Highlights
-
•
Eukaryotic protein O-mannosyltransferases (Pmts) are divided into three subfamilies (1, 2 and 4).
-
•
The filamentous fungus Aspergillus nidulans has three Pmts, one from each subfamily.
-
•
PmtA, PmtB and PmtC form heterodimers in all pairwise combinations.
-
•
PmtC (subfamily 4) forms homodimers.
Abbreviations: HA, hemagglutinin; HOG, high osmolarity glycerol; Pmt/PMT, protein O-mannosyltransferase
Keywords: Aspergillus nidulans, Protein O-mannosyltransferase, O-mannosylation, Msb2, PMT
Abstract
Eukaryotic protein O-mannosyltransferases (Pmts) are divided into three subfamilies (Pmt1, Pmt2, and Pmt4) and activity of Pmts in yeasts and animals requires assembly into complexes. In Saccharomyces cerevisiae, Pmt1 and Pmt2 form a heteromeric complex and Pmt 4 forms a homomeric complex. The filamentous fungus Aspergillus nidulans has three Pmts: PmtA (subfamily 2), PmtB (subfamily 1), and PmtC (subfamily 4). In this study we show that A. nidulans Pmts form heteromeric complexes in all possible pairwise combinations and that PmtC forms homomeric complexes. We also show that MsbA, an ortholog of a Pmt4-modified protein, is not modified by PmtC.
1. Introduction
Protein O-mannosylation is a type of protein glycosylation found in prokaryotes and eukaryotes [1–3]. In eukaryotes, protein O-mannosyltransferases (Pmts) are integral membrane proteins localized in the endoplasmic reticulum (ER) [4,5]. Pmts transfer a mannose residue from dolichyl phosphate mannose to the hydroxyl residue of serine or threonine in secreted proteins facing the luminal side of the ER [6,7]. Further elongation of O-mannosyl glycans takes place in the Golgi using a different set of enzymes and GDP-sugar [1]. O-mannosylation is important for stability, localization and function of secreted proteins [8–10] and a total lack of O-mannosylation is lethal in eukaryotes. Pmts in eukaryotes are grouped into 3 subfamilies with names based on the Saccharomyces cerevisiae enzymes Pmt1, Pmt2 and Pmt4 [11]. S. cerevisiae has 7 Pmts; Candida albicans has 5 [12]; and Aspergillus nidulans and other filamentous fungi have 3, one from each subfamily [2,13].
In S. cerevisiae Ser/Thr-rich domains of secreted proteins are likely to be mannosylated and membrane-associated proteins are mannosylated by Pmt4 [14,15]. But beyond these generalizations, the consensus sequence directing O-mannosylation is not known [3]. Nonetheless, some targets of Pmts have been found empirically and these exhibit specificity toward individual target proteins [8,16]. However, there is evidence of limited substrate overlap. For example, the S. cerevisiae Pmt4 and Pmt1/Pmt2 complex mannosylate different domains of Ccw5p [17].
O-mannosyltransferase activity requires a Pmt complex [18–20]. In S. cerevisiae and S. pombe, members of the Pmt1 subfamily form heteromeric complexes with members of the Pmt2 subfamily [13,21]. The Pmt4 subfamily of S. cerevisiae forms a homomeric complex [13].
In previous work, both our group and another group independently showed that the filamentous fungus A. nidulans has three pmts each representing a different subfamily: PmtA from subfamily 2, PmtB from subfamily 1, and PmtC from subfamily 4. Both groups also showed that ΔpmtA, ΔpmtB, ΔpmtC and the double ΔpmtA ΔpmtB were viable and that each null mutant had a distinctive phenotype [22,23]. These results strongly suggested that either PmtA and Pmt B do not form complexes in A. nidulans as the orthologous Pmt2 and Pmt1 do in S. cerevisiae, or that such subfamily 1/subfamily 2 complexes are not required for viability in A. nidulans as they are in S. cerevisiae. In this study we tested the three A. nidulans Pmts for the ability to form heteromeric and homomeric complexes. We also examined modifications of the A. nidulans ortholog of S. cerevisiae Msb2, a HOG pathway osmosensor modified by Pmt4 [24,25].
2. Materials and methods
2.1. Aspergillus strains and media
The A. nidulans strains used in this study (Table 1) were incubated in complete and minimal media (CM and MM) with supplements as previously described [22]. Genetic manipulations were carried out using standard A. nidulans protocols as previously described [22].
Table 1.
