Time-resolved quantitative proteomics was performed to elucidate the cellular response of
R. toruloides to copper stress. With this method, proteins (significance ≥ 2, fold change ≥ 2, detected in at least one sample per group, LFQ by PEAKS Studio,
Tables S1 and S2 Supplementary Materials) could be detected as differentially regulated at time point 96 h and 144 h in Cu(I) and Cu(II) stress. Different databases were used for IFO0559 and IFO0880, as the genomes vary between those two haplotypes. These strain-specific databases differ in protein annotation, which makes the comparison of the two haplotypes challenging as proteins can be detected under different names or are only included in one of the two databases. The highest number of differentially expressed proteins was detected in the comparison of IFO05559 at 0 mM to 0.5 mM Cu(I) after 96 h with 954 proteins, including 585 upregulated and 369 downregulated proteins. For IFO0880, the comparison of 0 and 0.5 mM Cu(II) treatment at 144 h resulted in a maximum of 946 differentially expressed proteins, with 593 upregulated and 353 downregulated proteins (
Figure 5).
3.6.1. Effect of Copper on Oxidative Stress
As copper toxicity is often attributed to oxidative stress [
15,
18], all proteomic data was analyzed for typical markers of oxidative stress (
Table 1), to validate oxidative stress as a mechanism of copper toxicity in
R. toruloides IFO0559 and IFO0880. Catalases, peroxidases, and superoxide dismutases are among the most common protein responses to oxidative stress, as they are specialized in inactivating reactive oxygen species such as peroxides and superoxides [
12,
19,
40,
41]. Additionally, the data were examined for redoxins, antioxidant enzymes, capable of interacting with respective transcription factors [
19,
40]. No increased amounts of proteins related to carotenoid synthesis were detected, which is in accordance with no significant increase in carotenoid titers.
Among those groups, two catalases were 2 to 3-fold downregulated, while all peroxidases, 4 different proteins in total, with the exception of glutathione peroxidase and phosphatidic acid phosphatase type 2/haloperoxidase, increased by 2 to 3-fold. Interestingly, for both haplotypes, catalase decline was only detected at 96 h (IFO0559 Cu(II) 96 h and IFO0880 Cu(I) 96 h). In contrast, peroxidases were only detected in strain IFO0559 at both time points. While the concentration of glutathione peroxidase and phosphatidic acid phosphatase type 2/haloperoxidase declined by about 2-fold in Cu(I) 144 h and Cu(II) 96 h samples, increased levels of peroxidase were observed in Cu(II) treated samples at 96 h and 144 h. Higher levels of phosphatidic acid phosphatase type 2/haloperoxidase were observed in Cu(I) samples at 144 h, as they increased 2.61-fold compared to control samples with only ammonia supplementation. In most samples, superoxide dismutase (SOD) and manganese SOD were not detected at differential levels to their control, except in IFO0880 when supplemented with Cu(II). At both time points, SOD was upregulated by about 4-fold, while manganese SOD was only detected at 144 h with about a 2-fold increase.
In the case of redoxin production, an increase in the respective redoxin amount was observed for 31 out of 41 hits (significance and fold change higher 2), with the highest increase in thioredoxin and thioredoxin h of about 6-fold (Cu(II) IFO880 96 h) and 8-fold (Cu(II) IFO880 144 h), respectively. Other types of upregulated redoxins include glutaredoxin and peroxiredoxin.
In general, more differentially regulated redoxins were detected for IFO0880 than for IFO0559; however, most downregulated redoxins were detected in samples of IFO0559 after 144 h Cu(I) supplementation followed by samples at 96 h in IFO0880 for both Cu(I) and Cu(II) treatment.
Another component in protection against oxidative stress is glutathione S-transferases (GST). Glutathione itself can act as an antioxidant and when conjugated to toxic molecules such as peroxidized lipids, resulting from ROS and their subsequent reactions with cellular components, it can act in detoxification. Moreover, GSTs act in detoxification by transporting toxins, resulting in excretion or their import into detoxification organelles such as the vacuole. In several publications, GSTs were also shown to bind free copper ions, therefore protecting cells against copper toxicity [
18,
42,
43].
