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

This Article Abstract Full Text (PDF) All Versions of this Article: 292/6/L1526    most recent 00463.2006v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Sehgal, P. B. Articles by Shah, M. Search for Related Content PubMed PubMed Citation Articles by Sehgal, P. B. Articles by Shah, M.

Dysfunction of Golgi tethers, SNAREs, and SNAPs in monocrotaline-induced pulmonary hypertension

Departments of ¹Cell Biology and Anatomy and ²Medicine, New York Medical College, Valhalla, New York

Submitted 27 November 2006 ; accepted in final form 27 February 2007

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Monocrotaline (MCT)-induced pulmonary hypertension (PH) in the rat is a widely used experimental model. We have previously shown that MCT pyrrole (MCTP) produces loss of caveolin-1 (cav-1) and endothelial nitric oxide synthase from plasma membrane raft microdomains in pulmonary arterial endothelial cells (PAEC) with the trapping of these proteins in the Golgi organelle (the Golgi blockade hypothesis). In the present study, we investigated the mechanisms underlying this intracellular trafficking block in experiments in cell culture and in the MCT-treated rat. In cell culture, PAEC showed trapping of cav-1 in Golgi membranes as early as 6 h after exposure to MCTP. Phenotypic megalocytosis and a reduction in anterograde trafficking (assayed in terms of the secretion of horseradish peroxidase derived from exogenously transfected expression constructs) were evident within 12 h after MCTP. Cell fractionation and immunofluorescence techniques revealed the marked accumulation of diverse Golgi tethers, soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein receptors (SNAREs), and soluble NSF attachment proteins (SNAPs), which mediate membrane fusion during vesicular trafficking (GM130, p115, giantin, golgin 84, clathrin heavy chain, syntaxin-4, -6, Vti1a, Vti1b, GS15, GS27, GS28, SNAP23, and -SNAP) in the enlarged/circumnuclear Golgi in MCTP-treated PAEC and A549 lung epithelial cells. Moreover, NSF, an ATPase required for the "disassembly" of SNARE complexes subsequent to membrane fusion, was increasingly sequestered in non-Golgi membranes. Immunofluorescence studies of lung tissue from MCT-treated rats confirmed enlargement of perinuclear Golgi elements in lung arterial endothelial and parenchymal cells as early as 4 days after MCT. Thus MCT-induced PH represents a disease state characterized by dysfunction of Golgi tethers, SNAREs, and SNAPs and of intracellular vesicular trafficking.

soluble N-ethylmaleimide-sensitive factor attachment protein receptors; endothelium; intracellular membrane trafficking; Golgi blockade

We previously suggested that the initiating event in MCT-induced PH in vivo was a disruption of caveolin-1 (cav-1)/raft function at the level of the PAEC leading to hyperactivation of promitogenic STAT3 and ERK1/2 signaling (29). The fact that the loss of cav-1 was evident within 48 h of MCT administration, namely before the development of clinical PH, suggested that this loss of cav-1 from the plasma membrane was part of the initiating mechanism of this disease. This loss of cav-1 (and cav-2) in PAEC in lungs of MCT-treated rats and the hyperactivation of STAT3 has now been confirmed by Jasmin et al. (18). Mechanistically, we have related this loss of plasma membrane cav-1 to a trafficking block through the Golgi in both lung endothelial and epithelial cells (the Golgi blockade hypothesis) (37, 38, 55). This Golgi blockade in megalocytosis affected not only cav-1 but also affected the subcellular localization and function of platelet/endothelial cell adhesion molecule-1 (PECAM-1), endothelial nitric oxide synthase (eNOS), and E-cadherin (29, 36, 37, 38, 55). Thus the key defect in MCTP-induced PH at the subcellular level appeared to be an inhibition of membrane-associated trafficking to/through the Golgi apparatus, which likely led to the mislocalization of diverse vasorelevant proteins away from the cell surface (29, 36, 37, 38, 55).

In the present study, we have investigated the mechanisms underlying this block in anterograde trafficking to/through the Golgi in MCTP-treated lung cells in cell culture and in vivo in the rat. The data obtained show that the Golgi blockade in lung cells exposed to MCTP is characterized by a marked dysfunction of Golgi tethers, soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein receptors (SNAREs), and soluble NSF attachment proteins (SNAPs), which mediate membrane fusion events during intracellular vesicular trafficking.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Cell culture, monocrotaline treatment, and cell fractionation. Growth of bovine PAEC and the human pulmonary type II-like alveolar epithelial cell line A549 in T-75 flasks, 10-cm Petri dishes, or six-well plates was done as described by us previously (29, 36, 37, 38, 55). PAEC were used between passages 4 and 20.

The active pyrrolic derivative of MCT was prepared by the method of Mattocks et al. (31) as described previously. Briefly, MCT (Transworld Chemicals, Rockville, MD) in chloroform (0.4% wt/vol) was mixed with o-chloranil (0.5% in chloroform wt/vol) (Sigma-Aldrich, St. Louis, MO) followed by the addition of 70% potassium hydroxide (9% vol/vol) and separation of the organic phase from the aqueous (31). The organic phase was dried with sodium sulfate and concentrated under nitrogen (31). Mass spectrometric analyses showed that 25–30% of the input MCT was converted to the pyrrolic derivative by this method (data not shown). The concentration of MCTP selected for use in the present experiments was the equivalent of 200 µM of the starting MCT; thus effectively it was 50 µM equivalent of the active MCTP in accordance with our mass spectrometric analyses. (This corresponds to 17 µg/ml when expressed in the manner of Ref. 67 in which those authors used a range between 5 and 35 µg/ml MCTP and far less than the 105 µM purified MCTP used in Ref. 22.) In preliminary dose-response analyses using sequential phase-contrast microscopy, we (29, 36, 55, unpublished data) observed maximal megalocytosis with extended cell survival of PAEC and stimulation of DNA synthesis in the range 100–200 µM equivalent of MCT (with 25% fractional conversion to MCTP); thus in the present cell culture experiments aimed at elucidating the underlying biochemistry, we selected the highest tolerated concentration (200 µM equivalent of MCT corresponding approximately to 50 µM or 17 µg/ml of the active pyrrole). This MCTP in dimethyl formamide (DMF) was added to almost confluent cultures of PAEC or A549 cells in 10-cm Petri dishes or six-well plates, and megalocytosis was allowed to develop. Control cultures received equivalent volume of DMF alone (0.4% vol/vol). Megalocytosis began to develop within 12–18 h and was fully evident by 24–48 h. Most experiments were performed 4 days after addition of MCTP. In cell fractionation experiments, to avoid detecting proteins that were merely transiting through the Golgi at the time of harvesting the cells, and to thus isolate only the Golgi resident proteins, cultures were pretreated with cycloheximide (10 µg/ml) for 2 h (48, 53). Cycloheximide inhibits protein synthesis and subsequent entry of any new proteins into the ER-Golgi and thus flushes away secretable proteins from the Golgi (48, 53).