A. nidulans strains and plasmids.
| Strain | Genotype/phenotype | Source or reference |
|---|---|---|
| A850 | argB2::trpC_B methG | FGSC⁎ |
| ATK08 | pyrG89 argB2::trpC_B pyroA4 ΔpmtA::AfpyrG | [22] |
| ATK16 | pyrG89 ΔpmtB::AfpyrG argB2 pyroA4 | [22] |
| ATK38 | pyrG89 wA3 argB2 pyroA4 ΔpmtC::AfpyrG | [22] |
| ATK42 | pyrG89 wA3 argB::trpC_B pyroA4 | This study |
| ATK45 | pyrG89 argB::trpC_B nkuA::Afpyro pyroA4 | [22] |
| ATK89 | pyrG89 argB::trpC_B nkuA::Afpyro pyroA4 pmtA::S-tag-AfpyrG | This study |
| ATK95 | pyrG89 argB2::trpC_B pyroA4 ΔpmtA::Afpyro | This study |
| ATK103 | ATK38::AfargB-gpd(P)- ANID_01359-S-tag | This study |
| ATK104 | pyrG89 wA3 ΔpmtB::Afpyro argB2 pyroA4 | [22] |
| ATK154 | pyrG89 argB::trpC_B nkuA::Afpyro pyroA4 pmtC::HA-tag-AfargB | This study |
| ATK165 | pyrG89 argB::trpC_B pyroA4 pmtA::S-tag-AfpyrG pmtC::HA-tag-AfargB | This study |
| ATK168 | pyrG89 ΔpmtB::Afpyro argB::trpC_B pyroA4 pmtA::S-tag-AfpyrG pmtC::HA-tag-AfargB | This study |
| ATK172 | A850::AfargB-gpd(P)-ANID_07041(MsbA)-S-tag | This study |
| ATK177 | ATK16::AfargB-gpd(P)-AN ID_07041(MsbA)-S-tag | This study |
| ATK179 | ATK08::AfargB-gpd(P)-AN ID_07041(MsbA)-S-tag | This study |
| ATK184 | ATK38::AfargB-gpd(P)-AN ID_07041(MsbA)-S-tag | This study |
| ATK187 | pyrG89 pmtB::HA-tag-AfargB argB::trpC_B nkuA::Afpyro pyroA4 | This study |
| ATK192 | pyrG89 pmtB::HA-tag-AfargB argB::trpC_B pyroA4 pmtA::S-tag-AfpyrG | This study |
| ATK193 | pyrG89 pmtB::HA-tag-AfargB argB::trpC_B pyroA4 pmtC::S-tag-AfpyrG | This study |
| ATK195 | ATK149 X ATK104 | This study |
| ATK200 | pyrG89 pmtB::HA-tag-AfargB argB::trpC_B pyroA4ΔpmtA::Afpyro pmtC::S-tag-AfpyrG | This study |
| ATK208 | pyrG89 argB::trpC_B pyroA4 pmtC::S-tag-AfpyrG | This study |
| ATK211 | ATK177 X ATK95 | This study |
| ATK217 | pyrG89/pyrG89 argB::trpC_B/argB::trpC_B pyroA4/pyroA4 pmtC::S-tag-AfpyrG/pmtC::HA-tag-AfargB | This study |
| Plasmids | ||
| pAfargB2 | Ampr argB2 | G. S. May |
| pAO81 | GA4-S-Tag AfpyrG | [26] |
| pDV2 | Ampr argB-gpd(p)-ccdB-sgfp | [28] |
| pMT-3xHA | Ampr argB-alcA(p)-ccdB-3xHA | [28] |
| pFNO3 | Kanr GA5-GFP AfpyrG | [26] |
| pTK74 | pDV2::ANID_07041-S-tag | This study |
Fungal Genetics Stock Center School of Biological Sciences, University of Missouri, Kansas City, Kansas City, Missouri, USA
2.2. Construction of tagged Pmts
Strains bearing single copy epitope tagged Pmts were constructed by fusion PCR [26] using primers listed in Table 2. Primer names in Table 2 indicate tag identity and whether primers are upstream or downstream of the designated pmt gene. Amplicons were purified and transformed into ATK45. Homologous integration resulting in strains bearing a single tagged Pmt replacing the original Pmt was verified by PCR and Southern. All strains constructed along with detailed genotypes are shown in Table 1.
Table 2.
Primers.