In our proteomic data set, 32 hits were identified for differentially regulated GSTs, of which 10 were detected in IFO0559 samples and 22 in IFO0880 strains. A total of 27 hits represent upregulated GST enzymes, while 5 were downregulated. Of those 5, 3 were detected in IFO0559. While Cu(II) treatment resulted in only 4 GST hits in IFO0559, Cu(I) treatment resulted in 6 hits. For IFO0880, 13 hits were detected in Cu(II) treated condition, while 9 GST hits were detected in Cu(I) samples. Significant upregulation was observed in several IFO0880 samples. In Cu(II) 96 h and Cu(I) 144 h 10- and 9.28-fold upregulation was detected. For Cu(I) 144 h, another glutathione transferase, glutathione transferase omega-1, accumulated to 12.94-fold increased levels compared to ammonia control settings. The highest accumulation of glutathione S-transferases in IFO0559 was measured for Cu(I) 96 h with 4.52-fold accumulation compared to the control.
In addition to glutathione transferases, 4 genes involved in glutathione synthesis were identified. While glutamate-cysteine ligase was detected in all IFO0880 samples, glutamate synthase was only detected in IFO0559 Cu(II) at 96 h, IFO0880 Cu(II) at 96 h, and IFO0880 Cu(I) at 96 h conditions. For these hits, upregulation between 2.86- and 6.67-fold was observed. Glutathione synthase itself was only detected in IFO0880 Cu(II) at 96 h samples with a 2-fold increase.
Glutathione-mediated detoxification of formaldehydes, a product of lipid peroxidation, is dependent on S-formylglutathione hydrolase to recycle the glutathione from the formed S-formylglutathione [
44]. Other enzymes partitioning in glutathione recovery from different detoxification reactions include glutathione hydrolase, hydroxyacylglutathione hydrolase, and lactoylglutathione lyase, the latter two being part of the methylglyoxal detoxification, also known as glyoxalase I and II. With methylglyoxal being induced by oxidation of lipids (among other factors), these enzymes might be involved in the response to oxidative stress in
R. toruloides, as was shown for different microorganisms, in which mutants defective in methylglyoxal detoxification underwent severe oxidative stress with rapid loss of viability [
45,
46]. All four enzymes were upregulated in different IFO0880 samples, with S-formylglutathione hydrolase showing the highest fold-change of 9.09 in Cu(II) 96 h condition.
3.6.2. Effect of Copper on Iron-Sulfur Proteins
As Irazusta et al. demonstrated in their study for
R. mucilaginosa [
19],
R. toruloides also exhibited increased levels of enzymes involved in oxidative stress response, indicating that copper toxicity in
R. toruloides is indeed mediated through oxidative stress. However, several publications discussed an additional aspect of copper toxicity. In their work, the effect of free copper ions on iron–sulfur (Fe–S) clusters was investigated. It was postulated, that copper can inactivate iron–sulfur cluster proteins by directly damaging the iron-sulfur proteins as well as interfering with Fe–S cluster formation [
47,
48,
49]. Especially Cu(I) competes with iron, destabilizing Fe–S cluster. Chillappagari et al. further established an increase in protein levels involved in the iron-sulfur assembly machinery under excess copper. Additionally, they found copper stress to increase the abundance of proteins involved in iron uptake [
49].
In our proteomics data, 14 hits were identified for proteins of the iron–sulfur assembly machinery (
Table 2). Almost all upregulated components were detected in Cu(II) samples of IFO05559 at both time points and one in Cu(I) treated IFO0880 at 144 h. In general, most hits were identified in IFO0559 (10 hits). Overall, more iron–sulfur assembly proteins were downregulated (10 hits) than upregulated (4 hits).
Furthermore, 20 protein hits involved in iron transport were identified (
Table 2), of which 14 constitute upregulated proteins (2.38- to 33.33-fold). For Cu(I) and Cu(II) treatment, 11 and 9 hits were detected, respectively. All except one downregulated protein (0.44- to 0.25-fold) were detected in IFO0559, with a distribution of overall hits between IFO0559 and IFO0880 of 12 to 8. The protein with the highest and most consistent upregulation is a zip-like iron-zinc transporter, which was detected upregulated in all sample sets except for IFO0880 Cu(II), where it was not detected at all.
In summary, these results imply a measurable effect on the discussed enzymes; however, the effect is not as distinct and uniform as that described for other organisms.
Table 2.
Differentially expressed proteins involved in iron-sulfur cluster assembly and iron transport discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Further fold changes at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
Table 2.
