Ablation of a critical surfactant protein B intramolecular disulfide bond in transgenic mice.

The 79-amino acid, mature SP-B peptide contains three intramolecular disulfide bonds shared by all saposin-like proteins. This study tested the hypothesis that the disulfide bond formed between cysteine residues 35 and 46 (residues 235 and 246 of the SP-B proprotein) is essential for proper function of SP-B. To test the role of this bridge in SP-B function in vivo, a construct was generated in which cysteine residues 235 and 246 of the human SP-B proprotein were mutated to serine and cloned under the control of the 3.7-kilobase hSP-C promoter (hSP-B(C235S/C246S)). In two transgenic mouse lines, expression of the mutant peptide in the wild-type murine SP-B background was invariably lethal in the neonatal period. In four additional lines, survival was inversely related to the level of transgene expression. To test the ability of the mutant peptide to functionally replace the wild-type protein, transgenic mice were crossed into the SP-B null background. No animals that expressed hSP-B(C235S/C246S) in the murine SP-B-/- background survived the neonatal period. hSP-B(C235S/C246S) proprotein accumulated in the endoplasmic reticulum and was not processed to the mature, biologically active peptide. The results of these studies demonstrate that the intramolecular bridge between residues 235 and 246 is critical for intracellular trafficking of SP-B and suggest that overexpression of mutant SP-B in the wild-type background may be lethal.

Pulmonary surfactant is a complex mixture of phospholipids and surfactant proteins (SP-) 1 A, B, and C that is synthesized and secreted by the Type II epithelial cell into the alveolar airspace. Surfactant forms a film along the alveolar epithelium that reduces surface tension to very low levels during lung deflation. Deficiency of pulmonary surfactant results in alveolar collapse, leading to respiratory distress syndrome, a leading cause of morbidity and mortality among neonates worldwide. Substantial benefit is derived from treating these infants with surfactant replacement preparations, particularly those con-taining SP-B or SP-C. Human infants with mutations in the SP-B gene that result in complete absence of SP-B protein develop severe respiratory distress syndrome at birth and, despite intensive respiratory therapy, ultimately succumb to the disease (1). Similarly, SP-B null mice die of acute respiratory distress syndrome within minutes of birth, have reduced lung volumes, decreased concentrations of mature SP-C peptide, and highly disorganized lamellar bodies, the intracellular storage form of surfactant (2,3). Collectively, these results indicate that SP-B is critical for lung function.
SP-B is translated as a 381-amino acid preproprotein that is co-translationally cleaved to generate the SP-B proprotein. The SP-B proprotein is sorted to the multivesicular body where the 102-amino acid carboxyl and 177-residue amino-terminal peptides are sequentially cleaved to generate the 79-amino acid mature SP-B peptide (4). The hydrophobic SP-B peptide is stored with surfactant phospholipids in specialized secretory granules (lamellar bodies) in which phospholipids are arranged as concentric membrane lamellae. In the absence of SP-B, lamellar bodies contain numerous vesicles, but few or no lamellae, consistent with a key role for SP-B in the packaging of lamellar body phospholipids. The contents of the lamellar body are secreted into the alveolar space where SP-B and SP-C facilitate the rapid formation of a surfactant film.
The intra-alveolar form of mature SP-B exists in an oxidized state in which six cysteine residues participate in three intramolecular sulfhydryl bridges, while the seventh cysteine forms an intermolecular bridge that results in SP-B homodimers (5,6). The precise arrangement of the three intramolecular disulfide bridges places SP-B in the saposin-like protein (SAPLIP) family, which also includes saposins A-D, amoebapore, and NK-lysin. SAPLIP proteins serve a variety of functions including glycosphingolipid metabolism, host defense, and in the case of SP-B, surfactant function in the alveoli. The mature SP-B peptide differs from the other family members in that it is more hydrophobic and forms sulfhydryl-mediated dimers.