| Primer name | Sequence |
|---|---|
| Pmt2FWent | CACCATGGCTGAAATTGGCTTTG |
| Pmt2RVnostop | GTTAGCGATTCGCCAACCG |
| pmt2-GA4Fw | GAGCGGTTGGCGAATCGCTAACGGAGCTGGTGCAGGCGC |
| Stag-pmt2RV | GAACTGTAACGCTCTATCACGGTTCCTGTCTGAGAGGAGGCACTGATGC |
| pmt2DnFW | GAACCGTGATAGAGCGTTACAGTTCCCGTTG |
| pmt2DNRV | GCGATCCCAATCCTTCCTATCTCTGTCATC |
| PMT2F301 | CCGAATCGGCCGCTCAAACATCGTG |
| PMT2R6311 | TGCGACTGGGCGAGAAAGCGTGAGG |
| up177PMT2F | TCCGCAGTCGCCCGCCAGATATGAG |
| StagR | GCGCCAATTGCTGTTGCCAGGTGAGG |
| PMT11093F | GTGCTTGGTCAATCCGGGAGAGGG |
| Pmt1RVnostop | ACGGGTCGCCTCGTTGCCTTC |
| PMT1-GA5F | GCAACGAGGCGACCCGTGGAGCTGGTGCAGGCGC |
| argBAf-PMT1DnR | CAACAGACCGACATATTATCTCATAGGATTTTCCCCTTG |
| Pmt1dn-FW | GATAATATGTCGGTCTGTTGAACTACCTGCC |
| PmtBdnR2 | CTGAGAGGGAGTTCGGAAACGTGC |
| 463Pmt4F | GTCCCCTATGTCGCCCTCCG |
| Pmt4RVnostop | TTTCGCGAAGTGCAAGTCATAGC |
| Pmt4GA4F | CTATGACTTGCACTTCGCGAAAGGAGCTGGTGCAGGCGCTGGAG |
| pyrGAFRV-Pmt4SPRV | AGAACGAGATGAGATCAGATGCTCCTGTCTGAGAGGAGGCACTGATGCG |
| Pmt4dnFW | GAGCATCTGATCTCATCTCGTTCTCTCCC |
| Pmt4dnRV | CTATCCACGGTATGAGCTGAGCGAGTAATG |
| argBpmt4RV | AGAACGAGATGAGATCAGATGCTCCGATTTTCATAGGATTTTCCCCTTG |
| StagR | GCGCCAATTGCTGTTGCCAGGTGAGG |
| MSB2entF | CACCATGGTTTCCCAGACGGCTC |
| MSB2-StagR | GCGCCTGCACCAGCTCCGTTCCATCCCAGAGAGTTC |
| StagF | GGAGCTGGTGCAGGCGCTGGAGC |
| StagR | GCGCCAATTGCTGTTGCCAGGTGAGG |
2.3. Construction of S-tagged Pmt target proteins
The GA4 S-tag fragment with stop codon was amplified from pAO81 and the gene of interest was amplified from the start codon to one codon before the stop codon. Amplicons were fused by PCR, ligated into the pENTR/D-TOPO vector using the pENTR/D-TOPO Cloning Kit (Invitrogen Co., CA) and transferred into the pMT-DV2 destination vector using Gateway LR Clonase II (Invitrogen Corp., CA) and transformed into A850 and Δpmt strains. All strains constructed along with detailed genotypes are shown in Table 1. All primer sequences are shown in Table 2. Primer names in Table 2 indicate tag identity and whether primers are upstream or downstream of the designated pmt gene.
2.4. Membrane fraction preparation
1 × 108 conidia/ml of the specified tagged-PMT strain were inoculated to CM (50 mL for target protein extraction and 1 L for immunoprecipitation) and shaken at 200 rpm and 30°C for 8 h. Mycelia were filtered, washed with cold stop buffer (0.9% NaCl, 1 mM NaN3, 10 mM EDTA, 50 mM NaF, pH 7.0), and ground in liquid nitrogen. Two milliliters of cold extraction buffer (50 mM Tris–HCl, pH 7.5, 0.3 mM MgCl2 plus protease Inhibitors (Complete, Mini, EDTA-free; Protease Inhibitor Cocktail Tablets, Roche) were added to 1 g ground mycelia and vortexed for 10 min at 4 °C. The cell suspension was centrifuged at 500×g for 10 min at 4°C. The supernatant was collected and centrifuged for 30 min at 20,000 rpm at 4 °C (Sorvall SS34 rotor). One mL buffer containing 50 mM Tris–HCl, pH 7.5, 7.5 mM MgCl2, and 15% glycerol was added per 1 mL of pellet and stored at −80°C. Protein was quantified with RC DC Protein Assay Kit (Bio-Rad Laboratories, CA) using bovine serum albumin as a standard.
2.5. Immunoprecipitation
Immunoprecipitation methods were adapted from Girrbach and colleagues [7]. Twenty milligrams of membrane fraction was solubilized in 4 mL of lysis buffer (20 mM Tris–HCl, pH 7.5, 140 mM NaCl, 0.3 mM MgCl2, 10% glycerol, 0.35% sodium deoxycholate, 0.5% Triton X-100 plus Protease Inhibitor (Complete, Mini, EDTA-free; Protease Inhibitor Cocktail Tablets, Roche). One hundred μL of agarose immobilized anti-epitope tag antibody slurry was added per 20 mg of protein. Agarose immobilized rabbit anti-S tag or rabbit anti-HA antibody was used for immunoprecipitation (Immunology Consultants Laboratory, Inc., Newberg, OR). Incubation with the solubilized membrane fraction was carried out at 4°C on a rocker for 2 h followed by 5 washes at 4°C with equal volume of cold lysis buffer and one wash with 1 mL cold Tris-buffered saline. Agarose beads were resuspended in 50 μL cold Tris-buffered saline. The bound proteins were eluted with 3 × SDS loading dye by 5 min at 95°C.
2.6. Western blot analyses
20 μL per lane of epitope-tagged Pmt eluent was loaded into 2 gels to be further probed with anti-S tag or anti-HA antibody. The proteins were resolved on 4–20% gradient Tris–HEPES SDS gels (Pierce Protein Gels, Thermo Fisher Scientific Inc., IL) and transferred to a nylon membrane. S-tagged Pmts were detected with mouse anti-S-tag antibody (1:5000) (Abcam Inc., MA), followed by Sheep anti-Mouse IgG ECL antibody coupled to horseradish peroxidase (1:5000) (GE Healthcare, NJ). HA-tagged proteins were detected with mouse anti-HA-tag antibody (1:5000) (Invitrogen Co., CA), followed by Sheep anti-Mouse IgG ECL antibody coupled to horseradish peroxidase (1:5000) (GE Healthcare, NJ).