Differentially expressed proteins involved in iron-sulfur cluster assembly and iron transport discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Further fold changes at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
IFO0559 | IFO0880 |
---|
| | Cu(I) | Cu(II) | | | Cu(I) | Cu(II) |
---|
| Accession | 96 h | 144 h | 96 h | 144 h | | Accession | 96 h | 144 h | 96 h | 144 h |
---|
Iron–sulfur cluster assembly |
Iron–sulfur cluster assembly protein | M7X0I3 | 0.42 | 0.44 | n.d. | down | Iron–sulfur cluster assembly protein | A0A0K3C953 | n.d. | up | 0.28 | n.d. |
Iron–sulfur cluster assembly accessory protein Isa2 | M7WS85 | n.d. | 0.47 | 3.70 | 5.44 | Probable cytosolic iron-sulfur protein assembly protein 1 | A0A0K3C8P4 | up | n.d. | 0.46 | 0.48 |
Fe–S cluster assembly protein DRE2 | M7WV82 | n.d. | 0.24 | n.d. | up | Fe–S cluster assembly protein DRE2 | A0A2T0AFS9 | n.d. | 3.54 | n.d. | up |
Cytosolic Fe–S cluster assembly factor NBP35 | M7WL71 | up | 0.28 | n.d. | 0.18 | | | | | | |
Cytosolic Fe–S cluster assembly factor CFD1 | M7X358 | down | 0.37 | down | 2.35 | | | | | | |
Iron transporter |
Zip-like iron-zinc transporter | M7X8Q2 | 10.14 | 4.85 | 33.33 | 4.31 | Zip-like iron-zinc transporter | A0A0K3CEY7 | 18.26 | 7.18 | n.d. | n.d. |
Zip-like iron-zinc transporter | M7WSK3 | n.d. | 0.4 | up | n.d. | | | | | | |
Iron permease | M7X6M4 | n.d. | up | 0.34 | up | Iron permease FTR1/Fip1/EfeU | A0A2T0A4Z9 | 3.94 | up | n.d. | 2.38 |
MFS transporter siderophore-iron: H+ symporter | M7WNH3 | 0.26 | n.d. | 0.30 | down | MFS transporter siderochrome-iron transporter | A0A0K3C6P7 | up | n.d. | n.d. | 3.70 |
Iron/copper transporter Atx1 | M7X0X4 | down | 0.44 | up | 2.81 | | | | | | |
Iron complex transport system ATP-binding protein | M7X547 | 3.12 | up | n.d. | down | Iron complex transport system ATP-binding protein | A0A0K3CJI3 | up | 2.47 | down | down |
Siderophore iron transporter mirC | M7X1M8 | up | 5.07 | down | n.d. | Siderophore iron transporter mirC | A0A0K3CCV2 | down | up | 0.25 | 3.33 |
3.6.3. Effect of Copper on Zinc Proteins
Barber et al. concluded in their review that copper also impacts zinc-dependent processes, including zinc cofactor enzymes and zinc-containing transcription factors [
50]. Interestingly, both zinc finger transcription factors and zinc-containing enzymes such as zinc-type alcohol dehydrogenase were detected at differential levels compared to their controls (
Table S8 Supplementary Materials). In total, 77 hits for zinc finger proteins were identified (46 in IFO0559 samples and 31 in IFO0880 samples), of which 42 are downregulated.
Additionally, proteins required for zinc homeostasis were identified (
Table 3), namely cation efflux protein zinc transporter, protein of cation efflux protein family zinc transporter, zip-like iron-zinc transporter, mitochondrial zinc maintenance protein, 1 and solute carrier family 30 (Zinc transporter) member 1.
While mitochondrial zinc maintenance protein 1 is downregulated in IFO0880 Cu(II) 144 h, a 3.41-fold increase compared to the control was detected in IFO0880 Cu(I) 96 h, as well as another protein with the same designation in IFO0880 Cu(I) 144 h with a 2.06-fold upregulation.
Both the cation efflux protein zinc transporter and protein of cation efflux protein family zinc transporter were downregulated in IFO0559 during Cu(II) exposure; however, Cu(I) appears to exhibit the opposite effect increasing the production of cation efflux protein zinc transporter by 3.13-fold compared to control in IFO0559 at 144 h. Similarly to iron transporters, zip-like iron-zinc transporter was consistently upregulated in all samples except IFO0880 Cu(II).