Purification of Golgi membranes was done by using either one (Fig. 2) or two (Figs. 3 and 4) sequential sucrose flotation gradients as per the methods of Xu and Shields (69) and Fries and Rothman (12), respectively. PAEC or A549 cells grown in 10-cm plates and treated with MCTP in DMF or with DMF alone were harvested directly in an isotonic buffer (0.25 M sucrose-containing) that had 10 mM Tris·Cl, pH 7.4, and 1 mM MgCl₂ (69). Cells were broken with a Tekmar Tissumiser (15 s). A postnuclear supernatant (PNS) was obtained by low-speed centrifugation (2,500 rpm for 15 min). For assays based on protein matching, total protein was estimated in the PNS using the Bradford reagent (Bio-Rad). For assays based on cell counts, cells were counted upon harvest of control and MCTP-treated cultures using a hemocytometer. Protein- or cell-matched volumes of the PNS derived from control or MCTP-treated cultures were floated through the first sucrose gradient as per Xu and Shields (69). This was a step gradient with 6 ml of cell lysate adjusted to 1.4 M sucrose at the bottom, overlaid with 4 ml of 1.2 M sucrose and 1.5 ml of 0.8 M sucrose centrifuged in a Beckman SW41Ti rotor at 35,000 rpm for 4.5 h. The band between 1.2 M and 0.8 M sucrose represented the Golgi membranes, whereas the ER fractionated at the bottom of the gradient (38, 69). The Golgi band between 1.2 M and 0.8 M sucrose was diluted 10-fold in 0.25 M sucrose buffer and resedimented by centrifugation, and the pellet was resuspended in 100 µl of 0.25 M sucrose buffer. Before Western assays in these experiments (Fig. 2), samples from the purified Golgi membranes were matched for total protein using the Bradford reagent to establish changes in specific vesicle tethers, SNAREs, and SNAPs per unit of Golgi protein. For further separation of the cis-, medial-, and trans-Golgi cisternae, the Golgi band from the 0.8/1.2 M sucrose interface of the Xu and Shields (69) gradient was collected, adjusted to 1.28 M sucrose, and floated up a second, shallower gradient per Fries and Rothman (12). This second gradient was a step gradient with the sample adjusted to 1.28 M sucrose at the bottom overlaid with 0.66 ml each of 1.13 M, 1.05 M, 0.96 M, 0.85 M, and 0.5 M sucrose, respectively, and centrifuged in SW50.1Ti rotor at 35,000 rpm for 18 h. The cis-Golgi elements fractionate in the heavier part of such a gradient, whereas the trans-Golgi elements fractionate in the lighter part of the gradient (12). For analyses, fractions (500 µl) were collected from the top of the second gradient and assayed directly by Western blot analyses (aliquots of 25–30 µl) or were concentrated by precipitation with 10% TCA (usually 300 µl/fraction) before Western blotting. Western blotting was carried out as described earlier (29, 36, 37, 38, 54, 55). Image quantification was performed using NIH Image J as described previously (36–38, 54, 55).

View larger version (35K): [in this window] [in a new window]   Fig. 2. Time-course analyses of proteins in Golgi membranes after MCTP. Golgi membranes were prepared from PAEC or A549 cells in 100-mm plates at different times after MCTP using the 1-step gradient method of Xu and Shields (69). To obtain an approximate match in protein amounts in the postnuclear supernatant, we typically used 3, 5, and 8 100-mm cultures for the control 1 day after MCTP and 2 days after MCTP groups, respectively. Cell numbers in each harvested group were evaluated before homogenization using a hemocytometer chamber; total protein was estimated in the postnuclear supernatant. Approximately protein-matched amounts of the latter were used to prepare Golgi membranes by sucrose gradient flotation. The Golgi band between 1.2 M and 0.8 M sucrose was diluted 10-fold in 0.25 M sucrose buffer and resedimented by centrifugation, and the pellet was resuspended in 100 µl of 0.25 M sucrose buffer. Total protein was then estimated in the washed Golgi membrane pellet using the Bradford reagent (Bio-Rad). A: Western blotting for caveolin-1 (cav-1) in protein-matched samples of the washed Golgi membrane pellets derived from PAEC. B: Western blotting for cav-1 and various Golgi-resident proteins in respective cell number-matched (left) or protein-matched (right) samples of the washed Golgi membrane pellets derived from PAEC and A549 cells. Double arrowhead points to slower mobility golgin 84 bands observed in A549 cells 2 days after MCTP (also see Fig. 3). Two different antibodies for BMPR2 (a rabbit polyclonal, pAb, and a mouse monoclonal, mAb) were used. The pAb detected 120- and 75-kDa bands in bovine PAEC, whereas the mAb detected the 25-kDa band in human A549 cells (see Ref. 47 for a discussion of antibody- and cell type-specific differences in the performance of different anti-BMPR2 antibodies).

View larger version (36K): [in this window] [in a new window]   Fig. 3. Marked accumulation of cav-1, endothelial nitric oxide synthase (eNOS), vesicle tethers, SNAREs, and SNAPs in purified Golgi from MCTP-treated endothelial cells. Golgi membranes were purified from cultures of PAEC grown in 100-mm plates as described in MATERIALS AND METHODS [2 sequential flotation gradients (12, 69)] using protein-matched aliquots derived from cultures treated 4 days earlier with MCTP and then chased for 2 h with cycloheximide (10 µg/ml) to flush out any secretable proteins (48, 53). A and B illustrate a composite of several Western blot analyses of the respective gradient fractions probed for various markers.

View larger version (57K): [in this window] [in a new window]   Fig. 4. Marked accumulation of cav-1, clathrin heavy chain (CHC), vesicle tethers, SNAREs, and SNAPs in purified Golgi from MCTP-treated lung epithelial cells. Golgi membranes were purified from cultures of A549 cells grown in 100-mm plates as described in MATERIALS AND METHODS [2 sequential flotation gradients (12, 69)] using protein-matched aliquots derived from cultures treated 4 days earlier with MCTP and then chased for 2 h with cycloheximide (10 µg/ml) to flush out any secretable proteins (48, 53). A and B illustrate a composite of several Western blot analyses of the respective gradient fractions probed for various markers. C: whole mount negative-stain imaging of Golgi cisternae in the peak GM130-positive fractions in A.

Immunofluorescence studies. Immunofluorescence analyses were performed using PAEC and A549 cells cultured in six-well plates, as described previously (29, 36–38, 54, 55). Almost confluent cultures of PAEC and A549 cells were treated with MCTP in DMF or DMF alone, and megalocytosis was allowed to develop over the next 4 days. Cells were then fixed using 4% cold paraformaldehyde and permeabilized with 0.1% Triton X-100. Fixed cultures were stained using a combination of rabbit polyclonal and murine monoclonal antibodies and respective Alexa Fluor 488- or Alexa Fluor 594-tagged secondary antibodies (Molecular Probes, Eugene, OR). Nuclei were demarcated using 4',6-diamidino-2-phenylindole. Images were collected using a Leitz epifluorescence microscope system equipped with a black and white charge-coupled device (CCD) camera or alternatively using epifluorescence microscopy and an RGB camera system (especially for lung sections). Additionally, immunofluorescence data were also collected using the Bio-Rad MRC 1024 ES confocal imaging system with a black and white CCD camera. All data within each experiment were collected at identical imaging settings for the control and experimental groups, and images were subjected to iterative deconvolution using the NIH Image J software (36, 38, 54).

