCHOR DOMAIN^*

Wen-Jing Wang, Scott J. Russo, Surafel Mulugeta, and Michael F. Beers

From the Lung Epithelial Cell Biology Laboratories, Pulmonary and Critical Care Division, Department of Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104

Received for publication, February 14, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Rat surfactant protein C (SP-C) is synthesized as a 194-amino acid propeptide (SP-C-(1-194)) that is directed to the distal secretory pathway and proteolytically processed as an integral membrane protein to yield its mature form. We had shown previously that trafficking of proSP-C is mediated both by a signal anchor domain contained within the mature SP-C sequence and by a targeting domain in the NH₂-flanking propeptide. Based on evidence from other integral membrane proteins, we hypothesized that proSP-C targeting is effected by oligomerization of proSP-C monomers. To evaluate this in vitro, cDNA constructs encoding for either wild type proSP-C (pcDNA3/SP-C-(1-194)) or heterologous fusion proteins containing green fluorescent protein (EGFP) linked to SP-C-(1-194) (EGFP/SP-C-(1-194)), to mutant proSP-C lacking the NH₂ targeting domain (EGFP/SP-C-(24-194)), or to mature SP-C alone (EGFP/SP-C-(24-58)) were produced. In transfected A549 cells, fluorescence microscopy revealed that pcDNA3/SP-C-(1-194) and EGFP/SP-C-(1-194) were each expressed in CD63 (+), EEA1 () cytoplasmic vesicles. Expression of EGFP/SP-C-(24-194) or EGFP/SP-C-(24-58) resulted in translocation but retention in early compartments. When co-transfected with pcDNA3/SP-C-(1-194), both EGFP/SP-C-(24-194) and EGFP/SP-C-(24-58) were directed to CD63 (+) vesicles that also contained SP-C-(1-194). In contrast, trafficking of a folding mutant that forms juxtanuclear aggregates, EGFP/SP-C^C122/186G, was not corrected by cotransfection with pcDNA3/SP-C-(1-194). Chemical cross-linking studies of transfected cell lysates with bismaleimidohexane produced multimeric forms of both EGFP/SP-C-(1-194) and EGFP/SP-C-(24-58). These results indicate that sorting involves oligomeric association of proSP-C monomers mediated by the mature SP-C domain. Heteromeric assembly allows wild type proSP-C to facilitate trafficking of SP-C mutants with intact transmembrane domains but lacking targeting signals. We speculate that heterotypic oligomerization of wild type with SP-C folding mutants produces a dominant negative thus contributing to the pathology of chronic lung disease associated with patients heterozygous for mutant SP-C alleles.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Surfactant protein C (SP-C),¹ isolated from bronchoalveolar lavage (BAL) of a variety of mammalian species represents one of the most hydrophobic secretory proteins characterized (1-3). The alveolar form of SP-C (mature SP-C) is a 35-amino acid peptide (M_r 3,500) that contains an extremely high content of branched aliphatic side-chains as well as covalently attached palmitic acid at two adjacent cysteine residues in its NH₂ terminus (2). Mature SP-C partitions with surfactant phospholipid in organic extracts of BAL, and both purified natural or recombinant forms impart properties of rapid surface absorption and low surface tension to reconstituted synthetic lipids (4). In a lipid environment, the secondary structure of mature SP-C is predominantly -helical (5) and exhibits an extended lipophilic surface with positive charges located near the NH₂ terminus. The valyl-rich monomeric -helical state transforms into insoluble -sheet aggregates upon incubation in aqueous solvents (6). A dimeric form of SP-C has been recovered from BAL of several species, although the functional significance of this variant is unknown (7).

Synthesized exclusively by the alveolar type 2 cell as a 21-kDa proprotein precursor (proSP-C₂₁), the mature SP-C domain is positioned asymmetrically within the primary translation product flanked by propeptides of 23 amino acids at the NH₂ terminus and 136 residues at the COOH terminus (1-3, 8). Production of SP-C from this larger precursor is a multistep process requiring four discrete proteolytic cleavages, all occurring in subcellular compartments located distal to the medial-Golgi (9-14). Thus, translocation of synthesized proSP-C₂₁ to the ER and targeting of proprotein from the Golgi to distal vesicular organelles (i.e. multivesicular bodies and lamellar bodies) are crucial for biogenesis and secretion of the mature alveolar form.

Several important structural domains that influence proSP-C trafficking have been defined (9, 15). By deletion analyses, the mature SP-C domain (Phe²⁴-Leu⁵⁸) acting as a functional signal-anchor sequence was shown to mediate ER translocation (9). In vitro, the primary translation product has been shown to be a bitopic, transmembrane protein, anchored by the mature SP-C region in a type II (NH₂, cytosol/COOH, lumen) orientation (15, 16). ProSP-C remains in an integral membrane form during trafficking from the ER to the lamellar body, while undergoing proteolytic cleavage of the extramembrane propeptide regions (9, 13). Using both in vitro and in vivo models, the intracellular targeting motif has been localized to the cytosolic (NH₂-terminal) portion of proSP-C (9, 15). Fusion proteins containing EGFP and proSP-C mutants lacking conserved amino acid residues in the NH₂ domain (Glu¹¹-Thr¹⁹) are retained in calnexin-positive (ER) compartments (9, 10). Intratracheal injection of mice with adenovirus constructs encoding hemagglutinin (HA)-tagged SP-C proteins has confirmed that the NH₂ domain (Phe¹-Glu²³), in conjunction with the mature SP-C domain, is sufficient for secretion of SP-C (15).

Although not a functional targeting domain, the COOH-flanking propeptide (His⁵⁹-Ile¹⁹⁴) is also important to SP-C biosynthesis (11, 17). In contrast to NH₂ propeptide (ER retention) mutants, expression either of truncated forms or of site-directed mutants lacking one or both conserved cysteine in this COOH-flanking domain results in formation of unprocessed, ubiquitinated forms that aggregate and are instead directed to a newly defined subcellular compartment, the aggresome (18). The importance of disulfide-mediated folding of proSP-C has recently been underscored in vivo by the observation of chronic lung disease in a full-term infant with a mutation in the SP-C gene (19). The mutation c.460 + 1 G  A identified on one allele in both the patient and mother produced alternate splicing of the SP-C mRNA and exclusion of exon 4. Despite preservation of the mature protein sequence and NH₂ targeting motif, the resultant foreshortened mutant deleted one of two conserved cysteine residues in the COOH-flanking propeptide. The absence of detectable mature SP-C in this patient suggested that expression of the abnormal protein could have dominant negative effects on the function or metabolism of wild type SP-C produced from the normal allele.