For Pmt target protein analyses, an equal volume of 2 × SDS–PAGE sample buffer [27] was added to membrane fractions and heated for 5 min at 95°C. Forty μg solubilized protein were loaded per lane and detected as described above.
3. Results and discussion
To investigate Pmt complex formation in A. nidulans, we constructed the following strains with C-terminal purification tags: ATK89 (PmtAS), ATK187 (PmtBHA) and ATK154 (PmtCHA). Because we were concerned that over-expression of individual Pmts might alter interactions within complexes, we constructed these strains by replacing native Pmts with the corresponding tagged protein using homologous integration. In these strains, the tagged Pmt fusion constructs were integrated at native Pmt loci behind native Pmt promoters resulting in expression levels as close to wildtype as possible. Correct integrations were confirmed by PCR and Southern hybridization. All tagged Pmt strains showed wildtype phenotypes at 30°C and 42°C (data not shown). The S and HA tags are both small, with deduced molecular mass of 2.26 kDa and 3.98 kDa, respectively. The predicted molecular mass of PmtAS is 86.86 kDa, of PmtBHA is 108.98 kDa, and of PmtCHA is 92.28 kDa.
Attempts to detect tagged Pmts by western blot of total membrane fractions were unsuccessful, suggesting low levels of fusion protein. Therefore, immunoprecipitation of solubilized membrane fractions with agarose-immobilized anti-HA or anti-S tag antibodies was used, followed by SDS–PAGE and immunoblot (Fig. 1). The PmtAS strain yielded two bands with apparent molecular mass of approximately of 68 kDa and 165 kDa. The PmtBHA strain yielded two bands with apparent molecular mass of approximately 120 kDa and 180 kDa. The PmtCHA strain yielded two bands with apparent molecular mass of approximately 80 kDa and 160 kDa. In all cases, the lower band is likely the Pmt based on predicted molecular mass, strength of the signal and reproducibility. Pmts isolated from yeasts and animals frequently show apparent molecular masses that are smaller or larger than predicted [13,18]. Girrbach and Strahl reported similar double band patterns in S. cerevisiae blue native PAGE experiments with the higher bands resulting from heteromeric PMT complex formation [13]. However, our experiments included SDS so that proteins are expected to be denatured. Further these bands always appear at the same molecular mass for specific Pmts and their presence in strains carrying a Pmt deletion along with a tagged Pmt suggests that they are not heterodimers. A similar double band pattern was reported when the human PMT ortholog Pomt1 was expressed in HEK293T cells. In this case the authors speculated that the upper band resulted from protein aggregation due to Pomt1 over-expression [18]. Though the Pmts in our study were all expressed from endogenous promoters rather than over-expressed, it is possible that the higher bands seen in our immunoblots resulted from aggregation of these hydrophobic proteins during processing. It is also possible that the upper bands represent homodimers of tagged Pmts.
Fig. 1.

Immunoprecipitation of Epitope-tagged Pmts. Western blot analyses of immunoprecipitates from ATK89 (PmtAS), ATK187 (PmtBHA) and ATK154 (PmtCHA). Proteins were expressed from their native loci under the control of their endogenous promoters. Solubilized membrane-enriched fractions were immunoprecipitated with the corresponding agarose immobilized antibody against the epitope tag. Immunoprecipitates were resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (two blots on right). Arrows indicate Pmts.
3.1. PmtA (subfamily 2) forms heteromeric complexes with PmtB (subfamily 1)
To investigate whether A. nidulans PmtA and PmtB form a heteromeric complex like their S. cerevisiae orthologs, Pmt2 and Pmt1 [13], we derived a PmtAS PmtBHA strain (ATK192) from sexual crosses. Immunoprecipitation was performed with agarose-immobilized anti-HA antibodies and agarose-immobilized anti-S tag antibody. The immunoprecipitate was divided into two aliquots. One aliquot was probed with anti-S tag antibody and the other was probed with anti-HA antibody. When probed with anti-S tag antibody, the same bands that were seen in the PmtAS strain were visible (approximately 68 kDa and 165 kDa) (Fig. 2). When probed with anti-HA antibody, the same bands that were seen in the PmtBHA strain were visible (approximately 120 kDa and 180 kDa) (Fig. 2). Reciprocal experiments performed using anti-HA antibodies for immunoprecipitation before immunoblotting showed very high background, though the results were consistent with heteromeric complex formation by PmtA and PmB (data not shown). Our results show that PmtA and PmtB form heteromeric complexes in A. nidulans.
Fig. 2.

Immunoprecipitation of PmtA-PmtB heteromeric complexes. Solubilized membrane-enriched fractions from ATK192 (PmtAS PmtBHA) were immunoprecipitated with agarose immobilized anti-S tag antibody. Co-immunoprecipitates were resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (right). Arrows indicate Pmts.
