Crosstalk between protein N-glycosylation and lipid metabolism in Saccharomyces cerevisiae

The endoplasmic reticulum (ER) is a multi functional organelle and plays a crucial role in protein folding and lipid biosynthesis. The SEC59 gene encodes dolichol kinase, required for protein glycosylation in the ER. The mutation of sec59-1 caused a protein N-glycosylation defect mediated ER stress resulting in increased levels of phospholipid, neutral lipid and sterol, whereas growth was reduced. In the sec59-1∆ cell, the N-glycosylation of vacuolar carboxy peptidase-Y (CPY) was significantly reduced; whereas the ER stress marker Kar2p and unfolded protein response (UPR) were significantly increased. Increased levels of Triacylglycerol (TAG), sterol ester (SE), and lipid droplets (LD) could be attributed to up-regulation of DPP1, LRO1, and ARE2 in the sec 59-1∆ cell. Also, the diacylglycerol (DAG), sterol (STE), and free fatty acids (FFA) levels were significantly increased, whereas the genes involved in peroxisome biogenesis and Pex3-EGFP levels were reduced when compared to the wild-type. The microarray data also revealed increased expression of genes involved in phospholipid, TAG, fatty acid, sterol synthesis, and phospholipid transport resulting in dysregulation of lipid homeostasis in the sec59-1∆ cell. We conclude that SEC59 dependent N-glycosylation is required for lipid homeostasis, peroxisome biogenesis, and ER protein quality control.

Role of SEC59 in glycerophospholipid metabolism. The SEC59-1 mutation causes the accumulation of misfolded or unfolded proteins induced ER stress. Also, the membrane lipids and storage lipids are majorly synthesized in the ER. Hence we sought to identify the impact of Sec59p defect in lipid metabolism. The phospholipids, phosphatidyl choline (PC), phosphatidyl ethanolamine (PE), phosphatidyl inositol (PI), phosphatidyl serine (PS) and minor phospholipids, lysophosphatidylcholine (LPC), phosphatidic acid (PA), and cardiolipin (CL) levels were significantly increased in sec59-1Δ cell compared to the wild-type (Fig. 3A). The time lapse study also shows that, PC, and PE were increased at 36 h and 48 h (Fig. 3B), also, minor phospholipids were significantly increased in the sec59-1∆ cell until 48 h, whereas they were reduced in the wild-type cell.
Choline is a precursor for phosphocholine synthesis 7,8 . The microarray data revealed that genes involved in PC, and PE synthesis (CHO2, OPI3, CPT1, EPT, and CKI1) were up-regulated, and rest of the genes (CDS1, CHO1, PSD1, PSD2, PCT1, and EKI1) were not significantly altered in sec59-1Δ cell (Fig. 3C). However, the SAM1 and SAM2 (S-adenosylmethionine synthetase) were up-regulated by ∼2.0 fold in sec59-1Δ cell, which acts as a methyl donor, and is required for the PC synthesis by the activation of Cho2p and Opi3p. The URA7 gene was down regulated by 1.22 fold, whereas URA8 (CTP synthase) was up-regulated in sec591-Δ cell, which are required for the synthesis of PC through the activation of the phosphocholine cytidyltransferase in CDP-choline pathway 9,10 . The phospholipid precursors (LPA and PA) are required for synthesis of the phospholipid through CDP-DAG pathway. The genes involved in LPA and PA synthesis (GPT2, AYR1, LOA1, and ICT1) were up-regulated (Fig. 3D); and the phospholipid levels were increased in the sec59-1∆ cell until 48 h even though the phospholipases (SPO14, NTE1, and PLB2) were up-regulated.