Interestingly, downregulation higher than the cut-off (significance ≥ 2, fold change ≥ 2) of both cation efflux transporters was only detected in IFO0559, while such upregulation of mitochondrial zinc maintenance proteins was only detected in IFO0880, indicating, that both strains might have adapted differently to copper stress. However, differences might also occur through the different annotations of both strains, which would distort a direct comparison.
The presented results here are in line with the copper effects described by Barber et al. Although in a different organism, copper appears to disrupt zinc proteins, likely by displacement of zinc ions, similar to its effect on iron-sulfur clusters described above. They concluded that zinc-dependent transcriptional regulation is likely affected by copper [
50], which is in accordance with our findings of zinc finger proteins being impacted by copper excess.
In the work of Hassan et al., they were able to determine that zinc stress induces copper-specific depletion, as there was no significant effect on other transition metal ions (mangan, cobalt, nickel, and iron) [
51]. Interestingly, our data might indicate the opposite scenario, where copper depletes zinc likely by displacement.
Table 3.
Differentially expressed proteins involved in zinc homeostasis discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
Table 3.
Differentially expressed proteins involved in zinc homeostasis discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
IFO0559 | IFO0880 |
---|
| | Cu(I) | Cu(II) | | | Cu(I) | Cu(II) |
---|
| Accession | 96 h | 144 h | 96 h | 144 h | | Accession | 96 h | 144 h | 96 h | 144 h |
---|
Zinc homeostasis |
Protein of cation efflux protein family zinc transporter | M7XE18 | 0.44 | n.d. | 0.37 | 0.46 | Mitochondrial zinc maintenance protein 1 mitochondria | A0A0K3CHC4 | up | n.d. | down | 0.34 |
Cation efflux protein zinc transporter | M7XHN8 | up | 3.13 | 0.30 | n.d. | Mitochondrial zinc maintenance protein 1 mitochondria | A0A0K3CUM1 | 3.41 | up | up | n.d. |
Solute carrier family 30 (Zinc transporter) member 1 | M7WTR0 | up | 2.82 | n.d. | down | Mitochondrial zinc maintenance protein 1 mitochondria | A0A2T0AG28 | n.d. | 2.06 | n.d. | n.d. |
Zip-like iron-zinc transporter | M7X8Q2 | 10.14 | 4.85 | 33.33 | 4.31 | Zip-like iron-zinc transporter | A0A0K3CEY7 | 18.26 | 7.18 | n.d. | n.d. |
Zip-like iron-zinc transporter | M7WSK3 | n.d. | 0.4 | up | n.d. | | | | | | |
3.6.4. Storage of Copper Ions
As stated earlier, copper ions are stored in organelles in
S. cerevisiae to avoid excessive accumulation [
16]. About 25% of copper ions were stored in lipid droplets, with vacuoles and mitochondria also representing major storage compartments. Additionally, mitochondria and the Golgi apparatus are labile intracellular copper pools, which allow the cell to maintain the rapid kinetics of copper uptake and release [
52]. Lipid droplets might facilitate the transport of copper ions as they interact with the Golgi apparatus, mitochondria, and vacuoles through physical contact as well as proteins present on their surface [
16]. A new class of perilipin proteins was discovered in
R. toruloides NP11, the expression of which correlates with lipid quantity, indicating their importance in lipid accumulation [
53]. Fungal perilipin-like proteins were characterized to function in protecting lipid droplets from degradation [
54,
55]. In accordance with those publications, lipid droplet protein 1 (Perilipin-like protein) was detected in increased amounts in IFO0559 samples stressed with Cu(II) and IFO0880 samples stressed with Cu(I) and Cu(II) at 144 h.
