Human surfactant protein C (hSP-C) is synthesized by the alveolar type 2 cell as a 197 amino acid integral membrane proprotein and proteolytically processed to a secreted 3.7 kDa mature form. Although the SP-C null mouse possesses a non-lethal phenotype, a heterozygous substitution of A for G in the first base of intron 4 of the human SP-C gene (c.460+1A>G) has been reported in association with familial interstitial lung disease and absence of mature protein. This mutation produces a splice deletion of exon 4 (ΔExon4) resulting in removal of a positionally conserved cysteine in the C-terminal flanking propeptide. Based on a prior study showing that an identical deletion in the rat isoform diverted mutant protein to stable aggregates, we hypothesized that expression of the ΔExon4 mutation would result in disruption of intracellular trafficking of both mutant and wild-type proSP-C. We tested this in vitro using fusion proteins of EGFP conjugated either to wild-type SP-C (EGFP/hSP-C1-197) or to SP-C deleted of Exon4 (EGFP/hSP-CΔExon4). Fluorescence microscopy showed that EGFP/hSP-C1-197 transfected into A549 cells was expressed in a punctuate pattern in CD63 (+) cytoplasmic vesicles, whereas EGFP/hSP-CΔExon4 accumulated in ubiquitinated perinuclear inclusions linked to the microtubule organizing center. A similar juxtanuclear pattern was observed following transfection of SP-C cDNA lacking only cysteine residues in the C-terminal propeptide encoded by Exon 4 (EGFP/hSP-CC120/121G). To evaluate whether mutant proSP-C could function as a dominant negative, EGFP/hSP-CΔExon4 was cotransfected with HA-tagged hSP-C1-197 and resulted in the restriction of both forms to perinuclear compartments. Addition of Na+ 4-phenylbutyrate, a facilitator of trafficking of other misfolded proteins, attenuated the aggregation of EGFP/hSP-CΔExon4. We conclude that c.460+1A>G mutation of human SP-C results in disruption of disulfide-mediated folding encoded by Exon 4 leading to diversion of unprocessed proSP-C to aggresomes. The heterotypic oligomerization of hSP-C1-197 and hSP-CΔExon4 provides a molecular mechanism for the dominant-negative effect observed in vivo.
Many inherited diseases are caused by the homozygous expression of mutant protein that induces a loss-of-function through interference with protein trafficking or function. In contrast, the heterozygous expression of some mutations appears to be sufficient to generate pathology (Kopito and Ron, 2000). Such diverse disorders as Alzheimer's and Parkinson's diseases (Carrell and Lomas, 1997), spinocerebeller ataxia (Chai et al., 1999), Charcot-Marie-Tooth disease (Notterpek et al., 1999; Vanslyke et al., 2000), prion encephalopathies (Prusiner and Hsiao, 1994), inflammatory liver disease associated with α1 antitrypsin deficiency (Carrell and Lomas, 2002), systemic amyloidosis (Perutz, 1997) and cystic fibrosis (Johnston et al., 1998) are part of a growing list of inherited diseases known as `conformational' diseases that arise when a mutant protein undergoes structural rearrangement that endows an ability to aggregate. At the recent international conference `Alpha-1 antitrypsin deficiency and other conformational diseases', the mechanisms by which cells respond to production of abnormal protein conformers and how misfolded proteins cause cytotoxicity were highlighted as important emerging questions (Kopito and Ron, 2000).
Surfactant protein C (SP-C) is a 35 amino acid hydrophobic lung-specific protein produced exclusively by alveolar type 2 cells (Weaver and Conkright, 2001; Weaver, 1998; Solarin et al., 2001). The extreme hydrophobicity of the alveolar form (SP-C3.7) is due to a 23 amino acid, polyvaline domain. Modeling of its secondary structure and biophysical analyses of purified SP-C3.7 indicate that in a lipid environment, monomeric SP-C (positions 9-34) exists as a stable valyl-rich helix oriented parallel to lipid acyl chains and able to span phospholipid bilayers. In solution, monomeric SP-C transforms from α-helix toβ -sheet aggregates reminiscent of amyloid fibril formation. Amyloid SP-C fibrils were isolated from bronchoalveolar lavage of patients with alveolar proteinosis demonstrating a propensity for this motif to undergo conformational changes (Gustafsson et al., 1999; Johansson, 2001; Kallberg et al., 2001).