Evidence is accruing that many disease-causing mutations and modifications that alter protein folding result in defective targeting of both the mutant protein and co-expressed native forms (20-22). Accumulation of mutant protein in the ER is a major component of the pathogenesis of several neurological diseases (23-25). Plasma membrane transport of the Trembler-J (TrJ) mutant of peripheral myelin protein 22 (PMP22) in Schwann cells is blocked in an intermediate compartment and affects the transport of wild type protein (23, 24). In co-transfection studies, direct interaction (heteromeric association) between the two isoforms was demonstrated using co-immunoprecipitation and resulted in reduced plasma membrane expression of wild type PMP22 imparting a molecular mechanism to the neuropathy observed in the TrJ mouse and in human Charcot-Marie-Tooth disease. In Pelizaeus-Merzbacher leukodystrophy, disease severity is linked to relative expression levels of isoforms of one or both intrinsic membrane proteins of myelin encoded by the Plp gene (PLP and DM20). In contrast to ER retention noted during expression of mutant DM20 and PLP together, co-expression of a transport-competent DM20 reduces accumulation of mutant PLP in the ER of oligodendrocytes through formation of a heterodimer (25). Together these models suggest that, for some hydrophobic membrane associated proteins, homodimerization is an essential component for normal biosynthesis and that wild type/mutant heterodimers formation can occur in disease.

Recently, targeting and secretion of HA-tagged mature SP-C lacking the NH₂ targeting motif did not occur in SP-C / mice but was facilitated by co-expression with endogenous wild type proSP-C present in control mice (15). This finding, coupled with the recently described exon 4 mutation, and accumulating evidence for oligomeric sorting of secreted integral membrane proteins, led us to hypothesize that sorting and trafficking of SP-C from proximal compartments involves homomeric association of proSP-C. The present study investigates the association of proSP-C monomers in vitro using both functional characterization by intracellular targeting as well as chemical cross-linking techniques in lung epithelial cells. We demonstrate that, via the mature SP-C (signal anchor) domain, proSP-C is capable both of homodimer formation during normal trafficking and of heterotypic interactions with SP-C mutants that alter trafficking patterns of both mutant and wild type isoforms.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials

pEGFP-C₁ plasmid was purchased form CLONTECH (Palo Alto, CA). The pcDNA3 eukaryotic expression plasmid was obtained from Invitrogen (San Diego, CA). Polyclonal anti-green fluorescent protein antiserum was purchased form Molecular Probes (Eugene, OR). Texas Red-conjugated monoclonal and polyclonal antibodies were obtained from Jackson Immunoresearch Laboratories, Inc. (West Grove, PA). Anti-CD63 was purchased from Immunotech, Inc. (Marseille, France). Anti-calnexin polyclonal antibody was purchased from StressGen (Victoria, British Columbia, Canada). Anti-ubiquitin monoclonal antibody was purchased from Chemicon (Temecula, CA). Monoclonal anti-HA (clone 12CA5 recognizing a peptide epitope from the hemagglutinin protein of human influenza virus) was obtained from Roche Molecular Biochemicals. Bismaleimidohexane (BMH) was purchased from Pierce. Except where noted, other reagents were of electrophoretic grade and were purchased from Bio-Rad or Sigma.

ProSP-C Antiserum-- The monospecific polyclonal antisera (NPROSP-C) directed against the NH₂ propeptide-flanking domain of rat proSP-C was previously produced in rabbits using a synthetic peptide immunogen (14). Anti-NPROSP-C (epitope = Met¹⁰-Glu²³) recognizes proSP-C₂₁ and all major intermediates but does not recognize mature SP-C or COOH propeptide fragments.

SP-C cDNA Expression Constructs

The cDNA constructs utilized in this study are schematically illustrated in Fig. 1. All procedures involving oligonucleotide and cDNA manipulations were performed essentially as described in Ref. 26.

pcDNA3-SP-C-(1-194)-- The vector for the expression of untagged wild type proSP-C in eukaryotic cells (Fig. 1E) was previously produced by subcloning of a full-length rat SP-C-(1-194) cDNA (816 bp) (8) insert into the pcDNA3 expression vector at the EcoRI site of the polylinker as described (11).

EGFP/SP-C Fusion Proteins-- Chimeric fusion proteins consisting of EGFP and wild type rat SP-C (EGFP/SP-C-(1-194); Fig. 1A) or mutant SP-C cDNA containing either a truncation removal of the NH₂-terminal-flanking (EGFP-C₁/SP-C-(24-194); Fig. 1B) or mature SP-C alone (EGFP-C₁/SP-C-(24-58); Fig. 1C) were generated by single-step PCR. pcDNA3-SP-C-(1-194) was used as a template. Production of the double cysteine mutant, EGFP/SP-C^C122/186G (Fig. 1D) was done with a single PCR performed using EGFP/SP-C^C122G as previously described in Ref. 18.

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Constructs utilized for analysis of proSP-C trafficking. A-D, EGFP/SP-C fusion constructs. A, EGFP/SP-C-(1-194) fusion construct containing full-length rat SP-C; B, deletional construct consisting of truncation of NH2 terminus; C, mature SP-C alone produced by truncation of both NH2- and COOH-flanking propeptides; D, double cysteine folding mutant containing Gly for Cys at residues 122 and 186 in the COOH-flanking propeptide; E, pcDNA3-SP-C-(1-194), which is a full-length rat SP-C cDNA as described (11); F, the HA sequence attached to the amino terminus of pcDNA3/SP-C-(1-194). All SP-C inserts were generated by PCR with pcDNA3-SP-C-(1-194) as template and subcloned into either the EGFPC1 or pcDNA3 vectors as described under "Experimental Procedures." Amino acid nomenclature is based on published rat SP-C sequence (8). CC represents putative palmitoylation sites at residues 28 and 29.

