Inhibition of cellular processing of surfactant protein C by drugs affecting intracellular pH gradients.

Surfactant protein C (SP-C) is a hydrophobic protein synthesized and secreted exclusively by alveolar type II cells through proteolysis of a 21-kDa propeptide (SP-C21) to produce the 3.7-kDa surface active form. Previous studies from this laboratory have demonstrated that early processing of proSP-C involves extensive intracellular proteolysis of the COOH terminus of proSP-C21 in subcellular compartments, which include the acidic type II cell-specific subcellular organelle, the lamellar body. (Beers, M. F., Kim, C. Y., Dodia, C., and Fisher, A. B.(1994) J. Biol. Chem. 269, 20318-20328). The role of intracellular pH gradients in SP-C processing was studied in freshly isolated rat type II cells. Using vital fluorescence microscopy, the pH indicator acridine orange (AO) identified intense fluorescence staining of acidic cytoplasmic vesicles within fresh type II cells. The AO vesicular staining pattern was similar in cells labeled with the lamellar body marker phosphine 3R and the phospholipid dye nile red. AO fluorescence was quenched by the addition of a membrane-permeable weak base, methylamine. Immunoprecipitation of cell lysates with anti-proSP-C antisera following pulse-chase labeling (0-2 h) with 35S-Translabel demonstrated rapid synthesis of 35S-proSP-C21 with a time-dependent appearance of 16- and 6-kDa intermediates (SP-C16 and SP-C6). Tricine polyacrylamide gel electrophoresis analysis of organic extracts of cell lysates showed time-dependent appearance of mature SP-C3.7. The addition of 5 mM methylamine significantly blocked the post-translational processing of proSP-C resulting in disruption of normal precursor-product relationships and inhibition of SP-C3.7 formation. Methylamine-treated cells exhibited slow accumulation of SP-C16 and SP-C6, a persistence of SP-C21, and an absence of SP-C3.7 for the duration of the chase period. The lysosomotropic agent chloroquine, the proton ionophore monensin, and bafilomycin A1, a specific vacuolar H+-ATPase inhibitor, each caused inhibition of proSP-C processing in a similar manner. These results demonstrate that normal post-translational proteolysis of proSP-C occurs in acidic intracellular compartments, which include the lamellar body, and that complete processing to SP-C3.7 is dependent upon maintenance of transmembrane pH gradients by a vacuolar H+-ATPase.

Pulmonary surfactant is a heterogeneous surface active complex composed of approximately 80% phospholipid, 10% other lipids, and 10% proteins synthesized and secreted exclusively by the alveolar type II cell Hawgood, 1989). The surface tension lowering properties of surfactant are primarily due to the phospholipid components (mainly dipalmitoylphosphatidylcholine); however, enhancement of the biophysical properties of these lipids has been attributed to the presence of specific surfactant-associated proteins, SP-A, 1 SP-B, and SP-C (Hawgood, 1989;Mathalgian and Possmayer, 1990;Weaver and Whitsett, 1991). Significant evidence exists to support the notion that the type II cell plays a central role in the metabolism of all surfactant components participating in synthesis, secretion, and clearance of both lipids and proteins (Baritussio et al., 1992;Breslin and Weaver, 1992;Weaver and Whitsett, 1991;Young et al., 1989).
Morphologically, type II cells are readily distinguished from other lung cell types using transmission electron microscopy to show the presence of characteristic osmiophilic, concentric laminar organelles 1-2 m in diameter located in the cytoplasm (Phizackerly et al., 1979). At the light microscopy level, these lamellar bodies demonstrate intense fluorescence when isolated type II cells or lung tissue slices are incubated with phosphine 3R (Mason et al., 1977). Lamellar bodies can be recovered intact from whole lung as well as from isolated type II pneumocytes and contain all of the surfactant phospholipids as well as SP-A, SP-B, and SP-C (Chander, 1989;Oosterlaken-Dijksterhuis et al., 1991;Phizackerly et al., 1979). Using ultrastructural morphology, immunocytochemistry, autoradiography, and biochemical methodologies, it has been demonstrated that surfactant protein and lipid components are stored in lamellar bodies prior to release into the alveolar space (Oosterlaken-Dijksterhuis et al., 1991;Phelps and Floros, 1991;Voorhout et al., 1993;Weaver and Whitsett, 1989). Conversely, by the same methods, alveolar surfactant components have been shown to be endocytosed and routed back to lamellar bodies (Baritussio et al., 1992;Breslin and Weaver, 1992;Young et al., 1989). Isolated lamellar bodies also contain lysosomal acid hydrolases, and using immunological techniques, the lysosomal glycoprotein marker CD63 has been detected on the limiting membrane (Voorhout et al., 1993). Collectively, these data show that the lamellar body functions both as a classic secretory granule and also intersects with the endocytic pathways of the central vacuolar system.
Organic extracts of surfactant contain the two smaller lipophilic proteins, SP-B and SP-C, which alone or in combination are sufficient to confer properties of rapid surface adsorption and surface tension lowering to mixtures of synthetic phospholipids (Mathalgian and Possmayer, 1990). SP-C is a 33-35-amino acid extremely hydrophobic peptide migrating with a M r of 3700 under reducing conditions. Alveolar SP-C ("mature" SP-C 3.7 ) is generated by synthesis of a larger primary translation product, proSP-C (M r ϭ 21,000), post-translational addition of covalent palmitic acid residues, and intracellular proteolysis involving cleavage of NH 2 -and COOHterminal flanking domains of the precursor molecule to yield the secreted surface active form Beers et al., , 1994aBeers and Lomax, 1995;Fisher et al., 1989;Glasser et al., 1988;Vorbroker et al., 1992Vorbroker et al., , 1995b. Previously, this laboratory had reported the production and characterization of several epitope-specific polyclonal antibodies directed against rat proSP-C using synthetic peptides corresponding to spatially distinct regions of the proSP-C primary sequence (Beers et al., , 1994aBeers and Lomax, 1995). In two different experimental models labeled with [ 35 S]methionine, immunoprecipitations of both homogenates from a perfused rat lung preparation and lysates from freshly isolated type II cells with these antisera identified the 35 S-labeled 21-kDa SP-C primary translation product (Beers et al., 1994a;Beers and Lomax, 1995). Pulse-chase experiments demonstrated processing of proSP-C 21 through 16-and 6-kDa intermediate forms  and proSP-C 6 ). In both experimental systems, proteolysis of proSP-C 21 was blocked by brefeldin A, demonstrating that intracellular processing of proSP-C in the exocytic pathway was occurring in subcellular compartments located distal to the trans-Golgi network. Western blotting of lamellar bodies isolated from adult rat lungs showed that proSP-C 6 was enriched in this subcellular fraction, and immunoprecipitation analysis of 35 S-labeled lamellar bodies showed time-dependent appearance of SP-C 6 and production of mature SP-C 3.7 within this compartment (Beers et al., , 1994a(Beers et al., , 1994bBeers and Lomax, 1995).
In other cell systems, it is well recognized that there is a progressive decrease in pH along the exocytic pathway, whereas some organellar components of the vacuolar systems (both exo-and endocytic) contain a membrane proton pump (H ϩ -ATPase) that is responsible for generation and maintenance of the internal acidic environment (Beers et al., 1982;Mellman et al., 1986). The pH of most secretory organelles (including chromaffin granules, mast cell granules, and islet cell granules) as well as endocytic vesicles and lysosomes has been measured in the range of 4.5-6.5 and can be neutralized with the use of basic amines and proton ionophores (Beers et al., 1982;Mellman et al., 1986). Chander et al. (1986) have used [ 14 C]methylamine accumulation by isolated lamellar bodies to estimate that the pH of the lamellar body matrix is 5.6 -6.1. Additionally, the transmembrane pH gradient (⌬pH) was dependent upon external energy (ATP), and both lamellar body acidification and ATPase activity in isolated limiting membranes was inhibitable by N-ethylmaleimide and dicyclohexylcarbodiimide, indicative of the presence of a vacuolar H ϩ -ATPase (v-ATPase) (Chander, 1992).
Although the role of acidification of the central vacuolar system is not well understood, recent experiments in a variety of cell types have used novel, cell-permeable nontoxic specific inhibitors of the v-ATPase (bafilomycin A 1 and concanamycin A) to show that ⌬pH can play a role in the regulation of sorting, transport, and degradation of proteins as well as in modulation of ligand-receptor interactions and cell effector functions such as cytotoxic killing (Bidani and Heming, 1995;Bowman et al., 1988;Mellman et al., 1986;Orci et al., 1986).
In the type II cell, there have been no studies to date defining the role of the ⌬pH in metabolism of surfactant components. Based on the emerging understanding of the SP-C biosynthetic pathway, we hypothesized that normal acidification of the secretory pathway including the lamellar body is crucial for proper proteolytic processing of proSP-C. In this study, we utilized an epitope-specific proSP-C antiserum (Beers et al., 1994a) and a well characterized, metabolically labeled type II cell system (Beers and Lomax, 1995) to examine the role of organellar acidification in the processing of SP-C. Using both lipophilic basic amines and carboxylic ionophores, the disruption of transorganellar pH gradients resulted in a blockade of proSP-C processing that could be mimicked with bafilomycin A 1 , a specific inhibitor of v-ATPases. Collapse of the intracellular pH gradient was also associated with complete blockade of SP-C 3.7 production. Together, the data show that proteolytic processing of SP-C proprotein involves multiple proteolytic cleavages occurring in acidic subcellular compartments dependent upon vacuolar ATPase activity.