Vacuoles are utilized as storage for copper ions. The overexpression of
hmt1, coding for a vacuolar ABS heavy metal transporter (Htm1), in
Schizosaccharomyces pombe enhanced the metal tolerance of the yeast. Hmt1 is associated with the vacuolar membrane and could mediate compartmentalization of heavy metals [
56]. Htm1 was detected in significantly increased quantities between 2.74-fold when stressed with Cu(II) (144 h) and 17.38-fold when stressed with Cu(I) (96 h) in IFO0559 and between 3.70-fold when stressed with Cu(II) (96 h) and 9.22-fold when stressed with Cu(I) (144 h) in IFO0880 (
Table 4). The fold change in quantity was higher in both strains when stressed with Cu(I). Further hint on copper compartmentalization in the vacuole is the upregulation of the vacuoler transporter chaperones 2 and 4, which both exhibited increased accumulation in IFO0559 samples treated with Cu(I), for chaperone 2 at 96 h even to a 64-fold increase. In
S. cerevisiae, increased cytoplasmic calcium signals could also mitigate aluminum toxicity, as a mutant lacking the vacuolar calcium ion ATPase pump PMC1p exhibited a higher sensitivity to aluminum [
57]. In IFO0559, a vacuolar calcium ion transporter H
+ exchanger (M7X2U4) was detected in decreased concentrations under both Cu(I) and Cu(II) stress at 96 h, and in increased concentrations at 144 h. For cells stressed with Cu(II), a 34.66-fold increase in protein amount was detected. Additionally, a second vacuolar calcium ion transporter (M7XNE5), was increased in IFO0559 cultures stressed with Cu(I). Therefore, calcium homeostasis could similarly be able to reduce copper toxicity.
When stressed with 0.8 mM copper, a copper-resistant
thmea1 knockout mutant strain compared to the wild-type
Trichoderma harzianum Th-33 exhibited increased expression of proteins related to the transport of copper ions into the Golgi secretory pathway. This indicates a higher number of copper ions entering the Golgi vesicles [
58]. CCC2, a P-type ATPase, exports cytosolic copper ions into late or post-Golgi compartments [
59]. While none of the proteins in IFO0559 could be matched to this annotation, a copper P-type ATPase in IFO0880 was upregulated 3.19-fold when cultures were stressed with Cu(I) and 2.17-fold when cultures were stressed with Cu(II) after 144 h.
Another major storage compartment for copper ions is the mitochondria [
16]. These essential organelles play key roles in pathways such as ATP production, ß-oxidation, and oxidative stress clearance [
60]. The mitochondrial inner membrane metallopeptidase Oma1 is part of the quality control system in mitochondria. This protein is reported to be activated by various stressors such as mitochondrial, oxidative, and heat stress [
61,
62,
63]. Interestingly, in IFO0880 and IFO0559, Oma1 was detected in lower amounts when exposed to copper stress (
Table 4). The mitochondria can exist in multiple morphologies; under oxidative stress, mitochondria could fuse into a tubular morphology. When mitochondrial fusion was blocked, sensitivity to oxidative stress was elevated in the fungus
Cryptococcus neoformans [
60]. The detected levels of maintenance of mitochondrial morphology protein 1 (MMM1), mitochondrial distribution and morphology protein 10 (MDM10), mitochondrial sensitive to high expression protein 9, and mitochondrial inner membrane protein 1 decreased when exposed to copper, with the exception of MDM10 and mitochondrial inner membrane protein 1 in IFO0880 when exposed to Cu(I). Levels of the mitochondrial outer membrane protein (IML2) and mitochondrial distribution and morphology protein 12 (MDM12) were elevated in response to Cu(II) exposure in IFO0559 and IFO0880 after 144 h, although fold changes were lower than 2.
ATM1 is part of the iron metabolism in the mitochondria of
C. neoformans, the protein is important for the iron-sulfur cluster synthesis and heme metabolism [
64]. Copper toxicity is induced through the alteration of iron-sulfur cluster homeostasis as the cofactor binding to target proteins is disrupted [
65]. Sensitivity to oxidative stress was shown to be elevated in
atm1 mutants [
64]. To cope with copper stress,
C. neoformans increases the expression of ATM1, as increased export of iron–sulfur cluster precursors from the mitochondrial matrix to the cytosol, enabling iron–sulfur proteins to remain active and therefore ensure essential cellular processes [
65]. The mitochondrial ABC transporter
atm was only annotated in the genome of IFO0559, and elevated levels of the protein were detected upon exposure to copper stress (
Table 4). For Cu(I) stress, a fold change of around 4 was detected, while for Cu(II) stress protein amount increased, but no significant amounts of protein were detected.
Table 4.
Differentially expressed proteins involved in copper ion storage discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
Table 4.