NK-lysin is the only SAPLIP family member for which the three-dimensional structure has been solved, although sequence homology among family members suggests that all members share the saposin fold of NK-lysin (7). Studies of SP-B secondary structure and comparisons with the NMR structure of NK-lysin predict that SP-B contains five amphipathic ␣-helices folded into a globular protein domain (7). A structural model of the SP-B mature peptide has been proposed in which a hairpin turn in the mature peptide results in antiparallel alignment of the helices, bringing cysteine residues 235 and 246 into close apposition (8). It is likely that the cysteine bridge formed between residues 235 and 246 stabilizes this turn and contributes to the structural integrity of mature SP-B. The amino acid sequence located between cysteines 235 and 246 in the SP-B mature peptide is highly hydrophobic, suggesting that this region of the peptide interacts with surfactant lipids and might be critical for SP-B function. The current study was undertaken to test the hypothesis that the intramolecular cysteine bridge formed between residues 235 and 246 of SP-B is critical for SP-B structure and function in vivo.

MATERIALS AND METHODS
Generation of DNA Constructs-To generate the hSP-B C235S/C246S construct, site-directed mutagenesis was employed using a sequential PCR protocol (Current Protocols in Molecular Biology 8.5) to substitute serine for the cysteine residues at positions 235 and 246 in the SP-B preproprotein. Primers were chosen which would amplify a 1.6-kb fragment of human SP-B cDNA that includes the endogenous Kozak sequence and sequence encoding the 381-amino acid SP-B preproprotein (5Ј-GCGGAATTCGAGGTGCCATGGCTGAGTCAC and 5Ј-CAGGTTC-CGCGGAAGGTCGGGGCTGTGGATACACT). Internal primers used in the first round of PCR resulted in the substitution of serine for cysteine 246 (upstream primer 5Ј-CATCTCCCAGTGCCTGGCTGAG and downstream primer 5Ј-CACTGGGAGATGCCGCCCGCCAC). The second round of mutagenesis resulted in the substitution of serine for cysteine 235 (upstream primer 5Ј-GGTGTCCCGCGTGGTACCTCT and downstream primer 5Ј-CGCGGGACACCTGGGCCACTGCCACAGCT-A). This construct, hSP-B C235S/C246S , was cloned into the EcoRI and SstII sites of pEGFP-N1 (CLONTECH, Palo Alto, CA) which resulted in a fusion protein consisting of SP-B with GFP protein at the carboxyl terminus. To generate the transgene construct, primers (5Ј-GCGGAA-TTCGAGGTGCCATGGCTGAGTCAC and 5Ј-CGGAATTCTCATCAAA-GGTCGGGGCTGTGGATACA) were used to amplify hSP-B C235S/C246s using the SP-B/GFP fusion construct as template. This sequence was cloned into the EcoRI site of plasmid pCC10kbpA which contained the mouse CCSP promoter, rabbit ␤-globin intron 2 and bovine growth hormone polyadenylation signal (kind gift of Francesco DeMayo, Baylor University). To generate transgenic mice in which hSP-B C235S/C246S expression was restricted to the distal respiratory epithelium, the 3.7-kb human SP-C promoter fragment (9) was substituted for the CCSP promoter. The final vector is referred to as the BGI-hSP-C expression vector.
Intracellular Trafficking of hSP-B C235S/C246S in PC12 Cells-PC12 cells (American Type Culture Collection, Rockville, MD) were cultured as described (10). Twenty-four hours prior to transfection, cells were plated onto coverslips coated with mouse type IV collagen (Becton Dickinson, Cockeysville, MD). Coverslips with 70% confluent cells were transiently transfected with 2 g of plasmid DNA in 20 l of Lipo-fectAMINE (Life Technologies, Inc., Gaithersburg, MD) according to the manufacturer's recommendations. Twenty-four hours post-transfection, the cells were fixed, permeabilized, and incubated with primary antibody for 1 h followed by secondary antibody for 30 min at room temperature. Cells were viewed with a Leica DMIRB/E inverted microscope and scanned with a Leica 3-laser scanning confocal microscope system. Antibody directed to Chromogranin A was purchased from Incstar (Stillwater, MN) and the Texas Red-conjugated goat anti-rabbit secondary antibody was purchased from Vector Labs (Burlingame, CA). Metabolic labeling was performed essentially as described (11) using 70% confluent PC12 cells grown in 10-cm 2 tissue culture dishes that were transfected with 20 g of DNA in 246 l of LipofectAMINE. Immunoprecipitations were performed using anti-mature SP-B antiserum (number 28031) exactly as described (11).