Anti-NPROSP-C Antiserum
For these studies, we utilized an epitope-specific rat proSP-C antiserum, anti-NPROSP-C, previously produced in rabbits using a synthetic peptide antigen (Beers et al., 1994a). Anti-NPROSP-C recognizes a spatially distinct region on the linearized proSP-C molecule near the amino terminus (Met 10 -Glu 23 ). This domain forms the adjacent flanking sequence of the mature SP-C molecule. The polyclonal antiserum has been previously been shown to recognize all major synthetic precursors, and proSP-C intermediates but does not recognize mature SP-C and does not cross-react with SP-A, SP-B, proSP-B, or rat serum proteins.

Isolation of Type II Cells
Type II pneumocytes were isolated using elastase digestion of lungs from adult Sprague-Dawley rats by the method of Dobbs et al. (1986). The preparation obtained after panning on IgG-coated plates (i.e. fresh type II cells) contained greater than 80% type II cells.

S Metabolic Labeling
For metabolic labeling studies, we utilized freshly isolated cells maintained in suspension as published previously (Beers and Lomax, 1995). Cells harvested from the IgG plates (3-5 ϫ 10 6 cells/ml) were resuspended in serum-free, methionine-free Dulbecco's modified Eagle's medium (DMEM-Met), aliquoted into 15-ml conical centrifuge tubes, gassed with 5% CO 2 /95% air, capped, and equilibrated in a temperature controlled shaker bath at 37°C for 1 h prior to the addition of radiolabel.
For the duration of the metabolic labeling (pulse-chase) studies, the isolated type II cells were maintained in suspension culture by continuous shaking as previously published (Beers and Lomax, 1995). At the start of the equilibration/starvation period in DMEM-Met, inhibitors or control vehicle were added as indicated. Following substrate depletion (1 h), suspended cells were metabolically labeled by the addition of 100 Ci/ml Trans-35 S-label for a 30-min pulse-labeling period. Labeled cells were pelleted by centrifugation at 130 ϫ g for 10 min and immediately resuspended in prewarmed methionine replete medium (time ϭ 0) for varying lengths of chase time.
Harvesting of cells and media was achieved by centrifugation at 130 ϫ g for 10 min. Cell pellets were lysed by resuspension in buffer containing 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 6 mM EDTA, 2% Triton X-100, 1 mM phenylmethylsulfonyl flouride, and 5 g/ml each of aprotinin, leupeptin, and pepstatin. Cell lysates were immunoprecipitated as described below.

Analytical Methods Immunoprecipitation for Identification of 35 S-proSP-C Forms
Lysates from labeled type II cells and cell culture media were immunoprecipitated using anti-NPROSP-C antiserum as previously published (Beers and Lomax, 1995). Solubilized, radiolabeled cell lysate samples normalized to constant protein (containing 0.3-1.0 ϫ 10 6 total trichloroacetic acid precipitable counts) underwent pre-clearance by incubation with nonimmune rabbit serum. Specific precipitation of labeled proSP-C forms was achieved by incubation overnight at 4°C with anti-NPROSP-C immune serum and capture of immune complexes with protein A-agarose. Precipitated proSP-C proteins were liberated by heating to 100°C in sample buffer and analyzed by SDS-PAGE as described below. Counts in specific proSP-C forms were quantitated by direct ␤-counting of individual bands using an Ambis 4000 radioimaging detector and analysis software (Ambis, Inc.; San Diego, CA).

