The surfactant protein C mutation A116D alters cellular processing, stress tolerance, surfactant lipid composition, and immune cell activation

Eukaryotic expression vectors containing the full human SFTPC gene fused to either EGFP-tag (pEGFP-N1/hSP-C1-197 and pEGFP-C1/hSP-C1-197 to obtain proSP-C with EGFP fused to the C- or N-terminus, respectively) or hemagglutinin (HA)-tag (proSP-C with N-terminal HA-tag) were obtained as previously described [12]. The A116D point mutation was introduced into the wild-type (WT) SFTPC gene in these vectors using the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, USA) and the following primers: A116D_forward: 5'-GCC TAC AAG CCA GAC CCT GGC ACC TGC-3', A116D_reverse: 5'-GCA GGT GCC AGG GTC TGG CTT GTA GGC-3', following the manufacturer's protocol. The successful mutagenesis was confirmed by Sanger sequencing.

The mouse MLE-12 lung epithelial cell line (CRL-2119) [29] was obtained from the American Type Culture Collection (ATCC) and maintained in RPMI medium supplemented with 10% fetal bovine serum. Cells were transfected using FuGene 6 (Roche, Penzberg, Germany) according to the manufacturer's instructions. MLE-12 cells stably transfected with either pcDNA3/HA-hSP-C1-197 or pcDNA3/HA-hSP-CA116D vectors were obtained by selecting transfected cells in the presence of 600 μg/ml G418 in RPMI medium for four weeks. For drug exposure experiments, stable cells were grown for 24 hours in the presence of 10 μM of cyclophosphamide, azathioprine, methylprednisolone or hydroxychloroquine. Untreated cells that were cultured in parallel served as controls.

Total cell proteins were obtained by lysing the cells in lysis buffer [PBS, 20 mM EDTA, 1% v/v Elugent (Calbiochem, Bad Soden, Germany), protease inhibitor (Complete; Roche, Manheim, Germany)]. For immunoblotting, 30 μg protein was separated under reducing conditions using 10% NuPage Bis-Tris gels (Invitrogen, Karlsruhe, Germany) and transferred to a PVDF membrane. The following primary antibodies were used: monoclonal rat anti-HA-tag (1:1000; Roche), monoclonal mouse anti-GFP (1:500; Clontech, Heidelberg, Germany), and polyclonal goat anti-calnexin (1:500), polyclonal goat anti-calreticulin (1:500), monoclonal mouse anti-HSP90α/β, polyclonal goat anti-HSP70 (1:1000), and monoclonal anti-β-actin HRP conjugate (1:10000) (all from Santa Cruz Biotechnology, Santa Cruz, CA). Signal was detected using chemiluminescent labeling with Amersham ECL Detection Reagents (GE Healthcare), densitometrically quantified and normalized to the β-actin signal.

24 hours after transfection, cells grown on coverslips were fixed with 4% paraformaldehyde, permeabilised with 10% Triton X-100, and blocked for 30 min in PBS with 5% FBS. The following primary antibodies were used in 1:200 dilution: polyclonal rabbit anti-mouse LAMP3 (Santa Cruz), monoclonal mouse anti-human CD63/LAMP3 (Chemicon, Schwalbach, Germany), polyclonal rabbit anti-EEA1 (Acris Antibodies, Herford, Germany), monoclonal mouse anti-ubiquitin (Biomol, Hamburg, Germany) and polyclonal rabbit anti-syntaxin 2 (Synaptic Systems, Berlin, Germany). Species specific Alexa Fluor 488 or Alexa Fluor 555 secondary antibodies (Invitrogen) were used at 1:200. Samples were mounted and Alexa Fluor or GFP fluorescence was examined with Axiovert 135 fluorescent microscope and evaluated with AxioVision 4.7.1 software (Carl Zeiss, Jena, Germany).

LDH activity in cell lysates and culture supernatants was determined using the method of Decker and Lohmann-Matthes [30]. Briefly, 100 μl sample was mixed with 30 μl dye solution (18 mg/ml L-lactate, 1 mg/ml iodonitrotetrazolium in PBS). After adding 15 μl of the catalyst (3 mg/ml NAD⁺, 2.3 mg/ml diaphorase, 0.03% BSA, 1.2% sucrose in PBS), absorbance at 492 nm was determined at one minute intervals for 15 minutes at 37°C. Absolute LDH activity was calculated from a standard curve, using purified LDH (Sigma, Munich, Germany). The lower limit of detection was 20 Units/l; the assay was linear to 2500 Units/l.