Horseradish peroxidase secretion studies in transient transfection assays. Gene transfer assays for horseradish peroxidase (HRP) secretion were carried out essentially as described by Connolly et al. (8). Briefly, transfections of the constitutive expression vectors for secreted HRP species (pSR.ssHRP and pSR.ssHRP^KDEL of Ref. 8) (25 µg/well) together with the constitutive -galactosidase expression vector pCH110 (1 µg/well) were carried out into PAEC or A549 cell cultures in six-well plates, in triplicate wells for each experimental variable, using the Lipofectamine reagent (Polyfect; Qiagen, Valencia, CA) and the manufacturer's protocol (36). Two different protocols were used: 1) cells were first transfected and then treated with MCTP 12 h later, or 2) cells were first treated with MCTP and then transfected 1 day later. Culture medium was harvested at the times indicated in the respective experiments and assayed for HRP (8). Cell extracts were harvested for determination of -galactosidase activity. HRP activity in culture medium was normalized to the peroxidase activity in the same medium samples before MCTP treatment (Fig. 1C) or to the respective -galactosidase activities (Fig. 1D). The respective normalized data were evaluated using the Student's two-tailed t-test.

View larger version (53K): [in this window] [in a new window]   Fig. 1. Early (by 12 h) and late (48–96 h) phenotypic effects of monocrotaline pyrrole (MCTP). A: rapid development of megalocytosis in cultures of pulmonary arterial endothelial cells (PAEC). Freshly confluent cultures of PAEC in 6-well plates were treated with MCTP, and the phenotype was assessed by phase-contrast microscopy at different times. Scale bar, 50 µm. B: at 96 h after MCTP, there was marked enlargement of the Golgi in magalocytotic PAEC and A549 cells displayed using an immunofluorescence assay for the cis-Golgi marker GM130. Nuclei are demarcated using DAPI. Scale bar, 50 µm. C: reduction of gene-transferred HRP secretion within 12 h after MCTP. Cells (A549 in the experiment illustrated) in replicate 6-well cultures were first transfected with either of 2 HRP vectors in triplicate per variable. Twelve hours after transfection, the culture medium was collected, MCTP was added to one-half of the wells, and the cultures were incubated for another 12 h. HRP activity in the culture medium in each respective well after MCTP was normalized to the HRP activity in the medium of the same well before MCTP (thus each culture served as its own transfection control). Data are expressed in terms of this normalized ratio (means ± SE); *P < 0.05. D: marked reduction of gene-transferred HRP secretion by 48 h after MCTP. Cells (PAEC in a and A549 in b and c) in 6-well plates were treated with MCTP and then 24 h later transfected with either of 2 HRP vectors together with the -galactosidase expression vector pCH110. The culture medium was collected 16 or 48 h after transfection and for HRP activity. Data (means ± SE) are expressed in arbitrary peroxidase activity units normalized to -galactosidase activity in respective cell extracts. E: intracellular accumulation of gene-transferred HRP in MCTP-treated PAECs. PAECs in 6-well plates were treated with MCTP or dimethyl formamide (DMF). Twenty-four hours after MCTP treatment, cultures were transfected with either of the 2 HRP vectors or mock transfected. Forty-eight hours after transfection, cultures were fixed, and immunofluorescence analysis was performed using an anti-HRP MAb. Mean pixel intensity per cell of HRP staining of cultures was quantitated using NIH Image J software (expressed as means ± SE; n = 10 cells/group). Mean fluorescence intensity of mock-transfected cells was subtracted from that of cultures transfected with the 2 HRP vectors to control for background fluorescence. Mean fluorescence intensity of MCTP-treated PAECs transfected with pSR.ssHRP was 72.4% greater than that of control cells transfected with the same vector (P < 0.001), whereas mean fluorescence intensity of MCTP-treated PAECs transfected with the pSR.ssHRPKDEL vector was 117% greater than control cultures transfected with the same vector (P < 0.05). *P < 0.05.

Immunofluorescence studies of the Golgi in rat lung after MCT. These were carried out as described previously (29), and all protocols were approved by the Institutional Animal Care and Use Committee. Briefly, juvenile (4- to 6-wk-old) male Sprague-Dawley rats were subcutaneously administered MCT (60 mg/kg), and the lungs were harvested from control vehicle-treated rats as well as from those receiving MCT at 2 days, 4 days, and 4 wk later [PH is evident in such animals by 2 wk (29)]. The formaldehyde-fixed lungs were embedded in paraffin, and sections (4-µm-thick) were processed for immunofluorescence microscopy. In each of the different experiments (for example, Fig. 11, A and B, derives from different experiments carried out several months apart), lung tissue from control and experimental animals was obtained and fixed side by side, all embedding, sectioning, and processing for immunofluorescence was carried out side by side, and imaging data were collected under identical microscopy and camera settings. Thus evaluation of lung tissue from the experimental group was only in terms of the simultaneously processed controls.

View larger version (57K): [in this window] [in a new window]   Fig. 11. High magnification panels showing changes in the Golgi in the rat lung at 4 wk and 4 days (before the development of pulmonary hypertension and right ventricular hypertrophy) after MCT administration. Immunofluorescence studies showing alterations in Golgi markers in the pulmonary arterial endothelium and alveolar epithelium of rats administered MCT 4 wk (A) or 4 days (B) earlier. A and B illustrate 2 different experiments. At 4 wk after MCT, there was marked right ventricular hypertrophy in the subject animal illustrated. Sections of arterial endothelium are oriented with the lumen towards the top and are derived from arterial vessels of diameter between 50 and 100 µm. Scale bar, 20 µm.

Antibody reagents. Monoclonal antibodies to GM130, clathrin heavy chain, syntaxin-4, syntaxin-6, GS15, GS27, GS28, Vti1a, Vti1b, -SNAP, golgin 84, eNOS, LAMP1, and BMPR2 were purchased from BD Biosciences (Transduction Laboratory, San Diego, CA), whereas that to the macrophage marker ED-1 was from Serotec (Oxford, UK). Rabbit antibodies to cav-1 (sc-894) and eNOS and goat antibodies to N-ethylmaleimide sensitive factor (NSF) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit antibody to BMPR2 was from Abgent (San Diego, CA), that to giantin was from Abcam (Cambridge, MA), and that to SNAP23 was from Synaptic Systems (Goettingen, Germany). Rabbit antibody to p115 was a gift from Dr. Dennis Shields (Albert Einstein College of Medicine, New York, NY).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Earliest phenotypic changes produced by MCTP. MCTP-induced megalocytosis of PAEC, PASMC, and alveolar epithelial cells is typically depicted as developing by 48–96 h after exposure to the alkaloid (29, 36–38, 55). We were interested in determining the earliest time point after addition of MCTP at which we could observe 1) a phenotypic change at the cellular level and 2) a functional change in protein trafficking.

Figure 1A shows that by phase-contrast microscopy, phenotypic megalocytosis was evident in PAEC in culture within 9–12 h after MCTP exposure. Eventually, by 96 h, there was marked megalocytosis and an enlargement of the Golgi in both PAEC and A549 cells (Fig. 1B).

To determine whether there was a functional change in trafficking through the Golgi, we assayed anterograde secretion of HRP derived from exogenously transfected expression constructs. This HRP gene transfer method using respective soluble secreted HRP (ssHRP) cDNA expression constructs has been extensively characterized by Connolly and colleagues (8) and used extensively by cell biologists as a quantitative assay for changes in protein secretion. These cDNA constructs have a secretory signal sequence (of human growth hormone) added to the amino-terminal end of an artificially constructed HRP plasmid encoding HRP isoenzyme c (on the basis of its published amino acid sequence) to allow for entry into the ER and thus into the anterograde exocytic pathway via the Golgi. Thus the transcription and translation of this cDNA after transient transfection results in the generation of enzymatically active HRP, which is initially localized within the ER-Golgi. This HRP is then efficiently secreted from transfected cells into the medium where it remains enzymatically active. A block in trafficking through the Golgi would therefore be expected to reduce the HRP activity in the medium due to inefficient secretion/exocytosis as expressed after normalization with respect to the cellular synthesis of HRP.

Figure 1C shows that within 12 h of MCTP administration (i.e., at the same time when phenotypic megalocytosis becomes evident), there was a 40–50% reduction in anterograde trafficking as assayed by HRP secretion into the culture medium. In this series of experiments, MCTP was added 12 h posttransfection. The HRP activity was assayed by collecting medium both before and after MCTP treatment from the same culture. The enzymatic activity observed 12 h after MCTP treatment was normalized to that observed before addition of MCTP, but after transfection. Thus each culture served as its own "transfection" control. Under our experimental conditions, the ssHRP vector expressed with the KDEL tag was also secreted into the medium. Although KDEL is customarily viewed as an ER retention signal, secretion of this ssHRP^KDEL into the culture medium has been reported previously in HEp2 cells (8). Similarly, the present results showing secretion of ssHRP^KDEL may be a cell type-dependent phenotype indicative of a weak efficacy of KDEL-mediated retention of ssHRP in the PAEC and A549 cells used in the present experiments. In any case, the data in Fig. 1C are consistent with a functional reduction in protein secretion within 12 h of MCTP treatment.

To exclude the effect of MCTP on protein translation per se, we performed additional cotransfection experiments using the HRP vectors in combination with the -galactosidase expression vector pCH110. (-galactosidase is a cytosolic protein, and a block in trafficking through the Golgi would not be expected to change its activity or levels.) To control for the potential inhibition of cellular protein synthesis per se after MCTP, in these experiments we normalized the HRP activity in culture medium from control and MCTP-treated cultures in terms of the -galactosidase activity observed in the cell extracts of the respective cultures. This represents a control for both transfection efficiency and for translation. Figure 1D shows that when culture medium from PAEC or A549 cells was collected 16 or 48 h after transfection with MCTP added 24 h before transfection, there was a dramatic and sustained decrease in the HRP secretion in both cell types.

Is there increased intracellular accumulation of HRP in MCTP-treated cells pursuant to this secretory block? Figure 1E summarizes quantitative immunofluorescence assays of cell-associated HRP that demonstrate the increased intracellular retention of HRP in cells transfected with the respective vectors 1 day after MCTP exposure under experimental conditions corresponding to the secretion data in Fig. 1D. [Mean pixel intensity of MCTP-treated PAECs transfected with pSR.ssHRP was 72.4% greater than that of control cells transfected with the same vector (P < 0.001), and mean pixel intensity of MCTP-treated PAECs transfected with the pSR.ssHRP^KDEL vector was 117% greater than that of its controls (P < 0.05). The mean pixel intensity of mock-transfected cells was subtracted from that of the HRP-transfected cells to control for background fluorescence; n = 10 cells/group.] Together, the data in Fig. 1, C–E, provide convincing evidence for inhibition of anterograde secretion of HRP after exposure to MCTP.

To confirm the observations above derived from transient transfection and immunofluorescence experiments using a different approach, we performed cell fractionation experiments at different time points after addition of MCTP to PAECs and prepared Golgi membranes using the method of Xu and Shields (69). The purified Golgi fraction was washed and resedimented before Western blotting. Figure 2A shows that within 6 h after MCTP treatment, there was an increase in the amount of cav-1 associated with the Golgi membranes in Western blots carried out after matching for total protein in the respective Golgi fractions. Cav-1 associated with purified Golgi cisternae increased progressively over the next several hours (12 h in Fig. 2A) and days (1 and 2 days in Fig. 2B) after MCTP treatment. The fact that the amount of cav-1 associated with the purified Golgi cisternae increased in MCTP-treated samples after normalizing for total protein in the Golgi fraction argues against an increase in cav-1 in MCTP-treated Golgi membranes merely due to an increase in the size of the Golgi apparatus per se on a per cell basis.

Thus, a functional change in protein trafficking through the Golgi apparatus is observed within 6–12 h after exposure to MCTP concomitant with phenotypic enlargement of the affected cells. These are the earliest that effects of MCTP on cells have been reported; these temporal parameters suggest that inhibition of trafficking through the Golgi is likely an initiating event in the development of phenotypic megalocytosis.

Trapping of vesicle tethers, SNAREs, and SNAPs in purified Golgi cisternae. The above data were suggestive of a functional defect in one or more mediators of vesicular trafficking through the Golgi. We therefore investigated changes in specific Golgi tethers (GM130, p115, giantin, golgin 84, clathrin heavy chain), SNAREs (syntaxin-4, syntaxin-6, Vti1a, GS28, and SNAP23), and SNAPs (-SNAP) that regulate trafficking to and through the Golgi apparatus.

Golgi cisternae were purified from untreated PAEC and A549 cells or those treated with MCTP using the purification technique of Xu and Shields (69). Western blot analyses using respective Golgi membrane aliquots were carried out after matching either for cell number (i.e., matching numbers of cells from which the fractions were derived) or for protein (i.e., per unit of protein in the Golgi membranes). Figure 2B, left, shows that in a per cell basis comparison, increased amounts of Golgi tethers and scaffolding proteins GM130, p115, giantin, and golgin 84, and the SNAREs Vti1a and GS28 were associated with Golgi cisternae within 2 days after MCTP. The increased association of the tethers GM130, giantin, and p115 and the SNARE GS28 with Golgi cisternae was also evident when compared after normalizing for total protein in the purified Golgi band (Fig. 2B, right). Moreover, in both PAEC and A549 cells, a fraction of the golgin 84 was shifted to a slower mobility band 2 days after MCTP, consistent with increased phosphorylation. Remarkably, when compared on a per cell basis, there was an increase in BMPR2 receptor [120- and 75-kDa bands (47)] in Golgi membranes from PAEC 1–2 days after MCTP and an increase in the 25-kDa BMPR2 fragment (47) in the Golgi membranes from A549 cells, underlining the significance of this altered trafficking through the Golgi in MCTP-induced megalocytosis.

To further analyze the trapping of Golgi tethers, SNAREs, and SNAPs in MCTP-induced megalocytosis, we used two sequential gradients to separate the subpopulations of Golgi cisternae using the method of Fries and Rothman (12). These gradients were run such that the PNS from the control and MCTP-treated cultures were matched for total protein before running the first gradient. Figure 3 summarizes the results obtained from PAECs (Fig. 3, A and B, are independent experiments). Figure 3, A and B, shows that in PAECs, 4 days after MCTP treatment (when megalocytosis was very clear), there were large increases in the amounts of the Golgi tethers GM130, p115, giantin, and golgin 84 associated with purified Golgi membranes, and these appeared to largely cofractionate with each other even though GM130 is nominally considered to be a cis-Golgi marker and syntaxin-6 is nominally considered to be a trans-Golgi marker. Moreover, as expected, there were large increases in the amounts of eNOS and cav-1 associated with the Golgi membranes 4 days after MCTP in the PAECs, confirming our earlier findings. Figure 3B shows that the levels of the Golgi SNAREs syntaxin-6, GS28, and Vti1a also increased in the MCTP-treated purified Golgi membranes of PAEC. Interestingly, there was an increased association of syntaxin-4 (considered to be a plasma membrane SNARE protein) in the Golgi after MCTP as evidenced in Fig. 3B. Loss of syntaxin-4 from the plasma membrane and its increased association in Golgi membranes would be expected to decrease effectiveness of Golgi-plasma membrane anterograde transport. Figure 3B also shows clathrin, an important mediator of export from the Golgi (46), is also trapped in MCTP-treated PAEC Golgi membranes.

Figure 4 demonstrates that the trapping of Golgi tethers, SNAREs, and SNAPs is not cell type specific but also occurs in the A549 epithelial cells. Figure 4A shows the large increase in association of the tethers GM130, p115, giantin, and golgin 84 with purified Golgi membranes after MCTP treatment. Additionally, in A549 cells, GM130-positive Golgi membranes shifted to a lower density. Thus GM130, which is usually present in the heavier cis-Golgi cisternae, was shifted to a lighter trans-Golgi density consistent with the retention of lighter cargo in the cis-Golgi. Additionally, as in the data in Fig. 2B, golgin 84 was shifted to a slower mobility band after MCTP (Fig. 4A), indicative of increased phosphorylation. Figure 4B extends the observations regarding trapping of the SNAREs syntaxin-6, GS28, and Vti1a, the plasma membrane SNAREs syntaxin-4 and SNAP23, and additionally the SNAP protein -SNAP in Golgi cisternae to lung epithelial A549 cells treated with MCTP. Whole mount negative-stain electron microscopy of purified GM130-enriched Golgi membranes confirmed a four- to fivefold increase in the size of Golgi cisternae after MCTP (Fig. 4C). Thus, in MCTP-induced megalocytosis, there was a functional defect in the Golgi with trapping of Golgi tethers, SNAREs, and SNAPs and a decrease in trafficking through and to the Golgi apparatus. A block in the recycling of the SNARE proteins may explain why diverse tethers, SNAREs, and SNAPs investigated showed marked increases in the Golgi fractions.

Immunofluorescence studies: marked accumulation of vesicle tethers, SNAREs, and SNAPs in the Golgi in megalocytotic cells in culture. We used a second approach to confirm the cell fractionation findings reported in Figs. 2–4 by carrying out immunofluorescence analyses of PAEC and A549 cells fixed 4 days after MCTP treatment using the cold paraformaldehyde-Triton fixation protocol. Figures 5 and 6 summarize these immunofluorescence data.

View larger version (109K): [in this window] [in a new window]   Fig. 5. The trapping of vesicle tethers, SNAREs, and SNAPs in the enlarged/circumnuclear Golgi in MCTP-treated (4 day) megalocytotic PAEC. Immunofluorescence studies were carried out in PAEC grown in 6-well plates with and without MCTP treatment (4 day) using antibody probes for vesicle tethers (GM130, p115, giantin, golgin 84), SNAREs (syntaxin-6, GS28, Vti1a), and SNAP23. The additional antibody probes enumerated in Fig. 6 were also investigated in PAEC but provided only weak/negligible immunofluorescence (not shown). Scale bar, 50 µm.

View larger version (93K): [in this window] [in a new window]   Fig. 6. The trapping of vesicle tethers, SNAREs, and SNAPs in the enlarged/circumnuclear Golgi in MCTP-treated (4 day) megalocytotic lung alveolar epithelial cells. Immunofluorescence studies were carried out in A549 cells grown in 6-well plates with and without MCTP treatment (4 day) using antibody probes for vesicle tethers (GM130, p115, giantin, golgin 84), SNAREs (syntaxin-4, syntaxin-6, GS15, GS27, GS28, Vti1a, Vti1b), and SNAPs (-SNAP and SNAP23). Scale bar, 50 µm.

Figure 6 extends the immunofluorescence findings to A549 cells. The Golgi tethers GM130, p115, giantin, and golgin84, the SNAREs syntaxin-6, syntaxin-4, GS15, GS27, GS28, Vti1a, and Vti1b, and SNAP23 and the SNAP protein -SNAP all assume a circumnuclear localization after MCTP treatment. Specifically, there is a loss of plasma membrane/cell surface syntaxin-4 and SNAP23 after MCTP. The high degree of colocalization between p115 and the other markers used in these immunofluorescence studies is consistent with their cofractionation in Fig. 4.

That MCTP affected diverse Golgi tethers, SNAREs, and SNAPs in a similar manner (Figs. 2–6) was suggestive of the disruption of a common underlying process in vesicular trafficking through this organelle. The disassembly of the quaternary SNARE complexes subsequent to completion of the membrane fusion step into its constituent components requires the soluble cytosolic Cys-rich ATPase dubbed NSF. NSF is a common underlying requirement in vesicular trafficking irrespective of the particular tethers, SNAREs, or SNAPs involved (5, 30, 58). We therefore investigated whether NSF was affected in MCTP-induced megalocytosis. Figure 7 confirms that in untreated PAEC "soluble" NSF was present as a diffusely cytoplasmic protein. However, following MCTP exposure, cytoplasmic NSF was clearly sequestered away from the Golgi compartment. A similar sequestration was observed in A549 cells (data not shown). These data suggest that the disassembly of SNARE complexes may be the specific step inhibited after MCTP, thus providing a basis for understanding the accumulation of diverse tethers, SNAREs, and SNAPs in the Golgi after exposure to MCTP.

