Multicopy suppressor analysis of thermosensitive YIP1 alleles implicates GOT1 in transport from the ER

Abstract

Yip1p belongs to a conserved family of membrane-spanning proteins that are involved in intracellular trafficking. Studies have shown that Yip1p forms a heteromeric integral membrane complex, is required for biogenesis of ER-derived COPII vesicles, and can interact with Rab GTPases. However, the role of the Yip1 complex in vesicle budding is not well understood. To gain further insight, we isolated multicopy suppressors of the thermosensitive yip1-2 allele. This screen identified GOT1, FYV8 and TSC3 as novel high-copy suppressors. The strongest suppressor, GOT1, also displayed moderate suppressor activity toward temperature-sensitive mutations in the SEC23 and SEC31 genes, which encode subunits of the COPII coat. Further characterization of Got1p revealed that this protein was efficiently packaged into COPII vesicles and cycled rapidly between the ER and Golgi compartments. Based on the findings we propose that Got1p has an unexpected role in vesicle formation from the ER by influencing membrane properties.

Fig. 1.

Multicopy suppressors of the yip1-2 strain. Transformed wild-type (MSUC-3D) and yip1-2 (YXY12α) strains were grown to saturation in YMD-URA to maintain selection of plasmids, adjusted to an OD(600) of 1.0, and 5 μl of tenfold dilutions were spotted onto YPD plates. Plates were incubated at 25°C or 36°C for 48 hours. (A) Comparison of multicopy suppression of yip1-2 by FYV8 and TSC3. (B) Comparison of multicopy suppression of yip1-2 by GOT1. The dilution series experiments in A and B were conducted on separate occasions.

Fig. 2.

GOT1 is a multicopy suppressor of the yip1-1 and yip1-4 strains. The experiment was carried out as in Fig. 1. (A) Multicopy GOT1 suppression of yip1-1 (YXY11a) compared with the wild type (MSUC-3D). (B) The yip1-4 strain (RCY1764) compared with the wild type (RCY1768) at restrictive temperatures.

Fig. 3.

Got1p cycles between the ER and Golgi compartments. (A) Got1p co-fractionates with ER and Golgi markers. A whole-cell lysate from wild-type cells (CBY740) separated on an 18-60% sucrose density gradient, and fractions collected from the top. The top panel shows the fractionation pattern of Sec61p (ER marker), Och1p (Golgi marker) and Yip1p (ER-Golgi cycling marker). The bottom panel shows the fractionation pattern of Got1p and Yip1p with % sucrose based on refractometry readings indicated on the right axis. (B) Blocking ER export in the sec12-4 strain (37°C for 45 minutes) causes Got1p to accumulate in the ER fraction (P13). Erv25p and Erv41p, proteins that cycle between the ER and Golgi, also shift to the ER fraction. The first four lanes compare the wild type (RSY248) with sec12-4 (RSY263) and the last four lanes compare wild type (CBY740) with got1Δ (CBY1574).

Fig. 4.

Got1p is efficiently packaged into COPII vesicles. Large scale budding reactions were performed on wild-type (CBY740) and got1Δ (CBY1574) microsomes as described in the Materials and Methods. Total membranes (T), budded vesicles from minus (–) and plus (+) COPII incubations were immunoblotted for Sec12p (ER marker), Erv41p and Erv25p (ER-Golgi cycling markers) and Got1p. The total lane represents 25% of a budding reaction. Note that the packaging of proteins into COPII vesicles was not affected in got1Δ membranes.

Fig. 5.