For lipid analysis, cells grown in Petri dishes were harvested by scraping off in 2 ml PBS supplemented with protease inhibitor (Complete, Roche). The cell suspension was then sonicated (four strokes, 10 seconds; Branson Digital Sonifier S450D). Lipid classes and subspecies were determined by electrospray ionization tandem mass spectrometry (ESI-MS/MS) using direct flow injection analysis, as described previously [31, 32]. Cells were extracted according to the method described by Bligh and Dyer [33] in the presence of non-naturally occurring lipid species used as internal standards. A precursor ion scan of m/z 184 specific for phosphocholine containing lipids was used for phosphatidylcholine (PC), sphingomyelin (SPM) [32] and lysophosphatidylcholine (LPC) [31]. Neutral loss scans of m/z 141 and m/z 185 were used for phosphatidylethanolamine (PE) and phosphatidylserine (PS), respectively [34]. Phosphatidylglycerol (PG) was analyzed using a neutral loss scan of m/z 189 of ammonium adduct ions [35]. Ceramide and glucosylceramide were analyzed as previously described [36] using N-heptadecanoyl-sphingosine as internal standard. Quantification was achieved by calibration lines generated by addition of naturally occurring lipid species to pooled cell homogenate. All lipid classes were quantified with internal standards belonging to the same lipid class, except SM (PC internal standards). Each lipid class was calibrated with a variety of species covering chain lengths and number of double bonds of naturally occurring species. Correction of isotopic overlap of lipid species and data analysis was performed by self-programmed Excel macros for all lipid classes according to the described principles [32].

Human lymphocytes and neutrophils were isolated from whole blood using LeucoSep (Greiner Bio-One, Solingen-Wald, Germany) and Ficoll-Isopaque gradient density isolation method (GE Healthcare) according to the manufacturer's instructions. Cells were incubated for 6 hours (neutrophils) or 24 hours (lymphocytes) at 37°C with supernatants of MLE-12 cells expressing wild-type or mutant proSP-C. Cell-free supernatants were collected after 48 hours of growth and concentrated 7-fold, using Microsep 1 k centrifugal concentrators (Millipore, Schwalbach, Germany). Cells were analyzed by four-colour flow cytometry (FACSCalibur; BD Biosciences, Heidelberg, Germany) as described previously [37]. The following antibodies were used: PE-conjugated mouse anti-human CCR2-B (R&D Systems, Minneapolis, USA), FITC labeled anti-human CD8, FITC labeled anti-human CD4, PE-conjugated mouse anti-human CD11b/Mac-1, and PE-conjugated mouse anti-human CD181 (CXCR1, IL-8RA) (all BD Biosciences Pharmingen). Results are presented as mean fluorescence intensity (MFI) after subtracting background binding provided by non-specific isotypes. Calculations were performed with CellQuest analysis software (BD Biosciences).

The results are given as means ± standard error (SE) of the individual number of different subjects, each individual value representing the mean of 3-4 determinations or as indicated. In the case of lipid analysis, the results are presented as means ± standard deviation. For the comparison of two groups, unpaired t-test or Mann-Whitney test were used where appropriate. Comparisons of multiple groups were made using ANOVA followed by Dunnett's post hoc multiple comparisons test. For correlations, Spearman's non-parametric test was used. P-values of less than 0.05 were considered as statistically significant. All tests were performed using GraphPad Prism 4.0 (GraphPad Software).

To identify the intracellular processing intermediates of proSP-C, MLE-12 cells were transfected with eukaryotic expression vectors, allowing expression of fusion proteins between proSP-C and either an EGFP- or a HA-Tag. Stable expression of HA-tagged proSP-C^WT resulted in the appearance of a strong band at approximately 21 kDa and weaker bands at 22 kDa, 19 kDa, and 14 kDa (Figure 1A left). ProSP-C^A116D yielded bands similar to the wild type at 21 kDa and 14 kDa. We also observed a much stronger band at 22 kDa and a band at 15 kDa that was not seen in the wild type, indicating accumulation of proSP-C^A116D forms (Figure 1A, left). The postulated processing products are depicted in Figure 1B. Mature SP-C was never detectable because of the loss of the protein tag due to the final processing steps at the N-terminus.

Transient expression of N-terminal (C1) and C-terminal (N1) EGFP fusion products was detectable 24 hours post transfection. The primary N- and C-terminal fusion proteins were visible as bands at 48 kDa as expected. In the case of N-terminally tagged protein, a second band was seen at 36 kDa (Figure 1A, middle). There were no differences regarding band pattern between proSP-C^WT and proSP-C^A116D. Likewise, no differences in band pattern between proSP-C^WT and proSP-C^A116D were seen for processing intermediates containing the C-terminal EGFP-tag (Figure 1A right). This suggested that there was no change in the cleavage pattern or kinetics regarding the truncation of proSP-C from the C-terminus, which is supposed to be the first cleavage step [4]. The lowest band corresponded to the EGFP-tag, which has a size of approximately 27 kDa.

The intracellular localization of preprotein species, monitored by immunofluorescence, differed between MLE-12 cells stably expressing proSP-C^WT or proSP-C^A116D. While HA-tagged proSP-C^WT colocalized well with syntaxin 2, proSP-C^A116D did not (Figure 2A). In contrast, EGFP-tagged proSP-C^A116D was partially present in early endosomes detected as EEA1-positive vesicles, while proSP-C^WT was almost absent in those compartments (Figure 2B), thus confirming previous data [38]. Again, with this approach mature SP-C was not detectable because of the loss of the tag due to processing.