View larger version (89K): [in this window] [in a new window]   Fig. 7. N-ethylmaleimide sensitive factor (NSF) is sequestered away from the enlarged/circumnuclear Golgi in MCTP-induced megalocytosis in PAECs. PAEC cultures in 6-well plates were treated with MCTP or DMF (controls), and megalocytosis was allowed to develop over the next 4 days. Cultures were then fixed and stained for NSF using a goat pAb and the cis-Golgi marker GM130 using a MAb. Scale bar, 50 µm.

View larger version (25K): [in this window] [in a new window]   Fig. 8. MCTP does not affect enzymatic activities of Golgi-resident glycosidases. Golgi membranes were prepared from control and MCTP-treated PAECs grown in 10-cm plates 1 and 2 days after MCTP treatment using the method of Xu and Shields (69) as described in MATERIALS AND METHODS. Protein-matched aliquots from the purified Golgi band were used to assay for changes in the enzymatic activities of -D-mannosidase, -D-mannosidase, -D-glucosidase, and -D-glucosidase using the manufacturer's protocols. Enzymatic activity of MCTP-treated cultures is expressed in terms of that of control cultures normalized to 100. There was no significant difference between activities of the enzymes when assayed from control cultures and cultures pretreated with MCTP for 1 and 2 days (P 0.05; data expressed as means ± SE; n = 3 independent experiments).

View larger version (47K): [in this window] [in a new window]   Fig. 9. Marked increase in Golgi tethers in lungs of MCT-treated rats. Immunofluorescence analyses were carried out on lung sections of MCT- and vehicle-treated rats (4-wk) as described in MATERIALS AND METHODS. A: increased level of golgin 84 in a pulmonary arterial vessel (lumen is oriented towards the top of the section). B: increases in the tethers GM130, p115, and giantin in the lung parenchyma. Scale bar, 50 µm in A and 100 µm in B.

View larger version (84K): [in this window] [in a new window]   Fig. 10. Parenchymal cells in MCT-treated rat lungs showing changes in Golgi markers are distinct from infiltrating macrophages. Immunofluorescence analyses were carried out on lung sections obtained from rats treated with MCT 4 wk earlier using LAMP1, the macrophage marker ED-1, and Golgi marker p115 (A–C). A: marked infiltration of LAMP1-positive cells into the lung parenchyma 4 wk after MCT. B and C: although the LAMP1-positive cells are ED-1 positive (macrophages), they do not show stain for the Golgi marker p115, thus excluding macrophages from being the cells that show MCT-induced changes in the Golgi. Similar results were obtained using another Golgi marker, giantin (not shown). Scale bar, 50 µm; scale bars in the insets in B, 20 µm.

How early can the Golgi changes be detected in the rat lung? To address this question, we performed immunofluorescence analyses on rat lung sections 2 and 4 days after administration of MCT, i.e., before the onset of PH (29). Little or no differences were observed between 2-day MCT-treated and control rat lungs (data not shown). However, Fig. 11B shows that there were clear increases in circumnuclear GM130 and giantin within 4 days after MCT administration both in the arterial endothelium and parenchymal cells in MCT-treated rat lung compared with the simultaneously processed "controls." (Figure 11B control sections clearly show juxtanuclear puncta corresponding to normal Golgi elements; these are less apparent in Fig. 11A, which summarizes sections processed and imaged several months earlier). Together, the data in Figs. 9–11 are consistent with the occurrence of Golgi blockade in the rat lung following exposure to MCT.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
To investigate the underlying cause of a block in trafficking of diverse proteins through the Golgi apparatus in MCTP-induced megalocytosis (Fig. 12), we investigated possible changes in the underlying mediators of vesicular trafficking. To this end, we investigated changes in Golgi tethers, SNAREs, and SNAPs following exposure of lung endothelial and epithelial cells to MCTP in cell culture and in lungs of the MCT-treated rat.

View larger version (27K): [in this window] [in a new window]   Fig. 12. Schematic of the subcellular targets of MCTP and the subsequent phenotypic changes. UPR, unfolded protein response.

In the present study, we investigated changes in different v- and t-SNAREs involved in trafficking to and through the Golgi apparatus to look for an underlying cause for the trafficking defect in MCTP-induced megalocytosis. Of the proteins selected for study, GS28, GS27, and GS15 are resident SNARE proteins of the Golgi apparatus that mediate ER to Golgi and intra-Golgi trafficking (25, 59, 70); Vti1a is a Golgi-localized putative t-SNARE (28), whereas Vti1b is believed to be localized in the Golgi and endosomes (3, 40). Syntaxin-6 is a trans-Golgi-localized t-SNARE (7), whereas syntaxin-4 (which is the syntaxin-1 isoform in endothelial cells) is considered to be a plasma membrane-associated t-SNARE as is SNAP23 (44). Additionally, because the process of membrane fusion is not dependent exclusively on SNAREs but also requires tether proteins, which are believed to act before SNARE interactions, we also investigated MCTP-induced changes in the Golgi tethers p115, GM130, giantin, and golgin 84, which play a critical role in association of coatomer protein complex I (COP-I) vesicles with the Golgi apparatus and ER-Golgi and intra-Golgi transport (4, 51, 52). Additionally, as transport of proteins from the trans-Golgi network is dependent on clathrin, we included clathrin in our investigations (46). Although defects in these critical mediators of vesicular trafficking through the Golgi apparatus have been described individually and have widespread effects on subcellular trafficking [as examples, mutant syntaxin-6 isoforms reduce the amount of cav-1 present at the plasma membrane in human skin fibroblasts (7), dominant negative GM130 inhibits COP-I vesicle fusion with the Golgi and decreases VSV-G cell surface trafficking (52), knockdown of GS15 largely leads to redistribution of GS28 and syntaxin-5, but not p115, GM130, and syntaxin-6 (70)], we observed that in MCTP-induced megalocytosis, all tethers, SNAREs, and SNAPs examined were trapped in the Golgi as assayed using cell fractionation and assumed a circumnuclear/perinuclear localization by immunofluorescence (Fig. 12). Thus there was a profound defect in anterograde trafficking to and from the Golgi apparatus in MCTP-induced megalocytosis.

It is possible that the reason why diverse SNAREs and SNAPs were trapped in the Golgi is that their disassembly or recycling step was dysfunctional. The SNARE recycling defect may be due to sequestration and mislocalization of the ATPase NSF. This hypothesis is supported by data showing that the subcellular localization of NSF was discretely separable from all Golgi tethers, SNAREs, and SNAPs in cells treated with MCTP. Indeed, Iwakiri and colleagues (15) reported very recently that Golgi-targeted mutants of eNOS produced nitric oxide locally within this subcellular compartment and thus covalently modified the Cys-rich and redox-sensitive intracellular NSF by S-nitrosylation. They showed that a consequence of this covalent modification of NSF was the inhibition of its trafficking activity resulting in a block in trafficking through the Golgi of cargo proteins such as the vesicular stomatitis virus G protein (15). Thus, from the point of view of the present article, the aberrantly sequestered eNOS after MCTP, which generates intracellular nitric oxide in aberrant intracellular compartments (38), would further inhibit NSF and membrane trafficking through the Golgi in a self-reinforcing inhibitory loop.