Human SP-C (hSP-C1-197) is synthesized as a 197 amino acid proprotein and proteolytically processed as an integral membrane protein to the 3.7 kDa mature form that is subsequently transferred to the lumen of lamellar bodies for secretion into the alveolar space with surfactant phospholipids (Beers, 1998; Solarin et al., 2001). Complete biosynthesis requires four endoproteolytic cleavages of the SP-C propeptide and depends upon oligomeric sorting and targeting to subcellular processing compartments distal to the Golgi (Beers and Lomax, 1995; Johnson et al., 2001; Vorbroker et al., 1995; Wang et al., 2002). Studies using both the rat and human isoforms have demonstrated that the mature SP-C domain contained within the propeptide (residues 24-58) functions as a signal-anchor sequence effecting endoplasmic reticulum (ER) translocation, establishing a type II (NH2cytoplasm/COOHlumen) integral membrane orientation, and facilitating homomeric association during sorting (Russo et al., 1999; Keller et al., 1991; Wang et al., 2002). The NH2 flanking propeptide contains a functional targeting motif (E11-T19) for direction to late compartments; deletion or alteration of this region results in ER retention (Johnson et al., 2001). Conversely, removal or alteration of one or more cysteine residues from the COOH flanking propeptide results in mutant protein accumulation in a novel juxtanuclear compartment, the aggresome (Beers et al., 1998; Kabore et al., 2001).
Although the SP-C null mouse is viable (Glasser et al., 2001), recent reports in humans have shown heterozygous expression of over ten different mutant proSP-C forms in association with an absence of mature SP-C and interstitial lung disease (Nogee et al., 2001; Thomas et al., 2002; Nogee et al., 2002). The first SP-C mutation associated with interstitial lung disease was described in a term infant and mother both heterozygous for a G to A base substitution in the first codon of Intron 4 (Nogee et al., 2001). This mutation leads to abnormal splicing and a skip of Exon 4 (`hSP-CΔExon4') producing a foreshortened proSP-C lacking a conserved cysteine residue in the flanking COOH propeptide. Since cysteines in human proSP-C deleted by this mutation are positionally identical to a cysteine residue in the COOH flanking propeptide of the rat isoform that is essential for normal trafficking (Russo et al., 1999), we hypothesized that expression of the hSP-CΔExon4 isoform would result in disruption of COOH propeptide folding manifested in vitro by abnormal intracellular trafficking of mutant proSP-C, aggresome formation, and disruption of biosynthesis of the wild-type SP-C through creation of a dominant-negative effect.
To test this hypothesis, we expressed fusion proteins of EGFP and human proSP-C mutants lacking conserved cysteine residues, including the hSP-CΔExon4 mutation, and evaluated targeting and processing of expressed proteins in a transfected alveolar epithelial cell line. We report the first demonstration that expression of a clinically defined SP-C mutation results in accumulation of unprocessed, aberrant protein forms in a juxtanuclear compartment associated with components of the microtubule organizing center (MTOC). In addition, co-expression of the hSP-CΔExon4 variant with hSP-C1-197 is shown to also disrupt trafficking of the wild-type isoform. Finally, the oral butyrate analog, sodium 4-phenylbutyrate (4-PBA), which has been shown to promote trafficking of other mutant proteins such as CFTR, corrected mutant proSP-C juxtanuclear protein accumulation in vitro.
Materials and Methods
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 antisera were purchased from Molecular Probes (Eugene, OR). Texas Red-conjugated monoclonal and polyclonal secondary antibodies were obtained from Jackson Immunoresearch Laboratories (West Grove, PA). Anti-CD63 was purchased from Immunotech, (Marseille Cedex, France). 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 (Indianapolis, IN). Sodium 4-phenylbutyrate (4-PBA; Buphenyl®), manufactured by Triple Crown America (Perkasie, PA), was the generous gift of Ronald Rubenstein (Children's Hospital, Philadelphia, PA). Except where noted, other reagents were of electrophoretic grade and were purchased from Bio-Rad (Melville, NY) or Sigma (St Louis, MO).