Generation and Characterization of Transgenic Mice Expressing hSP-B C235S/C246S -The transgene was excised from the BGI-hSP-C expression vector by NotI/NdeI digestion, isolated by gel electrophoresis, and purified using Qiaex resin (Qiagen, Germany). The transgene was dialyzed for 48 h against 5 mM Tris (pH 7.5, .1 mM EDTA) and microinjected into fertilized FVB/N oocytes by the Children's Hospital Transgenic Core Facility. Founders were identified by a transgene specific PCR with primers that amplified a 336-base pair fragment of the hSP-B cDNA (upstream primer 5Ј-AGCAGCAATTCCCCATTCCT and downstream primer 5Ј-ATGGCCTGCTCGCTGCTGTTCC). Primers that amplified a 270-base pair fragment of the endogenous murine SP-B allele were included in the transgene reaction as a positive control (12). PCR conditions were 30 cycles, 58°C annealing temperature, 0.5 M transgene primers, 0.25 M mSP-B primers, 0.25 M dNTPs, 1 unit of display TAQ, and 1ϫ display TAQ buffer (PGC Scientific, Gaithersburg, MD). PCR results were confirmed by Southern analysis using a 32 P radiolabeled probe that hybridized to the 1.2-kb hSP-B EcoRI insert.
To identify transgenic lines that expressed hSP-B C235S/C246S protein, SDS-PAGE, and Western blotting was performed to detect the human SP-B proprotein (M r ϳ 42,000) which was detected in transgenic lung homogenate but not in wild-type lung homogenate. Lungs from 6-week-old mice from transgenic lines H, Ia, Ib, J, L, and M and neonatal lungs from offspring of transgenic founder mice N and K were isolated and homogenized in phosphate-buffered saline with 1% (v/v) protease inhibitor mixture (Sigma). Protein concentrations were determined by bicinchoninic acid protein assay (13) and equal amounts of total lung protein were analyzed by SDS-PAGE and Western blotting. Western blotting was performed with antibodies directed against the carboxyl terminus of SP-B proprotein (number 96189), mature SP-B (number 28031), and the amino terminus of SP-C proprotein (number 68514) (11,14). Mature SP-C peptide was detected with antibody generated against recombinant mature SP-C (provided by Byk Gulden, Konstanz, Germany).
The composition of SP-B oligomers was resolved by two-dimensional gel electrophoresis. Samples were run in the first dimension in 10% glycine SDS-PAGE gels under nonreducing electrophoretic conditions. Individual lanes were excised from the gel and incubated at room temperature for 2 h in 1ϫ glycine gel running buffer with 20 mM dithiothreitol (Sigma) in order to reduce sulfhydryl bridges. The excised lane was then placed in a large well of a second 10% glycine gel and subjected to SDS-PAGE in the second dimension under reducing electrophoretic conditions. Western blotting was performed using antibodies directed against mature SP-B (number 28031).
Lung Morphology in Transgenic Animals-Immunostaining for surfactant proteins was performed exactly as described with antisera directed against the carboxyl terminus of SP-B proprotein, mature SP-B, and the amino terminus of SP-C proprotein (15). Type II cell ultrastructure was examined by electron microscopy as described (16).
Surfactant Protein mRNA Expression-S1 nuclease mapping was performed similarly to that described (17). Briefly, the left lobe of four to six adult mouse lungs (transgenic lines H, Ia, Ib, J, L, and M) were homogenized in 4 M guanidine isothiocyanate with 0.1 M 2-mercaptoethanol and total lung RNA isolated. S1 probes specific for murine cytoplasmic ␤-actin, murine SP-C, murine SP-B, and human SP-B were radiolabeled with [␥-32 P]ATP (17,18) and hybridized with 3 g of RNA at 55°C overnight, followed by S1 nuclease (Life Technologies, Inc.) digestion at room temperature for 1 h. Protected fragments were separated in a 6% polyacrylamide, 8 M urea gel that was dried and subjected to PhosphorImaging (Molecular Dynamics, Sunnyvale, CA). All PhosphorImaging data were analyzed in ImageQuant (Molecular Dynamics) and hSP-B levels were normalized to cytoplasmic ␤-actin.