HA-tagged Wild Type ProSP-C-- using pcDNA-SP-C-(1-194) as a template, the HA tag (YPYDVPDYA) was added to the NH₂ terminus of rat SP-C-(1-194) using PCR. The 5' (forward) primer was an 82-mer and contained a KpnI site (bold), a Kozak sequence, and the HA coding sequence (underlined): TAC AAG GGT ACC ATG GAC TAC CCA TAC GAT GTT CCA GAT TAC GCT GCT GAC ATG GGT AGC AAA GAG GTA CTG ATG GAG AGC. The 3' (reverse) primer was a 21-mer that contained an XhoI site (bold): TCT AGA TGC ATG CTC GAG CG. Following amplification, the purified fusion insert was subcloned back into the pcDNA expression vector using digestion with KpnI and XhoI.

Cell Lines and Transfection

A549 Cell Line-- The lung epithelial cell line A549 utilized in transfection studies was originally obtained through the American Type Culture Collection (Manassas, VA) and has been used in prior studies (9-11, 18). A549 cells were grown at 37 °C in 5% CO₂ in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 100 IU/ml penicillin, and 100 µg/ml streptomycin.

Cell Transfection-- A549 cells, grown to 50% confluence on glass coverslips in 35-mm plastic dishes, were transiently transfected with single cDNA constructs (10 µg/dish) or co-transfected with two different constructs (5 µg/construct/dish) by CaPO₄ precipitation as previously described (9, 10, 18). The medium was replaced at 24 h, and transfected cells were maintained for up to 48 h.

Analytical Methods

Immunohistochemistry-- A549 cells grown on glass coverslips and transfected with EGFP chimeric constructs were prepared for indirect immunofluorescence microscopy as previously described (18). Following washing with phosphate-buffered saline (PBS) (Ca²⁺- and Mg²⁺-free), plated cells were fixed by immersion of coverslips in 4% paraformaldehyde at room temperature for 30 min, and then permeabilized by incubation for 30 min with 0.3% Triton X-100 in blocking buffer (5% bovine serum albumin and 10% normal goat serum in PBS). Immunostaining was performed by incubation with the primary antibody for 1 h at room temperature at the indicated dilutions. Coverslips were washed, incubated for 1 h at room temperature with a 1:200 dilution of secondary goat anti-mouse IgG monoclonal or secondary goat anti-rabbit IgG polyclonal antibodies, each conjugated to Texas Red. Cells were washed in PBS, air-dried at room temperature, and mounted on slides with Mowiol.

Fluorescence images were viewed on an Olympus I-70 inverted fluorescence microscope with filter packages High Q fluorescein isothiocyanate for EGFP (excitation at 480 nm, emission at 535/550 nm), and High Q TR for Texas Red (excitation at 560/555 nm, emission at 645/675 nm) obtained from Chroma Technology (Brattleboro, VT). Fluorescent and phase images were captured using a Hamamatsu 12-bit coupled-charge device camera. Image processing and overlay analysis were performed using IMAGE 1 software (Universal Imaging, West Chester, PA).

Polyacrylamide Gel Electrophoresis and Immunoblotting-- Cell pellets collected by scraping and centrifugation at 300 × g were solubilized with 40 µl of 50 mM Tris, 190 mM NaCl, 6 mM EDTA, 2% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, pH 7.4. Following centrifugation at 8,000 × g for 30 s to remove nuclei, proteins were separated by electrophoresis on a 12% polyacrylamide gel (27) and transferred to nitrocellulose (14).

Western blotting was done using successive incubations with primary polyclonal GFP antisera (1:5000) for 1 h and goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:10,000) for 1 h at room temperature. Bands were visualized by enhanced chemiluminescence using a commercially available kit (Amersham Biosciences). Fluorescence images were produced either by exposure to film or by direct acquisition using the Kodak 440 Image System. The intensity of each band was analyzed by either densitometric scanning of exposed film (Quantity 1 (PDI, Huntington Station, NY)) or by quantitation with Kodak1D software.

Chemical Cross-linking-- Chemical cross-linking of expressed proSP-C forms was performed as described by Millar et al. (28) for mitochondrial Mas70p. Transiently transfected A549 cells expressing either EGFP/SP-C-(1-194) or EGFP/SP-C-(24-58) were recovered from 35-mm² dishes 48 h after introduction of plasmid cDNA by scraping and centrifugation. The resulting cell pellets were resuspended in 10 µl of 0.25 M sucrose, 10 mM HEPES, pH 7.4, containing either BMH (10 mM), vehicle alone (Me₂SO), or reaction buffer alone. After incubation for 60 min at 4 °C, the reaction was stopped by addition of 100 mM dithiothreitol and the samples processed for analysis by SDS-PAGE and Western blotting as described above.

Statistical Analysis-- Mean data from multiple groups were compared with analysis of variance. All testing was performed with Excel Data Analysis software. Significance was accepted at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression of Wild Type and Mutant EGFP/SP-C Fusion Proteins-- Expression EGFP/SP-C fusion proteins was readily detected in transiently transfected A549 cells 24-48 h after introduction of plasmids. EGFP/SP-C-(1-194) was restricted to cytoplasmic vesicles of A549 cells consistent with successful translocation and export to distal compartments (Fig. 2, A, D, and G). EGFP/SP-C-(1-194) containing vesicles also stained positively with anti-NPROSP-C antiserum (Fig. 2, B and C), indicating that marker molecule and protein did not dissociate during trafficking. EGFP/SP-C-(1-194) containing vesicles also stained for CD63 (Fig. 2, E and F), a marker antigen associated with lamellar bodies and multivesicular bodies of type 2 cells (29, 30) but not with EEA1 (Fig. 2, H and I), an early endosome marker (31).