We suggest that MCTP-induced PH and megalocytosis might be looked upon as a disease characterized by Golgi dysfunction (Fig. 12). The present data highlight electron microscopic studies between 1950 and 1970 on the effects of pyrrolizidine alkaloids, including MCT on hepatocytes and PAECs (2, 32). Notably, Afzelius and Schoental (2), even then, based on electron microscopy alone, suggested that the effect on the Golgi (enlargement and "proliferation" of Golgi stacks) might well represent both a trafficking defect and a defect in regulation of cell division. Our present data are a tribute to those early electron microscopists.

From a vascular standpoint, our recent findings showing the trapping of eNOS with cav-1 in the Golgi after MCTP exposure with a subsequent decrease in cell surface nitric oxide production (38) likely accounts for the well-known observation of a marked reduction in nitric oxide production in lungs of MCT-treated rats (41). Eventually, there is a loss of both cav-1 and eNOS from lungs of MCT-treated rats (19, 29). A loss of cav-1 from PAEC has now been confirmed in human lesions in patients with primary PH (1) as well as in another model of experimental PH (62). Additionally, we have also recently shown that alteration in the function of the Golgi apparatus is not unique to the MCT model of PH, but similar changes are seen with hypoxia and senescence (i.e., the Golgi tethers and SNAREs GM130, giantin, p115, and syntaxin-6 assume a circumnuclear provenance in hypoxic and senescent PAEC, and there is a functional mislocalization of vasorelevant proteins like eNOS and cav-1 with loss from plasma membrane with perinuclear trapping partially colocalizing with the Golgi) (38). Moreover, there is extensive preexisting literature describing the presence of "plump" cells with increased ER-Golgi and Weibel-Palade bodies in both human PH and hypoxia-induced PH in the rat in vivo clearly indicative of a block in anterograde trafficking (11, 14, 16, 20, 33, 34, 39, 56, 65; see discussion in Ref. 38 for detailed description). These observations suggest a disruption of intracellular membrane trafficking and of vesicle tethers, SNAREs, and SNAPs within endothelial cells may be the common underlying mechanism contributing to the development of PH in diverse models such as MCT-induced and hypoxia-induced as well as familial and sporadic PH due to trafficking mutations in BMPR2 (27, 49). The observation that MCTP led to trapping of BMPR2 receptor in the Golgi (Fig. 2B) adds weight to this suggestion. That dominant negative mutations in BMPR2 produced PH in mice (66) and that chronic hypoxia decreased BMPR2 levels concomitant with development of PH (61) have been reported recently. The MCTP-induced block in anterograde trafficking and retention of BMPR2 in the Golgi (Fig. 2B) would be akin to a loss-of-function mutation. Moreover, it is now well established that Smad transcription factor signaling, which is a critical mechanism for the growth inhibitory effect of BMPs in PASMCs (24, 57, 63), transits the cytoplasm in a membrane-associated manner (such as along the early endosomal pathway for transcriptional activation and towards the late endosome/lysosome degradation) (reviewed in Ref. 54). Thus any disruption of intracellular membrane trafficking would affect this signaling. The limited penetrance (15%) of carriers of the autosomal dominant BMPR2 mutations that cause PH has spawned various "second-hit" hypotheses (26, 50, 71). It is noteworthy that even today consideration of BMPR2 signaling through the cytoplasm (a transforming growth factor- family/Smad pathway) fails to include membrane-associated endosomal signaling (as in the schematic in Ref. 71). We suggest that the likely limited ability of such BMPR2 mutations to elicit dominant negative effects in the face of redundant multiple vesicle tethers, SNARES, and SNAPs involved in membrane transport (through the Golgi or in Smad signaling) may account for this limited penetrance. Thus the present mechanistic insights (Fig. 12) are likely of broad clinical significance.

The observation that fluoxetine, which inhibits the serotonin transporter (5-HTT), prevented and reversed MCT-induced PH (13) and affects trafficking to/through the Golgi (9) highlights the need for a closer examination of altered intracellular trafficking mechanisms in this disease. The increase in 5-HTT in the vascular wall after MCT could reflect both increased synthesis (13) and decreased transcytoplasmic trafficking with accumulation in the Golgi (as in Figs. 2–4) and/or additional intracellular sites. Thus an MCTP-induced alteration in intracellular membrane trafficking provides for a unifying underlying mechanism accounting for alterations in diverse vasorelevant proteins in vascular cells (Fig. 12). Because of the diversity of vesicle tethers, SNAREs, and SNAPs affected, we do not exclude trafficking alterations to/through intracellular membrane compartments in addition to the Golgi. The amelioration of MCT-induced PH by vectors expressing eNOS, Ang-1, hepatocyte growth factor, and prostacyclin synthase (43, 72–74) may reflect procedures that overcome the signaling deficits caused by this altered subcellular trafficking. We suggest that strategies to discover small molecules that unblock the Golgi block [using the rapid assays described by us in Fig. 1C (peroxidase secretion) or Fig. 2A (cav-1 trapping in the Golgi) as discovery screens] represent a new rational approach towards therapeutic intervention in PH.

Our data demonstrated increased cav-1 in Golgi membranes in PAEC within 6 h after MCTP treatment (Fig. 2A), a visible phenotypic megalocytosis within 9–12 h, and an inhibition of functional anterograde secretion within 12 h (Fig. 1, C and D). These observations represent the earliest that the effects of MCTP on cells in culture have been reported to date. We suggest that MCTP, directly or indirectly, rapidly targets anterograde vesicular movement mediated by Golgi tethers, SNAREs, and SNAPs [such as syntaxin-6, which is required for anterograde trafficking of cav-1 (7)]. The subsequent slower accumulation of these trafficking mediator proteins in the Golgi by 2 days (Fig. 2B) would reflect their respective intracellular recycling times and the extent of the functional inhibition imposed by MCTP. Although the present study did not encompass vascular smooth muscle cells, we have shown previously that human PASMC in culture also develop megalocytosis upon exposure to MCTP (55).