3.2. PmtA (subfamily 2) forms heteromeric complexes with PmtC (subfamily 4)
In S. cerevisiae, Pmt2 and Pmt4 do not form complexes [13]. To investigate the interactions between the A. nidulans orthologs, strains carrying PmtAS and PmtCHA in the presence of PmtB (pmtA::S tag, pmtC::HA, ATK165) and absence of PmtB (pmtA::S tag, pmtC::HA, ΔpmtB; ATK168) were derived from sexual crosses. Immunoprecipitations were performed with agarose-immobilized anti-S tag antibody as described above. When probed with anti-HA antibody, Western blots of immunoprecipitates from both strains showed a band of approximately 80 kDa whether or not PmtB was present (Fig. 3). Reciprocal experiments performed using anti-HA antibodies for immunoprecipitation before immunoblotting were consistent with heteromeric complex formation by PmtA and PmtC (Fig. S1). Our results show that A. nidulans PmtA forms a heteromeric complex with PmtC in the presence and absence of PmtB.
Fig. 3.

Immunoprecipitation of PmtA-PmtC heteromeric complexes. Solubilized membrane-enriched fractions from ATK168 (PmtAS, PmtCHA, ΔpmtB; lane 1) and ATK165 (PmtAS, PmtCHA; lane 2) were immunoprecipitated with agarose immobilized anti-S-tag antibody. Co-immunoprecipitates were resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (right). Arrows indicate Pmts.
3.3. PmtB (subfamily 1) forms heteromeric complexes with PmtC (subfamily 4)
To investigate the interactions between PmtB and PmtC, a strain bearing PmtBHA and PmtCS in the presence of PmtA (pmtB::HA, pmtC::S tag; ATK193) and absence of PmtA (pmtB::HA, pmtC::S tag, ΔpmtA; ATK200) were derived from sexual crosses. Immunoprecipitation and Western blots were performed as described above. Immunoblots probed with anti-HA antibody showed a band of approximately 120 kDa whether or not PmtA was present (Fig. 4). Reciprocal experiments performed using anti-HA antibodies for immunoprecipitation before immunoblotting showed very high background, though the results were consistent with heteromeric complex formation by PmtB and PmtC (data not shown). Our results show that A. nidulans PmtB forms a heteromeric complex with PmtC in the presence and absence of PmtA.
Fig. 4.

Immunoprecipitation of PmtB-PmtC heteromeric complexes. Solubilized membrane-enriched fractions were immunoprecipitated with agarose immobilized anti-S-tag antibody. Lane 1 is co-immunoprecipitate from ATK200 (PmtBHA, PmtCS, ΔpmtA), lane 2 is co-immunoprecipitate from ATK193 (PmtBHA, PmtCS). Co-immunoprecipitates were resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (right).
3.4. PmtC (subfamily 4) forms homomeric complexes
In S. cerevisiae, Pmt4 functions as a dimer, not a monomer [13]. To investigate whether the A. nidulans ortholog, PmtC, is a monomer or forms homomeric complexes in A. nidulans, a stable diploid PmtCS PmtCHA strain (pmtC::S tag/pmtC::HA; ATK217) was constructed. Immunoprecipitation and Westerns were performed as described above and probed with anti-S tag antibody or anti-HA antibody. In both cases the same bands that were seen in the PmtCHA strain were visible (approximately 80 kDa and 160 kDa) (Fig. 5). Reciprocal experiments performed using anti-HA antibodies for immunoprecipitation before immunoblotting were consistent with homomeric complex formation by PmtC (Fig. S1). Our results show that A. nidulans forms a PmtC homomeric complex as is true for the orthologous S. cerevisiae Pmt4 [13].
Fig. 5.

Immunoprecipitation of PmtC-PmtC homomeric complexes. A solubilized membrane enriched fraction from diploid strain ATK217 (PmtCS coexpressed with PmtCHA) was immunoprecipitated with agarose immobilized anti-S tag antibody. Co-immunoprecipitate was resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (right).
Genetic evidence from our previous work suggests that this PmtC–PmtC homomeric complex is sufficient to support growth because the ΔpmtA ΔpmtB strain is viable [22]. Despite repeated attempts, we were unable to recover appropriately marked strains to investigate PmtA–PmtA and PmtB–PmtB homomeric complex formation. However, if PmtA–PmtA and PmtB–PmtB homomeric complexes exist, neither appears to be sufficient to support A. nidulans growth in the absence of the other Pmts since both the ΔpmtA ΔpmtC and ΔpmtB ΔpmtC double mutants are inviable.
