1. Introduction
A rigid cell wall (CW) is essential for protection of the yeast cells against negative environmental factors and plays a significant role in the stress response [
1,
2].
Complex of polysaccharides covalently linked to mannoproteins is a natural mega-glycoconjugate (mGC) that plays the central role as the CW structural component [
1,
2]. The main polysaccharide part of mGC consists of β-glucan (about 35–55%). The minor polysaccharide part of mGC is represented by chitin (about 1.5–6.0%), also being an important structural component [
1].
There are 30–50% of mannoproteins in the CW. They are highly mannosylated molecules with a protein part of 4–5% [
1]. Mannose chains are attached to asparagine (N) and serine/threonine (S/T) residues by way of N- and O-mannosylation respectively [
2]. CW mannoproteins can be divided into two groups: covalently attached to glucan as a part of mGC and non-covalently attached to mGC that can be extracted from the CW by treating with SDS under heating.
The group of non-covalently attached mannoproteins includes glucan remodeling enzymes [
1,
2,
3,
4] responsible for forming of mGC–glucanosyltransglycosylases (ncGTGs). These ncGTGs are essential for the protection of yeast cells against heating, drying and impact of inhibitors of CW assembly such as Congo Red, Calcofluor White and SDS [
5,
6,
7,
8,
9]. There are 4 ncGTGs in the CW: Bgl2, Scw4, Scw10 and Scw11. Among them the major ncGTGs are Bgl2 and Scw4 [
5,
7,
8,
9,
10].
Bgl2 is a major, constitutive and conservative mannoprotein not only in many species of yeast, but also in other fungi [
11]. Bgl2 isolated from
Saccharomyces cerevisiae CW demonstrated amyloid-like properties [
12,
13]. Bgl2 amino acid sequence has two potential sites of N-glycosylation (N-x-S/T), N202 and N284, and only one of them could be N-glycosylated [
14]. Still there are contradictory data on N-glycosylation in the literature. Authors of two unrelated studies found that N-glycosylation occurs on different sites [
15,
16].
Scw4 is also a constitutive and conservative mannoprotein among different fungi species [
11]. It is interesting that minor share of Scw4 molecules was recently identified as covalently linked mannoproteins [
17,
18]. Scw4 has also single N-glycosylation, determined at N89 position [
19].
In the absence of Bgl2 and Scw4 some GPI-anchored CW mannoproteins are not able to integrate into CW or attach correctly [
8,
10]. Also deletions of
BGL2 gene cause defects in CW assembly and the increase of the content of chitin in
S. cerevisiae CW [
20]. This compensatory response can be defined as “chitin repair” mechanism.
Ogataea parapolymorpha (
Hansenula polymorpha) cells, that have low level of chitin reparation after
BGL2 deletion, are characterized by disturbed formation of the bud scar [
21]. In mutant cells this area contains cytoplasmic cell material and numerous membrane invaginations [
21]. Simultaneous deletion of
SCW4 and
SCW10 or
SCW11 genes demonstrates slower growth rates and morphological abnormalities of the yeast cells [
6].
There is some data showing that various non-covalently bound proteins, ncGTGs in particular, can be extracted to varying degrees by various extractants used. At least part of Bgl2 can be extracted from the CW in Tris [
13] or in water under heating before [
12] and after [
22] lipid extraction. These facts allow us to conclude that ncGTGs have different mode of attachment to mGC of CW. However there is no information on the special association of Bgl2 as well as Scw4 molecules into different pools, for example, according to their peculiarities, post-translational modifications (PTMs) or mode of attachment.
There seem to be paradoxes in functioning of mGC-forming ncGTGs: these enzymes have to work on the whole area of the CW and remodel mGC to provide constant growth and division of the yeast cell, at the same time ncGTGs cannot move across mGC of the very rigid CW that is not fluid in contrast to plasma membrane, ncGTGs have to be in close contact with their substrate glucan by being firmly attached to it. Their uncontrolled activity can be hazardous for the cell leading to lytic phenotype caused by a weakened CW. In other words, ncGTGs are constantly sawing off the branch they are sitting on. Therefore, they have to reveal their mGC-forming glucan-remodeling activity only there and then, where and when it is necessary for the cell. It is still unclear how ncGTGs can carry out their functions being localized in the CW, outside of the cytoplasmic membrane on the border of the yeast cell and unstable environment characterized by a variable pH value and ion composition, which are, for example, metabolic factors participating in regulation activity of some enzymes inside the cell.
There is no hypothesis to explain these paradoxes. We hypothesized that the most attractive possible explanation could be different reversible post-translational modifications (PTMs) of ncGTGs’ molecules. Glutathionylation and phosphorylation, for example, are known as reversible modifications that alter or regulate functioning of proteins [
23,
24]. ncGTGs modified in a different way may have different properties—first of all different affinities for their substrate glucan and therefore have to be attached to it with varying degrees of strength.