The lipid transporters and aminophospholipid flippase of the plasma membrane provide the substrate for phospholipid synthesis from exogenous supplementation. The expression of plasma membrane permeases (GIT1, DNF1, DNF2, and DNF3), and aminophospholipid flippase (NEO1, SEC14, PDR16, PDR17, and DRS2) were increased in the sec59-1Δ cell. The GIT1 encodes a permease, which is involved in uptake of glycerophosphoinositol and glycerophosphocholine 11,12 , and its expression was up-regulated ∼2.5 fold in our study. The GDE1 (glycerophosphocholine phosphodiesterase) hydrolyzes the GPC to form choline and glycerol phosphate and its expression was up-regulated. The DNF1 and DNF2 involved in the transport of PC, PS, and PE from extracellular From the log base values greater than 0.6 fold is statistically significant (*p < 0.05). (D) Wild-type, wild + YEp352, sec59-1∆ and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 30 °C until mid-log phase. Equal amount of cells was serially diluted and three μl of cells were spotted on SC or SC-U agar plates and incubated for 48 h at 30 °C and the growth pattern was observed. (S1 A) For growth curve analysis, cell growth was studied by measuring the A 600 nm of cells at frequent time intervals until 24 h. (S1B) We have analyzed 50 fields for each sample, the cells with membrane defects were counted, and the values are expressed per 100 cells. (E) Membrane staining. Wild-type, sec59-1∆ and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 30 °C up to mid-log phase, and the cell membrane stained using the lipophilic dye DiOC6, as described in the methods section. Fluorescence imaging was performed on confocal microscope (LSM 710-Zeiss) equipped with a 100/1.40 oil objective and an AxioCam MRM camera (Zeiss). Bar, 5 μm. (F) Microarray expression of genes involved in the cell wall synthesis and assembly. The vector pRS314 bearing UPRE-GFP was expressed in the wild-type and sec59-1∆ cells, and grown at 30 °C in synthetic complete media devoid of tryptophan (SC-Trp). Equal amount of cells were collected every 12 h to 48 h and the activated UPRE-GFP was measured using F-4500 fluorescence spectro-photometer (excitation at 488 nm and emission at 510 nm). The wild-type control cells were compared with the sec59-1∆ cells and the data represents the mean ± SD (*P < 0.05, **P < 0.01 and ***P < 0.001) of triplicates from three separate experiments. (B) Glycosylation status of carboxypeptidase-Y. Wild-type and sec59-1∆ cells were grown in YPD medium at 30 °C. The cells were collected every 12 h, lysed and an equal amount of protein (50 μg) was separated by 8% SDS page as described in detail in the methods section. Anti-CPY was used to check the glycosylation status of CPY expression in a timedependent manner. (C) The expression of Kar2 protein. The wild-type, wild + YEp352, sec59-1∆ and sec591-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium until mid-log phase at 30 °C. The cells were lysed and equal amount of protein (50 μg) were separated by 8% SDS page as described in the methods section. Anti-Kar2 antibody was used to analysis the Kar2p expression and anti-Pgk1 was used as a loading control. (S1C) The original Western blot picture was scanned by Bio-Rad gel doc with a black-and-white gel tray, and the protein band was quantified with Image-J software. Kar2p expression was normalized with Pgk1p, and the data represent the means ± SD (*P < 0.05) of triplicates from three separate experiments (D) Microarray expression of genes involved in protein glycosylation, protein folding, and ER stress marker genes. www.nature.com/scientificreports www.nature.com/scientificreports/ to maintain the amino-phospholipid asymmetry and their expression was significantly increased in the sec591-Δ cell (Fig. 3C). The SEC14 involved in the transport and regulation of PC, and PI homeostasis and its expression was also up-regulated. The expression of NEO1 (putative aminophospholipid translocase) was significantly up-regulated and is known to be involved in the phospholipid asymmetry of plasma membrane.
The inositol, inositol phosphate, and inositol-containing phospholipids are signaling molecules required for the synthesis of phospholipid and complex sphingolipids. In the sec59-1∆ cell, the INO1 (Inositol 3-phosphate synthase) gene was up-regulated, whereas the OPI1 (Negative regulator of phospholipid synthesis) was significantly down regulated. The PIS1 (phosphatidyl inositol synthase) gene was up-regulated, and the expression of phosphatidylinositol phosphate phosphatases (INP52, INP53, and SAC1) were significantly increased, whereas the CST26 gene (LPI acyltransferase) was down regulated. The expression levels of INP52 and INP53 were higher than PIS1 in the sec59-1∆ cell (Fig. 3C). These results suggest that SEC59 mediated glycosylation is required for phospholipid homeostasis and turnover in S. cerevisiae. SEC59 plays a crucial role in sterol metabolism. The endoplasmic reticulum is a main site for sterol synthesis. Sterol homeostasis is regulated by ER-localized sensors at the transcriptional and post-transcriptional level 13 . Sterol is essential for the viability of eukaryotic cells and membrane integrity but the accumulation of sterol is detrimental to membrane function 14 . Sterol is synthesized through a complicated pathway involves about ∼30 Erg proteins. In the sec59-1∆ cell, sterol levels were increased by ∼2.0 fold, ∼4.0 fold and ∼8.0 fold at 24, 36 and 48 h, respectively (Fig. 4B). From the sterol metabolism, 23-genes involved in sterol synthesis were The wild-type and sec59-1∆ cells were grown in YPD medium at 30 °C. Equal amount of cells were harvested in a time dependent manner (12 h to 48 h), lipids were extracted and the individual phospholipids separated by two-dimensional TLC and the phosphorus was quantified as described in the methods section. (C) Expression of genes involved in phospholipid metabolism. The wild-type cells at different time intervals were compared with their respective sec59-1∆ cell and the data represents the mean ± SD (*P < 0.05, **P < 0.01 and ***P < 0.001) of triplicates from three separate experiments. ( (Fig. 5B). The sterol regulatory element binding proteins (Upc2p and Ecm22p) regulate the sterol biosynthetic genes (ERG2, ERG3) and SIP2 is a negative regulator of sterol synthesis. In the sec59-1∆ cell, the ERG2 gene expression was up-regulated, whereas the ECM22 and SIP2 were down regulated.