View larger version (45K): [in this window] [in a new window]   Fig. 2.   EGFP/SP-C-(1-194) chimeric protein is directed to CD63-positive, NPROSP-C-positive, EEA-1-negative cytoplasmic vesicles. A549 cells grown on glass coverslips at 80% confluence were transfected with 10 µg of EGFP-C1/SP-C-(1-194) using CaPO4 as described under "Experimental Procedures." At 48 h after transfection, cells were fixed, permeabilized, and stained with primary polyclonal NPROSP-C antiserum (A-C), primary monoclonal CD63 antiserum (D-F), or primary monoclonal EEA-1 antiserum (G-I) and IgG-specific secondary Texas Red-labeled antisera. Images acquired by video fluorescence microscopy using a High Q FITC filter for EGFP and a Texas Red filter are representative of four separate experiments and >50 cells for each construct. NPROSP-C staining was observed with Texas Red channel (B) and colocalized (yellow) with EGFP staining (A) on double-label color image assembled with green for EGFP and red for NPROSP-C (C). The majority of EGFP/SP-C-(1-194) (D) also colocalized with CD63 staining (yellow) (E) on color overlay (F). EGFP/SP-C-(1-194) green fluorescence (G) and EEA-1 staining (red) (H) failed to colocalize (I). N, nucleus.

In contrast to wild type proSP-C (Fig. 3A), EGFP-tagged SP-C mutants showed several distinct patterns of expression. EGFP/SP-C-(24-194), a construct lacking the NH₂-propeptide region, was expressed in a reticular pattern (Fig. 3B). This compartment also stained for the resident ER protein calnexin but not for ubiquitin (data not shown). EGFP/SP-C^C122/186G, a mutant containing the NH₂ targeting motif but lacking two conserved cysteine residues in the COOH propeptide region, was restricted to a perinuclear compartment (Fig. 3D) that was calnexin-negative, EEA-1-negative, but ubiquitin-positive (data not shown). EGFP/SP-C-(24-58), a fusion protein that contains only the mature SP-C sequence, translocated but formed perinuclear aggregates (Fig. 3C) that were ubiquitin-negative.

View larger version (49K): [in this window] [in a new window]   Fig. 3.   Subcellular distribution of EGFP/SP-C mutants. A549 cells were transiently transfected EGFP/SP-C-(1-194) (A), a mature SP-C fusion protein, EGFP/SP-C-(24-58) (B), the NH2-terminal truncated EGFP/SP-C-(24-194) (C), or the cysteine folding mutant EGFP/SP-CC122/186G (D) using CaPO4. Images for EGFP expression were acquired 48 h after transfection by fluorescence microscopy with a High Q FITC filter package. In contrast to EGFP/SP-C-(1-194) (A), EGFP/SP-C-(24-58) (B) and EGFP/SP-C-(24-194) (C) were translocated but retained in proximal compartments, whereas the cysteine folding mutant EGFP/SP-CC122/186G formed perinuclear aggregates (D). N, nucleus.

Wild Type ProSP-C Facilitates Trafficking of Mutant EGFP/SP-C Fusion Protein-- Functional evaluation of the interaction of proSP-C forms in the sorting process was studied by co-transfection studies. We hypothesized that if trafficking of proSP-C required oligomerization, then trafficking of mutant proSP-C forms lacking NH₂ propeptide domains normally retained in the ER could be facilitated by heterotypic interaction with wild type proSP-C containing complete targeting motifs. In Fig. 4, the construct pcDNA3-SP-C-(1-194) was used to express an untagged form of wild type proSP-C for co-expression with EGFP/SP-C-(24-194). At an equal molar ratio, both the truncated NH₂ mutant (detected by EGFP fluorescence; Fig. 4B) and the wild type form (detected immunostaining for proSP-C; Fig. 4C) co-localized within cytosolic vesicles (Fig. 4D). The co-transfection of EGFP/SP-C-(24-194) and wild type SP-C now resulted in delivery of mutant proSP-C to CD63-positive compartments (Fig. 5D). In control experiments, co-transfection with an irrelevant plasmid encoding for chloramphenicol acetyltransferase (pCAT) resulted in continued retention of EGFP/SP-C-(24-194) in the ER (Fig. 4A). Therefore, these data indicate that co-expression of wild type proSP-C can alter spatial expression of mutant SP-C forms lacking targeting motifs by facilitating transfer from proximal to distal compartments.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Wild type proSP-C facilitates transport of EGFP/SP-C-(24-194) to NPROSP-C-positive cytoplasmic vesicles. The EGFP/SP-C-(24-194) plasmid (5 µg) was co transfected at an equimolar ratio with either an empty plasmid (pCAT; panel A) or with pcDNA3/SP-C-(1-194) (panels B-D) into A549 cells by CaPO4. 48 h after transfection, cells were fixed, permeabilized, and stained with anti-NPROSP-C using TR-conjugated secondary antibody. Fluorescence images acquired using High Q FITC filter for EGFP (panels A and B) and a High Q TR filter package for NPROSP-C (panel C) revealed that EGFP/SP-C-(24-194) (B) colocalized with wild type proSP-C (C) in cytosolic vesicles (yellow) (D). N, nucleus.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   EGFP/SP-C-(24-194) is directed to CD63-positive vesicles following co-expression with wild type proSP-C. The fusion protein of EGFP/SP-C-(24-194) (5 µg) was transfected into A549 cells alone (panel A) or at an equimolar ratio with pcDNA3/SP-C-(1-194) (panels B-D) by CaPO4. 48 h after transfection, cotransfected cells were fixed, permeabilized, and stained with anti-CD63 using TR-conjugated secondary antibody. Fluorescence images acquired using High Q FITC filter for EGFP (panels A and B) and a High Q Texas Red filter package for CD63 staining (panel C) revealed that, although EGFP/SP-C-(24-194) alone was retained in ER compartments (A), cotransfection with wild type proSP-C resulted in colocalization (yellow) of EGFP/SP-C-(24-194) within CD63-positive compartments (D). N, nucleus.