Polyacrylamide Gel Electrophoresis and Quantitative Analysis
For precise resolution of low molecular weight proteins, one-dimensional SDS-PAGE was performed in 16.5% polyacrylamide gels using a Tris-Tricine buffer system as modified by our laboratory (Beers and Lomax, 1995;Beers et al., 1994b;Schagger and von Jagow, 1987). Gels containing electrophoresed samples were transferred to 0.2 M nitrocellulose at 60 mA/cm 2 for 12-18 h for subsequent analysis by ␤-scanning and autoradiography.

Electrophoretic Detection and Quantitation of SP-C 3.7
Because there are no available antisera against SP-C 3.7 , in order to measure the end product of SP-C processing, mature SP-C 3.7 was detected by its characteristic profile in Tricine gels. Freshly isolated type II cells were metabolically starved in methionine-cysteine-deficient media and pulse-chase labeled with 100 -150 Ci/ml Trans-35 S-label as described above. At indicated time points, cells were harvested into 1 ml of phosphate-buffered saline containing protease inhibitors, and the resulting suspension was probe sonicated at 60% power twice for 10 s to produce a cell lysate. 800 l of type II cell lysate was then extracted with chloroform/methanol (1:2) by the method of Bligh and Dyer (1959). The organic layer containing hydrophobic proteins was removed, dried under nitrogen, reconstituted into 25 l of gel electrophoresis loading buffer, and heated at 37°C for 15 min.
Following standard 16.5% Tricine gel electrophoresis and overnight transfer to nitrocellulose, the blots were both scanned for radioactivity in an Ambis 4000 radioanalyzer and autoradiographed for 2-4 days. Counts from the 35 S-band migrating at 4 kDa (identified using the M r of 14 C low molecular weight markers) were quantitated, and kinetic profiles for SP-C 3.7 generated.
Fluorescence Microscopy-Lamellar bodies in freshly isolated type II cells were identified using two fluorescent dyes. Using the method of Mason et al. (1977), type II cells suspended in DMEM were incubated with phosphine 3R (1 g/ml) and viewed through a fluorescence microscope ( ex ϭ 488 nm and em ϭ 520 nm).
Alternatively, some preparations were stained with the hydrophobic lipid dye, nile red (Brown et al., 1992). A stock solution prepared in absolute ethanol (1 mM) was subsequently diluted directly into fresh type II cell suspensions to a final concentration of 1 M. Viewed through fluorescence microscopy ( ex ϭ 563 nm and em ϭ 625 nm, long pass), nile red fluorescence in aqueous solution is quenched, but upon interaction with cellular phospholipid inclusions, orange-red fluorescent bodies appear (Brown et al., 1992). Cells labeled with either dye were each photographed in black and white using TMAX 3200 print film and exposure times of 1-30 s.
Type II cell vesicular transmembrane pH gradients were qualitatively visualized using the membrane-permeable fluorescent pH probe acridine orange (Chander et al., 1986;Mellman et al., 1986). Freshly isolated type II cells were resuspended in Hepes-buffered saline, pH 7.4, and incubated for 20 min in the presence or the absence of 5 mM methylamine. At the completion of the incubation period, 5 M acridine orange was added to the cell suspension, and dye was allowed to distribute for 10 min at 37°C. The cells were washed by centrifugation one time in Hepes-buffered saline, and 20 l of cell suspension was placed under a coverslip for examination by fluorescence microscopy. Cells were examined using the following filters: ex ϭ 488 nm and em ϭ 520 nm, and photographed in black and white as described above.
Statistical Analysis-Results were analyzed using SigmaPlot software (version 2.00; Jandel Scientific Corporation, San Rafael, CA) and included one-way analysis of variance followed by comparisons of group mean data using Student's t test. The level of statistical significance was taken as p Ͻ 0.05.

Vital Fluorescence Microscopy: Vesicular Acidification in In-
tact Type II Cells-Freshly isolated type II cells showed characteristic staining of their lamellar bodies with phosphine 3R. Suspensions of cells harvested by panning of the IgG-coated culture plates consistently contained greater than 80% type II cells, which exhibited bright yellow fluorescent cytoplasmic inclusions (Fig. 1A, arrow). In contrast, with this compound, macrophages showed a faint, diffuse green fluorescence pattern (Fig. 1A, arrowhead).
The neutral lipid dye, nile red, has also been shown to be useful in identifying phospholipid containing vesicles from a variety of nonpulmonary tissues (Brown et al., 1992). With an aliquot of the same type II cell preparation, nile red staining was performed. When viewed in the fluorescence microscope at 563 nm excitation and 625 nm emission, orange-red inclusions (arrows) of similar size and location as the phosphine 3R vesicles were observed exclusively within type II cells and indicated the presence of phospholipid-rich cytoplasmic vesicles in the freshly isolated cell population (Fig. 1B).
Using in situ techniques, quantitative measurement of the accumulation of 14 C-radiolabeled methylamine has previously shown that isolated lamellar bodies maintain a transmem-

FIG. 1. Fluorescence micrograph of phosphine 3R and nile red staining of type II cells. Fresh type II cells resuspended in DMEM
were incubated with phosphine 3R (1 g/ml) (A) or 1 M nile red prepared as described under "Experimental Procedures" (B). Aliquots were removed, coverslipped, and examined by fluorescence microscopy (for phosphine em ϭ 488 nm and em ϭ 520 nm; for nile red ex ϭ 563 nm and em ϭ 625 nm, long pass). Fresh type II cells exhibit punctate phosphine 3R staining in cytoplasmic vesicles (A, arrows) with vesicles of similar size and location also staining with nile red indicative of phospholipid accumulation (B, arrows). Macrophages stained with phosphine 3R fluoresce a diffuse green without vesicular structures (A, arrowhead). The pictures were photographed with TMAX ASA3200 black and white film using 2-s exposures.
brane pH gradient (acidic interior) (Chander et al., 1986). Freshly isolated type II cells were incubated with the pHsensitive dye acridine orange (AO) and examined by fluorescence microscopy. As shown in Fig. 2A, type II cells were found to contain AO fluorescent cytoplasmic vesicles (arrow). The appearance and distribution of AO vesicles in the cytoplasm was similar to that of the phosphine 3R/nile red-positive vesicles. No AO-stained vesicles were observed in alveolar macrophages. When type II cells were preincubated for 30 min with 5 mM methylamine, the acridine orange fluorescence was quenched consistent with a collapse of the organellar ⌬ pHmediated by the accumulation of the weak base (Fig. 2B).
Taken together, fluorescence microscopy using specialized stains indicated that preparations of freshly isolated type II cells used in these studies contained a phospholipid-rich compartment that maintained a transmembrane pH gradient (acidic interior) and that could be collapsed by incubation with methylamine. The data are consistent with the identification of at least a significant portion of these vesicles as lamellar bodies.
ProSP-C Processing by Type II Cells: Effect of Intracellular pH Gradients-The effect of intracellular pH gradients on synthesis and early processing events for proSP-C was studied using freshly isolated type II cells in suspension culture labeled with [ 35 S]methionine in a pulse-chase protocol. Autoradiographs of immunoprecipitates of control cell lysates with an epitope specific proSP-C antiserum previously shown to recognize all major proSP-C forms (anti-NPROSP-C)  demonstrated the appearance of the SP-C primary translation product (21-kDa band) with processing of proSP-C 21 protein through several intermediates (Fig. 3A). The time-dependent appearance of a 16-kDa intermediate followed by the appearance and subsequent disappearance of lower molecular weight proSP-C forms (6 -10,000) was consistently observed in all experiments. Specificity of the immunoprecipitation was confirmed by identical analyses using nonimmune serum (Fig.  3A, NIS).
When type II cells were preincubated with 5 mM methylamine for 30 min prior to pulse labeling, normal intracellular processing of proSP-C was markedly altered. Immunoprecipitation of lysates from methylamine-treated cells showed a persistence of the SP-C 21 primary translation product and a delayed accumulation of the SP-C 16-kDa and SP-C 6-kDa intermediates (Fig. 3B). The SP-C 6 -10 forms in methylaminetreated cells were only clearly detectable after 2 h of the chase using prolonged exposures of the film (not shown). Analysis of