In S. cerevisiae, six proteins with the StART (steroidogenic acute regulatory protein-related lipid transfer) like domain (YSP1, YSP2, SIP3, LAM4, LAM5, and LAM6) are involved in transport of sterol from plasma membrane to ER and conserved in eukaryotes 15 . The expression of YSP1, YSP2, and SIP3 were elevated; whereas the LAM4, LAM5, and LAM6 were not altered in the sec59-1∆ cell (Fig. 5C). The accumulation of sterol in the ER induces ER stress, influences protein misfolding, and activates sterol acetylation. The PRY1, PRY2, and SAY1 were involved in the export of the acetylated sterol and were up-regulated. Equal amounts of cells were harvested in a time-dependent manner (12 h to 48 h), lipids were extracted and the individual lipids were separated by TLC and quantified by densitometry scanning as described in the methods section. The wild-type cells at different time intervals were compared with their respective sec59-1∆ cell and the data represents the mean ± SD (*P < 0.05, **P < 0.01, and ***P < 0.001) of triplicates from three separate experiments.
Heme plays a central role in the transcriptional regulation of the genes involved in sterol homeostasis 16,17 , and are also required for enzyme activity of Erg3p (C-5 sterol desaturase), and Erg5p 18 . The expression of HEM1 and its transcription regulators (HAP2 and HAP3) were significantly reduced, whereas HEM15 (ferrochelatase) was elevated in the sec59-1∆ cell (Fig. 5C). The 5-aminolevulinate synthase (HEM1) is involved in heme biosynthesis and is homologous to human ALAS2, and its mutation causes X-linked sideroblastic anemia. The transcription factor HAP1 was up-regulated, whereas the ROX1 (oxygen responsive repressor) was down regulated (Fig. 5C). The transcription factor Hap1p positively regulates the ERG11 and negatively regulates the Rox1p. The HMG1 (HMG-CoA reductase) expression was significantly increased, and HMG2 was not altered in the sec59-1∆ cell. Finally, the over-expression of YEp352-SEC59 in sec59-1∆ strain restored the sterol levels as in the wild-type (Fig. 6B). These results suggest that SEC59 is involved in the sterol homeostasis.
Defect in SEC59 disrupts storage lipid homeostasis. The non-polar lipids TAG and SE are the energy reservoirs, and are stored as LD, which is surrounded by the phospholipid monolayer. The TAG is also required for membrane lipid synthesis. In the sec59-1 deletion strain, TAG (183%) and SE (178%) were increased, and subsequently, 1, 2-DAG, 1, 3-DAG, and free fatty acids (FFA) were also accumulated (Fig. 4B). The LD number and size were also significantly increased when compared to wild-type. The time-lapse study also showed that TAG, SE, and the LD number drastically increased in the sec59-1∆ strain (Fig. 4A,B), and subsequently, DAG, free fatty acids and sterol levels were also massively elevated when compared to wild type (Fig. 4B). The overexpression of YEp352-SEC59 in the sec59-1∆ cell restored the neutral lipids (TAG, SE, DAG, and STE), LDs, and FFA as in wild-type ( Fig. 6A-C). These results strongly reveal that Sec59p is involved indirectly in neutral lipid homeostasis and LD maintenance. The major LD component TAG is synthesized by diacylglycerol acyltransferase (DGA1) and lecithin cholesterol acyltransferase [LCAT/LRO1] 19,20 . In the sec59-1∆ cell, the DGA1 expression was significantly reduced whereas LRO1 and YJR098C expression were elevated. YJR098C has lecithin cholesterol acyltransferase (LCAT) motif (Accession ID: PF02450) but with an uncharacterized ORF and also could account for TAG accumulation in the sec59-1∆ cell (Fig. 5A). The TAG and lipid droplets are also involved in synthesis of membrane lipids and act as an energy source during starvation, and are hydrolyzed by TAG lipases. www.nature.com/scientificreports www.nature.com/scientificreports/ The major TAG lipases, TGL3, TGL5, and LPX1 (peroxisomal TAG lipase) were down regulated whereas as the minor lipases (TGL2 and TGL4) were up-regulated in the sec59-1∆ cell when compared to the wild-type (Fig. 5A). Another