Differentially expressed proteins involved in copper ion storage discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
| IFO0559 | IFO0880 |
---|
| | Cu(I) | Cu(II) | | Cu(I) | Cu(II) |
---|
| Accession | 96 h | 144 h | 96 h | 144 h | Accession | 96 h | 144 h | 96 h | 144 h |
---|
Copper ion storage |
Lipid droplet protein 1 (Perilipin-like protein) LDP1 | M7WE51 | n.d. | down | 2.56 | up | A0A2T0A369 | down | 2.23 | n.d. | 16.67 |
Vacuolar ABC heavy metal transporter Hmt1 | M7XIA9 | 17.38 | 7.36 | 5.26 | 2.74 | A0A0K3CL56 | 7.26 | 9.22 | 3.70 | 5.56 |
Vacuolar calcium ion transporter H+ exchanger | M7X2U4 | 0.28 | 2.58 | 0.22 | 34.66 | | | | | |
Vacuolar calcium ion transporter | M7XNE5 | 2.03 | 3.38 | n.d. | n.d. | A0A2T0A7V0 | n.d. | n.d. | down | n.d. |
Copper P-type ATPase | | | | | | A0A2T0A6Z7 | 2.81 | 3.19 | n.d. | 2.17 |
Vacuolar transporter chaperone 2 | M7WMK7 | 64.00 | 17.05 | n.d. | n.d. | | | | | |
Vacuolar transporter chaperone 4 | M7XMX5 | n.d. | 2.26 | n.d. | n.d. | | | | | |
Mitochondrial inner membrane metallopeptidase Oma1 | M7X2Q6 | down | n.d. | down | down | A0A0K3C9S0 | down | 0.46 | n.d. | n.d. |
Mitochondrial outer membrane protein IML2 | M7X0Q4 | n.d. | up | down | down | A0A2T0A031 | n.d. | up | 0.40 | down |
Maintenance of mitochondrial morphology protein 1 MMM1 | M7X412 | 0.38 | down | down | n.d. | A0A2T0AGR5 | down | down | n.d. | n.d. |
Mitochondrial distribution and morphology protein 12 MDM12 | M7X8M4 | n.d. | up | n.d. | n.d. | A0A0K3C7P5 | n.d. | up | n.d. | 0.50 |
Mitochondrial distribution and morphology protein 10 MDM10 | M7WSP8 | n.d. | 0.34 | n.d. | down | A0A0K3CJ28 | 0.5 | down | 2.33 | up |
Sensitive to high expression protein 9, mitochondrial | M7XMC6 | 0.41 | n.d. | 0.39 | n.d. | A0A2T0AFZ7 | n.d. | n.d. | n.d. | n.d. |
Mitochondrial inner membrane protein 1 | M7XNS0 | n.d. | n.d. | 0.35 | down | A0A0K3CI34 | 0.29 | down | up | n.d. |
Mitochondrial ABC transporter ATM | M7XAM6 | 4.62 | 4.85 | up | up | | | | | |
3.6.5. Effect of Copper on the Fatty Acid Profile
Zhu et al. (2012) presented a multi-omics analysis of lipid accumulation in
R. toruloides NP11 [
53], a close relative of IFO0559 [
10]. The annotation of this strain was used for the proteomics analysis of IFO0559 in the present study. Proteins involved in de novo fatty acid biosynthesis of NP11 under nitrogen-limited conditions were also upregulated in our study in IFO0559, when stressed with 0.5 mM Cu(I) and Cu(II) after 96 h, including ATP citrate synthase (Acl1), malic enzyme (Me1), acetyl-CoA carboxylase (Acc1), fatty acid synthase subunit beta (Fas1), and alpha (Fas2) for fatty acid synthesis and glycerol-3-phosphate dehydrogenase (Gpd) and glycerol-3-phosphate O-acyltransferase (Gat1) for glycerolipid synthesis (
Table 5). The highest fold changes were observed for Acc1 with 4.25-fold increase, and Fas2 with a 4.09-fold increase, when stressed with Cu(I). This likely correlates with the increase in lipid titers in IFO0559 when exposed to copper. The fold change of the differentially expressed proteins between Cu(I) and Cu(II) varied, with a higher fold change in proteins for cultures stressed with Cu(I), except for Me1. Moreover, Gdp and LDP1 were not detected in the cultures stressed by Cu(I). Cultures treated with Cu(I) exhibited a prolonged lag phase, and only entered the exponential phase when the respective control was already confined to the stationary phase. This could explain a stronger expression of proteins involved in de novo fatty acid biosynthesis after 96 h, as more biomass and lipid were produced in a shorter time frame. More importantly, a delta-9 fatty acid desaturase (9FAD) was increased 1.93-fold when comparing the proteome of samples treated with and without Cu(I). 9FAD synthesizes palmitoleic acid (C16:1) and oleic acid (C18:1) [
66]. In contrast, the control sample with 0.03% ammonia exhibited a 22.6% higher C18:1 content than cultures treated with Cu(I) at time point 96 h, which is the highest C18:1 content recorded in all samples. Between 96 and 144 h, the oleic acid content of control samples with ammonia decreased by 17.7%. In contrast, oleic acid levels in Cu(I) samples only slightly decreased (by 4.2%). This correlates with the observed higher levels of 9FAD in Cu(I) samples.