FIG. 2. Vacuolar acidification in intact type II cells and the effect of methylamine.
Fresh type II cells resuspended in Hank's balanced salt solution were preincubated with 5 mM NaCl (A) or 5 mM methylamine (B) for 20 min. Acridine orange (5 M) was added to each preparation for 10 min before aliquots were removed, coverslipped, and examined by fluorescence microscopy (E x ϭ 488 nm; E m ϭ 520 nm). In A, control type II cells contain punctate acridine orange staining in cytoplasmic vesicles (arrows). The fluorescence signal is abolished by the addition of methylamine (B). The pictures were photographed with TMAX ASA3200 black and white film using 2-s exposures.

FIG. 3. Disruption of SP-C synthetic processing in isolated type II cells by agents that modify intracellular pH gradients.
Representative pulse-chase experiment in which fresh type II cells were equilibrated for 1 h in methionine-deficient DMEM in the presence of 5 mM NaCl (A), 5 mM methylamine (B), or 2 m monensin (C) and then pulsed for 30 min with Trans-35 S-label (100 Ci/ml). Following the labeling period, cells were resuspended in methionine-replete, serumfree DMEM and chased for up to 2 h in the presence or the absence of inhibitors. At the indicated time points (post-pulse (P) ϭ time 0), cells and media were separated by centrifugation. Lysates (normalized to equal cell numbers) were prepared, immunoprecipitated using anti-NPROSP-C, and subjected to SDS-PAGE as described under "Experimental Procedures." In A, an autoradiograph of immunoprecipitated proteins from the control lysates demonstrates early appearance of primary SP-C 21 and subsequent detection of the time-dependent appearance of proSP-C 16 (arrowheads) and proSP-C 6 -10 (bracket) intermediates. 14 C Low molecular mass markers (lane MW; M r given in kDa at left) and control immunoprecipitation of a 1-h sample using nonimmune serum (lane NIS) are also shown. Both methylamine (B) and monensin (C) inhibit proteolysis of SP-C 21 and disrupt the normal progression of appearance of proSP-C intermediates. the medium of methylamine-treated cells by immunoprecipitation failed to detect the presence of proSP-C forms (not shown), indicating that neutralization of organellar pH did not induce constitutive secretion of proSP-C precursors.
In addition to methylamine, carboxylic ionophores are another class of "acidotropic" agents that have been used to elevate vacuolar pH (Bidani and Heming, 1995;Mellman et al., 1986). Monensin is a cell-permeable ionophore shown to mediate the exchange of monovalent cations for protons (i.e., cytoplasmic K ϩ for organellar H ϩ ). Monensin was tested for its effects on SP-C processing by preincubation with type II cells for 30 min prior to metabolic labeling. At a concentration of 2 M, monensin treatment markedly attenuated the proteolysis of SP-C 21 and SP-C 16 (Fig. 3C).
The effect of methylamine and monensin on proSP-C processing was quantitated using direct ␤-counting of the major bands. The net counts in the 35 S-labeled SP-C 21 , SP-C 16 , and SP-C 6 -10 forms appearing in the gels from the immunoprecipitated cell lysates were counted and then normalized for the number of methionine residues in each form as previously published (Beers and Lomax, 1995). Correction for interexperimental variability in absolute counts in the bands was done by expressing the sum of the total counts immunoprecipitated in the 3 major SP-C forms at the conclusion of the 30-min pulse (time 0) for each experiment as 100%. At subsequent time points within the chase, counts in each individual proSP-C band were then expressed as a percentage of the initial postpulse total. The normalized data for each time point from the individual experiments were combined and expressed as the means Ϯ S.E.
As Fig. 4 demonstrates, at the start of the chase period (time ϭ 0 h), almost 95% of the total counts found in immunoprecipitable proSP-C forms were in SP-C 21 in both control and treated cells. During the chase (0 -2 h), control lysates showed rapid disappearance of 35 S-SP-C 21 with a commensurate appearance of SP-C 16 and SP-C 6 forms and then subsequent disappearance of the SP-C 6 band consistent with conversion of proSP-C to the immunologically undetectable mature SP-C 3.7 (Fig. 4A). Under these conditions, only about 20% of the initial post-pulse amount of proSP-C 21 remained after 1 h, which was consistent with significant proteolytic processing. In contrast, administration of methylamine significantly blunted the proteolytic processing of SP-C 21 . As Fig. 4B shows, during the chase, methylamine-treated cells retained large amounts (90 -95% of postpulse levels) of proSP-C 21 and exhibited a delayed but significant accumulation of SP-C 16 and SP-C 6 .
The effect of monensin on the inhibition of initial proteolysis of the SP-C primary translation product was somewhat intermediate ( Fig. 4C) with the amount of 35 S-proSP-C 21 recovered after 1 h of the chase period approaching 70% that of the starting material. However, as for methylamine-treated cells, proSP-C 16 and proSP-C 6 forms accumulated in the monensin lysates at later time points, and the normal precursor-product kinetic relationships appeared to be significantly altered.
The effect of these acidotropic agents on proSP-C processing was not due to a generalized effect on proSP-C 21 translation and synthesis. At the end of the 30-min pulse-labeling period, the net raw counts incorporated into the 35 S-SP-C 21 primary translation product band were unchanged by the addition of methylamine as compared with matched controls (methylamine ϭ 102 Ϯ 3% of net counts in control; p ϭ 0.59; n ϭ 4). Likewise, monensin administration did not alter the initial rate of proSP-C synthesis (post-pulse monensin ϭ 90.6 Ϯ 7.2% net counts versus control; p ϭ 0.32, n ϭ 3). In addition, there was no difference in the relative amount of proSP-C 21 (as a percentage of total proSP-C forms) at the start of the chase period for methylamine-treated, monensin-treated, or control preparations (94.5 Ϯ 1.2, 95.0 Ϯ 1.0, and 93.9 Ϯ 0.8%, respectively).