major lipid droplet component is SE, which is synthesized by sterol esterification through the enzyme acyl-CoA: sterol acyltransferases (Are1p & Are2p) 21 . The ARE1 expression was down regulated, whereas ARE2 expression was up-regulated when compared to the wild-type. The SE is mainly synthesized through the Are2p (oxygen-dependent sterol esterification), and is hydrolyzed by YEH1 and YEH2 22 . In the sec59-1∆ cell, the YEH1 gene was down regulated whereas the TGL1 and YEH2 were up-regulated (Fig. 5A). However, the major MAG The wild-type, wild + YEp352, and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 30 °C up to mid-log phase, and stained with Nile Red as described in the methods section. Fluorescence imaging was performed on confocal microscope (LSM 710-Zeiss) equipped with a 100×/1.40 oil objective and an AxioCam MRM camera (Zeiss). Bar, 10 μm. (B) The wild-type, wild + YEp352, and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 30 °C. Equal amount of cells were harvested in mid-log phase, lipids were extracted and the individual lipids were separated by TLC and quantified by densitometry scanning as described in the methods section. (C) We analyzed 50 fields for each sample and counted the LD numbers, and the values are expressed per 100 cells. The wild-type cell lipids were compared to sec59-1∆ cell and the data represents the mean ± SD (*P < 0.05, **P < 0.01, and ***P < 0.001) of triplicates from three separate experiments. www.nature.com/scientificreports www.nature.com/scientificreports/ lipase YJU3 was not altered but there was repression of MAG lipase ROG1 whereas the expression of MGL2 (MAG lipase) was slightly increased in the sec59-1Δ compared to the wild-type. These results strongly suggest that, dolichol kinase dependent protein glycosylation defect dysregulates non-polar lipid metabolism.

SEC59 mutation affects lipid phosphatases and their gene regulation.
In S. cerevisiae, five lipid phosphatases (PAH1, DPP1, LPP1, APP1, and PHM8) are involved in the DAG synthesis. The PAH1 encodes a major Mg 2+ dependent phosphatidate phosphatase (PA), which dephosphorylates PA to yield DAG and is involved in de-novo synthesis of TAG, and LD [23][24][25] . In the sec59-1∆ cell, the expression of DPP1 was significantly increased, and PHM8 was slightly increased whereas the PAH1 expression was significantly reduced, and there were no changes observed in LPP1 and APP1 expression. The ZRT1 (Zinc transporter) was up-regulated, whereas the ZRT3 was down regulated, which led to defect in the transport of zinc from the vacuole to cytoplasm (Fig. 5A). The PHO4 (transcription factor of phm8) was repressed, but the PHM8 expression was slightly increased in the sec59-1 deletion strain when compared to the wild-type. The zinc dependent vacuolar PA phosphatase (DPP1) and magnesium dependent LPA phosphatase (Phm8p) provides a substrate for the TAG formation in the sec59-1∆ cell.
Effect of protein glycosylation defect on fatty acids of membrane and storage lipids. In the sec59-1∆ cell, both phospholipids and neutral lipids were increased. In order to investigate the influence of protein glycosylation defect on fatty acids, we subjected the individual phospholipids (PC, PE, PI, and PS) and neutral lipids (DAG, TAG, and SE) to Gas chromatography/Mass spectrometry (GC/MS) analysis (Fig. 6B).
In the sec59-1∆ cell, palmitoleic acid (C16:1) and oleic acid (C18:1) level were predominantly increased, whereas palmitic acid (C16:0) and C18:0 (Stearic acid) levels were reduced in PC, PE, and PI. In the PS, C16:1, C18:1, and C18:0 were increased but C16:0 was not altered in the sec59-1∆ cell when compared to the wild-type (Fig. 7B). In the non-polar lipid, the C16:1 and C18:1 levels were increased whereas the C16:0 and C18:0 were decreased in TAG, DAG, and SE of the sec59-1∆ compared to the wild-type (Fig. 7D). There was a shift towards the un-saturation of fatty acids in the phospholipids and neutral lipids during the protein glycosylation defect in the sec59-1∆ when compared to wild type cell.