Co-transfection with Wild Type SP-C Promotes Processing of Mutant SP-C-EGFP/SP-C-(24-- 194)The molecular forms of EGFP/SP-C-(24-194) produced in A549 cells were assessed by Western blotting. Using anti-GFP antibody, A549 cells transfected with EGFPC₁ alone showed expression of a major product matching the predicted size of 27 kDa (Fig. 6A, lane 1), which was not detected in cell lysates of mock-transfected cells (data not shown). When EGFP/SP-C-(1-194) was introduced, lysates contained three anti-GFP-positive bands. A form of 48 kDa, corresponding to the predicted size of the primary translation product of the fusion protein, as well as two smaller intermediates (doublet), were identified (Fig. 6A, lane 2). A band with M_r 27,000 (indicating liberation of free EGFP) was not found, showing that EGFP/SP-C-(1-194) was partially processed in this cell-type by two-step cleavage of the COOH terminus. In contrast, EGFP/SP-C-(24-194) was expressed predominantly as a single foreshortened primary translation product with M_r 45,000 (Fig. 6A, lane 3). However, if co-transfected with pcDNA3-SP-C-(1-194), processing of the EGFP fusion protein is significantly increased (Fig. 6A, lane 4) as compared with co-transfection with an equal amount of irrelevant plasmid, pCAT (Fig. 6A, lane 5).

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Cotransfection with pcDNA3/SP-C-(1-194) induces COOH-terminal processing of the SP-C mutant EGFP/SP-C-(24-194). A, Western blot analysis for detection of EGFP proteins. A549 cells (1 × 106) transfected with EGFPC1 (10 µg) (lane 1), EGFP/SP-C-(1-194) (10 µg) (lane 2), EGFP/SP-C-(24-194) (10 µg) (lane 3), pcDNA3/SP-C-(1-194) (5 µg) + EGFP/SP-C-(24-194) (5 µg) (lane 4), pCAT (5 µg) + EGFP/SP-C-(24-194) (5 µg) (lane 5) using CaPO4 were harvested 48 h after transfection by scraping and centrifugation for 10 min at 300 × g. Nuclear-free lysates were prepared from cell pellets as described under "Experimental Procedures." Fifty percent of each nuclear-free lysate was subjected to 12% SDS-PAGE, transferred to nitrocellulose and immunoblotted with primary polyclonal anti-GFP and bands visualized using ECL. EGFP/SP-C-(1-194) was expressed as a primary translation product with Mr of 48,000 with a lower molecular weight doublet (Mr = 33,000-40,000) consistent with COOH propeptide processing (lane 2). EGFP-C1 was expressed as a major product with Mr 27,000 (lane 1). EGFP/SP-C-(24-194) alone (lane 3) or co-transfected with pCAT (lane 5) was expressed predominantly as a foreshortened product of Mr 46,000 with minimal processing intermediates. Co-transfection with pcDNA3/SP-C-(1-194) resulted in an increase in EGFP/SP-C-(24-194) processing (lane 4). B, for each condition, the intensity of the primary translation product band was compared with that of the processing intermediates by densitometric scanning. The ratio of the processed intermediate doublet/proprotein for each condition was determined from three separate experiments and expressed as mean ± S.E. EGFP/SP-C-(1-194) generated ~50% processed intermediates (**, p < 0.05 versus other conditions by analysis of variance). Co-transfection with pcDNA3/SP-C-(1-194) quantitatively enhanced processing EGFP/SP-C-(24-194) compared with single transfection or pCAT co-transfection (*, p < 0.05).

Quantitative densitometry confirmed that the presence of wild type proSP-C could significantly enhance processing of EGFP/SP-C-(24-194) (Fig. 6B). When expressed as the ratio of the intensity of processing intermediates to primary translation product, cells transfected with EGFP/SP-C-(1-194) contained ~50% processed forms. In contrast, EGFP/SP-C-(24-194) remained almost 90% in the unprocessed state. Co-expression of wild type SP-C induced a 3-fold increase in the relative amount of processed proSP-C compared with either single transfection or in co-transfections containing the irrelevant plasmid pCAT.

Mature SP-C Can Be Directed to Cytosolic Compartments by Wild Type ProSP-C-- The in vitro or in vivo expression of mature SP-C in lung epithelial cells lacking endogenous SP-C results in retention of transfected protein in proximal compartments. To test whether the interaction of wild type proSP-C with mature SP-C could alter the sorting process, chimeric EGFP/SP-C-(24-58) and wild type proSP-C (pcDNA3-SP-C-(1-194)) were expressed together into A549 cells. Using an equimolar ratio, the presence of wild type proSP-C was sufficient to direct EGFP/SP-C-(24-58) to cytosolic vesicles. Using immunofluorescence labeling with an epitope-specific proSP-C antibody that does not recognize the mature SP-C domain, these vesicles were found to contain both proteins (Fig. 7, C and D). In contrast co-transfection with irrelevant pCAT plasmid did not affect spatial expression of the EGFP/SP-C-(24-58) fusion (Fig. 7A).

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Co-expression of wild type pcDNA3/SP-C-(1-194) facilitates transport of EGFP-tagged mature SP-C to proSP-C-positive cytoplasmic vesicles. The fusion protein of EGFP/SP-C-(24-58) (5 µg) was co transfected at an equimolar ratio with either an empty plasmid (pCAT; panel A) or with pcDNA3/SP-C-(1-194) (panels B-D) into A549 cells by CaPO4. 48 h after transfection, cells were fixed, permeabilized, and stained with anti-NPROSP-C and TR-conjugated secondary antibody. Fluorescence images were acquired using High Q FITC filter for EGFP (panels A and B) and a High Q TR filter package for NPROSP-C (panel C). Although EGFP/SP-C-(24-58) was retained in proximal compartments (A), overlay images (panel D) indicated that co-expression with wild type proSP-C resulted in direction of EGFP/SP-C-(24-58) to vesicles (B) that stained for wild type proSP-C (yellow) (C). N, nucleus.

These data indicate that co-trafficking is mediated by the close interaction of signal anchor domains (mature SP-C segment) and provide direct in vitro evidence extending recent in vivo data in mice suggesting that endogenous proSP-C can direct sorting of exogenous mature SP-C.

Cross-linking Studies-- To examine the possibility that proSP-C-(1-194) oligomerizes during trafficking, we took advantage of prior structural data indicating that cysteine residues located at residues 28 and 29 of the proprotein were positioned adjacent to the transmembrane segment of the signal anchor (mature SP-C) domain (16, 32, 33). In the event that dimers form, these cysteine residues, located on the cytosolic side of the membrane, would be expected to be in close proximity (Fig. 8A) and therefore available for chemical cross-linking by thiol-specific homobifunctional reagents such as BMH. Following transfection with EGFP/SP-C-(1-194), treatment of cell lysates with BMH resulted in formation of several cross-linked forms identified by SDS-PAGE and Western blotting with anti-GFP antiserum. BMH-treated lysates demonstrated bands of 96 kDa (consistent with a homodimer of the fusion protein) as well as two other higher molecular weight bands representing oligomers (Fig. 8B). The appearance of all larger bands was dependent on BMH and was not observed in lysates treated with solvent (Me₂SO) or incubated with buffer alone.