Inhibition of SP-C 3.7 Production by Methylamine-Although the kinetic profiles in Fig. 4 demonstrate precursor-product relationships between proSP-C 21 and other proSP-C intermediates, full processing to SP-C 3.7 cannot be directly shown using immunological methods. In order to demonstrate SP-C 3.7 production by this experimental system and to determine if FIG. 4. Quantitative analysis of inhibition of SP-C processing by acidotropic agents and proton ionophores. 35 S-Labeled SP-C 21 , SP-C 16 , and SP-C 10 -6 forms detected in cell lysates from experiments of fresh type II cells treated with control (A), methylamine (B), or monensin (C) as described in Fig. 3 were quantitatively analyzed by direct ␤-counting of radioactive bands using the Ambis 4000 Radioanalyzer. The net counts detected in each proSP-C band were normalized for the number of methionine residues in each intermediate based on approximate processing domains (Beers and Lomax, 1995) and predicted amino acid sequence as described in the text. The total counts immunoprecipitated in the three major SP-C forms at the conclusion of the 30-min pulse (time 0) for each experiment was expressed as 100%, and each band at subsequent time points was expressed as a percentage of the post-pulse total value. The double ordinate plot contains a left axis scale for SP-C 21 and an expanded scale on the right axis for SP-C 16 and SP-C 6 intermediates. The data points are expressed as the means Ϯ S.E. (n ϭ 7 for controls and n ϭ 3-4 for methylamine and monensin.) disruption of normal proSP-C processing kinetics affects production of mature SP-C, we identified and quantitated the amount of 35 S-labeled SP-C 3.7 in organic extracts of the type II cell lysates using high resolution Tricine gel electrophoresis.
In a previous study, we had shown that Tricine PAGE of lamellar body proteins could precisely resolve peptides of M r Ͻ10,000 into three separate bands migrating at 7-8, 6, and 4 kDa (Beers et al., 1994b). By Western blot, the 7-8-kDa band was identified as SP-B, and the 6-kDa band was identified as proSP-C 6 . By Coomassie Blue staining, the 4-kDa band comigrates with a synthetic SP-C 3.7 peptide standard, and using the method of Qanbar and Possmayer (1994), it is palmitoylated as assessed by [ 14 C]iodoacetamide labeling of palmitoylthioester linkages. 2 Following pulse-chase labeling in the presence and the absence of 5 mM methylamine, 35 S-SP-C 3.7 was identified in type II cell lysates by using the method of Bligh and Dyer (1959) to extract low molecular weight hydrophobic surfactant proteins into organic solvents and then separating them electrophoretically in SDS-PAGE. The amount of radioactivity in each of the 3.7-kDa bands was then directly quantitated from the gels by Ambis radioanalytic scanning.
In Fig. 5A, an autoradiograph representative of four separate experiments demonstrates the time-dependent appearance of the 3.7-kDa band in the control cells, which is blocked by the preincubation with methylamine. In Fig. 5B, group mean data demonstrate the qualitative kinetics of SP-C 3.7 production. To correct for interexperimental variability, the net counts measured in the 4-kDa band at each time point in both control and methylamine-treated lysates were normalized by setting the SP-C 3.7 counts in the 1 h control as 100%. Under control conditions, a small amount of SP-C 3.7 appeared as early as 30 min, but a significant increase in band intensity occurred at 1-2 h following the pulse. When compared with Figs. 3 and 4, SP-C 3.7 production was preceded by the synthesis of SP-C 21 as well as the appearance and disappearance of proSP-C 6 consistent with a precursor-product relationship for proSP-C 6 -mature SP-C 3.7 . In the presence of methylamine, commensurate with disruption of proSP-C processing and accumulation of intermediate forms, the production of SP-C 3.7 is completely blocked (Ͻ5% versus control at 1 h).
Chloroquine Also Inhibits proSP-C Metabolism-Chloroquine, a well known lysosomotropic agent, is a tertiary amine shown to accumulate in acidic organelles (Orci et al., 1986). As illustrated by the autoradiograph in Fig. 6A, chloroquine also caused inhibition of proSP-C processing. Pretreatment of type II cells with 10 M chloroquine resulted in disruption of the normal proteolysis of proSP-C 21 with an accumulation of significant amounts of SP-C 16 intermediate and a lesser accumulation of the lower molecular weight forms, which were detected on prolonged exposure of the blots to film. Processing kinetics from quantitation of individual bands from several experiments performed in the presence of chloroquine are shown in Fig. 6B. As for the other acidotropic agents, sequential proteolysis of each proSP-C form was blocked with chloroquine pretreatment. The use of chloroquine did not alter the initial rate of proSP-C synthesis (Post-pulse chloroquine ϭ 84.0 Ϯ 17.0% net counts versus control; p ϭ 0.63, n ϭ 3).
Inhibition of Vacuolar ATPase Blocks SP-C Processing-Although the effect of acidotropic agents on intracellular pH gradients is predicted to be most profound on the organelles with the highest degree of acidification (e.g. lamellar bodies, lysosomes, or multivesicular bodies), the additional effects of some of these reagents such as the alteration of the pH in other compartments (i.e., Golgi) (Orci et al., 1986) or the nonspecific inhibition of the proSP-C proteases by these drugs is possible. To exclude these factors and to more specifically define the effect of organellar pH on SP-C processing, experiments were carried out using a cell-permeable inhibitor of the v-ATPase, the H ϩ -ATPase previously shown to be responsible for acidification of secretory organelles. Freshly isolated type II cells were preincubated for 1 h with 0.2 M bafilomycin A 1 and analyzed for proSP-C synthesis and processing. As Fig. 7A demonstrates, bafilomycin A 1 produced a processing profile for proSP-C that was similar to that seen using methylamine, monensin, and chloroquine. On quantitation of individual intermediates (Fig. 7B), there was marked attenuation of proSP-C 21 proteolysis and a delayed appearance and accumulation of proSP-C 16 and SP-C 6 . Preincubation with vehicle alone (0.1% v/v Me 2 SO) had no effect of proSP-C processing profiles (data not shown).