The shift towards more unsaturated membrane lipids in the sec59-1∆ cell would be accompanied by major rearrangements in lipid classes for maintaining membrane shape and fluidity during the glycosylation defects.

SEC59-1 mutation attenuates peroxisome biogenesis.
In the sec59-1∆ cell, free fatty acid, neutral lipids (TAG and SE), and LDs were increased, in order to elucidate the impact of SEC59 defect on peroxisome biogenesis. Peroxisomes plays a crucial role in β-oxidation of fatty acids 26 and 32 genes (peroxins) have been involved in peroxisome biogenesis 27,28 . The Pex3p is a peroxisomal membrane protein and highly conserved from yeast to mammalian system that regulates peroxisome dynamics and peroxisome biogenesis by interacting with Pex19p 29 . The microarray results showed that more than 12 major "peroxin" genes (PEX2, PEX3, PEX4, PEX5, PEX19, PEX10, PEX13, PEX14, PEX15, PEX18, PEX22, and PEX28) were down regulated whereas only PEX1 was up-regulated, and rest of the peroxin genes (PEX6, PEX7, PEX8, PEX11, PEX12, PEX17, PEX21, PEX27, and PEX29) were not significantly altered in the sec59-1∆ cell (Fig. 8B,C). The peroxisome biogenesis defect was further confirmed using PEX3-EGFP (enhanced green fluorescence protein) overexpression. The biogenesis and maintenance of peroxisomes from ER-derived membrane components required Pex3p, which is independent of the oleic acid induced peroxisome biogenesis and proliferation 30,31 . In the sec59-1Δ cell, discrete punctate of Pex3-EGFP was reduced in the peroxisomal structures whereas the large and puncta peroxisome structure was observed in wild-type (Fig. 8A). The Pex3-EGFP fluorescence was also drastically reduced in the sec59-1∆ strain. In order to determine if the peroxisome proliferation defect was eventually recovered in the sec59-1∆ cells we looked at time lapse experiments which showed that the reduced peroxisome size and discrete structure in the sec59-1∆ cell even at 48 h. The peroxisomes appeared as dispersed and fragmented structures with various sizes at 24 h and 36 h resulting in a dramatic loss of peroxisomes structure and subsequently, PEX3-EGFP moved to vacuoles at 48 h (Fig. 8A). From these results, we can conclude that defect in dolichol kinase plays a major role in peroxisome biogenesis and maintenance.

Discussion
The present study reports the link between protein glycosylation and lipid metabolism, and its impact on the ER protein quality control, and peroxisome biogenesis in S. cerevisiae. The growth of the Sec59-1∆ strain was significantly reduced even in permissible temperature compared to wild-type (Figs 1D and S1). During the mutation of Sec59-1, the N-glycosylation of vacuolar carboxypeptidase-Y (CPY) was severely affected, and simultaneously UPR was constitutively induced as an adaptive response to protect the cells from stress. The ER stress was increased as noted by increased expression of Kar2p which could be due to a protein glycosylation defect in the sec59-1∆ strain ( Fig. 2A,B). In the sec59-1∆ strain, the intra-cellular membrane aggregation and fragmented membrane vesicles were increased (Fig. 1E). This pleiotropic phenotype could be due to a protein glycosylation defect and a similar phenotype has been reported in previous study 32,33 . The microarray results revealed that 14 genes involved in cell membrane synthesis, morphology maintenance, and cell wall glycan synthesis were up-regulated in the sec59-1Δ cell. Among them the SCS3 (involved in ER membrane proliferation) was up-regulated whereas RTN1 (involved in ER membrane integrity and morphogenesis) was down regulated (2019) 9:14485 | https://doi.org/10.1038/s41598-019-51054-7 www.nature.com/scientificreports www.nature.com/scientificreports/ in the sec59-1∆strain (Fig. 1F). These results could be attributed to increased ER membrane proliferation, and intercellular membrane vesiculation in the sec59-1Δ strain. The reduced membrane integrity and morphogenesis was depicted by an increases of fluorescence with DiOC6 implying the alteration of plasma membrane in the sec59-1Δ cell (Fig. 2E and S1B). To protect the cells from ER stress, the misfolded or unfolded proteins activate UPR as an adaptive response increasing the ER chaperone which could promote refolding of the protein. During the mutation of Sec59-1, the ER chaperones and UPR regulatory genes were increased in order to protect the cells from ER stress (Fig. 2D). Further, there was increased expression of genes involved in the ERAD pathway (DOA10, DFM1, UBX4, UFD1, HRD3, USA1, and DER1) which can eliminate the misfolded or unfolded proteins from the ER for the cell survival under ER stress. www.nature.com/scientificreports www.nature.com/scientificreports/ Apart from the protein synthesis and quality control maintenance, ER plays a major role in synthesis of membrane phospholipids and sterol. Lipid homeostasis is regulated by