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Chemical cross-linking of EGFP/SP-C fusion proteins in A549 cell membranes. A549 cells expressing either EGFP/SP-C-(1-194) (B) or EGFP/SP-C-(24-58) (C) were subjected to chemical cross-linking with BMH as schematically illustrated in panel A and detailed under "Experimental Procedures." Following resuspension of transfected cells in 0.25 M sucrose, 10 mM HEPES, pH 7.4, containing 10 mM BMH (BMH), vehicle (DMSO), or no additive (Buffer), mixtures were incubated for 60 min at 4 °C and the reaction terminated by addition of 100 mM dithiothreitol. The samples were analyzed by 12% SDS-PAGE and Western blotting with anti-EGFP antibody as described under "Experimental Procedures." Treatment of cells expressing EGFP/SP-C-(1-194) with Me2SO alone resulted in identification of an EGFP/monomer at Mr 48,000 as well as lower molecular weight processed forms. Addition of BMH produced cross-linked products migrating as dimers and higher molecular weight oligomeric forms. In an analogous fashion, analysis of cells expressing EGFP/SP-C-(24-58) (C) and treated with Me2SO identified an EGFP/SP-C monomer at Mr 30,000 without evidence of processing. Inclusion of BMH produced a major cross-linked product at Mr 60,000 with minor oligomeric forms. A nonspecific band (*) at ~ 55 kDa was observed in all lanes.

We next tested the hypothesis that homodimer formation was mediated by interactions between signal anchor domains of adjacent proSP-C molecules. To evaluate this possibility, we performed additional cross-linking studies on cell lysates from A549 cells transfected with the construct EGFP/SP-C-(24-58). Treatment with BMH resulted in production of a non-reducible band at M_r 60,000 consistent with the predicted size of a cross-linked, non-reducible fusion protein homodimer (Fig. 8C). These results indicate that the propeptide-flanking domains are not required for oligomerization.

COOH Cysteine Mutants Act as Dominant Negatives-- Because the transport of membrane-associated proteins such as PMP22 can be affected by interaction with mutant isoforms of these proteins (23, 24), we performed experiments to determine whether the expression of some SP-C mutants could alter trafficking of wild type SP-C. We have shown that mutation or deletion of one or both conserved cysteine residues at COOH terminus of proSP-C results in misfolding with mistargeting of unprocessed mutant protein, leading to formation of stable aggregates within aggresomes (18). To test if co-transfection could change the targeting of proSP-C aggregation mutants, wild type pcDNA3-SP-C-(1-194) was co-transfected into A549 cells together with the double cysteine mutant EGFP/SP-C^C122/186G. As shown in Fig. 9, in contrast to the NH₂-deleted mutants retained in the ER, co-transfection of wild type proSP-C did not change the steady state intracellular distribution of proSP-C folding mutants. EGFP/SP-C^C122/186G remained aggregated in a proximal compartment despite introduction of equimolar amounts of pcDNA3SP-C-(1-194) (Fig. 9, A and B). To examine whether these proSP-C folding mutants acted as dominant negatives, we examined the distribution of HA-tagged wild type proSP-C during co-transfection studies with GFP-tagged forms. Control experiments (Fig. 9, C and D) using co-transfection of both HA- and GFP-tagged wild type proSP-C demonstrated that the HA tagged form co-localized with EGFP/SP-C-(1-194) in cytoplasmic vesicles. In contrast, the COOH folding mutant EGFP/SP-C^C122G inhibited the delivery of HA tagged SP-C-(1-194) to cytosolic vesicles with both forms co-localizing within perinuclear aggregates (Fig. 9, E and F).

View larger version (40K): [in this window] [in a new window]   Fig. 9.   SP-C COOH folding mutants alter trafficking induce retention of wild type proSP-C. A and B, A549 cells were co-transfected with 5 µg of EGFP/SP-CC122/186G and 5 µg of either pCAT (A) or pcDNA3-SP-C-(1-194) (B) plasmids using CaPO4. EGFP expression localized 48 h after transfection by fluorescence microscopy with a High Q FITC filter demonstrates aggregation of EGFP fusion product despite the presence of wild type proSP-C. C and D, the fusion protein EGFP/SP-C-(1-194) (5 µg) was co-transfected at an equimolar ratio with HA/SP-C-(1-194). Fluorescence images of cells fixed 48 h after transfection were acquired using High Q FITC filter for EGFP (C) and a High Q TR filter for -HA staining (D) and demonstrate that EGFP- and HA tagged SP-C-(1-194) co-localize. E and F, the COOH SP-C folding mutant EGFP/SP-CC122G (5 µg) was co-transfected into A549 cells at an equimolar ratio with HA/SP-C-(1-194) by CaPO4. Cotransfected cells were fixed 48 h after transfection, permeabilized, and stained with primary anti-HA and secondary TR-conjugated antibody. Fluorescence images for EGFP (E) and HA staining (F) revealed colocalization of the tagged constructs in perinuclear aggregates. N, nucleus.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The surfactant protein C proprotein is a bitopic integral membrane protein that is trafficked to the multivesicular body and lamellar body of type 2 cells, where the mature protein is subsequently secreted with SP-B and surfactant lipids into the alveolar space via regulated exocytosis (1-3, 34). The biosynthetic pathway of SP-C comprises a number of discrete processing steps that are dependent upon sorting and delivery of the SP-C proprotein from proximal compartments to distal organellar sites. In previous studies, we defined two regions in the proSP-C sequence required for trafficking: the mature SP-C domain (Phe²⁴ to Leu⁵⁸), which functions as a signal/anchor sequence active in epithelial and non-epithelial cells, and the region Met¹⁰ to Thr¹⁹, contained within the NH₂-flanking propeptide, which is required for targeting to late vesicular compartments (9, 10). Using both functional and biochemical approaches, the present study extends these concepts by demonstrating that, in addition to membrane integration, biosynthetic processing of proSP-C also involves protein-protein interaction in the form of homotypic oligomerization. This event provides a mechanism whereby: 1) mutant SP-C proteins lacking targeting motifs can be targeted to distal compartments via heterotypic interactions with wild type SP-C; 2) misfolded SP-C mutants, diverted to ubiquitin-mediated degradative pathways, can interact with wild type SP-C, thereby functioning as dominant negatives. These additions to the molecular model for SP-C biosynthesis explain recent in vivo observations in which endogenous proSP-C in wild type mice (but not present in SP-C null littermates), facilitated the secretion of adenovirus-transfected HA-tagged mature SP-C (15) and in which humans with familial interstitial lung disease and heterozygous for a misfolded mutant proSP-C allele exhibited altered trafficking of wild type protein (19).