DISCUSSION
In the present work, synthesis and post-translational processing of the SP-C primary translation product was studied in isolated type II cells using pulse-chase labeling. Previously, we had utilized epitope-specific antisera and the macrolide antibiotic brefeldin A (BFA) in both a perfused lung preparation and isolated type II cells to characterize the initial post-translational proteolytic processing events for proSP-C 21 . These steps include cleavage of COOH-terminal propeptide domains in subcellular compartments distal to the trans-Golgi network followed by subsequent proteolysis of a vestigial NH 2 -terminal flanking domain in a particulate lamellar body fraction prior to secretion of SP-C 3.7 into the alveolar space (Beers et al., 1994a;Beers and Lomax, 1995). The present study extends these 2 M. F. Beers, unpublished observations.

FIG. 5. Detection and quantitation of mature SP-C 3.7 production in type II cell lysates. Type II cells labeled with 150 Ci/ml
Trans-35 S-label in a pulse-chase protocol as described for Fig. 3 were harvested at the indicated time points during the chase and lysed by sonication in phosphate-buffered saline. Hydrophobic proteins were extracted into chloroform/methanol (1:2) as described under "Experimental Procedures," separated by Tricine SDS-PAGE, and transferred overnight to nitrocellulose. A, representative autoradiograph showing the appearance of 35 S-SP-C 3.7 in control lysates, which is inhibited by methylamine. B, the net counts detected in the 3.7-kDa band by direct ␤-scanning of the nitrocellulose blots for each time point from controls (n ϭ 4) and methylamine-treated (n ϭ 3) lysates were normalized to the 1-h control value as described under "Results." The data points are expressed as the means Ϯ S.E. observations with new data demonstrating that in this isolated type II cell model, proSP-C is processed completely to SP-C 3.7 , that processing of SP-C occurs in acidic subcellular compartments, and that perturbation of these transmembrane proton gradients with weakly basic amines or ionophores results in disruption of normal processing kinetics and blockade of SP-C 3.7 production. The modification of intraorganellar pH on SP-C synthetic processing by acidotropic agents could be mimicked by bafilomycin A 1 , a cell-permeable specific inhibitor of vacuolar ATPases.
The experiments presented in this study were performed by metabolic labeling of freshly isolated type II cells in suspension in which acidotropic agents were used to disrupt proSP-C processing. This laboratory has previously shown that the use of this model for the dissection of the synthetic pathway for SP-C is physiologically relevant to events occurring in vivo. The proSP-C processing profile (time course and major intermediates) obtained by immunoprecipitation of fresh type II cells (Fig. 3A) was similar to that observed in a previously published analysis of lung homogenates from a 35 S-labeled isolated perfused lung (Beers et al., 1994a). Subcellular fractionation of the metabolically labeled perfused lung preparation showed that the initial proteolysis of SP-C 21 yielding a proSP-C 16 intermediate appears to occur within a vesicular compartment of the secretory pathway at a yet undefined site located between the trans-Golgi network and the lamellar body compartment (Beers et al., , 1994aBeers and Lomax, 1995). Based on immunoelectron micrographic data from Voorhout et al., the multivesicular body or a nonclathrin-coated Golgi budding vesicle appears to be the most likely site for this initial cleavage step (Voorhout et al., 1993). The second step in proSP-C processing, the conversion of SP-C 16 to SP-C 6 , is most likely also occurring in a late multivesicular body. Both Western analysis of subcellular lung fractions (which has shown a marked enrichment in SP-C 6 in lamellar bodies) as well as the metabolic labeling profile observed in 35 S-lamellar bodies indicate that the lamellar body fraction is the subcellular compartment for the final removal of the N-flanking domain in SP-C 6 (Beers et al., 1994a;Beers and Lomax, 1995).
A fundamental assumption of studies using epitope-specific proSP-C antibodies has been that under control conditions, proSP-C 21 and the intermediates that are generated in a precursor-product kinetic relationship lead to production of SP-C 3.7 . Because of its poor immunogenicity, detection of SP-C 3.7 has not been possible using conventional immunoprecipitation . Using an alternative approach, the unique solubility and low molecular weight of SP-C 3.7 were exploited to quantitate the intracellular production of 35 S-labeled SP-C 3.7 in organic extracts of type II cell lysates. In a previous report, we had shown the utility of Tricine PAGE to separate 35 S-radiolabeled bands of 7 (SP-B), 5.5 (proSP-C 6 ), and 4 (mature SP-C) kDa in lamellar body fractions (Beers et al., 1994b). This technique was modified for cell lysates by using the method of Bligh and Dyer (1959) to extract low molecular weight hydrophobic surfactant proteins into organic solvents followed by analysis with Tricine SDS-PAGE. The FIG. 6. Chloroquine blocks proteolytic processing of proSP-C 21 . A representative immunoprecipitation analysis using anti-NPROSP-C of cell lysates was obtained from a [ 35 S]methionine pulse-chase experiment using freshly isolated type II cells pretreated for 30 min with 10 M chloroquine prior to labeling. Labeling conditions were identical to those for Fig. 3. A, following a 30-min [ 35 S]methionine pulse, chloroquine-treated cell samples were harvested during the chase period and immunoprecipitated. The time points corresponding to each lane of the 16.5% Tricine gel are given in hours (post pulse (p) ϭ time 0). 14 C-Labeled molecular mass standards (29,18,14.7,6,and 3.5 kDa) are shown in lane MW. The position of the major intermediates are shown at right: SP-C 21 , SP-C 16 (arrowheads), and SP-C 6 (bracket) (detected at longer exposure times). B, quantitation of SP-C intermediates from chloroquine-treated cell lysates (n ϭ 3 separate experiments) analyzed as described in the legend to Fig.  4, showing that chloroquine-treated cells failed to exhibit normal processing kinetics for proSP-C. The double ordinate plot contains a left axis scale for SP-C 21 and an expanded scale on the right axis for SP-C 16 and SP-C 6 intermediates. The data points are expressed as the Means Ϯ S.E.