ER sensors through the transcriptional and post-transcriptional mechanisms 13 . In S. cerevisiae, membrane phospholipid and storage lipid classes recruit DAG from the precursor phosphatidic acid (PA). During the exponential growth phase, DAG is predominantly channeled to phospholipid synthesis, and required for membrane formation, cell viability, and growth, whereas the phospholipid biosynthesis is reduced during stationary phase 34 . In contrast, the levels of all major phospholipids (PC, PE, PI, and PS) were significantly increased from mid-log phase to stationary phase in the sec59-1∆ cell compared to the wild-type (Fig. 3A,B). Also, the microarray results revealed that increased expression of phospholipid genes due to down regulation of negative regulator (OPI1) of phospholipid synthesis gene (Fig. 3C). This result supports our previous studies, which suggest that increased phospholipid level in protein glycosylation defect and ER stress phenotype strain 32 . The plasma membrane flippase and permease (GIT1, DNF1, and DNF2) were www.nature.com/scientificreports www.nature.com/scientificreports/ up-regulated in the sec59-1∆ cell implying that phospholipid flippases provide the substrates for the CDP-choline and ethanolamine pathway. The derepression of PE-methylation and Kennedy pathway genes could account for the increased phospholipid synthesis in the sec59-1∆ cell. The increased phospholipid synthesis could be a protective response by increasing the hydrophobic surface of the ER, thereby sequestering misfolded proteins in response to ER stress 35 .
In the sec59-1∆ cell, the increased level of PI by the upregulation INO1 and PIS1, and increased expression of genes involved in dephosphorylation of phosphatidyl inositol phosphates (INP52, INP53, SAC1) ( Fig. 3A-C). PI is the most predominant lipid in the ER, which protects the cells from stress by increased expression of INO1.
Apart from the phospholipid, fatty acids are playing a crucial role in cellular stress response and have been reported in both S. cerevisiae and mammals 36 . In the sec59-1∆ cell, the C16:1 and C18:1 level were predominantly increased in PC, PE, PI, and PS, whereas C16:0 and C18:0 were reduced, except PS where it was not altered. Apart from the ER, phospholipids are synthesized in mitochondria, vacuoles, and Golgi apparatus. During the Sec59-1 mutation, PE, and CL level were significantly increased at mid-log phase (Fig. S1D), and the expression of genes involved in PE and CL synthesis (EPT1, ALE1, FMP30, and CRD1) were upregulated. This could be attributed to the increased levels of PE and CL observed in the sec59-1Δ cell (Fig. 3A-C). Furthermore, the major mitochondrial fusion and fission gene expression were also affected. The mitochondrial heat shock proteins are associated with mitochondrial unfolded response (mito-UPR), and the genes expressions were up-regulated. These cellular complications induced the apoptosis genes in the sec59-1Δ cell (Fig. S1F). These results depict the dysfunction of the mitochondrial genes in the sec59-1Δ cell.
In S. cerevisiae, during the stationary phase, the DAG and fatty acids are predominantly channeled to TAG synthesis and stored as LD in the cytosol. In the sec59-1∆ cell, TAG and SE levels were drastically increased. Also, the sterol, 1, 2-DAG, 1, 3-DAG, and free fatty acids were increased when compared to the wild-type (Fig. 6A-C). The genes involved in the sterol metabolism and its regulatory genes were increased (Fig. 5B,C). The accumulation of sterol in the sec59-1∆ cell could be due to reduced transport from the ER to the plasma membrane by the down-regulation of ARV1 and increased transport of sterol from the plasma membrane to the ER by the derepression of YSP1, YSP2, and SIP3 (Fig. 5B). The LD number and size were increased as measured by the increased intensity of Nile red fluorescence when compared to the wild-type (Fig. 6C). Also, the time lapse study revealed drastic increases of TAG, SE, and LDs in the sec59-1∆ strain (Fig. 4A,B) which is supported by increased expression of LRO1 and ARE2, and down-regulation of major TAG and SE lipases (TGL3, TGL5, and YEH2 and TGL1) (Fig. 5A). Our previous study shows that LD and neutral lipids were reduced in protein glycosylation defect phenotype (Cax4∆) due to specific N-glycosylation defect. The Sec59-1∆ phenotype results controversy to Cax4∆ strain may due to defect in multiple glycosylation defects including, N-glycosylation, O-glycosylation, and GPI anchor synthesis. DPP1 (Zn dependent phosphatidate phosphatase) was up-regulated in the sec59-1∆ could be account for reduced transport of zinc by the down regulation of vacuolar membrane zinc transporter (ZRT3). These results strongly revealed the vacuolar lipid phosphatase plays a role for the supply of substrate (DAG) to TAG synthesis when the Zn transport get affected in the sec59-1∆ cell (Fig. 5A).