A monospecific polyclonal rat proSP-C antiserum was produced from a synthetic peptide antigen and has been previously characterized (Beers et al., 1994). Anti-NPROSP-C (Met10 to Glu23) recognizes proSP-C21 and all major intermediates but does not recognize mature SP-C.
SP-C and EGFP/SP-C expression constructs
A full-length human SP-C cDNA insert of 875 bp corresponding to the published sequence of Warr et al. (Warr et al., 1987) was the generous gift of Philip Ballard. This insert was subcloned into the pcDNA3 eukaryotic expression vector polylinker at the EcoRI site as previously described for rat SP-C (Beers et al., 1998).
A family of chimeric fusion proteins consisting of EGFP and wild-type human SP-C (Met1 to Ile197) or SP-C cDNA either lacking Exon 4 or containing point mutations in cysteine residues in the COOH flanking propeptide were generated by polymerase chain reaction (PCR) (Fig. 1).
For EGFP/hSP-C1-197, pcDNA3-hSP-C (6+) served as template for a two primer single PCR reaction. The primers used were: (forward) 5′-dTACAAGTCCGGAATGGATGTGGGCAGCAAAGAGGTCCTG; (reverse) 3′-dTCTAGATGCATGCTCGAGCG. Amplification reactions containing 0.2 μM primers, 1.25 μM dNTP mixture, 1.5 μM MgCl2, 10 ng template, and 2.5 U VentR DNA polymerase (New England Biolabs, Beverly, MA) consisted of 30 cycles under previously published conditions (Johnson et al., 2001). An in-frame fusion protein was made through introduction of a BspEI site at the 5′ end and an XhoI site at the 3′ end for cloning into pEGFP-C1.
For mutant EGFP/hSP-CC189G, a two primer single reaction PCR technique was also used with pcDNA3-hSP-C (6+) serving as template. The primers used were: (forward) 5′-dTCCGGAC-TCAGATCTATGGATGTGGGCAGCAAAGAGGTCCTG; (reverse) 3′-dTCCGGTGGATCCCTAGATGTAGTAGAGCGGCACCTCGCC-ACCCAGGGTGTTCAC. Amplification produced a purified PCR insert containing the desired point mutation at codon 189 and lacking the 3′ untranslated region which was ligated into pEGFP-C1 following digestion with BamHI and BglII.
For construction of EGFP/hSP-CC121/122G, mutagenesis was performed by overlap extension PCR with a two round, four-primer technique as previously published (Russo et al., 1999). The primers are listed in Table 1. Purified intermediate segments (SP-C1-125X and SP-C117-197) generated in separate PCR reactions using pcDNA3-hSP-C (6+) as the template with primers A and B or primers C and D were fused with a second round of PCR using primers complementary to the 5′ -and 3′ -ends (primers A and D, respectively). The resulting mutant insert, hSP-CC121/122G containing the 3′ untranslated region was purified and ligated into pEGFP-C1 after digestion with BglII and BamHI.
For creation of EGFP/hSP-CΔExon4, deletion of exon 4 from pcDNA-hSP-C (6+) was also achieved by overlap extension PCR with a two round, four-primer technique. Exons 1-3 were amplified by single PCR using 2 primers: (forward) dTACAAGTCCGGAATGGATGTGGGCAGCAAAGAGGTCCTG; (reverse) dCTGCTGGTAGTCATACACCACGAGGCC. The other primary reaction amplified exon 5 through the 3′ untranslated Poly A tail using a forward primer containing complete overlap with the 3′ end of the reverse primer used to amplify Exon 3 (bold): (forward) dGGCCTCGTGGTGTATGACTACCAGCAGATGGAATGCTCT CTGCAGGCCAAGCCC; (reverse): dTCTAGATGCATGCTCGAGCG. The two overlapping primary products were fused by SOEing PCR using (forward) dTACAAGTCCGGAATGGATGTGGGCAGCAAAGAGGTCCTG and (reverse) dTCTAGATGCATGCTCGAGCG. Following digestion with BspEI and XhoI, the ΔExon4 insert was subcloned into pEGFPC1.