Surfactant Protein Synthesis in Lung Explant Culture-To determine if expression of the hSP-B C235S/C246S transgene affected expression of endogenous surfactant proteins, synthesis of SP-A and SP-C was assessed in lung explant cultures from day 18.5 mouse fetuses. Offspring from timed pregnancies were harvested by Caesarian section on the same day as natural delivery would occur (fetal day 18.5). Lungs were quickly isolated and minced into 1-mm 3 pieces with a McIlwain Tissue Chopper (Brinkman, Westbury, NY). Lung pieces were incubated (37°C, room air, rotating at 8 rpm) in cysteine and methioninedeficient Dulbecco's modified Eagle's medium (Life Technologies, Inc.) for 40 min and then supplemented with 0.5 mCi of Promix 35 S-labeled cysteine and methionine (Amersham Pharmacia Biotech). Four hours later tissues were harvested by gentle centrifugation (750 ϫ g), boiled for 4 min, and subjected to extensive sonication until all tissue had been dispersed. Tricarboxylic acid precipitable counts were estimated for tissue and media and equal disintegrations/min were immunoprecipitated sequentially with normal rabbit serum followed by antibodies directed against SP-C proprotein, mature SP-B and SP-A (SP-A antibody was a gift from Frank McCormack, University of Cincinnati). Immunoprecipitates were subjected to SDS-PAGE and dried gels were analyzed by PhosphorImaging.

Intracellular
Trafficking of hSP-B C235S/C246S PC12 Cells-Intramolecular sulfhydryl bridges stabilize protein structure and in some instances provide information for intracellular sorting (19 -21). To determine if the sulfhydryl bridge formed between SP-B residues 235 and 246 is essential for SP-B folding and/or sorting to the regulated secretory pathway, PC12 cells were transfected with the hSP-B C235S/C246S /GFP fusion construct in which both cysteine residues were mutated to serine and compared with cells transfected with wild type hSP-B/GFP. Transfected cells were radiolabeled with [ 35 S]cysteine and [ 35 S]methionine followed by immunoprecipitation of cell lystates and media with SP-B antiserum and treatment with endoglycosidase H (Endo H). Both wild-type and mutant SP-B/GFP fusion proteins were detected in the media as Endo H-resistant forms, confirming that loss of the Cys C235S/C246S sulfhydryl bridge did not prevent exit from the endoplasmic reticulum or secretion (Fig. 1A). To determine if hSP-B C235S/C246S /GFP was sorted to the regulated secretory pathway, SP-B/GFP fusion proteins were localized by confocal microscopy 24 h after transfection of PC12 cells. GFP fluorescence partially colocalized with chromogranin A in cells transfected with wild-type (not shown) or mutant (Fig. 1B) SP-B/GFP constructs, consistent with sorting of hSP-B C235S/C246S /GFP to dense core secretory granules of the regulated secretory pathway. For both wildtype and mutant constructs, there were populations of SP-B/ GFP and chromogranin A proteins which did not colocalize. In addition, the fluorescence intensity in dense core granules was consistently higher in cells transfected with the wild-type SP-B construct than with the mutant construct, while there appeared to be increased fluorescence in the Golgi region of cells transfected with hSP-B C235S/C246S /GFP. While the kinetics of SP-B/GFP trafficking may have been altered in cells transfected with the mutant peptide, detection of hSP-B C235S/C246S / GFP in dense core secretory granules and media strongly suggested that the Cys 235/246 sulfhydryl bridge was not critical for sorting and secretion of SP-B/GFP and that ablation of this cysteine bridge would not alter the intracellular transport of SP-B in vivo.