The Sec59-1 cells exhibit increases the unsaturated fatty acids, especially C16:1, and C18:1 in neutral lipids and phospholipids. The only gene that has been reported to involved in the synthesis of mono-unsaturated fatty acids is OLE1 (delta 9-fatty acid desaturase) 37 , but its expression was significantly reduced in the sec59-1∆ cell. This result suggests that there may be other desaturases in S. cerevisiae that are yet to be identified under the stress.
LD is a ubiquitous intracellular organelle, a storage depot that provides energy during starvation and involved in membrane lipid synthesis. LD biogenesis and growth are uniquely dependent on neutral lipid synthesis. LD's are often closely associated with various organelles, including peroxisomes, mitochondria, vacuole, and endoplasmic reticulum (ER). ER plays a potential role in peroxisome biogenesis and triggers the targeting signal of peroxisomal membrane proteins (PMPs) to peroxisomes. The increased lipolysis of TAG and SE increases free fatty acids, which undergo β-oxidation in the peroxisome. In sec59-1∆ cell, the TAG, SE, LD's, and FFA were increased and subsequently, the genes involved in β-oxidation (POX1, POT1, IDP2, ECI1, TES1, and DCI1) were also up-regulated (Fig. 8B).
The microarray result revealed that 12 genes involved in peroxisome biosynthesis were down regulated whereas the PEX1 was up-regulated, and the rest of the genes were not altered (Fig. 8B). Furthermore, the peroxisome biogenesis defect was confirmed by the reduced PEX3-EGFP fluorescence in the sec59-1∆ when compared to the wild cell. These results suggest that dolichol kinase dependent protein glycosylation directly or indirectly plays a role in peroxisome biogenesis and β-oxidation. Finally, we can conclude that impaired growth of the Sec59-1Δ cells is associated with constitutive unfolded protein response, ER stress, and dysregulation of lipid metabolism accounting for the pleiotropic phenotypes. Thus, the S. cerevisiae dolichol kinase controls membrane biogenesis by coordinating lipid homeostasis with protein quality control.

Materials and Methods
The yeast strains used in this study (WT-RSY252-MATa; his4-619; ura3-52 and RSY27 MATa; his4-493; ura3-52; suc-432; sec59-1∆ KanMAX4) were gifted by Prof. Randy Schekman (Howard Hughes Medical Institute and Department of Molecular and Cell Biology, University of California, USA) The plasmid YEp352-SEC59 and its control vector were gifted by Prof. Markus Aebi (Institute of microbiology, ETHZ, Zurich, Switzerland). The plasmid pIH024-PEX3-EGFP was a generous gift from Prof. Dr. Harald W. Platta (Institute of Biochemistry and Pathology, Department of Biochemistry, Germany).The anti-kar2 antibody was a gift from Prof. Jeffrey L. Brodsky (Department of Cell Biology, University of Pittsburgh, PA and USA), and anti-CPY antibody was gifted by Prof. Neta Dean (Department of Biochemistry and Cell Biology, Stony Brook University, NY and USA). Yeast nitrogenous base (YNB), yeast extract and peptone were obtained from Difco, the nitrocellulose membrane was purchased from Millipore (India Pvt. Ltd), and thin layer chromatography plates were obtained from Merck Growth study. The wild-type, wild + YEp352, sec59-1∆, and sec59-1∆ + YEp352-SEC59 cells were grown in YPD and SC or SC-U medium at 30 °C up to mid-log phase. Cell mass was measured at A 600 nm, and equal mass of cells were collected and serially diluted from 10 −1 to 10 −5 dilution and three μl of cells were spotted on (YPD and SC or SC-U) appropriate agar plates, and then incubated the plates for 48 h at 30 °C.