A triple mutant lacking all three cysteines in the COOH domain (EGFP/SP-CC120/122/189G) was created by a single PCR performed using EGFP/hSP-CC121/122G as template with a primer set identical to that used to generate EGFP/hSP-CC189G.
Hemagglutinin-A-tagged wild-type proSP-C
By using pcDNA-hSP-C1-197 as a template, the hemagglutinin (HA) tag (YPYDVPDYA) was added to the N-terminus of hSP-C1-197 by PCR. The 5′ (forward) primer was an 84mer and contained a KpnI site (bold), a Kozak sequence, and the HA-coding sequence (underlined): TCCGGACTCGGTACCATGGATTACCCATACGATGTTCCAGATTACGCTGCTGATGTGGGCAGCAAAGAGGTCCTGATGGAGAGC. The 3′ (reverse) primer was a 36mer that contained a BamHI site (bold): CTATAGGGATCCGCCCTCTAGATGCATCCTCGACCC. The purified insert was subcloned back into the pcDNA3 vector using digestion with KpnI and BamHI.
All other procedures involving oligonucleotide and cDNA manipulations were performed essentially as described (Ausbel et al., 1995). Automated DNA sequencing in both directions failed to detect nucleotide mutations in wild-type or mutant SP-C constructs.
A549 cell line and transfection
The lung epithelial cell line A549 used in transfection studies was originally obtained through the American Type Culture Collection (Manasas, VA) and has been used in prior studies (Beers et al., 1998; Russo et al., 1999; Kabore et al., 2001). A549 cells grown to 50% confluence on glass coverslips were transiently transfected with EGFP/SP-C constructs (10 μg/dish) or co-transfected with two different constructs (5 μg/construct/dish) by CaPO4 precipitation as previously described (Russo et al., 1999). Where indicated, 4-PBA was added at the time of transfection and continued until harvest.
For double label studies, immunostaining of permeabilized cells fixed by immersion of coverslips in 4% paraformaldehyde was performed by incubation with primary antisera for 1 hour at room temperature followed by either secondary goat anti-mouse IgG monoclonal or secondary goat anti-rabbit IgG polyclonal antisera (Kabore et al., 2000; Russo et al., 1999). Fluorescence images of air-dried and Mowiol mounted slides 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). Image acquisition, processing, and overlay analysis were performed using IMAGE1 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 solubilization buffer (50 mM Tris, 190 mM NaCl, 6 mM EDTA, 2% Triton X-100, 1 mM PMSF, pH 7.4). Following centrifugation at 8000 g for 30 seconds to remove nuclei, proteins were separated by electrophoresis on a 12% polyacrylamide gel and transferred to nitrocellulose (Laemmli, 1970).
Immunoblotting of transferred samples was done using successive incubations with primary polyclonal GFP antisera (1:5000) for 1 hour and goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:10,000) for 1 hour at room temperature. Bands were visualized by enhanced chemiluminescence using a commercially available kit (Amersham, Arlington Heights, IL).
EGFP/hSP-C1-197 is targeted to cytosolic vesicles of A549 cells
We have previously utilized the human alveolar epithelial cell line, A549, to characterize functional domains that mediate trafficking of the rat SP-C proprotein (Beers et al., 1998; Russo et al., 1999; Wang et al., 2002; Kabore et al., 2000). In the current study, expression of the EGFP/hSP-C1-197 fusion protein was readily detected in transiently transfected A549 cells within 24-48 hours after introduction by CaPO4 precipitation. Similar to that observed with rat proSP-C, EGFP/hSP-C1-197 was expressed in cytoplasmic vesicles of A549 cells (Fig. 2). This pattern was similar to that seen with A549 cells transfected with wild-type human proSP-C lacking the EGFP tag identified by immunostaining with anti-proSP-C antiserum (Fig. 2B) or with an HA-tagged wild-type proSP-C (Fig. 2C). These results indicate that the presence of EGFP does not interfere with trafficking patterns for human SP-C1-197, which contains motifs for ER translocation and direction of heterologous fusion proteins to vesicular compartments.