The role of oligomerization in SP-C biosynthesis is supported by four lines of evidence. 1) Co-transfection of wild type protein facilitates transfer of mutant forms lacking the NH₂ targeting domain (EGFP/SP-C-(24-194)) from proximal compartments to cytosolic vesicles (Figs. 4 and 5). 2) Double-labeling immunofluorescence co-localizes the mutant and wild type forms in the same intracellular compartments (Fig. 4). 3) Facilitated transfer of EGFP-(24-194) by co-transfected wild type proSP-C promotes intracellular processing similar to that of transfected wild type proSP-C (Fig. 6). 4) Chemical cross-linking with BMH demonstrates that SP-C-(1-194) is capable of forming homodimers (Fig. 8B), indicating close spatial interaction between monomers (Fig. 8A).

Intermolecular interaction via homodimerization has been shown to play a critical role during sorting of both luminal proteins and integral membrane proteins. Chromogranins A and B, found in the matrix of dense secretory granules of the adrenal medulla, each homotypically oligomerize shortly after synthesis in the endoplasmic reticulum. From a variety of structural analyses, maintenance of a folded conformation via a disulfide-bonded loop in the NH₂ terminus as well as other structure-dependent motifs in the COOH termini of these soluble, luminal molecules are prerequisites for their sorting in the trans-Golgi-network to secretory granules (35-38). Importantly, these studies have also shown that heterotypic interactions between transfected mutant forms and endogenous granins can facilitate normal anterograde trafficking of mutant proteins (39). Likewise, many integral membrane proteins also require oligomerization for sorting. Through the use of chimeric proteins, the integral mitochondrial membrane protein, Mas70p, has also been shown to undergo oligomer-dependent sorting facilitated by its signal anchor domain (28, 40).

In contrast to either of these proteins, SP-C and its proprotein precursor represent a hybrid molecule in that, within the biosynthetic pathway, the proprotein is transferred/transported as an integral membrane protein and then deposited into the lamellar body lumen and secreted in tight association with phospholipids and SP-B into the alveolar space (1-3, 34). Deletion analysis in this study indicates that the mature SP-C domain, which contains a transmembrane -helix (spanning positions 34-58), is a promoting structure for formation of oligomers. The trafficking pattern of tagged mature SP-C (EGFP/SP-C-(24-58)) can be redirected from retention in proximal compartments to CD63-positive acidic vesicles by wild type proSP-C (pcDNA3/SP-C-(1-194)) as detected by colocalization of EGFP fluorescence and wild type proSP-C using an epitope specific proSP-C antibody, which recognizes the NH₂-flanking propeptide present only in the unlabeled wild type proSP-C (Fig. 7). The use of the minimal construct containing only mature SP-C indicates that this is the most likely region mediating homodimeric association. Interestingly, similar commensurate signal anchoring and oligomerization properties have been attributed to Mas70P (28, 40).

Direct interactions between SP-C proteins were demonstrated via chemical cross-linking, complementing the functional evidence shown by co-transfection. The data in Fig. 8 indicate that oligomeric interactions are mediated by non-covalent interaction of the transmembrane domains. BMH is a well characterized homobifunctional molecule capable of cross linking Cys-SH residues of adjacent molecules that are within 16.1 Å (see Fig. 8A) (28). This technique has been used previously to study molecular determinants of oligomerization in Mas 70P. Because of the instability of the SP-C tertiary structure in aqueous environments, in situ chemical cross-linking offers advantages over other techniques for the detection of protein-protein interactions. The -helical form of SP-C seems to be thermodynamically most stable in a micellar environment, whereas upon removal from lipid, it transforms into insoluble -sheet aggregates as a result of a high kinetic barrier for unfolding in aqueous solvents (5). In fact, if left in aqueous solutions, it can go on to form insoluble amyloid fibrils (6, 41). Therefore, any conditions that compromise the integrity of the -helical structure of the transmembrane domain (e.g. extraction into aqueous detergent solutions, denaturation, and/or immunoprecipitation) would likely affect the direct detection of interaction of mutant and wild type forms of SP-C through the disruption of tertiary and quaternary structure. ProSP-C is a type II integral membrane protein (15, 16), and, in the event that homo-oligomers form via the signal anchor, cysteine residues on the cytosolic portion of adjacent proSP-C molecules would be located within 5 amino acids of the membrane -helix and in close proximity for available linkage via BMH (Fig. 8A). Under reducing conditions, SDS-PAGE identifies monomeric EGFP/SPC-(1-194) , which forms higher oligomers upon BMH-mediated cross-linking (Fig. 8B). Mature SP-C alone can also form homodimers when EGFP/SP-C-(24-58) was cross-linked with BMH (Fig. 8C). In both cases, oligomers captured with BMH treatment produced integer values for higher molecular mass forms of 96 kDa and higher (for EGFP/SP-C-(1-194)-BMH-EGFP/SP-C-(1-194)) and 60 kDa (for EGFP/SP-C-(24-58)-BMH-EGFP/SP-C-(24-58)), indicating that heterotypic oligomerization with another protein is unlikely.