3.5. Pmt complexes are formed in vivo
To eliminate the possibility that the A. nidulans Pmt complexes we detected might result from nonspecific aggregation of Pmts in solubilized membrane fractions, we performed the following control experiment. Membrane fractions were isolated from strains in which S tagged and HA tagged Pmts were co-expressed, and solubilized. Membrane fractions were also isolated separately from strains in which S tagged Pmts were expressed, and from strains in which HA tagged Pmts were expressed. Isolated membrane fractions were combined and solubilized. Agarose-immobilized anti-S tag antibody was used to immunoprecipitate protein from solubilized membranes of the strains co-expressing S tagged and HA tagged Pmts or from the combined and solubilized membranes of the strains separately expressing tagged Pmts. Identical aliquots were probed with anti-S tag antibody or anti-HA antibody. The co-expressed Pmts co-precipitated, but the separately expressed and combined Pmts did not, suggesting that the Pmt complexes we detected in earlier experiments were formed in vivo and not the result of nonspecific protein aggregation (Fig. 6).
Fig. 6.

Pmt complexes form in vivo. Solubilized membrane enriched fractions or combined solubilized membrane fractions were immunoprecipitated with agarose immobilized anti-S tag antibody. Lane 1 is co-immunoprecipitate from ATK192 (PmtAS, PmtBHA). Lane 2 is co-immunoprecipitate from membrane fractions of ATK89 (PmtAS) combined with membrane fractions of ATK187 (PmtBHA). Lane 3 is co-immunoprecipitate from ATK165 (PmtAS, PmtCHA). Lane 4 is co-immunoprecipitate from membrane fractions of ATK89 (PmtAS) combined with membrane fractions of ATK154 (PmtCHA). Co-immunoprecipitates were resuspended in SDS loading dye and resolved on 4–20% gradient Tris–HEPES–SDS–polyacrylamide gels, transferred to membranes, and probed with anti-S tag antibody (left) or anti-HA antibody (right).
3.6. MsbA, an ortholog of a Pmt4-modified protein, is not modified by PmtC
In S. cerevisiae, Msb2, an osmosensor in the HOG pathway, is modified by Pmt4 [24,25]. Using the S. cerevisiae Msb2 protein sequence to query the A. nidulans genome database at the Broad Institute (http://www.broad.mit.edu) yielded ANID_07041.1, which we named “MsbA”. MsbAS was expressed from the gpd promoter in wildtype and Δpmt mutants, membrane fractions were isolated and probed with anti-S tag antibodies in Westerns (Fig. 7). The apparent molecular mass of MsbA in the ΔpmtA and ΔpmtB mutants was 160 and 170 kDa respectively, while the molecular mass of MsbAS in ΔpmtAΔpmtB mutant appeared lower at 150 kDa. The apparent molecular mass of MsbA in the ΔpmtC mutant was approximately 240 kDa, which was the same as that in the wildtype. MsbA possesses a putative N-glycosylation site. The molecular mass of PNGase F treated MSbAS in ΔpmtA, ΔpmtB and ΔpmtC was unchanged (data not shown). Our results suggest that MsbA is modified by PmtA and PmtB, but not by PmtC, and that PmtA and PmtB make separate modifications to MsbA.
Fig. 7.

MsbA (AN7041) is modified by PmtA and PmtB, but not by PmtC. Crude membrane fractions containing MsbAS from Δpmt mutants and wildtype were treated with 2 × SDS loading dye and separated on 7% SDS–polyacrylamide gel, transferred to a membrane, and probed with anti-S tag antibody.
In S. cerevisiae both Pmt subfamily 1 and Pmt subfamily 2 have two members (Fig. 8). In immunoprecipitation experiments, the predominant interaction partner for Pmt1 is Pmt2, though low levels of Pmt3 (from subfamily 2) are also reproducibly precipitated. Similarly, the predominant interaction partner for Pmt 5 (from subfamily 1) is Pmt3, though low levels of Pmt2 are also precipitated [13]. The Pmt 4 subfamily has only one member and it only forms homodimers. Thus in S. cerevisiae five Pmt complexes exist and low level complexes can substitute if the predominant complexes are lost.
Fig. 8.

Diagram comparing Pmt complex formation in A. nidulans and S. cerevisiae (adapted from Girrbach and Strahl) [13]. In S. cerevisiae Pmt1 forms heteromeric complexes with Pmt2 and Pmt4 forms homomeric complexes. Pmt1 forms heteromeric complexes with Pmt3 (subfamily 2) in the absence of Pmt2, while Pmt2 forms heteromeric complexes with Pmt5 (subfamily 1) in the absence of Pmt1. In A. nidulans PmtB (subfamily 1) forms heteromeric complexes with PmtA (subfamily 2) and with PmtC (subfamily 4). PmtA (subfamily 2) forms heteromeric complexes with PmtC (subfamily 4). PmtC (subfamily 4) forms homomeric complexes. Ovals indicate Pmt1 subfamily. Diamonds indicate Pmt2 subfamily. Pentagons indicate Pmt4 subfamily.