The vesicular compartment targeted by wild-type proSP-C was identified using double-label fluorescence microscopy. Transfected EGFP/hSP-C1-197 expression was spatially restricted to CD-63-positive compartments, a marker antigen associated with lamellar bodies and to multivesicular bodies of type 2 cells (Fig. 3). Furthermore, this compartment was Calnexin-negative, EEA1-negative, and failed to stain for ubiquitin (data not shown). Collectively, these data indicate that, in this model, proSP-C is directed to a lysosomal like organelle that shares antigen markers with lamellar bodies of type 2 cells.
Altered distribution of EGFP/hSP-CΔExon4 expressed in lung epithelial cells
In contrast to wild-type human SP-C, EGFP fusion proteins containing the SP-CΔExon4 mutant were expressed in a time frame similar to that of wild-type proSP-C but were retained in proximal, juxtanuclear compartments (Fig. 4C).
Western blotting of A549 cell lysates using an anti-EGFP antisera demonstrated that the mutant proSP-C protein forms expressed by EGFP/hSP-CΔExon4 represented steady-state accumulation of an unprocessed, non-degraded fusion protein (Fig. 4E). The anti-EGFP antisera detected a major product of 27 kDa for cells transfected with EGFP alone. When wild-type EGFP/hSP-C1-197 was introduced, lysates contained two sets of anti-EGFP-positive bands. A doublet corresponding to the primary translation product of the fusion protein (Mr 48,000) and an accompanying higher molecular weight form (Mr 50,000) consistent with the known palmitoylated isoform is seen. A second set of two smaller intermediates was also identified; however, a band at 27 kDa (consistent with liberation of free EGFP) was not found. This profile is consistent with the notion that human SP-C-EGFP was palmitoylated and partially processed in this cell model by proteolysis of the C-terminus but without proteolytic modification of the NH2 propeptide. In contrast to wild-type proSP-C, expression of the hSP-CΔExon4 mutant resulted in formation of a single product representing a truncated primary translation product expressed without further modification. Together, the immunocytochemistry and western blotting data demonstrate that hSP-CΔExon4 accumulates in perinuclear inclusions without processing or degradation.
Human SP-CΔExon4 is directed to aggresomes
Since the recognition of abnormal and misfolded proteins is mediated by ubiquitin that promotes delivery of substrates to the 28S proteasome, we performed immunocytochemical staining of transfected A549 cells to further characterize the compartment. In contrast to wild-type SP-C, aggregates formed following transfection with EGFP/hSP-CΔExon4 were ubiquitinated (Fig. 5A-C). Furthermore, this compartment was CD63 and calnexin-negative (not shown) but in close approximation to tubulin-containing filaments, which suggested localization near the microtubule-organizing center (MTOC) (Fig. 5D-F). Taken together, the data suggest that the expression of the hSP-CΔExon4 mutant results in accumulation of mutant forms in a compartment compatible with aggresomes.
Alteration of EGFP/hSP-C expression is promoted by disruption of disulfide-mediated folding
To examine the mechanism underlying the accumulation of hSP-CΔExon4 in aggresome structures, we performed site-directed mutagenesis to effect conservative substitution of Cys120 and Cys121 encoded within the fourth exon of human SP-C. When coupled to EGFP, subcellular expression patterns for the fusion protein were similar to those seen with removal of the entire fourth exon (Fig. 6C). As for EGFP/hSP-CΔExon4, aggregate formation was accompanied by colocalization with ubiquitin and α-tubulin (data not shown). Western blotting demonstrated that all cysteine mutant SP-C isoforms were expressed as single bands, which indicates that, as for SP-CΔExon4, COOH propeptide mutations result in stable formation of unprocessed forms.
Human SP-CΔExon4 acts a dominant negative
Previously published in vitro evidence indicated that trafficking of SP-C occurred via formation of oligomeric complexes (Wang et al., 2002). Because cases of interstitial lung disease associated with heterozygous expression of the hSP-CΔExon4 mutation also manifested a lack of detectable mature SP-C (Nogee et al., 2001), we hypothesized that the mutant protein could act to produce a dominant-negative effect by trapping wild-type protein in proximal compartments. To study this interaction, cotransfection studies were performed using EGFP/hSP-CΔExon4 and an HA-tagged wild-type human SP-C expressed in the pcDNA3 vector (Fig. 7). Cotransfection of both tagged wild-type forms (EGFP/SP-C1-197 and HA/SP-C1-197) resulted in co-expression of each fusion protein in the same cytosolic vesicles (Fig. 7A-C). In contrast, the co-expression of EGFP/hSP-CΔExon4 and HA/SP-C1-197 was now associated with restriction of both forms to perinuclear compartments (Fig. 7D-F). These results are consistent with the notion that heteromeric sorting of hSP-CΔExon4 mutants and wild-type SP-C can produce a functional dominant negative to inhibit the trafficking of wild-type protein.