In order to determine the stress level of cells expressing SP-C^A116D, lactate dehydrogenase (LDH) release of stably transfected cells was measured. Expression of SP-C^A116D led to a marked overall increase of LDH release, compared to WT cells, suggesting a reduction of cell viability (Figure 3). Exposure of MLE-12 cells expressing SP-C^WT to pharmacological substances currently applied in ILD therapy significantly increased the release of LDH by the cells in the case of azathioprine. While azathioprine treatment resulted in a pronounced increase in LDH release also in cells expressing SP-C^A116D, hydroxychloroquine, methylprednisolone and cyclophosphamide did not significantly alter LDH release by these cells.

We determined the change in protein level of the two heat shock proteins Hsp90 and Hsp70, and the two ER-resident chaperones calreticulin and calnexin, in MLE-12 cells expressing SP-C^WT and SP-C^A116D, after exposure to pharmacological substances used in ILD therapy: cyclophosphamide, azathioprine, methylprednisolone or hydroxychloroquine (Figure 4A-D). While in the case of calnexin no significant alterations were seen, cells expressing SP-C^A116D showed an increase in the expression of calreticulin, Hsp70 and Hsp90 at baseline and after treatment with pharmaceuticals. However, none of the applied drugs resulted in a significant change in expression level of any chaperone, neither in cells expressing SP-C^WT nor in cells expressing SP-C^A116D. However, we noted a slight increase in the expression levels of Hsp70 and Hsp90 in SP-C^A116D cells treated with azathioprine (Figure 4C+D).

Mass spectrometric lipid analysis showed that total phospholipid amount was reduced in cell lysates from MLE-12 cells transfected with SP-C^A116D, compared to cells transfected with SP-C^WT (Table 1). Moreover, the phospholipid composition was significantly altered: the percentage of phosphatidylcholine (PC) and ceramide (Cer) were significantly decreased while the percentages of lyso-phosphatidylcholine (LPC), phosphatidylserine (PS), phosphatidylethanolamine (PE) and sphingomyelin (SPM) were significantly increased in cells expressing SP-C^A116D. Treatment with methylprednisolone or hydroxychloroquine did not correct the loss of PC in SP-C^A116D expressing cells, although a significant increase of PC was noted in the cells treated with hydroxychloroquine (Figure 5A, left chart). Treatment with methylprednisolone as well as with hydroxychloroquine led to a significant reduction of increased LPC levels in cells expressing SP-C^A116D (Figure 5A, right chart). The effect was more pronounced in the case of hydroxychloroquine.

The phospholipid secretion by MLE-12 cells was assessed in the supernatant (Table 1). Similar to the intracellular lipid pattern, PC and Cer were significantly decreased in the supernatant of SP-C^A116D cells, while the percentages of LPC and SPM were increased. In contrast to the findings in cell lysate, PE was not altered in the supernatant of SP-C^A116D cells. Remarkably, PS was significantly decreased in the supernatant while being increased in cell lysate of SP-C^A116D cells. PC levels in the supernatant of cells treated with methylprednisolone were elevated compared to untreated cells by tendency, in WT as well as in A116D cells. A significant increase in LPC secretion was noted in cells expressing SP-C^A116D (Table 1; Figure 5B, right chart). Treatment with hydroxychloroquine, but not with methylprednisolone, restored LPC level in the supernatant of mutant SP-C expressing cells to the level observed in WT cells. Interestingly, hydroxychloroquine also reduced LPC in the supernatant of WT cells, albeit not significantly. Taken together, hydroxychloroquine treatment seems to be able to counteract the alterations of PC and LPC levels seen in cells expressing SP-C^A116D.

We hypothesized that accumulation of misfolded SP-C might somehow lead to an enhanced accumulation of leukocytes and thus boost the inflammatory reaction. To test this hypothesis, we examined whether cells expressing SP-C^A116D stimulated the expression of CCR2 on lymphocytes and CXCR1 on neutrophils by incubating isolated neutrophils or lymphocytes with 7-fold concentrated supernatants of MLE-12 cells expressing either WT or A116D SP-C. While no difference in surface receptor expression between WT and mutant was observed in CD8+ lymphocytes, CD4+ lymphocytes showed a highly significant increase in the level of surface receptor CCR2 expression in response to the supernatant of SP-C^A116D expressing cells (Figure 6A). The same was true in the case of CXCR1, which was increased on CD4+ lymphocytes after incubation with the mutant cell supernatant, but remained unaltered on CD8+ lymphocytes (Figure 6B). We further analyzed the surface receptor expression on neutrophils. The supernatant of cells expressing SP-C^A116D increased CXCR1 expression on neutrophils, but did not affect CD11b levels (Figure 6C). Non-concentrated supernatants gave the same results by tendency, although less pronounced. A clear concentration dependency of the effects was observed (data not shown). This suggests that SP-C^A116D-expressing MLE-12 cells were able to modulate the surface receptor expression on the cells of immune system through the secretion of soluble factors into the medium.