Generation and Characterization of hSP-B C235S/C246S Transgenic Mice-To identify the function of the Cys 235/246 sulfhydryl bridge in vivo, the hSP-B C235S/C246S construct was expressed in transgenic mice. hSP-B C235S/C246S expression was directed to the distal respiratory epithelium using the 3.7-kb human SP-C promoter to recapitulate the endogenous expression pattern for SP-B ( Fig. 2A). Seven of 20 (35%) offspring from fertilized oocyte injections were transgene positive, as identified by both PCR and Southern blot analyses of tail DNA (not shown). These seven animals are referred to as transgenic founders H through N and were bred with wild-type FVB/N mice to establish eight separate transgenic lines. Transgenic line I carried two insertion sites that segregated independently during four consecutive crosses with wild-type littermates to produce transgenic lines Ia and Ib. Western analyses of total lung homogenates were performed on offspring from transgenic animals using antibodies that detected antigenic epitopes in the mature peptide (M r ϳ 16,000) or NH 2 -terminal propeptide of the SP-B proprotein (M r ϳ 42,000). SP-B proprotein was readily detected in transgene positive animals but was not detected in wild-type mice, consistent with expression of human SP-B (Fig. 2B). Transgenic RNA and protein was detected in seven of eight transgenic lines (lines H, Ia, Ib, K, L, M, and N).
To from these crosses should have expressed human SP-B in the absence of murine SP-B, no animals of this genotype (hSP-B C235S/C246S , mSP-BϪ/Ϫ) were identified in any of the five independent lines bred into the mSP-B null background (Table  I). This outcome suggested that the hSP-B C235S/C246S mutant protein could not functionally replace endogenous SP-B.
Intracellular Processing of hSP-B C235S/C246S -To determine if the lethality observed in hSP-B C235S/C246S , mSP-BϪ/Ϫ mice was related to an inability to process the hSP-B C235S/C246S proprotein to mature SP-B peptide in vivo, SP-B processing was analyzed in tissue homogenates prepared from fetal lung. In hSP-B C235S/C246S , mSP-BϪ/Ϫ mice, SP-B proprotein was readily detected while mature SP-B was not detected (Fig. 3,  lanes 3 and 4). One possible explanation for the failure to process the hSP-B C235S/C246S protein is that ablation of the Cys 235/246 sulfhydryl bridge resulted in retention of the proprotein in the endoplasmic reticulum. To test whether hSP-B C235C/ C246S proprotein was trapped in the endoplasmic reticulum, SP-B trafficking was characterized in explant tissue from fetal lungs from wild-type and transgenic littermates (Fig. 4). Newly synthesized SP-B proprotein was detected in wild-type lung explants, but was rapidly processed, resulting in low levels of SP-B, M r ϳ 42,000 (Fig. 4, lanes 1-4); in contrast, newly synthesized proprotein in hSP-B C235S/C246S lung explants was not processed and accumulated (Fig. 4, lanes 7-10). SP-B proprotein in hSP-B C235S/C246S mice was Endo H sensitive, whereas virtually all SP-B proprotein in wild-type mice was Endo H resistant. These findings are consistent with retention of hSP-B C235S/C246S protein in the endoplasmic reticulum and suggested that, unlike the hSP-B C235S/C246S /GFP fusion protein, hSP-B C235S/C246S failed to exit the endoplasmic reticulum.
Transgene-dependent Mortality in Mice Expressing hSP-B C235S/C246S -While characterizing the hSP-B C235S/C246S mice, we observed that transgenic lines with higher expression levels of mutant hSP-B protein had lower numbers of surviving transgenic offspring. To determine if transgene RNA levels were inversely correlated with survival, total lung RNA was prepared for S1 nuclease analysis of hSP-B C235S/C246S mRNA expression (Fig. 5).  Five separate hSP-B C235S/C246S transgenic lines were bred into the mSP-BϪ/Ϫ background to determine if hSP-B C235S/C246S could reverse the neonatal lethal SP-BϪ/Ϫ phenotype. One of every seven offspring was predicted to be hSP-B C235S/C246S , mSP-BϪ/Ϫ (33 of 231 mice). No surviving hSP-B C235S/C246S , mSP-BϪ/Ϫ offspring were identified, which indicated that hSP-B C235S/C246S was unable to functionally replace endogenous mSP-B.  2B and data not shown). Two transgenic offspring from the N line observed at birth suffered lethal neonatal respiratory distress within the first 30 min, while nine wild-type littermates survived the neonatal period, strongly suggesting that expression of the