In other words, PTMs intended to regulate the crosstalk between ncGTGs and their substrate glucan should be able to reduce or increase the strength of ncGTG attachment to glucan. Revealing of ncGTG pools extractable in different ways characterized by relevant set of PTMs could be an evidence of such correlations.
Such evidence is of a particular interest, as the data on the reversible PTMs of the yeast CW ncGTGs are insufficient in the literature as well as there is no information describing a correlation of PTMs of ncGTGs and the mode of attachment of these enzymes inside the CW.
Another possible explanation may be the location of ncGTGs inside the CW. We supposed that these enzymes have to be compartmentalized in the CW in local zones, for example, for better communications with plasma membrane corresponding microcompartments [
25] designated for targeted delivery of excretory vesicles. Such vesicles are migrating through the CW and containing different proteins including modifying enzymes that serve for insertion of reversible PTMs, for instance protein kinases and glutathione-S-transferases [
26,
27,
28].
Up until now there has been almost no information in the literature on the localization of ncGTGs in CW in general and especially the existence of microcompartments containing ncGTGs in CW. Our previously obtained results [
12] allow us to assume that at least the major conservative and constitutive yeast CW ncGTG Bgl2 should perform its functions in microcompartments on the entire surface of the cell. Arrangement of glucan-remodeling enzymes in the CW microcompartments could explain a poorly studied mechanism for increasing the surface area of yeast cells during growth similar to the principle of a “tortoise shell”, when individual parts of the surface increase in size along the edges, while the central part does not grow.
In this work we investigated whether the presence of PTMs in ncGTG molecules correlates with differences in strength and/or mode of attachment of these enzymes. We also made our efforts to reveal the localization and peculiarities of ncGTGs in the CW of S. cerevisiae with Bgl2 as a model and answer the question of whether it is diffusely distributed in the matrix of this organelle or assembled into microcompartments.
3. Discussion
Starting this work, we proceeded from the assertion that the ncGTGs involved in the formation of the new mGC fragments of the CW and in the remodeling of the existing one, are surrounded by their own substrate (mGC glucan) and are unable to move along it.
Being fixed in the matrix of the CW, ncGTGs hydrolyze glucan molecules, insert synthesized de novo oligosaccharides and glycosylated proteins and form mGC complex. Because of their essential functions these enzymes must be either active or inactive in strict accordance with the need. At the same time, mGC is both the object of the action of ncGTGs and a very important structural megamolecule ensuring stability, as well as taking part in dynamic changes of the cell envelope during cell growth, followed by continuous expansion of the CW. Therefore, it is obvious that ncGTGs should be active only when and where it is required.
In order to successfully achieve the co-ordinated work of the ncGTGs ensemble, two circumstances must be met. The first one is the precise regulation and the second is the coordinated control of functioning of ncGTGs localized in the CW. However, there is no recognized hypothesis on the issue.
We hypothesized that reversible modifications of ncGTG molecules may be the best way for such precise regulation. We also suggested that not diffuse, but compact localization of ncGTGs in the CW in local zones is important for the directed and synchronized operation of ncGTGs as it was demonstrated for yeast plasma membrane microcompartments [
25].
Compact localization of ncGTGs would also be suitable for the delivery of molecules that can regulate their functioning, for example, enzymes that carry out the reversible PTMs, and are transferred from the cells to the CW, most likely inside vesicles [
26,
27,
28].
The first significant step in testing our hypothesis was answering the question of whether there could be pools of ncGTG molecules with different PTMs in the CW. We also searched for a correlation of presence or absence of such modifications with the properties of ncGTG molecules. First of all, we studied whether the modifications were related to the strength and mode of attachment of ncGTG molecules to the CW.
In this work, we revealed 3 pools of ncGTGs in the CW of
S. cerevisiae cells that differ in the strength and mode of attachment. We called them “T”, “G” and “L” pools. In T pool as well as in L pool several ncGTGs were detected, but only in Bgl2 and Scw4 peptides were in sufficient quantities so as to allow us to estimate their PTMs (
Table 1). These results are supported by the data obtained earlier about predominance of these two ncGTGs in the CW [
5,
7,
8,
9,
10] and our results (
Figure 1 and
Table 1). It is important to note that some molecules of Scw4, unlike Bgl2, can covalently attach to mGC and integrate into it [
17,
18] thereby becoming invisible in our analysis. Other molecules of Scw4 exist in the CW as ncGTG. G pool is represented almost solely by one protein Bgl2 (
Table 1). There is a very slim chance that an “invisible” protein, non-digestible by trypsin, is present in G pool besides Bgl2 (according to amino acid sequences from Saccharomyces Genome Database [
33], most probably it is not ncGTG). The results obtained indicate that Bgl2 and Scw4 in these three pools have different PTMs (
Table 1).
In our work we did not find differences in the extent of N-glycosylation of Bgl2 extracted from the CW under various conditions. N-glycosylation is not considered a reversible modification. It was possible to assume that, while having the same extent of glycosylation (1 oligomannosyl N-glycan per 1 protein molecule), Bgl2 can be represented in the CW by the mix of N202 and N284 glycosylated molecules where N-glycosylation serves as an additional marker for its distribution either into T or G pools (
Figure 2). At the beginning of our work, the correlation between the glycosylation sites and distribution of Bgl2 between the pools was unclear because of contradictory N-glycosylation data in the literature [
15,
16]. In this work, it was shown that in Bgl2 only N202 is glycosylated. It is possible that N-glycosylation of this protein plays a very important role for its attachment to the CW, but this type of PTM is not related to the regulation of its incorporation into T or G pools.
We can conclude that phosphorylation together with a possible presence or absence of association of Bgl2 with other proteins seems to be a discriminating factor for inclusion of Bgl2 into T or G pools. Scw4 is not identified in G pool whereas in T pool multiple phosphorylation is characteristic of its molecules as well as for Bgl2. We suppose that Bgl2 enters the CW as L pool of PTM-free (except N-glycosylation) molecules, it then becomes phosphorylated to a varying extent and forms two pools: multi- and monophosphorylated ones. The first one includes, besides the phosphorylated molecules, the molecules modified and non-modified with glutathione, which are possibly conformationally labile. The second pool, according to the mode of its extraction and data from transmission and fluorescent microscopies, is represented by the molecules tending to fibrillate (
Figure 3) and can be approximated by the separate part of Bgl2.
We did not reveal Bgl2 fibrils in T pool where it is highly phosphorylated, as well as in L pool where it is phosphorylated to a small extent. One may say that its ability to fibrillate is affected by the presence of the other proteins because there are many different proteins in L pool (Ygp1, Tos1, Pry3, Hsp150, Eng1 in addition to proteins identified earlier [
22]), and some other proteins besides Bgl2 and Scw4 were identified in T pool (Cwp1, Tos1, Pho3, Eng1).
Nevertheless, taking into account a strong denaturing effect that the procedure of obtaining L pool causes, it is possible to consider the role of phosphorylation in the ability of Bgl2 to fibrillate similar to that for Rim4 protein [
34].
We found that Bgl2 in T pool contains three peptides with multiple phosphorylation, which probably did not allow Bgl2 from this fraction to demonstrate the ability to form fibrils. Therefore, this protein was extracted in relatively mild conditions. This assumption is also consistent with the fact that Bgl2 extracted in G pool by more rigorous treatment with GuHCl capable of dissolving amyloid proteins [
35,
36,
37] has only one phosphorylated peptide with one phosphorylated amino acid residue.
We also suppose that Scw4 enters the CW as a part of L pool with both modifications: C173-glutathionylated and T137-phosphorylated molecules, as well as without these modifications. At the same time, according to the literature there are two pools of Scw4 molecules in the CW that are covalently and non-covalently attached to glucan [
17,
18]. In our analysis we did not reveal Scw4 molecules being C173-glutathionylated and T137-phosphorylated in T pool. However, in T pool we revealed differently modified Scw4: C354-glutathionylated and multiphosphorylated molecules (
Table 1). The results obtained may suggest that different reversible modifications that we revealed for Scw4 as well as for Bgl2 may serve as a marker of their distribution in different pools.
Our results allow us to suggest that the lipid component in the CW is very similar to lipid droplets, previously described in yeast [
38,
39]. The outer layer of the lipid droplets is represented by phospholipids, the middle layer by ergosterol, and the inner layer contains neutral lipids [
39]. We identified all components of the lipid droplets in the CW before SDS treatment (
Table 2). It should be noted that the CW preparations before SDS treatment did not contain plasma membrane traces. After washing with SDS, phospholipids practically disappear from the CW, but ergosterol and neutral lipids remain (
Table 2). It is possible to suppose that treatment with SDS could lead to the removal of an outer layer of the lipid droplets.
Lipid droplets have been shown to be essential for the process of CW biogenesis in sporulating yeast [
40]. It is possible that lipid droplets also play an important role in the functioning of the CW of a vegetative cell. It is possible that lipid droplets are a lipid component that is associated with ncGTG, identified as L pool in this work.
We revealed that part of Bgl2 and Scw4 extracted in T pool had glutationylated C68 and C354 in their molecules respectively.
Glutathionylation plays an important role in a number of metabolic, signaling, and transcription processes, modulating the functioning of the proteins involved in these processes and protecting them from oxidative stress [
41]. However, relatively small amounts of glutathionylated proteins were found in bacteria and yeast compared to mammalian cells [
42]. C68 glutathionylation was reliably detected in some Bgl2 molecules in T pool, and the absence of this modification of C68 in Bgl2 in G pool may indicate different functions of these two pools of this protein. It is possible that Bgl2 obtained by extraction in Tris may undergo glutathionylation in response to oxidative stress or may be constitutive [
41]. The latter could be realized for protein–protein interaction and for regulation of enzymatic activity [
23,
41,
42].
We believe that such a modification may be important for the functioning of Bgl2, for example, it can stimulate or inhibit the interaction of Bgl2 molecules with each other or with other CW proteins. To understand the structural basis of the observed differences in the properties of Bgl2 from T and G pools, we simulated the structure of Bgl2 with and without glutathione (
Figure 4).
If the glutathione molecule is located inside the protein, significant changes in the structure of Bgl2 occur (
Figure 4A,B), part of the structure of the β-barrel is disrupted. The C-terminal portion of the Bgl2 molecule with the glutathione facing inwards is modified and becomes much more conformationally labile. This does not allow us to conclusively confirm the possibility of formation of a disulfide bond between C262 and C310 residues. Such an arrangement of glutathione with high probability can strongly influence the arrangement of amino acid residues in the zone of the active center of the enzyme or in the zone of its binding to the substrate and affect the activity of Bgl2. It was found that the Bgl2 fold with the external arrangement of glutathione is stable, as well as the Bgl2 molecule without this PTM (
Figure 4A,C,D). However, in this case, the formation of a disulfide bond between the C262 and C310 is definitely more difficult, since C262 is located inside the molecule, and this prevents C310 from approaching it. Probably the conformational mobility of the C-terminus plays a role in the formation of Bgl2 oligomers capable of enzymatic activity.
In our work we visualized Bgl2 on the surface of the whole cells, as well as in the isolated CW both with and without crosslinking of CW mannoproteins (
Figure 5). It is important to note that there has been no information available on the localization of this ncGTG in yeast CW until now.
It was crucially important that we identified Bgl2-zones in the CW, which we named microcompartments or patches, not only in isolated
wt cells (
Figure 5A,B), but also in the CW even after treatment of yeast cells with EDC crosslinker (
Figure 5D,E) that is able to connect protein molecules at a peptide bond distance. This assured us that consequent procedures aimed at visualization of these proteins did not affect their native localization. This fact allowed us to believe in the other results of visualization of the patches obtained without EDC pretreatment (
Figure 5G,H,J,K).
Using Bgl2 as an example, we showed that ncGTGs can be located irregularly on the cell surface: in separate patches (microcompartments). The presence of Bgl2 in large and small compact patches with inter-spot space before (
Figure 5D,E,G,H) and only small patches (in a larger amount on the CW) after (
Figure 5J,K) T pool extraction allows us to suppose that Bgl2 from G pool could be shielded by Bgl2 from T pool. Moreover, Bgl2 from G pool could be inside T pool.
The absence of possibility of lateral movement along the CW and abundant presence in close contact with its substrate—glucan of the mGC require special conditions of activity regulation and both serve as an evidence of rigid fixation of Bgl2 inside the CW.
In this work, we obtained convincing confirmation of the previously discovered fact that at least part of Bgl2 may be present in the CW in the form of molecules with a strong tendency to fibrillation. This part of the protein is extracted into GuHCl and, after extraction, tends to associate into fibrils.
In
Figure 6 we summarize our results based on the revealed correlations between the different PTMs and the strength/mode of attachment of Bgl2 in the CW also taking into consideration the results obtained in the experiments on visualization of the location of this ncGTG inside the CW.
The mosaic arrangement of GTGs (first of all, Bgl2) suggests that the growth and extension of the rigid lateral CW can be carried out according to the principle of growth of the tortoise shell, when the entire surface as a whole increases its area by increasing the area of individual sections while most of its surface remains unchanged. Our results give the first evidence of a mosaic arrangement of GTGs in the CW of yeast and reveal multiple post-translational modifications that correlate with their properties, primarily with the strength of attachment of GTGs in the CW. Together, these data confirm our assumptions and may be the key to further studies leading to an answer to the question of how the mGC remodeling enzymes ncGTGs function.
At the beginning of our experimental work, the mode of attachment and regulation of functioning of non-covalently bound glucanosyltransglycosylases of
Saccharomyces cerevisiae CW seemed to be an unsolved multipronged puzzle. We made assumptions regarding the key points—PTMs, localization, and attachment of these enzymes—and concentrated our efforts on that. We obtained the data that allow to fill being summarized with results obtained earlier [
5,
6,
7,
8,
9,
10,
18,
22,
43] in a part of the puzzle and are helpful in our planning further scientific investigations, namely, discovering the role of each modification for the yeast cells under different conditions, or whether enzymes can change localization during cell growth and division, as well as under stress, and, as well, whether other enzymes within the CW are arranged in microcompartments or not.
4. Materials and Methods
S. cerevisiae strains used in this work are presented in
Table 3.
4.1. Construction of Yeast Strains
To obtain potential Bgl2 N-glycosylation site mutants, PCR mutagenesis of plasmid-encoded
BGL2 gene [
7] was performed with Q5 site-directed mutagenesis kit (New England Biolabs, Moscow, Russia) according to manufacturer’s instructions. For construction of pBGL2-N202 plasmid with N202Q mutation 5′-CAAGGTCAAACCATGCAACAAGCTTCTTACTCATTCTTTGATGATATTATGC-3′ and 5′-CCAGTAGGAGAACGCGTTAG-3′ primers were used. For construction of pBGL2-N284 plasmid with S286A mutation 5′-GAAGATTGGAAGCCAAACACTGCAGGTACCTCTGATGTCG-AGAAG-3′ and 5′-ATCAAAGGCTTCAAAAACAATAAC-3′ primers were used. Obtained plasmids pBGL2-N202 and pBGL2-N284 were verified by sequencing and transformed into
bgl2Δ strain to produce N202-OE and N284-OE strains respectively.
4.2. Yeast Growth Conditions
BY4742 yeast strain was grown in liquid nutritious YPD medium (1% yeast extract, 2% peptone, 2% glucose). WT-OE, N202-OE, and N284-OE yeast strains were grown in an uracil-free synthetic medium [
45] supplemented with 2% D-galactose for the induction of
BGL2 under the control of the GAL10-CYC1 hybrid promoter. The cell cultures were grown for 19 h (log-phase) at 30 °C with agitation on an orbital shaker (New Brunswick, Moscow, Russia) at 200 rpm.
4.3. Yeast CW Isolation
Log-phase yeast cells were precipitated by centrifugation for 10 min at 1650×
g (Rotina, Moscow, Russia), washed twice with 0.05 M potassium-phosphate buffer pH 8.0 and disrupted in the shaker (Heidolph, Moscow, Russia) with glass beads (0.5 mm; Sigma, Moscow, Russia) under cooling. The extent of cell disruption was estimated with a light microscope (Opton, Moscow, Russia). CW preparations containing less than 0.1% of intact cells were used in the further work. CW were separated from the intracellular content by centrifugation at 2580×
g for 5 min. CW and cells formed the double-layer precipitate; CW forming the upper layer were carefully suspended in water and separated from the cells. CW were washed twice with water, twice with 1% sucrose, twice with 1 M NaCl, twice with 1% NaCl, and once with water. The amount of CW was estimated spectrophotometrically (absorbance at 540 nm, A540) [
13]. CW obtained by this procedure will be further named “untreated”.
4.4. Yeast CW Partial Deproteinization
For a partial deproteinization CW were treated with 1% SDS for 1 h at 37 °C. Deproteinized CW were separated by centrifugation at 2580×
g for 5 min and were washed five times with 0.2 M Na-Ac buffer pH 5.6, three times with n-butanol-water mixture 0.7:1 (
v/
v) and with water to remove traces of SDS [
13]. CW obtained by this procedure will be further named “purified”.
4.5. Glucanase Treatment of Cells and CW
Log-phase yeast cells were precipitated, washed twice with water, and incubated in a 3.7% paraformaldehyde in 1× PBS solution pH 7.4 for 20 min at room temperature and then overnight at 4 °C. Fixed cells were washed twice with 50 mM NH4Cl in 1× PBS solution pH 7.4 and twice with 1× PBS solution pH 7.4. After that yeast cells were treated with β-1,3-glucanase from Hordeum vulgare (Megazyme, Moscow, Russia) in 1.2 M mannitol for 4 h at 37 °C (in ratio 12.5 units of β-1,3-glucanase activity to 100 optical units of yeast cells A540), followed by washing twice with 1.2 M mannitol solution to remove β-1,3-glucanase and then were used as a preparation for immunofluorescence microscopy.
Purified CW were washed twice with 50 mM NH4Cl in 1× PBS solution pH 7.4 and twice with 1× PBS solution pH 7.4. Then CW were treated with β-1,3-glucanase from Hordeum vulgare (the same ratio) in water for 4 h at 37 °C, followed by twice washing with water to remove β-1,3-glucanase and then were used as a preparation for immunofluorescence microscopy.
4.6. CW Protein Crosslinking
Untreated CW were processed with 0.1% SDS for 15 min at room temperature, washed five times with 0.2 M Na-Ac buffer pH 5.6 and then with water. Crosslinker EDC (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride) (ThermoFisher Scientific, Moscow, Russia) was added to CW transferred to MES buffer in the ratio 1 optical unit CW (A540) in 10 µL buffer to the final concentration 2.5 or 5 mM. Obtained mixture was incubated for 2 h at room temperature. Then CW were separated by centrifugation and washed with water three times. Uncrosslinked CW proteins were removed with 3% SDS under boiling for 10 min. CW were washed five times with 0.2 M Na-Ac buffer pH 5.6 and with water. Obtained CW were treated with β-1,3-glucanase from Trichoderma sp. (Megazyme, Moscow, Russia) overnight at 37 °C (in ratio 0.9 units of β-1,3-glucanase activity to 100 optical units of yeast cells A540), followed by numerous washing off β-1,3-glucanase with water and then were used as a preparation for immunofluorescence microscopy.
4.7. Sequential Extraction of Non-Covalently Attached CW Proteins
4.7.1. Extraction of T Pool
Purified CW in a ratio of 1 optical unit (A540) to 10 µL of 0.1 M Tris solution pH 9.8 were incubated for 3.5 h at 30 °C. The extract was separated from the CW by centrifugation at 12,000× g (Eppendorf, Moscow, Russia) for 2 min.
4.7.2. Extraction of G Pool
CW after Tris extraction were washed with 0.1 M Tris twice and with Milli-Q H2O three times. Then CW were incubated in 6 M GuHCl pH 5.6 in a ratio of 1 optical unit of CW (A540) to 10 µL of 6 M GuHCl for 2 h at 30 °C and agitation 200 rpm. The extract was separated from the CW by centrifugation at 12,000× g for 5 min and dialyzed against water overnight. Dialysis tubes with a cut-off limit of 6–8 kDa (Serva, Moscow, Russia) were prepared by boiling in 10 mM EDTA (Sigma, Moscow, Russia) and then in water.
Tris extract was neutralized with 1.2 M Na-Ac buffer pH 5.6 to final molarity and pH of solution 0.8 M and 5.6, respectively. Tris and GuHCl extracts from CW of wt and bgl2Δ strains were treated with β-1,3-glucanase (0.125 units to 100 µL of extracts) from Hordeum vulgare (Megazyme, Moscow, Russia) overnight at 4 °C before immunofluorescence microscopy and for 7 h at 30 °C before TEM.
4.7.3. Extraction of L Pool
CW extracted with GuHCl were washed 8 times with Milli-Q H
2O and centrifuged for 5 min at 12,000×
g. Chloroform-methanol mixture (2:1;
v/
v) was added to the pellet of CW (20 optical units of CW A540 in 1 mL of mixture), thoroughly mixed and incubated for 1 h at 30 °C and agitation 200 rpm and then centrifuged at 12,000×
g, according to Bligh and Dyer [
46] with modifications. CW were washed with water until complete disappearance of chloroform in the supernatant. Proteins were extracted by adding Milli-Q H
2O (10 µL per 1 optical unit of CW A540) and heating at 100 °C for 5 min. The extracts were analyzed by LC-MS/MS.
4.8. Preparation of Cell Lysates
Cells of overexpressing strains WT-OE, N202-OE, and N284-OE were disrupted as described in the section “Yeast CW isolation.” Obtained preparations (cell lysates) of disrupted cells were equilibrated by total protein concentration according to Lowry [
47] with modifications. Samples were analyzed by polyacrylamide gel electrophoresis (PAGE) with and without endoglycosidase H (EndoH, Sigma, Moscow, Russia) treatment. Incubation with EndoH was carried out for 15 min at 37 °C in the ratio 0.0004 units of EndoH activity to 20 µL of sample.
4.9. Electrophoresis, Western Blot Analysis
PAGE was performed according to Laemmli [
48] with modifications (Laemmli buffer additionally contained 5% β-mercaptoethanol and 0.625 mM EDTA) in 4% concentrating and 12% resolving polyacrylamide gel [
13]. Various CW extracts were equalized by the optical density of CW at 540 nm. Cell lysates were equalized by the amount of proteins according to Lowry [
47]. PAGE was performed in the presence of prestained protein molecular weight markers (Fermentas, Moscow, Russia). Protein staining in gel was performed with Coomassie G-250 according to Peisker [
49] or with silver nitrate according to Gharahdaghi [
50] with modifications. To identify Bgl2 bands Western Blot analysis was performed according to Rekstina [
22] with modifications. Primary polyclonal antibodies against Bgl2 were raised in male BALB/c mice (SPF status) in the laboratory of Dr. O.S. Morenkov (Institute of Cell Biophysics, Russian Academy of Sciences, Pushchino, Russia) with PAGE-purified protein (40 μg per mouse) and were used in previous investigations [
12,
13,
22]. Secondary polyclonal rabbit anti-mouse IgG antibodies were labeled by horse radish peroxidase (Invitrogen, Moscow, Russia). Protein–antibody complexes were visualized by enhanced chemiluminescence using the ThermoFisher Scientific ECL system (ThermoFisher Scientific, Moscow, Russia).
4.10. LC-MS/MS Analysis
Sample preparation was performed as described previously with minor modifications [
51]. Sodium deoxycholate (SDC) reduction and alkylation buffer pH 8.5 were added to extracts containing 100 µg protein so that the final concentration of protein, Tris, SDC, TCEP, and 2-chloroacetamide were 1mg/mL, 100 mM, 1% (
w/
v), 10 mM, and 40 mM, respectively. This stage was omitted for samples that were prepared for the determination of post-translational modifications.
The solution was boiled for 10 min and after cooling down to room temperature, the equal volume of trypsin solution in 100 mM Tris-HCl pH 8.5 was added in a 1:100 (w/v) ratio. Digestion was performed at 37 °C overnight. Peptides were acidified to a final concentration of 1% trifluoroacetic acid (TFA) for SDB-RPS binding, and 20 µg was loaded on two 14-gauge StageTip plugs. Ethylacetate/1% TFA (125 mL) was added, and the StageTips were centrifuged at 200× g. After washing the StageTips using two wash steps of 100 µL ethylacetate/1% TFA and one of 100 µL 0.2% TFA consecutively, peptides were eluted by 60 µL of elution buffer (80% acetonitrile, 5% ammonia). The collected material was completely dried using a SpeedVac centrifuge (Savant, Moscow, Russia) and stored at −80 °C before LC-MS/MS analyses. Before analyses peptides were suspended in loading buffer (2% acetonitrile, 0.1% TFA) and sonicated for 2 min (Elmasonic S100, Elma, Moscow, Russia).
Approximately 1 µg of peptides was loaded for 2 h gradient. Peptides were separated on a 25-cm 75-µm inner diameter column packed in-house with Aeris Peptide XB-C18 2.6 µm resin (Phenomenex, Moscow, Russia). Reverse-phase chromatography was performed with an Ultimate 3000 Nano LC System (ThermoFisher Scientific, Moscow, Russia), which was coupled to the Q Exactive HF mass spectrometer (ThermoFisher Scientific, Moscow, Russia) via a nanoelectrospray source (ThermoFisher Scientific, Moscow, Russia). Peptides were loaded in buffer A (0.2% (v/v) formic acid) and eluted with a linear 120-min gradient of 4–45% buffer B (0.1% (v/v) formic acid, 80% (v/v) acetonitrile) at a flow rate of 350 nL/min. After each gradient, the column was washed with 95% buffer B for 5 min and reequilibrated with buffer A for 5 min. Column temperature was kept at 40 °C. Peptides were analyzed on a mass spectrometer, with one full scan (300–1400 m/z, R = 60,000 at 200 m/z) at a target of 3 × 106 ions, followed by up to 15 data-dependent MS/MS scans with higher-energy collisional dissociation (HCD) (target 105 ions, max ion fill time 60 ms, isolation window 1.4 m/z, normalized collision energy (NCE) 28%, underfill ratio 2%), detected in the Orbitrap (R = 15,000 at fixed first mass 100 m/z). Other settings: charge exclusion—unassigned, 1, >6; peptide match—preferred; exclude isotopes—on; dynamic exclusion—30 s was enabled. Each sample was analyzed by LC-MS/MS in the three biologic replicates.
4.11. Immunofluorescence Microscopy of Samples Stained with Antibodies
Images of yeast cells and CW treated in different ways were obtained with fluorescent confocal scanning microscope Leica TCS SP2 AOBS (Leica, Rostock, Germany). Images of material from protein extracts (T and G pools) were obtained with fluorescent confocal scanning microscope Carl Zeiss Axiovert 200M LSM 510 META (Zeiss, Moscow, Russia).
Yeast cells, CW, and protein extracts (T and G pools) were fixed on glass slides with 3.7% paraformaldehyde for 20 min at room temperature incubated in conditions preventing desiccation. Samples were stained with mouse primary polyclonal antibodies against Bgl2 (obtained as described in the
Section 4.9) and secondary polyclonal goat anti-mouse antibodies IgG labeled with Alexa-488 fluorophore (Invitrogen, Moscow, Russia) or with rabbit primary polyclonal antibodies against Gas1 and secondary polyclonal goat anti-rabbit antibodies IgG labeled with Alexa-647 fluorophore (Invitrogen, Moscow, Russia; Rostock, Germany). Antibodies to Gas1 were kindly provided by Dr. M.O. Agaphonov (Federal Research Center “Fundamentals of Biotechnology”, Russian Academy of Sciences, Moscow, Russia).
Cells, CW, and protein extracts of bgl2Δ strain were used as the antibody controls to cells and CW of wt strain. Untreated with EDC CW of wt strain were used as the control to EDC-crosslinked CW of wt strain.
4.12. Transmission Electron Microscopy (TEM)
Small volumes (2 µL) of protein extracts (T and G pools) from CW of wt and bgl2Δ strains were absorbed onto glow-discharged carbon-coated, Formvar-filmed 200-mesh copper grids overnight in conditions preventing desiccation. Negative-staining with 2% uranyl acetate solution was performed for 2 min. Grids were allowed to dry in a light-protected environment and were visualized on electron microscopes JEM-100B or JEM-1011 (JEOL, Moscow, Russia).
4.13. Determination of CW Lipids
Extraction of lipids from CW treated or untreated with SDS was performed according to Bligh and Dyer [
46] with modifications. CW were incubated in the mixture of chloroform-methanol mixture (2:1;
v/
v) for 2 h at 30 °C and 200 rpm. Obtained extract was separated from the CW by centrifugation at 12,000×
g (Minispin, Moscow, Russia) for 10 min. The pellet was incubated once more in the water-chloroform-methanol mixture (0.8:1:2;
v/
v/
v) and centrifuged. Both supernatants were combined. Chloroform and water in equal volumes were added to the total supernatant until the ratio of water-chloroform-methanol mixture (1.8:2:2;
v/
v/
v). The obtained mixture was intensively shaken and centrifuged. The lower chloroform phase was separated and evaporated. For removal of protein impurities, the resulting lipid fraction was dissolved in a small volume of benzene and centrifuged. The benzene fraction was evaporated and the resulting extract containing lipids was analyzed for the presence of phospholipids and neutral lipids.
Separation of phospholipids and neutral lipids was carried out by the method of thin layer chromatography (TLC) on glass plates for high-performance thin layer chromatography (HP-TLC) 10 × 10 and 10 × 5 cm with a fixed layer of silica gel (Merck, Moscow, Russia). Chromatography of phospholipids was carried out in a chloroform-methanol-formic acid solvent system (65:25:4, v/v/v). Neutral lipids were identified in a solvent system: hexane-diethyl ether-acetic acid (80:20:1, v/v/v). A double solvent system was used to separate the ceramides: chromatography was first carried out in diethyl ether and then in a chloroform-methanol-water system (40:10:1, v/v/v) 2/3 of the initial height. Phospholipids, ceramides, and neutral lipids (Sigma, Moscow, Russia) were used as standards.
Samples were applied as dots with a micro-syringe in volumes from 2 μL to 100 μL (lipid concentration in the mixture from ~2.5 mg/mL to ~5 mg/mL). After chromatography, the plates were dried. To identify spots on the plates, they were stained with 10% CuSO4 solution and 8% orthophosphoric acid. Then the plates were incubated at 160 °C during 10 min.
The amount of detected lipids was analyzed with Image J software. The spots corresponding to each detected lipid were calculated by intensity of their brightness. The amount of lipids was statistically estimated. The confidence interval was calculated using the Student’s t-test.
4.14. Bioinformatic Analysis
Initial Bgl2 model was placed into dodecahedral unit cell for use with periodic boundary conditions (cell parameters a = b = c = 64 Å, α = β = 60°, γ = 90°), solvated with TIP3P water [
53] and ionized with NaCl to 0.12M. Total energy was minimized for 5000 steps using steepest descent minimizer, then the system was equilibrated, with protein heavy atoms harmonically constrained, for 50,000 steps in NVT regime using Berendsen thermostat, followed by NPT equilibration for 500,000 steps. During the production run, temperature was controlled by Nose-Hoover method [
54] and pressure by isotropic Parrinello-Rahman coupling method [
55]. Temperature setpoint was 310 K. Data frames were recorded every 100 ps. Timestep used for equilibration and production was 2fs with LINKS method of H-bonds constraining. Total of 100 ns production trajectory was obtained.
Bgl2-glutathione conjugate model was built by docking. Glutathione conformations were clustered using DBSCAN method, and the most populated cluster was selected. Steered molecular dynamics was applied to assess the possibility of glutathione folding “inside” the Bgl2. Additional harmonic potential was applied to the glutathione center of mass to assess.
All simulations were performed with full-atom CHARMM36 force field [
56], using GROMACS software [
57] on 6-core 3.50GHz Intel(R) Xeon(R) CPU E5-1650 v3 workstation equipped with dual Quadro K2200 GPUs (Nvidia). VMD [
58] was used for MD trajectory analysis and visualization.