FIG. 7. Bafilomycin A 1 blocks proteolytic processing of proSP-C.
A, representative immunoprecipitation analysis using anti-NPROSP-C of cell lysates obtained from a [ 35 S]methionine pulse-chase experiment using freshly isolated type II cells pretreated for 30 min with bafilomycin A 1 (final concentration, 10 g/ml in Me 2 SO) prior to labeling as described in the legend to Fig. 3. Following a 30-min 35 S pulse, the time points at which cell samples were harvested during the chase period are given above the corresponding lane in hours (post pulse (p) ϭ time 0). 14 C-Labeled molecular mass standards (in kDa) are shown in lane MW. A 2-h control sample immunoprecipitated with nonimmune serum is shown in lane NIS. B, double ordinate plot showing quantitative expression of SP-C intermediates detected in bafilomycin-treated cell lysates (n ϭ 3 separate experiments) as a function of chase time. As for Fig. 4, the plot contains a left axis scale for SP-C 21 and an expanded scale on the right axis for SP-C 16 and SP-C 6 intermediates. The data points are expressed as the means Ϯ S.E. kinetic data generated by direct quantitation of the gels (Fig. 5) demonstrate time-dependent appearance of SP-C 3.7 during the chase period that follows both the initial proteolysis of proSP-C 21 and the appearance/disappearance of SP-C 16 and SP-C 6 . This is consistent with a precursor-product relationship for proSP-C-mature SP-C.
A direct quantitative comparison of the counts in SP-C 3.7 in the organic extracts with those counts in proSP-C forms obtained by immunoprecipitation of the lysates is not possible because of the differences in technique and the uncertainty in the total recovery. The efficiency of immunoprecipitation and the partition co-efficient for mature SP-C in the organic phase of the Bligh-Dyer extract make the calculation of the exact percentage of full conversion of SP-C 21 impossible. However, although stoichiometric quantitation is not possible, based on two independent investigations, it is unlikely that much of the synthesized SP-C 21 is degraded. Previously, we used brefeldin A to block anterograde transport of proSP-C from the endoplasmic reticulum-Golgi compartment of isolated type II cells (Beers and Lomax, 1995). In pulse-chase analysis using BFA, we saw no significant degradation of proSP-C 21 in immunoprecipitates of these cell lysates up to 4 h after labeling, indicating that proSP-C 21 retained in the endoplasmic reticulum in the BFA-treated state is not metabolized. Recently, work by Vorbroker et al. (1995b) has confirmed this finding using type II cells cooled to 20°C to mimic the BFA phenotype. Cells treated in this fashion also do not degrade proSP-C 21 (Vorbroker et al., 1995b). These results differ from those obtained for secretion of apolipoprotein B in HepG2 cells in which significant amounts of apo B are degraded in the endoplasmic reticulum rather than processed to mature lipoprotein (Dixon et al., 1991). Taken in toto, the BFA results, the documentation of SP-C 3.7 in cell lysates, and the precursor kinetics suggest that a degradative path in the exocytic pathway is minor.
Incubation of type II cells with acidotropic agents results in disruption of the normal processing sequence for proSP-C. Methylamine (Fig. 3B) altered the observed normal precursorproduct relationship between proSP-C 21 and proSP-C 16 as well as the kinetics of conversion of SP-C 16 to SP-C 6 and proteolysis of SP-C 6 (Fig. 4). Commensurate with the accumulation of intermediates, production of SP-C 3.7 , indicative of the final anterograde processing of SP-C 6 , was also completely blocked by methylamine (Fig. 5). Based on these results, it appears that the disruption of the normal precursor-product curves (control) leading to the accumulation of SP-C 16 /SP-C 6 (methylamine) is associated with a total blockade of SP-C 3.7 production. In addition to methylamine, other classes of acidotropic agents including chloroquine (Fig. 6) and the ionophore monensin (Fig.  3C) as well as bafilomycin A 1 , a selective inhibitor of H ϩ ATPases, all effectively disrupted much of the proteolytic processing of proSP-C 21 and resulted in the characteristic accumulation of SP-C 16 and varying amounts of SP-C 6 . Therefore, it is likely that all of these agents completely block production of SP-C 3.7 .
Concurrent with the observed alterations in metabolic processing of SP-C, fluorescence microscopy indicated that acidotropic agents also affect the normal acidification of type II cell subcellular organelles. Staining of freshly isolated type II cells with acridine orange demonstrated the presence of acidic subcellular vesicles ( Fig. 2A). Methylamine administration resulted in total quenching of the fluorescent signal indicative of complete collapse of all intracellular transmembrane pH gradients by the membrane-permeable basic amine. (Fig. 2B). The distribution and location of the acidic type II cell vesicles visualized with acridine orange were similar to yellow-gold cytoplasmic fluorescent vesicles identified with phosphine 3R (Fig.   1A) and orange-red vesicles stained with nile red (Fig. 1B). Phosphine-stained vesicles were shown by Mason et al., 1977) to correspond to lamellar bodies in electron micrographs of both isolated type II cells and intact lung, whereas nile red has been used to demonstrate intracellular phospholipid vesicles in other cells (Bidani and Heming, 1995;Greenspan et al., 1985).
Although the results obtained using fluorescent dyes are consistent with acridine orange staining of lamellar bodies, co-localization of the dyes within the same cells using double labeling would yield more definitive characterization of vesicular staining. However, the physical properties of nile red and acridine orange preclude their use with selective filter packages for such studies. Nile red is actually a dual purpose lipid dye. In the cytoplasm, nile red exhibits both yellow-gold fluorescence ( x ϭ 450 -500 nm; m ϭ 528 nm), indicative of neutral lipids as well as the orange-red fluorescence ( x ϭ 515-563 nm; m Ͼ 590 -625 nm) used in this study to delineate phospholipid containing organelles (Bidani and Heming, 1995;Greenspan et al., 1985). Likewise, for AO, in addition to qualitatively measuring intracellular pH gradients (using fluorescein type filter packages), the dye binds to both DNA and RNA. Upon binding to RNA, there is a spectral shift of RNA-AO complexes to red (E x ϭ 460 nm; E m ϭ 650 nm). Thus, both dyes exhibit additional overlapping spectral characteristics due to interactions with other intracellular moieties that cannot be remedied by the use of selective filters. Nonetheless, the quenching of acridine orange fluorescence in intact type II cells (Fig. 2B) clearly demonstrates that methylamine collapses all intracellular pH gradients and suggests that at least some of these vesicles are lamellar bodies. These new findings in intact cells are consistent with published data of Chander et al. (1986) in which the intralamellar body pH measured in vitro using isolated lamellar bodies was 5.5-6.0 and in which intense acridine orange fluorescence of purified lamellar bodies fractions was also demonstrated. Therefore, it is probable that lamellar bodies, the most acidic subcellular compartment of type II cells, are among the brightest fluorescing organelles visualized with acridine orange, but other acidic vesicles could contribute to the overall cytochemical staining pattern.