In contrast A. nidulans contains only 3 Pmts, one from each subfamily (Fig. 8). Not surprisingly, we found that PmtA (subfamily 2) interacts with Pmt B (subfamily 1) and PmtC (subfamily 4) interacts with itself. More surprisingly, we also found that both PmtA and Pmt B interact with Pmt C (subfamily 4). Because we were unable to generate appropriately marked diploid strains, we do not know if PmtA and Pmt B can also form homodimers and because single copy Pmts driven by native promoters were not produced in large quantities, we were unable to address quantitative issues such as which complexes predominate. However, it is clear that in A. nidulans at least four Pmt complexes exist, though it is not clear which complexes predominate. A. nidulans is the first fungus reported to form heteromeric complexes involving the Pmt4 subfamily, though such complexes have been reported in animals where POMT1 (subfamily 4) forms a heteromeric complex with POMT2 (subfamily 2) [18,19]. It is possible that the novel heterodimers we observed with Pmt C might be low level forms that can substitute when other complexes are lost allowing A. nidulans to survive with its reduced complement of single Pmts from each subfamily [22].
4. Conclusions
In this study we show that the three A. nidulans Pmts (PmtA from subfamily 2, PmtB from subfamily 1, and PmtC from subfamily 4) form heteromeric complexes in all pairwise combinations and that PmtC forms a homomeric complex. We also show that MsbA, the A. nidulans ortholog of S. cerevisiae Msb2, is modified by PmtA and PmtB, but not by PmtC.
Acknowledgments
This work was supported by DOE Biosciences award DE-FG02-97ER20275 and a grant from the Office of the Vice President for Research at the University of Georgia to M.M. T.K. was supported by the Royal Thai Government. The funding sources had no involvement in study design, collection, analysis, writing, or decision to submit. T.K. and M.M. conceived and designed the project. T.K. collected the data. T.K. and M.M. analyzed the data and wrote the paper.
Appendix A. Supplementary data
Supplementary Figure 1.

Supplementary Fig. S1
References
- 1.Strahl-Bolsinger S., Gentzsch M., Tanner W. Protein O-mannosylation. Biochim. Biophys. Acta. 1999;1426:297–307. doi: 10.1016/s0304-4165(98)00131-7. [DOI] [PubMed] [Google Scholar]
- 2.VanderVen B.C., Harder J.D., Crick D.C., Belisle J.T. Export-mediated assembly of mycobacterial glycoproteins parallels eukaryotic pathways. Science. 2005;309:941–943. doi: 10.1126/science.1114347. [DOI] [PubMed] [Google Scholar]
- 3.Loibl M., Strahl S. Protein O-mannosylation: what we have learned from baker’s yeast. Biochim. Biophys. Acta. 2013;1833:2438–2446. doi: 10.1016/j.bbamcr.2013.02.008. [DOI] [PubMed] [Google Scholar]
- 4.Haselbeck A., Tanner W. O-glycosylation in Saccharomyces cerevisiae is initiated at the endoplasmic reticulum. FEBS Lett. 1983;158:335–338. doi: 10.1016/0014-5793(83)80608-5. [DOI] [PubMed] [Google Scholar]
- 5.Strahl-Bolsinger S., Scheinost A. Transmembrane topology of pmt1p, a member of an evolutionarily conserved family of protein O-mannosyltransferases. J. Biol. Chem. 1999;274:9068–9075. doi: 10.1074/jbc.274.13.9068. [DOI] [PubMed] [Google Scholar]
- 6.Gentzsch M., Strahl-Bolsinger S., Tanner W. A new Dol-P-Man:protein O-D-mannosyltransferase activity from Saccharomyces cerevisiae. Glycobiology. 1995;5:77–82. doi: 10.1093/glycob/5.1.77. [DOI] [PubMed] [Google Scholar]
- 7.Girrbach V., Zeller T., Priesmeier M., Strahl-Bolsinger S. Structure-function analysis of the dolichyl phosphate-mannose: protein O-mannosyltransferase ScPmt1p. J. Biol. Chem. 2000;275:19288–19296. doi: 10.1074/jbc.M001771200. [DOI] [PubMed] [Google Scholar]
- 8.Lommel M., Bagnat M., Strahl S. Aberrant processing of the WSC family and Mid2p cell surface sensors results in cell death of Saccharomyces cerevisiae O-mannosylation mutants. Mol. Cell Biol. 2004;24:46–57. doi: 10.1128/MCB.24.1.46-57.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Weber Y., Prill S.K., Ernst J.F. Pmt-mediated O mannosylation stabilizes an essential component of the secretory apparatus, Sec20p, in Candida albicans. Eukaryotic Cell. 2004;3:1164–1168. doi: 10.1128/EC.3.5.1164-1168.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Timpel C., Strahl-Bolsinger S., Ziegelbauer K., Ernst J.F. Multiple functions of Pmt1p-mediated protein O-mannosylation in the fungal pathogen Candida albicans. J. Biol. Chem. 1998;273:20837–20846. doi: 10.1074/jbc.273.33.20837. [DOI] [PubMed] [Google Scholar]
- 11.Gentzsch M., Tanner W. The PMT gene family: protein O-glycosylation in Saccharomyces cerevisiae is vital. EMBO J. 1996;15:5752–5759. [PMC free article] [PubMed] [Google Scholar]