Recent in vivo observations lend further support for the notion of oligomerization in a molecular model for sorting and trafficking of proSP-C (depicted in Fig. 10). Adenovirus-mediated expression of HA-tagged mature human SP-C (hSP-C-(24-57)) resulted in secretion of the HA peptide by wild type mice but not by SP-C null (/) mice (15). The current model proposes that SP-C mutants lacking targeting signals (EGFP/SP-C-(24-194) and EGFP/SP-C-(24-58)), are translocated to the ER, anchor into the membrane, and adopt a functional conformation. However, absence of the NH₂ terminus prevents transport out of the ER, resulting in retention (Fig. 10A). Cotransfection of wild type SP-C or expression of these mutants in wild type mice leads to facilitation of transport via the NH₂ targeting motif contained within the wild type protein to cytosolic compartments for processing and secretion (Fig. 10B). Thus, both in vivo and in vitro, if folded properly, motif-deficient, HA-or EGFP-tagged SP-C constructs can literally be dragged through the biosynthetic pathway by wild type SP-C containing the missing targeting sequences.

View larger version (19K): [in this window] [in a new window]   Fig. 10.   Model of oligomeric association for sorting of SP-C: schematic presentation of possible fates of synthesized proSP-C. A, single transfection of NH2-terminal truncated EGFP/SP-C-(24-194). Mutants lacking this domain are translocated but retained in proximal compartment without further processing to cytoplasmic vesicles. B, co-transfection of EGFP/SP-C-(24-194) with wild type SP-C (pcDNA3-SP-C-(1-194)) results in heteromeric association and redirection of mutant EGFP/SP-C-(24-194) to CD63 (+), NPROSP-C-positive vesicles. C and D, in contrast, COOH folding mutants are retrotranslocated from the ER and directed preferentially to degradative pathways forming aggregates. Through heteromeric association and retrotranslocation prior to sorting in the Golgi (D), folding mutants can aggregate wild type SP-C producing dominant negatives. Not shown is the fate of wild type proSP-C, which, after homomeric dimerization, is sorted and targeted to the regulated secretory pathway.

This model also provides a mechanism for the creation of dominant negatives by misfolded SP-C mutants. We have previously demonstrated that proSP-C mutants lacking conserved cysteine residues on the proSP-C COOH-flanking domain are misfolded, mistargeting, ubiquitinated, and then directed to proteasome-dependent pathways (11, 18) (Fig. 10C). Importantly, the steady state expression of this type of mutant (EGFP/SP-C^C122/186G) could not be reversed by cotransfection with wild type proSP-C (Fig. 8), and HA-tagged wild type protein was redirected to proximal compartments (Fig. 9). An analogous human mutation has been reported recently (19). In this case, both mother and a full-term infant were heterozygous for a single base mutation (G to A base substitution) in the first codon of intron 4 (460 = 1 G  A), resulting in abnormal splicing and a skip of exon 4 leading to production of a foreshortened proSP-C isoform missing a conserved cysteine residue at position 120/121 in the COOH-flanking region. The index patient was heterozygous, proprotein was made, but little or no SP-C_3.7 was detected. Furthermore, there was evidence of alveolar inflammation and fibrosis. The functional consequence of this mutation, i.e. absence of mature SP-C, is consistent with formation of proprotein heterodimers and impairment of trafficking and thus processing of the wild type isoform. Mutation of this same conserved residue in the analogous region of the rat isoform (Cys¹²²) obliterates disulfide-mediated folding, promotes aggresome formation, and facilitates co-aggregation of wild type and mutant protein forms (11, 18) (Fig. 9).

Our model for oligomeric sorting of SP-C can also account for these observations. Expression of folding mutant molecules subsequently tagged for degradative pathways by ubiquitination can function as dominant negatives in which the recognition of misfolding in the ER and retrotranslocation of mutant protein to the cytosol precedes and preempts a sorting signal from wild type protein normally active at the trans-Golgi network (Fig. 10D). Oligomeric association thus diverts both mutant and wild type forms prior to entry into sorting compartments, thereby functioning as a dominant negative. This pattern observed in SP-C biosynthesis is similar to what has been described in some chronic degenerative neurological diseases. The abnormal trafficking of the PMP22 gene product to the plasma membrane has been linked to a set of dominantly inherited peripheral neuropathies including the Trembler J mouse and Charcot-Marie-Tooth I disease (23, 24). Wild type PMP22 protein, a homodimer, has been shown to be affected by the presence of mutant PMP22 protein in which the neuropathy appears to result both from decreased trafficking of wild type PMP22 and from a toxic gain of function via the accumulation of wild type and TrJ-PMP22 in the intermediate compartment.

In summary, protein-protein interactions have been shown to be required for normal protein sorting and have been associated with various pathophysiological conditions, including chronic degenerative disease in the central nervous system and other organs. Although somewhat unique in that it is both an integral membrane protein and a secreted peptide, surfactant protein C appears to represent another in a growing class of proteins that require homomeric association for sorting and trafficking. Further characterization of these events will allow for a more detailed understanding of the molecular basis for the increasingly recognized syndrome of chronic lung disease associated with heterozygous expression of mutant forms of proSP-C in humans (19), for the developmental lung toxicity recently described in transgenic mice overexpressing mature SP-C alone (42), and for the apparent lack of pathology found in homozygous SP-C null mice (43). Additional studies aimed at understanding SP-C biosynthesis in the context of the emerging concept of conformational disease are in progress.

	ACKNOWLEDGEMENTS

We thank Seth Thomas Scanlon for assistance with image processing. We thank Drs. Susan Guttentag and Michael Koval for editorial assistance.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants HL-19737 and P50-HL56401 (both to M. F. B.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Pulmonary and Critical Care Division, University of Pennsylvania School of Medicine, 807 BRB II/III Bldg., 421 Curie Blvd., Philadelphia, PA 19104-6160. Fax: 215-573-4469; E-mail: mfbeers@mail.med.upenn.edu.

Published, JBC Papers in Press, March 20, 2002, DOI 10.1074/jbc.M201537200

	ABBREVIATIONS

The abbreviations used are: SP-C, pulmonary surfactant protein C (3.7 kDa); BAL, bronchoalveolar lavage; BMH, bismaleimidohexane; EGFP, enhanced green fluorescent protein; ER, endoplasmic reticulum; FITC, fluorescein isothiocyanate; GFP, green fluorescent protein; HA, hemagglutinin antigen; PBS, phosphate-buffered saline; TR, Texas red.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES