P125, a COPII INTERACTING PROTEIN 3

44 CHAPTER Sec31A associates with p125A, a Sec23 interacting protein. 3.1 GST-Sec31A pulls down p125A from rat liver cytosol. The N-terminal region of Sec31A (amino acids 1-332) consists of repeating WD40 domains, which is known to interact with the Sec13 protein (Shugrue et al., 1999, Tang et al., 2000). The C terminal region of Sec31A contains a proline-rich region which contains several SH3-binding consensus sites (XPXXP) (Mongiovi et al., 1999) and a cluster of potential phosphorylation sites (Shugrue et al., 1999, Tang et al., 2000). A schematic diagram of the domain structure of full-length human Sec31A is shown in Figure 3.1 In Tang et al., (1999), it was shown that Sec31A antibodies, as well as the Glutathione S-transferase (GST) fusion protein of the C-terminal 180 amino acids of Sec31A (GST-Sec31A), inhibited ER to Golgi transport of VSVG in a semi-intact cell system. It was postulated that GST-Sec31A probably inhibits transport by sequestering proteins that are required for the functions of endogenous Sec31A. To identify possible Sec31A interacting proteins that are important for ER to Golgi transport, a GST-Sec31A large scale pulldown experiment was performed. schematic diagram of the GST-Sec31A construct is shown in Figure 3.2. A 45 WD-40 Domain N Proline-rich region 332 808 1145 Figure 3.1. Schematic diagram of full-length human Sec31A. Adapted from Tang et al al., 2000. 2000 Numbers indicate amino acid residues. residues Sec31A C-terminal 180 amino acids 1040 1220 Figure 3.2 Schematic diagram of GST-Sec31A construct. 1220 C 46 Rat liver cytosol was subjected to pulldown with GST-Sec31A bound beads. For control, GST bound beads were used. The beads were washed extensively, eluted with SDS sample buffer and subjected to SDS-PAGE analysis. The gel was then stained with coomassie blue and destained to the desired band intensity. As shown in Figure 3.3, several polypeptides from the total cytosol were retained by GST-Sec31A (lane 3) but not by GST (lane 2). The bands that were present in the GST-Sec31A, but not the GST pull down lane was excised and subjected to mass spectrometric analysis. Mass spectrometric analysis revealed the presence of p125A and other proteins. The fact that the p125A band was not present in the GST pulldown lane suggested that there was specific interaction with the C-terminal region of Sec31A. As p125A was also identified as an interacting protein for Sec23, we therefore focused the subsequent studies on the interaction of Sec31 with p125A and its physiological relevance. 3.2 p125A interacts with Sec31A To verify the interaction of p125A with Sec31A, co-immunoprecipitation experiments were performed with myc-tagged full length p125A (myc-p125AFL) and GFP-tagged full length Sec31A (GFP-Sec31AFL). The tagged proteins were exogenously expressed in HEK293 cells either singly or in combination. Cell lysates were prepared and subjected to immunoprecipitation (IP) with myc antibodies. If p125A does indeed associate with Sec31A, GFP-Sec31AFL would be co-immunoprecipitated by myc antibodies only in the presence of myc-p125AFL. As shown in Figure 3.4, the constructs were adequately expressed as indicated by the 5% input of each lysate that was loaded (lane 1-3). myc-p125AFL was specifically immunoprecipitated by the myc antibodies (lane 4) and GFP-Sec31AFL was not (lane 6). No bands were observed in the control IP lanes (lane - 9), indicating that the IP was specific. GFP- 47 M 220 p125A 357 PYTEEFS 363 319 YDVYLYDR 326 97 Fumarate dehydratase y GRP78 69 46 GST-Sec31A Figure 3.3. GST-Sec31A pulls down p125A from rat liver cytosol. Rat liver cytosol (50 mg) was subjected to pulldown with 500µg of GSTSec31A protein pulldown. The proteins were resolved by SDS-PAGE and stained with Coomassie blue. The proteins were analyzed by amino acid microsequencing. 48 5% Input myc-IP ctrl IP - + + - + + - + + GFP-Sec31AFL + + - + + - + + - myc-p125A FL 250 150 GFP-Sec31AFL WB : GFP 150 100 myc-p125AFL WB : myc Figure 3.4. myc-p125AFL can co-immunoprecipitate GFP-Sec31AFL. mycp125AFL and GFP-Sec31AFL were expressed exogenously in HEK293 cells either singly or in combination. Cell lysates were prepared and subjected to immunoprecipitation by either anti-myc (9E10) antibodies or control mouse IgG. 49 Sec31AFL was co-immunoprecipitated by myc-antibodies when expressed together with myc-p125AFL (lane 5). These results confirm that Sec31A does indeed interact with p125A, at least in the context of overexpressed co-transfections. To define the region of p125A that is responsible for its interaction with Sec31A, myc-tagged constructs of eight different regions of p125A were generated (Figure 3.5). A brief summary of the constructs is as follows. The N-terminal (NT) portion of p125A is made up of amino acids 1-304, and the C-terminal (CT) region contains amino acids 305-1000. NT600 and NT650 consist of residues 1-600 and 1-650, respectively. The CTΔ100 and CTΔ200 are mutants with 100 or 200 amino acids deleted from the N-terminus of p125A, in that order. myc-p125A 200-600 and mycp125A 260-600 contains residues 200-650 and 260-600 respectively. All deletion mutants were generated as described in Section 2.2 using primers listed in Table 2.2. The amplified fragments were then digested and cloned into their respective sites in pDmyc-neo and confirmed by sequencing. Together with constructs expressing fulllength p125A and KIAA0725p (the p125 homologue, p125B) (Tani et al., 1999), pulldown experiments were performed. They were transiently expressed in HEK293 cells and the resulting lysates were incubated with either GST-Sec31A or GST. After extensive washing, proteins retained specifically by these beads were then examined by SDS-PAGE and Western blotting. As shown in Figure 3.6 (panel a), protein expression levels of transfected constructs were of similar amounts except for myc-p125A200-650, myc-p125A260-600 and myc-p125ACT (panel a, lane 5, and 7), which had lower expressions. The red asterisks indicate the expected sizes of the respective constructs. Full-length p125A, myc-p125A200-600, myc-p125A260-600, myc-p125ACTΔ100 and myc- Figure 3.5. Schematic of p125A deletion constructs. cDNA fragments encoding for the mutants were generated by PCR. The fragments were digested with SalI and NotI and then subsequently cloned into pDmycneo. myc-p125BFL, the homologue of p125A, also known as KIAA0725p (Nakajima et al., 2002). myc-p125B FL myc p125CT∆200 myc-p125CT∆200 myc-p125CT∆100 myc-p125CT myc-p125 201-650 myc-p125 260-600 myc-p125NT600 myc-p125NT650 myc-p125NT myc-p125AFL 50 51 150 100 75 * * 50 37 * * * * * 10 11 10 11 * * * 25 a 5% Input, WB: myc b 5% Input, Ponceau 150 100 75 50 37 # 25 c GST-Sec31A pulldown, WB : myc d GST-Sec31A pulldown, Ponceau 150 100 75 50 37 25 e GST pulldown, WB : myc f GST pulldown, WB : Ponceau Figure 3.6. Residues 260-600 of p125A are sufficient for binding to Sec31A. myc-p125A deletion constructs were subjected to pulldown analysis usingg GST-Sec31A. The different deletion constructs were expressed p in HEK293 cells. * indicates the expected sizes of the various mutants. Solid triangle indicates GST-Sec31A and the empty triangle, GST. # indicates the smallest fragment that could be pulled down by GST-Sec31A. Panels on the right (b, d and f) are Ponceau staining for the immunoblots. 52 p125ACTΔ200 proteins were all efficiently pulled-down by GST-Sec31A beads (panel c, lane 1, 5, 6, 8, 9), while less NT600 and NT650 proteins (panel c, lane and 4) were pulled-down. Detection of the myc-p125ANT protein was found to be almost negligible (panel c, lane 2) and myc-p125ACT interaction with GST-Sec31A was not present. p125B (KIAA0725), a homologue of p125A (Tani et al., 1999) which exhibits significant sequence similarity with p125A throughout the entire sequence, was found not to interact with Sec31A. It was also noted that when all the cell lysates were subjected to pull-down concurrently with GST protein control, none of the tagged proteins were detected (panel e). These results confirm that the interactions between p125A and