In vitro analysis of Got1p function in ER-Golgi transport assays. (A) Multicopy GOT1 partially restores COPII budding from yip1-4 membranes. Washed semi-intact cell membranes containing [³⁵S]gpαf were prepared from wild-type (RCY1768 with pRS426), yip1-4 (RCY1764 with pRS426) and yip1-4/GOT1 (RCY1764 with pRS426-GOT1) strains. Membranes were mock treated (NA) or incubated with COPII proteins at 29°C for 20 minutes and [³⁵S]gpαf packaged into budded vesicles measured as described in the Materials and Methods. (B) Membranes lacking Got1p do not display budding or tethering defects in vitro. Washed semi-intact cell membranes as above were prepared from wild-type (CBY740) and got1Δ (CBY1574) strains. Membranes were mock treated (NA), incubated with COPII proteins or COPII proteins plus Uso1p at 23°C for 30 minutes and freely diffusible vesicles quantified to assess levels of vesicle budding and tethering. (C) Membranes lacking Got1p do not display in vitro transport defects. Washed semi-intact cell membranes as above were mock treated (NA) or incubated with COPII, Uso1p and LMA1 proteins to reconstitute (Recon) transport. After 70 minutes at 23°C, the amount of Golgi-modified [³⁵S]gpαf was measured to determine transport efficiency. Error bars represent the range of duplicate determinations.

Fig. 6.

Immunoprecipitation analysis of Got1p. (A) Got1p was not recovered with the Yip1 complex. Digitonin solubilized microsomes from wild type (FY834) or Yif1p-3HA (CBY801) strains were mock precipitated (–Ab) or precipitated with anti-HA antibody (+Ab). Total (T) and solubilized (S) samples represent 1.5% of the starting material. Note specific co-immunoprecipitation of Yip1p and Yos1p with Yif1p-3HA. (B and C) Got1p forms a low molecular weight homo-oligomer. (B) Triton-X-100-solubilized microsomes from Yif1p-3HA (CBY801) or Yif1p-3HA/Got1-3MYC (CBY1373) strains were mock precipitated (–Ab) or precipitated with anti-cMyc antibody (+Ab). Total (T) and solubilized (S) samples represent 1.0% of the starting material, whereas the unbound (U) lane represents 1.7% of the total unbound material. Note that the strain expressing Got1p-3MYC also expresses endogenous Got1p. The arrow indicates Got1p-3MYC, the asterisk indicates position of anti-Myc antibody light chain and the arrowhead indicates a crossreacting species. (C) Detergent-solubilized membrane proteins were resolved on a 1-12% sucrose velocity gradient containing 1% Triton X-100. Globular molecular mass markers were run on parallel gradients (arrows above graphs); Erv25 complex (∼100 kDa) and Sec22p were internal controls.

Fig. 7.

GAL1 overexpression of 3HA-Got1p inhibits growth and causes a secretory block. Wild-type (CBY740) and GAL1-3HA-GOT1 (CBY2577) strains were pre-cultured in medium with 1% galactose and 1% glucose and then shifted to medium with 2% galactose. (A) Cultures were analyzed for growth over time by absorbance. (B) At the 5 hour time-point, cell extracts were prepared and proteins monitored by immunoblot. Arrowheads identify the position of 3HA-Got1p and arrows indicate untagged Got1p. The panel labeled `(dark)' shows a saturated exposure of the anti-Got1p blot to visualize endogenous levels of protein. The ER (p1), Golgi (p2) and mature (m) forms of CPY are indicated. Note the reduced growth rate and accumulation of the ER form of CPY in the GAL1-3HA-GOT1 strain.

Fig. 8.

GAL1 overexpression of 3HA-Got1p causes abnormal ER and Golgi morphologies. (A) Wild-type and GAL1-3HA-GOT1 strains expressing Sec63-GFP (CBY2537 and CBY2542, respectively) were cultured in 2% galactose as in Fig. 7 and observed by fluorescence and bright field microscopy at the 5 hour time point. Note the elaboration of ER membranes and appearance of membrane gaps in the GAL1-3HA-GOT1 strain. (B) Wild-type and GAL1-3HA-GOT1 strains expressing Sec7-GFP (CBY2538 and CBY2543, respectively) were treated as described in A. Note that GAL1-3HA-GOT1 cells display a diffuse Sec7-GFP localization pattern with punctate Golgi structures reduced or absent. Scale bars: 5 μm. A quantitative analysis of these abnormal ER and Golgi morphologies in a population of cells is provided in supplementary material Fig. S5.