- 12.Timpel C., Zink S., Strahl-Bolsinger S., Schroppel K., Ernst J. Morphogenesis, adhesive properties, and antifungal resistance depend on the Pmt6 protein mannosyltransferase in the fungal pathogen Candida albicans. J. Bacteriol. 2000;182:3063–3071. doi: 10.1128/jb.182.11.3063-3071.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Girrbach V., Strahl S. Members of the evolutionarily conserved PMT family of protein O-mannosyltransferases form distinct protein complexes among themselves. J. Biol. Chem. 2003;278:12554–12562. doi: 10.1074/jbc.M212582200. [DOI] [PubMed] [Google Scholar]
- 14.Strahl-Bolsinger S., Tanner W. Protein O-glycosylation in Saccharomyces cerevisiae. Purification and characterization of the dolichyl-phosphate-D-mannose-protein O-D-mannosyltransferase. Eur. J. Biochem. 1991;196:185–190. doi: 10.1111/j.1432-1033.1991.tb15802.x. [DOI] [PubMed] [Google Scholar]
- 15.Hutzler J., Schmid M., Bernard T., Henrissat B., Strahl S. Membrane association is a determinant for substrate recognition by PMT4 protein O-mannosyltransferases. Proc. Natl. Acad. Sci. U.S.A. 2007;104:7827–7832. doi: 10.1073/pnas.0700374104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gentzsch M., Tanner W. Protein-O-glycosylation in yeast: protein-specific mannosyltransferases. Glycobiology. 1997;7:481–486. doi: 10.1093/glycob/7.4.481. [DOI] [PubMed] [Google Scholar]
- 17.Ecker M., Mrsa V., Hagen I., Deutzmann R., Strahl S. O-mannosylation precedes and potentially controls the N-glycosylation of a yeast cell wall glycoprotein. EMBO Rep. 2003;4:628–632. doi: 10.1038/sj.embor.embor864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Manya H., Chiba A., Yoshida A., Wang X., Chiba Y. Demonstration of mammalian protein O-mannosyltransferase activity: coexpression of POMT1 and POMT2 required for enzymatic activity. Proc. Natl. Acad. Sci. U.S.A. 2004;101:500–505. doi: 10.1073/pnas.0307228101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ichimiya T., Manya H., Ohmae Y., Yoshida H., Takahashi K. The twisted abdomen phenotype of Drosophila POMT1 and POMT2 mutants coincides with their heterophilic protein O-mannosyltransferase activity. J. Biol. Chem. 2004;279:42638–42647. doi: 10.1074/jbc.M404900200. [DOI] [PubMed] [Google Scholar]
- 20.Gentzsch M., Immervoll T., Tanner W. Protein O-glycosylation in Saccharomyces cerevisiae: the protein O-mannosyltransferases Pmt1p and Pmt2p function as heterodimer. FEBS Lett. 1995;377:128–130. doi: 10.1016/0014-5793(95)01324-5. [DOI] [PubMed] [Google Scholar]
- 21.Willer T., Brandl M., Sipiczki M., Strahl S. Protein O-mannosylation is crucial for cell wall integrity, septation and viability in fission yeast. Mol. Microbiol. 2005;57:156–170. doi: 10.1111/j.1365-2958.2005.04692.x. [DOI] [PubMed] [Google Scholar]
- 22.Kriangkripipat T., Momany M. Aspergillus nidulans protein O-mannosyltransferases play roles in cell wall integrity and developmental patterning. Eukaryotic Cell. 2009;8:1475–1485. doi: 10.1128/EC.00040-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Goto M., Harada Y., Oka T., Matsumoto S., Takegawa K. Protein O-mannosyltransferases B and C support hyphal development and differentiation in Aspergillus nidulans. Eukaryotic Cell. 2009;8:1465–1474. doi: 10.1128/EC.00371-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.O’Rourke S.M., Herskowitz I. A third osmosensing branch in Saccharomyces cerevisiae requires the Msb2 protein and functions in parallel with the Sho1 branch. Mol. Cell Biol. 2002;22:4739–4749. doi: 10.1128/MCB.22.13.4739-4749.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yang H.Y., Tatebayashi K., Yamamoto K., Saito H. Glycosylation defects activate filamentous growth Kss1 MAPK and inhibit osmoregulatory Hog1 MAPK. EMBO J. 2009;28:1380–1391. doi: 10.1038/emboj.2009.104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Yang L., Ukil L., Osmani A., Nahm F., Davies J. Rapid production of gene replacement constructs and generation of a green fluorescent protein-tagged centromeric marker in Aspergillus nidulans. Eukaryotic Cell. 2004;3:1359–1362. doi: 10.1128/EC.3.5.1359-1362.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ausubel F.M., Brent R., Kingston R.E., Moore D.D., Seidman J.G., editors. Current Protocols in Molecular Biology. J. Wiley; New York: 2001. [Google Scholar]
- 28.Toews M.W., Warmbold J., Konzack S., Rischitor P., Veith D. Establishment of mRFP1 as a fluorescent marker in Aspergillus nidulans and construction of expression vectors for high-throughput protein tagging using recombination in vitro (GATEWAY) Curr. Genet. 2004;45:383–389. doi: 10.1007/s00294-004-0495-7. [DOI] [PubMed] [Google Scholar]