Lipid extraction and separation. Lipids were extracted from the yeast cells by the method of Bligh and Dyer 38 . Briefly, to the cell pellet 400 µl of methanol and 200 µl of chloroform were added and vortexed. Further, 400 µl of acidified water (2% orthophosphoric acid) was added and vigorously vortexed. The lipids were extracted with Chloroform and subjected to two-dimensional thin layer chromatography (TLC) with Silica Gel 60F254 TLC plates 39 . The phospholipid were separated using the following solvents: for first dimension chloroform/ methanol/ammonia (65:25:5, v/v) and for the second dimension, chloroform/methanol/acetone/acetic acid/water (50:10:20:15:5, v/v). Individual lipids were located by comparing the Rf value of the unknown with the Rf value of the standard. Spots corresponding to individual phospholipids were scraped off and quantified 40 . For analyzing the neutral lipids (TAG, and SE), the lipids were separated using petroleum ether/diethyl ether/acetic acid (70:30:1; v/v) with triolein and cholesteryl oleate as standards, and lipids were quantified with slight modifications 41 . Briefly, the plates were dipped in methanolic MnCl 2 solution (0.63 g MnCl 2 ·4H 2 O, 60 ml of water, 60 ml of methanol, and 4 ml of concentrated sulfuric acid), dried, and heated at 120 °C for 15 min. Densitometric scanning was performed at 500 nm with a CAMAG TLC Scanner. fatty acids analysis. The phospholipids and neutral lipids separated in the TLC were extracted using chloroform/methanol (2:1, v/v), and subjected to methanolysis form conversion to methyl esters using BF3/methanol 42 . Fatty acid methyl esters (FAME) were separated and quantified by gas chromatography mass-spectrometry (GC/ MS). The heptadecanoic acid methyl ester (C17:0) was used as an internal standard for fatty acid quantification.
nile Red staining. The wild-type, sec59-1∆, and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 37 °C and 180 rpm. The cells were harvested at mid-log phase and washed with 1x PBS and stained with lipophilic dye Nile Red. Fluorescence imaging was performed on the confocal laser scanning microscope (LSM 710-Zeiss) equipped with a 100/1.40 oil objective and an AxioCam MRM camera. DiOC6 Staining (3,3′dihexyloxacarbocyanine iodide). The wild-type, sec59-1∆ and sec59-1∆ + YEp352-SEC59 cells were grown in SC or SC-U medium at 180 rpm. Cells were harvested at mid-log phase and washed with 1x PBS then cells were stained with one μl of lipophilic dye DiOC6 (1 mg/ml ethanol). Fluorescence imaging was performed immediately after staining on the confocal laser scanning microscope (LSM 710-Zeiss) equipped with a 100/1.40 oil objective and an AxioCam MRM camera.
Fluorescence measurement of UPRE-GFP. The wild-type + pRS314-UPRE-GFP and sec59-1∆ + pRS314-UPRE-GFP plasmid bearing cells were grown in SC-Try medium at 180 rpm in 30 °C. The cells were collected every 12 h intervals (12 h to 48 h), then cell mass was measured at A 600 nm and equal amount (1.0 OD) of cells were resuspended with 1 ml 1x PBS buffer (8 g/L NaCl, 1.44 g/L Na 2 HPO 4 , 0.24 g/L KH 2 PO 4 , 0.2 g/L KCl). The GFP intensity was measured on an F-4500 Fluorescence Spectro-photometer (Hitachi). The measurements were carried out at 488 nm (excitation) and 510 nm for the emission wavelength.

Microarray analysis.
For the preparation of RNA, the wild-type and sec59-1∆ cells (0.1 OD 600 ) were grown until 48 h in YPD medium and harvested by centrifugation. Cell pellets were shock frozen in liquid nitrogen and stored at −70 °C until further use. Total RNA was isolated using an RNeasy Kit (Qiagen) including DNase I treatment according to the manufacturer's instructions. Yeast lysate was prepared by mechanical disruption of cells using a bead mill. The integrity of RNA was tested by agarose gel electrophoresis. The RNA concentration was estimated by the A 260 nm and the purity determined A 260 nm A 280 nm ratio. For expression profiling, 1 μg of total RNA was linearly amplified and biotinylated using the One-Cycle Target Labeling Kit (Affymetrix, Genotypic, Bangalore, India) according to the manufacturer's instructions. The labeled and fragmented cRNA (15 μg) were hybridized to Affymetrix Yeast 8.0 Gene Chip ® arrays (Affymetrix). After hybridization, the arrays were washed and stained in a Fluidics Station 450 (Affymetrix) with the recommended washing procedure. Biotinylated cRNA bound to target molecules was detected with streptavidin-coupled phycoerithrin, biotinylated anti-streptavidin IgG antibodies and again streptavidin-coupled phycoerithrin according to the manufacturer's protocol. Arrays were scanned using the GCS3000 Gene Chip scanner (Affymetrix) and GCOS 1.4 software. Scanned images