Treatment with chemical chaperones alters intracellular distribution of EGFP/hSP-CΔExon4
4-Phenylbutyric acid has been shown to facilitate the trafficking of mutantΔ 508CFTR both in vitro and in vivo (Rubenstein et al., 1997; Rubenstein and Zeitlin, 2000). To investigate whether 4-PBA could alter aggregation of mutant SP-C, A549 cells transfected with EGFP/hSP-CΔExon4 were treated with increasing concentrations of 4-PBA for 48 hours. In contrast to cells treated with saline alone, A549 cells expressing mutant EGFP/hSP-CΔExon4 showed redirection of mutant fusion protein to cytosolic vesicles (Fig. 8). The in vitro pharmacological correction of juxtanuclear accumulation of mutant SP-C in A549 epithelial cells occurred at a range of doses from 1 to 5 mM. Doses greater than 5 mM were associated with cellular toxicity.
The correct folding and deployment of biologically active proteins throughout the cell is an important step in gene expression (Kopito and Ron, 2000). Encoded proteins destined for delivery to the cell surface and for secretion are translocated into the endoplasmic reticulum (ER) during biosynthesis. Anterograde trafficking is tightly linked with the acquisition of biological function, and misfolded, potentially defective proteins are selectively tagged for degradation either in the ER or in proteasomes by a multi-step pathway (Kopito, 1997; Kopito and Ron, 2000; Gelman et al., 2002). While protein folding and quality control systems have been attributed to the molecular pathogenesis associated with several rare respiratory diseases, including cystic fibrosis and α1-antitrypsin deficiency (Carrell and Lomas, 2002; Carrell and Lomas, 1997), the recent association of mutant forms of surfactant protein C with familial forms of interstitial lung disease has raised the possibility that a similar pathophysiological mechanism is resulting from expression of these mutant forms of SP-C (Nogee, 2002). Previously we had shown that mutations in the COOH flanking propeptide encoded by the rat SP-C cDNA that promote misfolding result in the formation of protein aggregates and inhibition of wild-type protein trafficking in vitro (Kabore et al., 2001). The present report extends these observations by examining trafficking of a human SP-C mutation, hSP-CΔExon4, recently identified in a mother and daughter each heterozygous for the mutation and both with a biopsy diagnosis of interstitial lung disease (Nogee et al., 2001). We show that the expression of the hSP-CΔExon4 mutation results in juxtanculear accumulation of unprocessed mutant protein forms that produce a dominant-negative effect on the trafficking of co-expressed wild-type proSP-C. Importantly, as has been observed with CFTR, treatment with sodium 4-phenylbutyrate (4-PBA), which is implicated in the regulation of endogenous chaperones, is shown to attenuate the altered expression pattern.
The expression of mutant hSP-CΔExon4 is characterized by disruption of trafficking in the earliest compartments. Previously, ten Brinke et al., has demonstrated that a mutation in the N-terminal portion of human SP-C (hSP-CK34R35) resulted in blockade of egress of proSP-C from the ER and inhibited its palmitoylation, a Golgi-mediated event (ten Brinke et al., 2001). On SDS/PAGE analysis, we found that the EGFP fusion protein containing wild-type SP-C produces a high molecular weight doublet (Fig. 6E) that has been show to correlate with generation of a primary translation product and a higher molecular weight dipalmitoylated form (Vorbroker et al., 1992). A single band was observed when mutant SP-C forms containing deletions or mutations of COOH cysteine residues (hSP-CΔExon4, hSP-CC120/121G and hSP-CC189G) were expressed. The lack of a doublet is consistent with inhibition of delivery of these isoforms to the Golgi. However, unlike the persistent ER retention noted by ten Brinke et al., for the non-palmitoylated hSP-CK34R35 mutant, the fluorescence microscopy for EGFP/hSP-CΔExon4 protein instead demonstrates accumulation in a cytosolic compartment. Coupled with immunocytochemistry showing the presence of ubiquitination (a cytosolic event) and colocalization with α-tubulin, trafficking of the hSP-CΔExon4 mutant is consistent with ER translocation, misfolding, retrotranslocation, ubiquitination and clustering near the MTOC, prerequisites for the formation of a novel cellular structure termed the aggresome (Johnston et al., 1998; Kopito and Ron, 2000; Kopito, 2000).
A further consequence of mutant protein expression is an inhibition of cellular trafficking of wild-type SP-C. Recently, we have shown using crosslinking of transfected A549 cell lysates with bis-malemeidohexane that multimeric forms of wild-type rat SP-C fusion proteins (EGFP/SP-C1-194) are generated indicating that SP-C biosynthesis involves oligomeric association of proSP-C monomers (Wang et al., 2002). Thus, the results presented here are consistent with a dominant-negative effect induced by heteromeric sorting of wild-type and mutant protein to the same compartment. A schematic of the molecular modeling for the heteromeric interaction of wild-type and mutant SP-C is shown in Fig. 9. In this model, the early aggregation of heteromers of wild-type and mutant SP-C inhibits both the trafficking and processing of the wild-type form. A similar series of events has been observed for other mutant integral membrane proteins suggesting that this represents a common metabolic pathway for the handling of mutant proteins (Tobler et al., 1999). The heterozygous expression of hSP-CΔExon4 in humans with interstitial lung disease is associated with a lack of detectable mature SP-C (Nogee et al., 2001) that could be accounted for by this mechanism.
It is unlikely that the pathology associated with the hSP-CΔExon4 is due to a lack of SP-C. SP-C null mice have been produced (Glasser et al., 2001). The resulting phenotype produces viable mice at birth that grow to adulthood without apparent early structural pulmonary abnormalities. This is in contrast to a recent study from Weaver's group showing that lung-specific overexpression of the mature SP-C protein alone (lacking the propeptide flanking domains) in transgenic mice is neonatal lethal (Conkright et al., 2002). Thus, the consequence of expression of hSP-CΔExon4 cannot be explained simply by the lack of SP-C (dominant-negative effect) but suggests that this mutation is associated with a toxic gain of function.
There are several potential mechanisms whereby proSP-C folding mutants could affect cellular function. Formation of aggregates of misfolded proteins within specialized cells has been linked to a number of pathological states in both animal models and in humans (Kopito and Ron, 2000; Kopito, 2000). Perinuclear inclusions composed of aggregated, ubiquitinated protein and intermediate filament proteins are present in amyloidosis and in several neurodegenerative diseases including Alzheimer's disease as well as the peripheral myelin protein 22 (PMP-22)-associated polyneuropathies (Tobler et al., 1999). Similar inclusions have been associated with mutant PMP-22 expression in Schwann cells from the Trembler-J mouse (Notterpek et al., 1999). The exact role of aggresome formation in the pathogenesis of lung disease has not yet been investigated in paradigms of naturally occurring mutant proSP-C expression or under conditions of abnormal proSP-C trafficking. Recently, we used chimeric proteins containing EGFP and rat proSP-C to demonstrate that intraluminal disulfide-mediated folding of the C-terminus of rat proSP-C is required for intracellular trafficking of the intact propeptide (Kabore et al., 2001). Deletional mutants lacking cysteine residues at positions 122 and/or 186 of rat proSP-C were retained in a juxtanuclear compartment that stained for ubiquitin, α-tubulin and vimentin. Treatment of cells transfected with these mutants with the proteasome inihibitor lactacysteine enhanced formation of the juxtanuclear inclusions. Misfolding of unprocessed mutant proteins led to the formation of stable pool of unprocessed protein.