GST-Sec31A are specific and molecularly dissectible. mycp125A260-600 was the smallest fragment that was efficiently pulled down by GSTSec31A (panel c, lane 7, #) indicating that it likely contains the minimal Sec31A binding sites. This fragment does not include the known Sec23A interacting domain, (residues 135-259, Mizoguchi et al., 2002), suggesting that p125A’s interaction with Sec31A is likely direct and independent of Sec23A. 3.3 p125A’s interaction with Sec31A is independent of Sec23A To determine if p125A’s interaction with Sec31A is independent of Sec23A, myctagged constructs expressing different fragments of p125A, together with constructs expressing full-length p125A, were transfected into HEK293 cells. The resulting cell lysates were incubated with myc antibodies. After extensive washing, proteins retained specifically by the myc-antibodies were analyzed. The rational would be that p125A containing binding domains to Sec23A and Sec31A should coimmunoprecipitate these two proteins. Furthermore, The N-terminal region (mycp125ANT), containing only Sec23A interacting domain should only coimmunoprecipitate Sec23A but not Sec31A; while myc-p125A 260-600, which was 53 mapped as the smallest fragment to bind Sec31A (Figure 3.6), should only coimmunoprecipitate Sec31A, but not Sec23A. As shown in Figure 3.7, the protein levels of myc-p125ANT and myc-p125A200-600 (5% Input, lane and 5) was found to be much lower than that of myc-p125AFL, NT600 and NT650 (lane 1, and 4). Amount of endogenous Sec31A and Sec23A protein present were of similar amount in the input. Detection of the myc antibody showed that the various myc-tagged p125A proteins were effectively immunoprecipitated. No bands were observed in the control IP using mouse IgG, indicating that immunopreciptations by myc antibodies were specific. As expected, the myc-tagged full-length p125A co-immunoprecipitated both Sec31A and Sec23A from the cell lysate (lane 6). myc-p125ANT600 and myc-p125ANT650 which contained interacting domains for both Sec23A and Sec31A could also coimmunoprecipitate both proteins (lane and 9). However, myc-p125ANT, containing only the Sec23A interacting domain, could only co-immunoprecipitate Sec23A (lane 7). myc-p125A260-600 did not co-immunoprecipitate Sec23A and was also not efficient in co-immunoprecipitation of Sec31A from the lysate (lane 10). This may be due to the competition from endogenous full length p125A for Sec31A binding. No bands were observed in the control experiments using mouse IgG, indicating that immunopreciptations by myc antibodies were specific. 3.4 p125A associates with Sec13/Sec31 complex in the cytosol p125A was pulled-down by the GST fusion protein of Sec31A C-terminal 180 residues from rat liver cytosol (Figure 3.3). This implies that p125A may be in a complex with Sec31A in the cytosol. Sec13 (33 kDa) and Sec31 (140 kDa) exists in the cytosol as a heterotetramer and was found at high molecular weight fractions of 92 all the three knockdown cells redistributes into smaller punctate structures scattered throughout the cell (panel b, d and f). This suggests that the fragmented structures are indeed Golgi and could be further broken down by treatment with nocodazole. 3.13 Rate of ER export is reduced in p125A depleted cells p125A is known to interact with two COPII proteins, Sec23A (Tani, et al., 1999) and Sec31A (current work). Both Sec23A and Sec31A are involved in COPII-mediated ER export. Therefore, to determine if p125A plays a role in this pathway, the integrity of ER export of cells treated with p125A RNAi was examined. HeLa cells stably expressing GT-GFP were used for the ER export assay. Brefeldin A (BFA) is a fungal metabolite that is widely used in the study of protein transport. It causes reversible redistribution of Golgi proteins into the ER and thus blocks protein transport from the ER to the Golgi (Klausner et al., 1992). When BFA is washed out, the Golgi is reformed by proteins exiting the ER. The cells were incubated in the presence of BFA for thirty minutes. BFA was