Zhang et al. detected a two-fold increase in lipid titers in IFO0880 as a result of Acc1 and Diacylglycerol O-acyltransferase (DGA1) overexpression [
10]. The overexpression of 9FAD was associated with increased lipid titers [
67]. In our work, Acc1 was upregulated for cultures stressed with Cu(I), but it was downregulated when exposed to Cu(II) (
Table 5). DGA1, which catalyzes the terminal step of triacylglycerol (TAG) formation, was downregulated in cultures stressed with Cu(I) and could not be detected in cultures stressed with Cu(II). Acl1 was upregulated in IFO0880 with a 2.2-fold increase. Moreover, Fas2 was overexpressed with a fold change of 2.76 for cultures stressed with Cu(I). In accordance with the lower lipid amount detected for IFO0880 when stressed with Cu(I) (
Figure 3b), a decrease in the protein amounts of DGA1, Me1, Gpd, Gat1, and LDP1 was detected after 96 h. Increased lipid titers were detected for cultures stressed with Cu(II); however, only Acl1 and 9FAD upregulation was detected. Higher amounts of 9FAD could explain the increase in C18:1 for IFO0880 stressed with Cu(II) when comparing the fatty acid profiles at 96 h and 144 h. Furthermore, Gat1 and Gpd, which are responsible for glycerolipid synthesis, were detected in increased amounts.
Table 5.
Differentially expressed proteins involved in fatty acid biosynthesis discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
Table 5.
Differentially expressed proteins involved in fatty acid biosynthesis discussed here. For IFO0559 and IFO0880, fold change of samples grown under Cu(I) and Cu(II) stress was compared to respective control. Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
| IFO0559 | IFO0880 |
---|
| | Cu(I) | Cu(II) | | Cu(I) | Cu(II) |
---|
| Accession | 96 h | 144 h | 96 h | 144 h | Accession | 96 h | 144 h | 96 h | 144 h |
---|
Fatty acid biosynthesis |
ATP citrate synthase Acl1 | M7WHC9 | 2.61 | up | up | 3.33 | A0A0K3CJ29 | 2.2 | up | up | up |
Malic enzyme Me1 | M7WHN9 | up | up | 2.44 | up | A0A0K3CAF6 | 0.35 | down | n.d. | 0.49 |
Acetyl-CoA carboxylase Acc1 | M7XLR4 | 4.25 | 3.93 | n.d. | up | A0A0K3C6V6 | up | up | down | down |
Fatty acid synthase subunit beta, fungi type Fas1 | M7WSW5 | 3.47 | 2.29 | up | n.d. | | | | | |
Fatty acid synthase subunit alpha, fungi type Fas2 | M7XM89 | 4.09 | 2.95 | up | up | A0A0K3C4G6 | 2.76 | up | n.d. | down |
Glycerol-3-phosphate dehydrogenase (NAD+) Gpd | M7WSY9 | n.d. | n.d. | up | n.d. | A0A2S9ZYP9 | 0.27 | 0.48 | down | down |
Glycerol-3-phosphate dehydrogenase (NAD+) Gpd | | | | | | A0A2T0A892 | down | up | up | n.d. |
Glycerol-3-phosphate dehydrogenase (NAD+) Gpd | | | | | | A0A0K3CDD5 | 0.48 | down | up | up |
Glycerol-3-phosphate O-acyltransferase Gat1 | M7X5G5 | 2.66 | up | n.d. | down | A0A0K3CHG4 | down | up | down | 4.35 |
Delta-9 fatty acid desaturase 9FAD | M7XI95 | up | up | down | n.d. | A0A191UMV5 | n.d. | 2.03 | up | up |
Diacylglycerol O-acyltransferase DGA1 | M7WKS9 | up | n.d. | n.d. | n.d. | A0A191UMW0 | down | 3.13 | n.d. | n.d. |
3.6.6. Effect of Ammonia Supplementation on the Protein Expression
To understand the effect of ammonia on the biomass and lipid formation of IFO0559 and IFO0880, the differences in protein production between samples without supplementation (0 mM) and samples with 0.03% ammonia were analyzed.