The precise mechanism(s) by which changes in intracellular pH gradients are affecting the processing of proSP-C and production of SP-C 3.7 are not defined. The SP-C processing pathway is schematically illustrated in Fig. 8. The major block in SP-C synthetic processing induced by these reagents could exist primarily at the level of the lamellar body (blockade of final processing), which would lead to the accumulation of proSP-C 6 and the absence of mature SP-C. From published studies, the exocytic pathway demonstrates a progressive acidification from the endoplasmic reticulum to the secretory organelles (Orci et al., 1986). The most significant alteration of cellular function described for both lipophilic amines (e.g., methylamine and chloroquine) and proton ionophores (e.g., monensin) is related to their effects of vacuolar acidification (Mellman et al., 1986). For type II cells, this would most likely include the acidic lamellar body compartment (pH ϭ 5.5-6.0) (Chander et al., 1986) but could also include the nonclathrincoated multivesicular body, shown to possess an acidic pH (pH 6.5) (Mellman et al., 1986). The defect(s) in proSP-C processing induced by alkalinization of these acidic compartments could have several possible mechanisms, occurring alone or in combination. These include: (i) drug-induced alterations in organellar pH that indirectly influence protease function (through changes in pH away from the enzyme optimum); (ii) direct inhibition of the protease(s); and (iii) pH-induced alterations either in targetting to or in uptake of proSP-C intermediates by the processing organelle(s).
Secretory granules in the neurohyphophysis and pancreas each utilize an acidic intragranular pH that provides optimal conditions for the proteolytic processing enzymes that cleave prohormones into mature species (Loh et al., 1984;Orci et al., 1986). Chloroquine has previously been shown to directly inhibit cathepsin B activity in rat fibroblasts (Wibo and Poole, 1974). At present, none of the proteases that process SP-C are currently defined, nor are the pH profiles for these events known; however, the potential role for pH in the processing of another hydrophobic surfactant protein SP-B has been partially elucidated. Processing of the 42-kDa SP-B primary translation product has been shown to mediated by a cathepsin D-like protease, which has a pH optimum of 5.0. At pH 7.0, enzyme activity was inhibited over 85% . Further studies characterizing specific SP-C proteases, their enzymatic properties, and rates of proSP-C intermediate packaging will be required.
On the basis of the data presented, disruption of SP-C processing at the level of the conversion of SP-C 6 to mature SP-C by acidotropic agents could account for the absence of mature SP-C and accumulation of SP-C 6 seen with administration of all the drugs tested. However, the accumulation of SP-C 16 consistently induced by each reagent as well as variations in the relative amount of proSP-C 6 (e.g., methylamine and monensin versus chloroquine) and in levels of SP-C 21 seen in the cell lysates treated with different classes of reagents also indicates that these compounds may be having effects in addition to changing vacuolar pH (Figs. 3, 4, and 5). Both lipophilic amines as well as carboxylic ionophores have also been shown to have other secondary cellular effects including alterations in targetting of proteins in the secretory pathway, changes in vesicular membrane trafficking, and induction of vacuolar and Golgi swelling (Mellman et al., 1986;Radons et al., 1994). We have previously shown that chromaffin granules utilize transmembrane pH gradients to regulate the uptake and storage of biogenic amines (Beers et al., 1982). Additionally, the use of methylamine and chloroquine was consistently associated with a small increase in total label incorporation in these samples during the chase, which was somewhat higher than at the end of the pulse, and raises the possibility that these agents could also affect the efficiency of the chase.
To exclude some of these nonspecific effects, additional experiments were done using an inhibitor of the vacuolar H ϩ ATPase. It has previously been shown that the lamellar body limiting membrane is markedly enriched in H ϩ -ATPase activity, which was inhibited by known blockers of vacuolar ATPases and distinguishable from other H ϩ -ATPases (F 0 F 1 , E 1 E 2 , and P type) (Chander, 1992). Bafilomycin A 1 , a cellpermeable, specific v-ATPase inhibitor disrupted the processing profiles resulting in a secondary accumulation of SP-C 16 and SP-C 6 in the cell lysates (Fig. 7). This result suggests that the observed impairments in proSP-C processing are not due solely to a direct inhibition of protease activity by reagents such as chloroquine and methylamine. Because azide insensitive H ϩ -ATPase activity has been found in clathrin-coated vesicles isolated from the brain (Xiao-Song and Stone, 1986), the role of pH in prelamellar body compartments will require the localization of v-ATPase in type II cells and intracellular pH measurements using ultrastructural techniques.
Recently, additional evidence for the importance of proper intracellular targetting and proteolysis in the biosynthesis and post-translational processing of SP-C has been underscored by the phenotypic characterization of congenital SP-B deficiency resulting from mutational disruption of the SP-B gene (Nogee et al., 1993). In both the ⌬121ins2 (deMello et al., 1994) and R236C mutations (Ballard et al., 1995), the SP-B-deficient phenotype is associated with abnormal post-translational processing of SP-C manifested as an abnormal accumulation of a 6 -12-kDa proSP-C intermediate with a concomitant lack of detectable SP-C 3.7 (Ballard et al., 1995;Vorbroker et al., 1995a). Electron microscopy of SP-B-deficient lung tissue has shown absence of well formed lamellar bodies in type II cells and loss of directional secretion (deMello et al., 1994). The disruption of exocytic organellar acidification, SP-C post-translational processing, and intracellular trafficking mediated by the collapse of transmembrane proton gradients in type II cells are consistent with this notion that the normal intracellular processing and packaging of SP-C protein is dependent upon the presence of phenotypically normal type II cells.
In conclusion, synthetic processing of SP-C studied in metabolically labeled adult alveolar type II cells using epitope-specific proSP-C antisera has demonstrated extensive proteolytic remodeling of the SP-C primary translation product leading to production of mature SP-C 3.7 . The normal post-translational cleavage pattern of the propeptide-flanking regions, which can be blocked by basic amines and ionophores, leads to inhibition of SP-C 3.7 formation and accumulation of metabolic intermediates. Together these results indicate that complete processing is dependent, in part, upon maintenance of intracellular transmembrane pH gradients in the exocytic pathway by a vacuolar proton ATPase in vesicular compartments, which include the lamellar body.