transgene in this line was incompatible with life. The N and K founders were likely mosaic animals whose lungs were comprised largely of non-transgenic cells. This conclusion was supported by Western and immunohistochemical analyses on lungs from these founders, which demonstrated low levels of hSP-B C235S/C246S protein and very few positively staining cells in the distal respiratory epithelium, respectively (not shown). In contrast, fetal lungs from line N transgenic offspring stained intensely throughout the distal respiratory epithelium for human SP-B (not shown). The inverse correlation between transgene expression and survival suggested that hSP-B C235S/C246S proprotein might interfere with the normal processing of wild-type SP-B proprotein. To determine if expression of the hSP-B C235S/C246S transgene altered levels of endogenous SP-B proprotein or mature peptide, tissue homogenates from fetal lungs were analyzed by immunoblotting. Mice that expressed both hSP-B C235S/C246S and wild-type protein (hSP-B C235S/C246S , mSP-Bϩ/ϩ or hSP-B C235S/C246S , mSP-Bϩ/Ϫ) had much lower levels of mature mSP-B peptide than their non-transgenic littermates (Fig. 6). These results, coupled with the accumulation of Endo H-sensitive hSP-B C235S/C246S in the endoplasmic reticulum (Fig. 4), suggested that the mutant protein might impede trafficking of wild-type proprotein to the multivesicular body where proteolytic processing to the mature peptide occurs. Consistent with this hypothesis, higher molecular weight species of SP-B proprotein were consistently detected in transgenic lungs but were never detected in wild-type lungs (Fig. 7). Two-dimensional gel electrophoresis, in which the first dimension was run under nonreducing conditions (Fig. 7, lanes 2-4) and the second under reducing conditions, revealed that the higher molecular weight forms of pro-SP-B (M r ϳ 80,000 and 120,000) in nonreduced samples migrated with M r ϳ 42,000 under reducing conditions (not shown). It is conceivable that oligomers may form between wild-type and mutant SP-B proproteins, thereby preventing the wild-type protein from exiting the endoplasmic reticulum.
Alternatively, the dominant negative phenotype may have resulted from a generalized decrease in surfactant protein synthesis in transgenic animals or from changes in lung morphogenesis. To test the first possibility, SP-A and SP-C protein synthesis levels were assessed by metabolic labeling of fetal lungs. No differences were detected in SP-A and SP-C protein synthesis rates between transgenic and non-transgenic lungs (not shown). It is therefore unlikely that expression of the transgene resulted in a global decrease in surfactant protein synthesis. To determine if lung structure was altered in transgenic mice, lungs of transgenic and wild-type fetal day 18.5 littermates were analyzed. Histological analyses and immunohistochemistry with anti-SP-C antibody revealed no overt changes in lung structure (not shown). Taken together, these data suggest that hSP-B C235S/C246S protein interferes with processing of wild-type SP-B protein but does not affect the synthesis of other surfactant proteins or overall lung structure.

DISCUSSION
The SP-B mature peptide contains three intramolecular sulfhydryl bridges which are conserved in all members of the SAPLIP family. To test the function of the bridge between cysteine residues 235 and 246 of SP-B, hSP-B C235S/C246S was first expressed as a fusion protein with GFP in PC12 cells. The fusion protein was secreted in an Endo H-resistant form and was detected in the dense core granules of PC12 cells by confocal microscopy, consistent with targeting of hSP-B C235S/C246S to the regulated secretory pathway. To test the function of Cys 235/246 bridge in vivo, hSP-B C235S/C246S was expressed in the distal respiratory epithelium of SP-BϪ/Ϫ mice. Replacement of endogenous mSP-B with hSP-B C235S/C246S failed to reverse the neonatal lethality of SP-B null mice. This result indicates that the cysteine 235/246 bridge is essential for normal SP-B function. Closer analyses revealed that mutation of the Cys 235/246 bridge resulted in an SP-B molecule that was unable to exit the endoplasmic reticulum in vivo. The proposed function of this bridge is to stabilize a hairpin turn motif in the SP-B mature peptide. Thus, it is likely that stabilization of this bridge is necessary for the correct folding of SP-B in vivo.