The central nitrogen metabolism is composed of glutamate dehydrogenase (Gdh1), glutamine synthase (Gln1/2), and glutamate synthase (Glt1). Under nitrogen-limited conditions, these proteins were upregulated in NP11 [
53]. In accordance, elevated levels of Gln2 and Gdh1 were detected in control samples (0 mM) in IFO0559 after 96 h, and lower levels were observed in samples after 144 h (
Table 6). While Gdh1 was not detected in IFO0880, Gln 2 and Glt1 were downregulated nearly 2-fold. In contrast, Gln1 was detected in higher amounts for samples with higher nitrogen concentrations in both IFO0559 and IFO0880. The enzyme catalyzes the condensation of glutamate and ammonia to form glutamine.
The urea cycle converts ammonia to urea, in
S. cerevisiae the urea can afterwards be exported into the medium [
68]. In this study, proteins related to the metabolism of the urea cycle were detected in increased amounts, including arginosuccinase, argininosuccinate synthase, arginase, and ornithine carbamoyltransferase in both strains (
Table 6). After 96 h, the level of these proteins increased in IFO0880, with the exception of argininosuccinate synthase. Jagtap et al. hypothesized an increased concentration of ornithine, which is produced by arginase in the urea cycle, reflecting the utilization of amino acids for the production of energy in IFO0880 [
69]. This might contribute to the better growth of IFO0880 reported in our work. However, expression of carbamoyl-phosphate synthase, the rate-limiting step of the urea cycle, is downregulated in both strains, with the exception of IFO0559 after 144 h. Awad et al. investigated the effect of different nitrogen sources on biomass and lipid production in the oleaginous yeast
Cutaneotrichosporon oleaginosus. They observed increased biomass as well as lipid accumulation when cells were grown on urea instead of ammonium sulphate as a nitrogen source [
70]. As YNB only contains ammonium sulphate, the supplemented ammonium was converted into urea, which could explain the better growth observed in this study, including the higher lipid accumulation.
Table 6.
Differentially expressed proteins involved in ammonia metabolism discussed here. For IFO0559 and IFO0880, fold change of samples grown with ammonia supplementation compared to respective control (0 mM). Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
Table 6.
Differentially expressed proteins involved in ammonia metabolism discussed here. For IFO0559 and IFO0880, fold change of samples grown with ammonia supplementation compared to respective control (0 mM). Protein names and UniProt accession numbers are listed. Abbreviations are written in bold. Further fold change at 96 h and 144 h of samples are stated. Fractions below 1 depict downregulation. n.d. = not detected; up = upregulated below cut-off score; down = downregulated below cut-off score. Cut-off score: significance ≤ 2 or fold change ≤ 2.
| IFO0559 | IFO0880 |
---|
| Accession | 96 h | 144 h | Accession | 96 h | 144 h |
---|
Ammonia and Urea Cylce |
Glutamate dehydrogenase (NADP+) Gdh1 | M7X2B5 | down | up | A0A2T0AGW3 | n.d. | n.d. |
Glutamine synthetase Gln1 | M7XEY4 | n.d. | up | A0A0K3C755 | up | n.d. |
Glutamine synthetase Gln2 | M7Y051 | 0.49 | n.d. | A0A0K3CSP6 | n.d. | 0.44 |
Glutamate synthase (NADPH/NADH) Glt1 | M7WY92 | n.d. | n.d. | A0A0K3CF34 | down | n.d. |
Arginosuccinase | M7X7Q5 | up | n.d. | A0A0K3CRK9 | up | 2.63 |
Argininosuccinate synthase | M7WMY0 | 2.08 | up | A0A0K3CJ95 | up | up |
Arginase | M7XGF4 | 2.17 | n.d. | A0A0K3CJ80 | 3.57 | up |
Ornithine carbamoyltransferase | M7WUR1 | up | n.d. | A0A2T0AE55 | 2.44 | up |
Carbamoyl-phosphate synthase (glutamine-hydrolyzing) | M7WUQ8 | down | up | A0A0K3C8U1 | 0.39 | 0.33 |