then removed and the cells were allowed to recover in the presence of cyclohexamide (to block new protein synthesis) for various periods of time before they were fixed. Prior to BFA treatment, GT-GFP in control cells was found in compact structures located at the perinuclear region (Figure 3.30, panel a). Upon treatment with BFA, GT-GFP was redistributed to the ER (panel b). Thirty minutes after BFA removal, some GT-GFP was found concentrated in post ER compartments (panel c). By fortyfive minutes, the ER staining of GT-GFP was substantially reduced and most of the protein was concentrated at the perinuclear region (panel d). Sixty minutes after BFA washout, most GT-GFP has redistributed back to the Golgi in loosely compact non-target siRNA p125 5A siRNA b h a g 0’ 30’ i c 45’ j d 60’ k e 90’ l f Figure 3.30 ER export is delayed in p125A silenced cells. HeLa cells stably expressing GT-GFP were transfected either with non-target or p125A siRNA. siRNA At 72 hours after transfection, transfection the cells were treated with 10 μg/ml BFA for 30 to redistribute GT-GFP GT GFP into the ER. ER The cells were then washed in BFA-free medium, and at the indicated times after BFA removal, samples were fixed for fluorescence microscopy to assay for export of GT-GFP to post-ER compartments. All of the images in this figure were taken at the same exposure and processed in parallel. Bar = 10 μm. No BFA Time after BFA Washout 93 non-target siRNA p1225A siRNA b h a g 15’ 30’ i c 60’ j d 90’ k e 120’ l f Figure 3.31. Rate of ER export is delayed in p125A silenced cells (Movie). HeLa cells stably expressing GT-GFP were transfected either with non-target g or pp125A siRNA. At 72 hours after transfection,, the cells were treated with 10 μg μg/ml BFA for 30 to redistribute GT-GFP into the ER. The cells were then washed in BFA-free medium, and subjected to live cell confocal microscopy. Bar =10 μm. These movies are available in the attached CD ROM. 0’ Time after BFA Washout 94 95 structures (panel e). By ninety minutes, the GT-GFP has reassembled back to the normal Golgi staining pattern. In p125A depleted cells, GT-GFP was localized in punctated fragments distributed around the perinuclear region (panel g). After BFA treatment, GT-GFP from these cells also redistribute back into the ER (panel h), similar to the control cells. However, thirty minutes after BFA was removed, the majority of GT-GFP was still found in the ER, with very little distributed back to punctate structures near the perinuclear region (panel i). Forty-five mintues after BFA washout, some GT-GFP was observed in aggregates around the perinuclear region, however, the majority of the GT-GFP was still found in the ER. By sixty minutes, more GT-GFP was distributed to larger punctate structures around the perinuclear region. Ninety minutes after BFA removal, GT-GFP was distributed to punctate structures around the perinuclear region. Thus, protein export was delayed but not completely inhibited in p125A depleted cells. Similar results were observed by live cell imaging (Figure 3.31). These results imply that p125A is important for efficient transport of GT-GFP out of the ER. 3.14 p125A is involved in the transport of VSVG-tsO45-YFP out of ER. The rate of ER export of GT-GFP after BFA treatment was reduced in p125A depleted cells. Therefore, to test whether the reduced level of p125A affects the anterograde transport of other cargoes, the export of VSVG (Vesicular Stomatitis Virus G) protein was monitored. VSVG is one of the most commonly used markers in membrane trafficking studies. A temperature sensitive mutant of the VSVG (VSVG-tsO45) tagged with YFP (Yellow Fluorescent Protein) was employed. At the non-permissive temperature, VSVG-tsO45 is misfolded and accumulates in the ER f e Sec31A g c VSVG-tsO45-GFP h d merge Figure 3.32. Fi 32 p125A 125A and d Sec31A S 31A are involved i l d in i VSVG-tsO45-YFP VSVG t O45 YFP transport t t outt off the th ER. ER HeLa H L cells ll were transfected f d o with VSVG-tsO45-EGFP and incubated at non-permissive temperature at 40 C for 24 hours. after release from nonpermissive temperature, the cells were fixed with 4% paraformaldehyde and then incubated with mouse antibody against Sec31A (panel b) and rabbit polyclonal antibodies against p125A (panel a), followed by incubation with Alexa Fluor® 647 goat anti-mouse IgG g and Alexa Fluor® 555 ggoat anti-rabbit IgG. g The lower panels represents the enlarged g view of the inset in the upper panels. The merged images are shown in panels d and h. Arrowheads indicate colocalization of p125A, Sec31A and VSVG-tsO45-GFP. Bar = 10 μM b a p125A 96 97 (Bergmann et al., 1981; Pelham, 1997; Gougeon et al., 2002). This allows the synchronizing of transport events pertaining to the ER export of the VSVG. Upon shifting cells to the permissive temperature, VSVG-tsO45 exits the ER and is transported to the Golgi, and then to the plasma membrane (Bergmann et al., 1981; Pelham, 1997; Gougeon et al., 2002) HeLa cells were transfected with VSVG-tsO45-YFP-expressing plasmid and incubated at 40°C for 24 hours to accumulate VSVG-tsO45-YFP protein in the ER. After 24 hours, the transfected cells were shifted to 32oC for min. Cells were then fixed and double-labeled with anti-p125A and anti-Sec31A and then analysed by fluorescence microscopy. Upon shifting to the permissive temperature, VSVG- tsO45-YFP was transported towards the Golgi, en route to the plasma membrane. Some of the VSV-G-tsO45-YFP was found to be associated with p125A and Sec31A containing structures (Figure 3.32, white arrows) suggesting that p125A and Sec31A is involved in the exit of VSV-G-tsO45-YFP out of the ER. As shown, p125A can be found colocalizing with VSVG in the ERES marked by Sec31A at five minutes after release to the permissive temperature. Thus, it is important to determine if the depletion of p125A affects the export of VSVG from the ER. To address this, p125A depleted cells were transfected with VSVG-tsO45-YFP and shifted to the non-permissive temperature of 40oC for 24 hours to trap the YFPtagged protein in the ER. Protein synthesis was blocked with cyclohexamide before the cells were allowed to return to the permissive temperature at 32oC. The cells were fixed with paraformaldehyde at various times, after release from the non-permissive temperature and imaged by fluorescence microscopy. 98 In control cells, VSVG-tsO45-YFP was found prominently in dispersed perinuclear structures at fifteen minutes after release (Figure 3.33, panel c). By thirty minutes, VSVG-tsO45-YFP has reached the Golgi complex (panel d). After sixty minutes, VSVG-tsO45-YFP was observed to accumulate at the plasma membrane (panel e). The majority of VSVG-tsO45-YFP was observed to be at the plasma membrane after ninety minutes (panel f). In contrast, cells treated with either Sec13 or p125A siRNA showed a marked delay in the transport of VSVG-tsO45-YFP to the Golgi complex and plasma membrane (panel g-p). Some VSVG-tsO45-YFP could be observed in large aggregates at the perinuclear region; however, a significant amount of the protein remained in the ER at thirty minutes after temperature release (panel h and m). VSVG-tsO45-YFP was distributed to the Golgi-like structures at sixty minutes (panel e and o). By ninety minutes, some of the VSVG-tsO45-YFP protein was transported to the plasma membrane while the main bulk of the protein was accumulated in the ER or perinuclear structures (panel k and p). In the morphological study of VSVG-tsO45-YFP, the transport of this glycoprotein through the Golgi, to the plasma membrane was impaired (Figure 3.33). To determine if the slowdown is due to the delay of entry of the protein into the Golgi, a biochemical VSVG-tsO45-YFP transport assay based on the acquisition of Endoglycosidase H (EndoH) resistance was performed (Figure 3.34) EndoH is an enzyme that specifically cleaves the high-mannose oligosaccharide side chains that are added to a peptide backbone in the ER. The conversion of EndoH sensitive highmannose N-linked glycans into complex-type sugars confers EndoH resistance to the glycoprotein. This conversion occurs in the cis-medial Golgi (Alberts et al., 2007). Thus, the rate of acquiring EndoH resistance is a measure of VSVG transport from the ER to the Golgi. non-target siR n RNA Sec13 ssiRNA p125A siRNA c h m a g 15’ 30’ n i d 60’ j o e 90’ p k f Figue 3.33 VSVG-tsO45-YFP transport is delayed in p125A silenced cells. Hela cells were transected twice with 400 nM siRNA At 48 hrs, siRNA. hrs the cells were transfected with 1μg of VSVG-EYFP and incubated at 40oC overnight. overnight At 72 hrs, hrs the cells o o were released from 40 C and incubated at 37 C for various times before fixing with paraformaldehyde. Bar = 10μM 0’ 99 100 A Time (mins) EndoH Non-target siRNA 100 p125A siRNA 100 30 + 60 + 90 + 120 + + 10 R (Golgi) S ((ER)) R (Golgi) S (ER) WB: VSVG B Ratio [R/(R+S)] 0.8 0.6 0.4 0.2 0.0 Non-target siRNA p125A siRNA 30 60 90 120 Time (min) Figure 3.34. VSVG-tsO45-YFP acquisition of EndoH resistance is delayed in p125A knockdown cells. HeLa cells transfected with siRNA p125A and non-target control and with the plasmid for VSVGGYP as described d ib d under d Materials M i l andd Methods. M h d To T allow ll VSVG GYP VSVG-GYP transport, the cells were shifted to 32 °C. At the indicated times, the cells lysed and subjected to EndoH treatment. R and S denote EndoHresistant and –sensitive forms, respectively (A). The ratio (percentage) of the amount of EndoH-resistant form to that of the total amount (EndoH-resistant + EndoH-sensitive form) is plotted. Green solid line represents non-target siRNA treated cells and red dashed lines indicated p125A depleted cells (B). 101 p125A depleted cells were transfected with VSVG-tsO45-YFP and shifted to the nonpermissive temperature of 40oC for 24 hours to trap the YFP-tagged protein in the ER. Protein synthesis was then blocked with cyclohexamide, and the cells were returned to the permissive temperature at 32oC for various time lengths. The cell lysate was treated with EndoH. The proteins were then resolved on SDS-PAGE and subjected to immunoblot analysis for VSVG-tsO45-YFP. The lower band corresponds to the EndoH-sensitive ER form of VSVG-tsO45-YFP, while the upper band corresponds to the EndoH-resistant Golgi form. The bands were quantified using the Quantity One analysis software (BioRad). The ratio of the intensities of the upper band and lower band is a measure of transport of VSVG from the ER to the Golgi. Immediately upon release from the non-permissive temperature, all of the VSVGtsO45-YFP proteins were still accumulated in the ER and are EndoH sensitive in both control and p125A depleted cells (lane 2). EndoH resistant form of VSVG-tsO45YFP was clearly visible after transport for sixty minutes. About half of the glycoprotein in the control cells acquired EndoH resistance at this time (lane 6, upper panel). By two hours, almost 80% of VSVG was converted to the resistant form (lane 10, upper panel). However, in p125A depleted cells, the rate of conversion to EndoH resistant form was reduced. At sixty minutes, only about 20% of VSVG-tsO45-YFP was resistant to EndoH (lane 6, lower panel). Even though the levels of EndoH resistant proteins increased with time in p125A depleted cells, the amounts were less than that of the control cells (lane and 10, lower panel) The corresponding values of the intensities were plotted graphically in Figure 3.34. As observed from the graph, Endo H resistance increases with time. However, in p125A depleted cells, the rate of conversion of the VSVG-tsO45-YFP into the EndoH 102 resistant form was significantly retarded. This reduction is probably caused by delay of transport of VSVG-tsO45-YFP into the Golgi where EndoH resistance was acquired. These results, taken together with the morphological data in Figure 3.33, strongly suggest that p125A depletion impairs the rate and efficiency of ER to Golgi transport of VSVG. 3.15 p125A’s localization to the membrane is dependent on Sec31A Immunofluoresence data showed that when Sec31A was depleted in HeLa cells, p125A localization to punctate structures at the membrane were noticeably reduced (Figure 3.23), while its protein level was not affected (Figure 3.22). This suggests that some p125A has lost its membrane association and redistributed to the cytosol. If the association of p125A with the membrane is indeed dependent on Sec31A, depletion of Sec31A should therefore cause an increase of p125A protein levels in the cytosol and a corresponding decrease in the levels on the membrane. The p125A association with the membrane in Sec31A-depleted cells was examined by subcellular fractionation. Cytosol and membrane fractions from Sec31A depleted cells were obtained as described in Section 2.15. Equivalent amounts of both fractions were analyzed by immunoblotting. Syntaxin (Syn6), a SNARE that functions in the trans-Golgi Network (TGN), was used as the marker for the membrane fraction (Bock et al., 1997); and the Rho GDP dissociation inhibitor (RhoGDI), a cytosolic protein that inhibits the dissociation of GDP from Rho GTPases (Fukumoto et al., 1990), was used as the marker for the cytosolic fraction. The bands were quantified using the Quantity One analysis software (BioRad). The ratio of the intensities of the membrane and cytosolic bands to the total is a measure of the distribution of the protein to the membrane and cytosol respectively. 103 B A Non-target siRNA m 150 100 37 c Sec31A siRNA m c Sec31A p125A C Sec13 37 Syn 25 Rho GDI Figure 3.35. Fi 35 The Th majority j it off p125A 125A is i redistributed di t ib t d to t the th cytosol t l in i Sec31A S 31A silenced cells. Membrane (m) and cytosol (c) fractions were obtained from cells transfected with the control or Sec31A siRNA . Equivalent amounts of each fraction were immunoblotted with antibodies against Sec31A, p125A, Sec13, Syn6 and Rho GDI (A). The experiment shown in A was repeated three times. The intensity of the bands were quantified using Quantity One software (Biorad). The ratio of distribution is relative to the total . Total = m (membrane fraction ) + c (cytosol fraction). Ratio = m or c/(m+c). The distribution of p125A (B) and Sec13 (C) to membrane and cytosol in non-target control and Sec31A knockdown cells. *P 0.1) when compared to the control cells. This suggests that p125A may work downstream of Sec23A’s membrane association and is not required for Sec23A’s membrane association. Interestingly, distribution of Sec13 and Sec31A to the membrane and cytosolic fractions were noticeably changed 105 p125A siRNA m c A 100 non-target siRNA m c B p125A 150 Sec31A 37 Sec13 75 Sec23A 37 Syn6 Rh GDI RhoGDI 25 C D Figure 3.36. Knock-down of p125A causes Sec31A and Sec13 but not Sec23A to redistribute to the cytosol. (A) Membrane (m) and cytosol (c) fractions were obtained from cells transfected with the control or p125A siRNA . Equivalent amounts of each fraction were immunoblotted with antibodies against Sec31A, p125A, Sec13, Sec23A, Syn6 and Rho GDI. The experiment shown in A was repeated three times. times The intensity of the bands were quantified using Quantity One software (Biorad). The ratio of distribution is relative to the total . Total = m (membrane fraction ) + c (cytosol fraction). Ratio = m or c/(m+c). The distribution of (B) Sec23A, (C) Sec31A and (D) Sec13 to membrane and cytosol in non-target control and p125A knockdown cells. #P>0.1, *P[...]... previously shown to interact with p12 5A (Tani et al., 1999) was not depleted as much as Sec3 1A and p12 5A This may suggest that Sec2 3a and p12 5A interaction may occur only on the ER membrane, where the COPII 58 coat proteins are functionally assembled Rab8 is a small GTPase that is involved in the vesicular transport between the TGN and the basolateral plasma membrane (Huber et al., 19 93) It was not expected... giantin (panel b), GM 130 (panel e) and golgin97 (panel h) and rabbit polyclonal antibodies against p12 5A (panel a, d and g), followed by incubation with Alexa Fluor® 488 goat anti-mouse IgG and Alexa Fluor® 555 goat anti-rabbit IgG The merged images are shown in panels c, f and i Bar = 10 μM 69 surprising that p12 5A