The discrepancy between the in vitro and in vivo trafficking of the hSP-B mutant could be attributed to fusion with the GFP reporter protein. GFP has been widely used to study the intracellular localization of proteins, including several studies of proteins targeted to the regulated secretory pathway (22)(23)(24). Although GFP did not prevent correct intracellular sorting in these studies, a recent report indicated that the fusion of yeast secretory proteins to GFP resulted in aberrant targeting of constitutively secreted molecules to the yeast vacuole (25). In the present study mutant SP-B/GFP may have circumvented the protein quality control mechanisms in the endoplasmic reticulum due to a chaperone effect of the GFP molecule. Given these results, caution should be exercised in utilizing GFP as a marker for mutant secreted proteins.
The observation that the hSP-B C235S/C246S mutant protein failed to reverse the neonatal lethal SP-B null phenotype is consistent with the recent clinical finding that a mutation in this bridge in a compound heterozygote human infant was incompatible with life. 2 A single base pair change that resulted in substitution of arginine for cysteine at position 235 was described in an infant who was a compound heterozygote at the SP-B locus. The second SP-B allele carried the most common mutation in hereditary SP-B deficiency, which involves an insertion of 2 base pairs in codon 121 (121ins2), resulting in a frameshift and no SP-B proprotein in the lungs (26). This infant suffered from neonatal respiratory distress that was not responsive to surfactant replacement therapy and succumbed in the perinatal period. SP-B proprotein was detected in the lungs by immunohistochemistry using an antibody to the COOH-terminal domain of the SP-B propeptide, indicating that full-length SP-B proprotein was synthesized. 2 The results of the current study suggest that respiratory distress syndrome in this infant was the consequence of an inability to generate mature SP-B peptide related to trapping of the proprotein in the endoplasmic reticulum.
Production of mutant hSP-B C235S/C246S protein resulted in dose-dependent lethality. Transgenic lines that produced low levels of hSP-B C235S/C246S had near-normal survival rates, whereas lines that expressed intermediate levels of hSP-B C235S/C246S had significantly decreased survival rates, and lines that expressed high levels of hSP-B C235S/C246S had low levels or no surviving offspring. This is the first example in which overproduction of mutant SP-B protein in wild-type mice resulted in phenotypic abnormalities. In transgenic mice expressing high levels of hSP-B C235S/C246S protein, there was a marked reduction in the amount of wild-type SP-B mature peptide. Sulfhydryl-dependent oligomers of SP-B proprotein were also detected in these animals, indicating that the mutant protein contained exposed cysteine residues not normally available for intermolecular disulfide bonding of SP-B protein. It is possible that the dominant negative phenotype in moderate and high transgene expressing mice was caused by the formation of heterodimers between wild-type and mutant SP-B peptides. This finding could have important clinical implications if heterodimerization of mutant and wild-type SP-B results in decreased levels of SP-B in the alveolus. Individuals with mutations that result in misfolded SP-B may be at particular risk during lung injury such as hyperoxia or inflammation, which have been shown to be associated with decreased SP-B production (27,28).
A recent study by Chi et al. (29) identified two forms of SP-B mRNA in lung tissues from multiple strains of mice as well as rat and rabbit. The larger SP-B mRNA corresponded to the full-length wild-type transcript whereas the shorter form represented an alternatively spliced mRNA in which 69 base pairs were deleted from the beginning of exon 7. The 69-base pair sequence encodes 23 amino acids of the mature SP-B peptide, including the Cys 235/246 intramolecular sulfhydryl bridge. This form of SP-B mRNA accounted for approximately 30% of total SP-B mRNA in mouse and rabbit lung. Despite the relative abundance of this mRNA form, the results of the current study suggest that the protein encoded by this transcript would be unable to function in the absence of the Cys 235/246 sulfhydryl bridge; however, the phenotype associated with the deletion cannot be predicted with certainty and may differ from the sulfhydryl bridge mutation.
In summary, hSP-B C235C/S246S protein did not reverse the neonatal lethal respiratory failure in SP-B gene targeted mice. hSP-B C235S/C246S protein failed to exit the endoplasmic reticulum and was likely misfolded. Overproduction of hSP-B C235S/ C246S protein in wild-type mice decreased the concentration of mature SP-B protein and resulted in diminished survival in the neonatal period. These data suggest that mutations in the cysteine residues that make up the intramolecular sulfhydryl bridges of SP-B may predispose individuals to clinical disease even in the presence of a second wild-type allele.