The pioneering work of Lamé and colleagues (22, 23) has shown that MCTP selectively derivatizes glucose-regulated protein 58 (also called ERp57 or ER-60), protein disulfide isomerase, endothelial cell-specific protein disulfide isomerase, and reticulocalbin species in endothelial cells in culture (Fig. 12). There is a selective targeting of proteins with the Cys-X-X-Cys motif. In as much as these proteins all reside in the lumen of the cellular secretory pathway, including specifically the ER and the Golgi (23), this covalent derivatization might lead to an inhibition of their luminal protein-folding function and contribute to Golgi blockade (Fig. 12). Indeed, we have previously shown that MCTP triggers aspects of the unfolded protein response in megalocytotic endothelial and lung alveolar cells (36, 37). Additionally, we note that NSF is a Cys-rich protein and that its biological activity is specifically inhibited by N-ethylmaleimide (30). Modification of cysteine residues of NSF by nitric oxide-mediated S-nitrosylation inhibits its ability to disassemble SNARE complexes without affecting its ATPase function and prevents exocytosis of Weibel-Palade bodies in human aortic endothelial cells in culture (30). Also, mutations of cysteine residues at position 11, 21, 334, 568, and 582 of NSF block its ability to interact with (associate with) SNARE proteins, and mutations in cysteine residues 91 and 264 permit association but block NSF-mediated SNARE disassembly in vitro (30). Thus the biological activity of NSF is dependent on functional cysteine residues. It is possible that NSF may represent a target protein directly derivatized by MCTP.

The observations that pyrrolizidine alkaloids such as australine and indolizidine alkaloids such as swainsonine target specific Golgi-resident glycosidases (42, 64) suggested the possibility that MCT might also target specific glycosidases in the Golgi. The report that swainsonine caused PH and right ventricular hypertrophy in calves (17) was intriguing. However, we observed only minimal changes in the specific activities of -D- or -D-mannosidases or -D- or -D-glucosidases in Golgi membranes purified from cells 2 days after MCTP exposure.

Endothelial cell apoptosis is an important component of the pathogenesis of PH (35, 62). Chiu and colleagues (6) have demonstrated the ability of a COOH-terminal fragment of the Golgi protein p115 to initiate and amplify apoptosis. That MCTP causes sequestration of this apoptosis initiator protein (p115) in the Golgi (Figs. 3–6) is intriguing in that megalocytotic endothelial cells resist apoptosis (55) as do other vascular cells in the development of PH (35).

To summarize, we suggest that MCT-induced PH involves dysfunction of Golgi tethers, SNAREs, and SNAPs leading to reduced disruption of trafficking of diverse vasorelevant proteins (such as cav-1, eNOS, BMPR2, PECAM-1, Tie-2) to the cell surface of vascular cells (Fig. 12). This Golgi blockade may represent a unifying underlying mechanism that accounts for blocks in intracellular trafficking and cell cycle progression and resistance to apoptosis. Discovery screens for small molecules that unblock the Golgi block provide a new approach to finding ways to treat PH.

	GRANTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
This work was supported by National Heart, Lung, and Blood Institute Grant HL-073301.

	ACKNOWLEDGMENTS

 
We thank Dr. Dennis Shields for helpful discussion and for the anti-p115 antibody.

	FOOTNOTES

 

Address for reprint requests and other correspondence: P. B. Sehgal, Rm. 201, Basic Sciences Bldg., Dept. of Cell Biology and Anatomy, New York Medical College, Valhalla, NY 10595 (e-mail: pravin_sehgal{at}nymc.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

	REFERENCES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 

1. Achcar RO, Demura Y, Rai PR, Taraseviciene-Stewart L, Kasper M, Voelkel NF, Cool CD. Loss of caveolin and heme oxygenase expression in severe pulmonary hypertension. Chest 129: 696–705, 2006.[CrossRef][ISI][Medline]
2. Afzelius BA, Schoental R. The ultrastructure of the enlarged hepatocytes induced in rats with a single oral dose of retrorsine, a pyrrolozidine (Senecio) alkaloid. J Ultrastruct Res 20: 328–345, 1967.[CrossRef][ISI][Medline]
3. Antonin W, Holroyd C, Fasshauer D, Pabst S, Von Mollard GF, Jahn R. A SNARE complex mediating fusion of late endosomes defines conserved properties of SNARE structure and function. EMBO J 19: 6453–6464, 2000.[CrossRef][ISI][Medline]
4. Beard M, Satoh A, Shorter J, Warren G. A cryptic Rab1-binding site in the p115 tethering protein. J Biol Chem 280: 25840–25848, 2005.[Abstract/Free Full Text]
5. Bonifacino JS, Glick BS. The mechanism of vesicle budding and fusion. Cell 116: 153–166, 2004.[CrossRef][ISI][Medline]
6. Chiu R, Novikov L, Mukherjee S, Shields D. A caspase cleavage fragment of p115 induces fragmentation of the Golgi apparatus and apoptosis. J Cell Biol 159: 637–648, 2002.[Abstract/Free Full Text]
7. Choudhry A, Marks DL, Proctor KM, Gould GW, Pagano RE. Regulation of caveolar endocytosis by syntaxin 6-dependent delivery of membrane components to the cell surface. Nat Cell Biol 8: 317–328, 2006.[CrossRef][ISI][Medline]
8. Connolly CN, Futter CE, Gibson A, Hopkins CR, Cutler DF. Transport into and out of the Golgi complex studied by transfecting cells with cDNAs encoding horseradish peroxidase. J Cell Biol 127: 641–652, 1994.[Abstract/Free Full Text]
9. Donati RJ, Thukral C, Rasenick MM. Chronic treatment of C6 glioma cells with antidepressant drugs results in a redistribution of G_s. Mol Pharmacol 59: 1426–1432, 2001.[Abstract/Free Full Text]
10. Eddahibi S, Morrell N, d'Ortho MP, Naeije R, Adnot S. Pathobiology of pulmonary arterial hypertension. Eur Respir J 20: 1559–1572, 2002.[Abstract/Free Full Text]
11. Eunhee SY, Hojoong K, Hyekyung A, Strother J, Morris T, Masliah E, Hansen LA, Kyungjoo P, Friedman PJ. Distribution of obstructive intimal lesions and their cellular phenotypes in chronic pulmonary hypertension. A morphometric and immunohistochemical study. Am J Respir Crit Care Med 162: 1577–1586, 2000.[Abstract/Free Full Text]
12. Fries E, Rothman JE. Transport of vesicular stomatitis virus glycoprotein in a cell-free extract. Proc Natl Acad Sci USA 77: 3870–3874, 1980.[Abstract/Free Full Text]
13. Guignabert C, Raffestin B, Benferhat R, Raoul W, Zadigue P, Rideau D, Hamon M, Adnot S, Eddahibi S. Serotonin transporter inhibition prevents and reverses monocrotaline-induced pulmonary hypertension. Circulation 111: 2812–2819, 2005.[Abstract/Free Full Text]
14. Heath D, Smith P, Gosney J, Mulcahy D, Fox K, Yacoub M, Harris P. The pathology of early and late stages of primary pulmonary hypertension. Br Heart J 58: 204–213, 1987.[Abstract/Free Full Text]
15. Iwakiri Y, Satoh A, Chatterjee S, Toomre DK, Chalouni CM, Fulton D, Groszmann RJ, Shah VH, Sessa WC. Nitric oxide synthase generates nitric oxide locally to regulate compartmentalized protein S-nitrosylation and protein trafficking. Proc Natl Acad Sci USA 103: 19777–19782, 2006.[Abstract/Free Full Text]
16. Jaenke RS, Alexander AF. Fine structural alteratio