staining overlays moderately to the cis-Golgi markers, giantin (panel c) and GM 130 ... Sec3 1A as observed by double labeling with mouse anti-Sec3 1A (panel b and e) The panels at the bottom (d, e and f) show a magnified view of the inset in the panels on the top p12 5A was shown to interact with Sec3 1A (current work) and Sec2 3A (Tani et al., 1999) Sec3 1A is a component of COPII and marks the ER exit sites (Tang et al., 2000) The ability of p12 5A to bind to both Sec2 3A and Sec3 1A of COPII. .. of anti-p12 5A antibody can immunoprecipitate about 10-15% of p12 5A from the cell lysate Interestingly, Sec3 1A, which is in the same complex as p12 5A, was not co-immunoprecipitated In parallel, HEK2 93 cell lysate was also incubated with either anti-Sec3 1A antibody or control rabbit IgG Sec3 1A was efficiently immunoprecipitated and Sec 13 was co-immunoprecipitated Sec3 1A and Sec 13 were shown to interact... is found at the trans-Golgi network (TGN) As many of the ER exit sites are apposed to the cis-Golgi stack (Kirk and Ward, 2007), it is not 67 p12 5A Sec3 1A merge a b c d e f Figure 3. 14 p12 5A colocalizes with Sec3 1A HeLa cells were fixed with 4% paraformaldehyde and then incubated with mouse antibody against Sec3 1A and rabbit polyclonal antibodies against p12 5A, followed by incubation with Alexa Fluor®... GST-Sec3 1A AntiSec3 1A antibody therefore recognizes the p12 5A binding domain on Sec3 1A and could potentially block any p12 5A protein from binding to Sec3 1A Thus, it was not surprising that anti-Sec3 1A antibody was not able to co-immunoprecipitate p12 5A 64 p12 5A DAPI merge b c d e f g h i Pre-adsorption with 0.5mg GST-p12 5A Pre-adsorption with 0.5mg GST-Bet3 a Figure 3. 12 p12 5A localizes to punctate structures... anti-Sec3 1A and anti-p12 5A antibodies 56 Thyroglobulin 669 kDa Apoferritin 4 43 kDa 150 kDa WB : Sec3 1A 150 kDa 100 kDa WB : p12 5A 75 kDa WB : Sec2 3A 37 kDa WB : Sec 13 Figure 3. 8 The majority of p12 5A fractionates together with the Sec 13/ Sec31 complex in the cytosol HeLa cytosol was subjected to gel filtration in Superose 6™ medium at a flow rate of 0 .3 ml/min Fractions were collected at 0.6 ml each and then... knockdown greatly reduced the levels of Sec3 1A This observation implies that the Sec 13 may be important for the protein stability of Sec3 1A and vice versa p12 5A protein levels were not affected in both Sec 13 and Sec3 1A depleted cells while Sec2 3A protein levels were slightly elevated in Sec 13 knockdown Non-target control did not affect the protein levels of p12 5A, Sec3 1A, Sec 13 and Sec2 3A, indicating that the... 62 1 2 3 4 5 6 250 150 WB : p12 5A 100 75 50 37 Blocked with GST-Bet3 Blocked with GST-p12 5A Figure 3. 11 Anti-p12 5A antibody recognizes a polypeptide of 125 kDa GST fusion protein corresponding to residues 500-758 was used to raise antibody against p12 5A in rabbits Antibody was affinity purified 30 µg of Hela lysate (Lanes 1 and 3) and HEK2 93 lysate (Lanes 2 and 4) were loaded per lane Sample was resolved... immunoprecipitates were resolved by SDS-PAGE and analyzed by immnoblotting with anti- myc, anti-Sec3 1A and and y ( ) anti-Sec2 3A antibodies 5% of each lysate was loaded (lanes 1-5) 55 about 600-700 kDa (Tang et al., 2000) Therefore, if endogenous p12 5A is indeed in a complex with the Sec 13- Sec31 heterotetramer in the cytosol, it should elute in similar fractions with Sec 13 and Sec31 by gel filtration assay Gel . 44 CHAPTER 3 Sec3 1A associates with p12 5A, a Sec 23 interacting protein. 3. 1 GST-Sec3 1A pulls down p12 5A from rat liver cytosol. The N-terminal region of Sec3 1A (amino acids 1 -33 2) consists. depleted as much as Sec3 1A and p12 5A. This may suggest that Sec2 3a and p12 5A interaction may occur only on the ER membrane, where the COPII 58 coat proteins are functionally assembled. Rab8 is a. kDa Apoferritin 4 43 kDa 150 kDa 150 kDa WB : Sec3 1A 100 kDa 75 kDa WB : p12 5A WB : Sec2 3A 37 kDa WB : Sec2 3A WB : Sec 13 Figure 3. 8. The majority of p12 5A fractionates together with the Sec 13/ Sec31 complex