The present study predominantly concerns the formation of aggresomes in lung epithelial cells expressing mutant proSP-C proteins. It does not address the sequellae of the long-term effect of the aggresome on type-2 cell function. The experimental conditions that fostered the formation of aggresomes in this study were overexpression that probably imposes a significant stress on the degradative capacity of the cell. Our experiments raise the possibility of a homeostatic mechanism in the lung epithelium for the handling of mutant proteins. We speculate that the transport of ubiquinated proSP-C mutants to aggresomes functions to clear the cytoplasm of potentially toxic aggregates or may serve as a staging ground for eventual removal by incorporation into autophagocytic structures. However, western blots of lysates (Figs 4, 6) confirm the accumulation of a single form of mutant EGFP/proSP-C without evidence of smaller degradation products, which suggests that the half-life of these structures could be long. Thus, although aggresome formation could provide a cytoprotective role, the toxicity of the long-term accumulation of non-degraded aggregates is unclear. Therefore, given the recent in vitro results presented in this report, the expression of mutant forms of surfactant protein could contribute to lung pathogenesis through a toxic gain-of-function from either alterations in type-2 cell viability, function and/or homeostasis induced by protein aggregation. This concept of cellular toxicity from expression of mutant SP-C has been suggested by the recent work of Thomas et al., who have described a kindred of patients with interstitial lung disease extending over five generations, all expressing another SP-C mutation who developed interstitial lung diseases in both adults and children (Thomas et al., 2002). This mutation involves a glutamine for leucine substitution at codon 188 (hSP-CL188Q), which is located directly adjacent to Cys189. When transiently transfected in the MLE-12 lung epithelial cell line, hSP-CL188Q induced cell necrosis as assessed by release of LDH into the media, which indicates that SP-C mutants are capable of inducing cellular toxicity. It is interesting to note that in the present study, when Cys189 was mutated (EGFP/hSP-CC189G), the mutant fusion protein was directed to aggresomes (Fig. 6), which suggests a possible mechanism for these findings.
Chemical chaperones provide a potential therapy for conformational lung diseases by altering the intracellular fate of mutant proteins though intervention in their biosynthetic processing, degradation, or association with chaperones. Several agents have shown promise in interrupting `downstream' pathology in a number of systems (Sato et al., 1996; Rubenstein and Zeitlin, 2000; Fischer et al., 2001). 4-phenylbutyric acid is a short chain fatty acid scavenger that functions as an ammonia scavenger for the treatment of urea cycle enzyme deficiencies. It has also been used to induce production of fetal hemoglobin and is being evaluated for the treatment of sickle cell disease. In cystic fibrosis, 4-PBA has been shown to promote trafficking of the mutant protein and functional correction of transport defects. Part of the mechanism for this effect is related to the regulation of endogenous chaperones by this compound. When A549 cells expressing hSP-CΔExon4 were treated with 4-PBA, perinuclear aggregation was inhibited and increases in anterograde transport to cytosolic compartments were seen (Fig. 8). Although Hsc-70 expression was dramatically altered by 4-PBA in cells expressing the Δ508 mutant of CFTR, we were unable to find consistent changes in expression of hsp70 or hsc70 in A549 cells transfected with hSP-CΔExon4 (M.F.B., unpublished). The modulation of a different chaperone, changes in microtubule function, or direct effects of 4-PBA on SP-C folding remain mechanistic possibilities that will require additional studies.
In summary, we have shown that expression of hSP-CΔExon4 is associated with marked alterations in protein trafficking, which can also affect the targeting of wild-type SP-C in a dominant-negative fashion. In addition to hSP-CΔExon4 and hSP-CL188Q, Nogee has reported the presence of ten different SP-C mutations in 34 term infants and children with chronic lung disease (Nogee et al., 2002). In addition, two other mutations in the SP-C gene have been identified in patients with interstitial lung disease (M.F.B., A. Hamvas and P. A. Stevens, unpublished). Similar to hSP-CΔExon4 and hSP-CL188Q, most of these are clustered in the COOH flanking propeptide. Understanding the trafficking patterns of these mutants as well as the long term consequences of their expression through the use of stably transfected cell lines as well as transgenic models will be critical to the overall characterization of the pathogenesis of interstitial lung disease associated with mutant isoforms of SP-C.
We thank Viviane Martin for assistance with construction of the HA tagged construct and Susan Guttentag for helpful discussions. This work was supported by NIH HL-19737 and P50-HL56401 (both to M.F.B.).
- Accepted November 9, 2002.
- © The Company of Biologists Limited 2003