Adaptive changes in the fungal cell wall mediate copper homeostasis

Copper homeostasis mechanisms are essential for microbial adaption to changing copper levels within the host during infection. In the opportunistic fungal pathogen Cryptococcus neoformans (Cn), the Cn Cbi1/Bim1 protein is a newly identified copper binding and release protein that is highly induced during copper limitation. Recent studies demonstrated that Cbi1 functions in copper uptake through the Ctr1 copper transporter during copper limitation. However, the mechanism of Cbi1 action is unknown. The fungal cell wall is a dynamic structure primarily composed of carbohydrate polymers, such as chitin and chitosan, polymers known to strongly bind copper ions. We demonstrated that Cbi1 depletion affects cell wall integrity and architecture, connecting copper homeostasis with adaptive changes within the fungal cell wall. The cbi1Δ mutant strain possesses an aberrant cell wall gene transcriptional signature as well as defects in chitin and chitosan deposition. These changes are reflected in altered macrophage activation and changes in the expression of specific virulence-associated phenotypes. Furthermore, using Cn strains defective in chitosan biosynthesis, we demonstrated that cell wall chitosan modulates the ability of the fungal cell to withstand copper stress. In conclusion, our data suggest a dual role for the fungal cell wall, in particular the inner chitin / chitosan layer, in protection against toxic levels of copper and providing a source of metal ion availability during copper starvation. Given the previously described role for Cbi1 in copper uptake, we propose that this copper-binding protein is involved in shuttling copper from the cell wall to the copper transporter Ctr1 for regulated microbial copper uptake. Author summary Microorganisms must be equipped to readily acquire essential micro-nutrients like copper from nutritionally poor environments while simultaneously shielding themselves from conditions of metal excess. We explored mechanisms of microbial copper homeostasis in the human opportunistic fungal pathogen Cryptococcus neoformans (Cn) by defining physiological roles of the newly described copper-binding and release protein Cn Cbi1/Bim1. Highly induced during copper limitation, Cbi1 has been shown to interact with the high-affinity copper transporter Ctr1. We defined Cbi1-regulated changes in the fungal cell wall, including controlling levels of the structural carbohydrates chitin and chitosan. These polysaccharides are embedded deeply in the cell wall and are known to avidly bind copper. We also defined the host immunological alterations in response to these cell wall changes. Our data suggest a model in which the fungal cell wall, especially the chito-oligomer layer, serves as a copper-binding structure to shield the cell from states of excess copper, while also serving as a copper storage site during conditions of extracellular copper depletion. Given its ability to bind and release copper, the Cbi1 protein likely shuttles copper from the cell wall to copper transporters for regulated copper acquisition.

cell wall integrity to strains with a ctr1 mutation. These results are consistent with prior studies 160 suggesting that Cbi1 and Ctr1 are independent components of a copper transporter complex [5]. 161 Additionally, these findings indicate that defective Cbi1 function is likely responsible for much of the loss 162 of cell wall integrity in the cuf1 mutant. 163 We also assessed Cn copper-dependent cell wall integrity phenotypes using an alternative 164 method of copper limitation to extracellular copper chelation by BCS. We incubated the Cn strains in 165 media containing ethanol and glycerol as sole carbon sources. Growth on these non-fermentable 166 carbohydrates is only supported by cellular respiration, effectively shunting intracellular copper into the 167 mitochondria to the copper-dependent enzymes required for oxidative phosphorylation. Incubation of 168 the WT and cbi1∆ C-WT strains in YPEG + 0.01% SDS caused a ~40% growth reduction, and an even more 169 severe reduction of growth (~90%) in the cbi1∆ strain ( Fig 1D). This growth impairment was 170 complemented in all strains by supplementation with CuSO 4 , suggesting that shuttling of copper from 171 other pathways towards respiration influences the ability of Cn to withstand cell surface stress. Depletion 172 of Cbi1 further decreased cell fitness under these conditions. 173 We also tested the cell wall integrity of the Cn ccc2∆ mutant, a strain defective in copper transport 174 withinin the secretory pathway and subsequent altered metalation of secreted proteins [2,23]. A modest 175 growth defect on Congo red was observed for the ccc2∆ strain, and the defect developed independent of 176 copper availability (Fig. 1E). These results suggest that defective copper loading of enzymes in the 177 secretory compartment is likely not the cause of the Cn cell wall phenotypes observed during copper 178 deficiency.

179
The Cfo1 ferroxidase is a copper-dependent enzyme involved in high-affinity iron acquisition [24].
180 Therefore, loss-of-function mutations in Cuf1 or other components of the Cn copper uptake machinery 181 would be predicted to affect intracellular iron concentrations as well as copper levels. Additionally,9 182 previous studies demonstrated that iron homeostasis is important for proper fungal cell wall and 183 membrane architecture [25,26]. We therefore analyzed the effects of exogenous copper or iron on the 184 BCS-induced cell wall phenotypes of the cbi1∆ and cuf1∆ mutants. Individual supplementation of the 185 growth medium with copper,but not iron, restored growth to the cbi1∆ and cuf1∆ strains in the presence 186 of cell wall stress and copper depletion ( Fig 1F). 187

Changes in Cn cell wall composition in response to defective copper homeostasis
188 To further characterize the specific role of Cbi1 in cell wall homeostasis during copper stress, we 189 used transmission electron microscopy (TEM) to characterize the cell wall architecture of the wildtype 190 (WT) and cbi1∆ strains incubated in both copper-sufficient and copper-deficient growth conditions (Fig 2).

191
In copper-sufficient conditions the Cn WT cell wall consists of two layers characterized by differing 192 electron density [27][28][29]. Extracellular copper sequestration by the highly copper-specific chelator BCS 193 resulted in decreased electron density in the innermost cell wall layer composed primarily of chitin and 194 chitosan (Fig 2 A, C). These chito-oligomers efficiently bind bivalent metals (including copper ions), 195 consistent with the higher electron density of this cell wall layer during copper sufficiency [14,15]. The 196 cell wall of the cbi1 mutant strain was similar to WT during copper sufficiency, displaying distinct layers 197 based on electron density. However, the copper-starved cbi1 strain demonstrated a reduction in total 198 cell wall thickness compared to both the WT strain and the copper-sufficient cbi1∆ cells ( Fig 2B). Also in 199 contrast to WT, there was no reduction in the inner cell wall electron density in the cbi1 mutant strains 200 during copper deficiency (Fig 2A, C). These results suggest a model in which Cbi1 is required for the release 201 of cell wall-bound metals during copper starvation.  2D).

232
To explore the functional relevance of changes in Cn cell wall gene transcript abundance as a 233 function of copper availability, we quantified β-glucan and chitin/chitosan levels in the WT, cbi1∆ mutant, 234 and cbi1∆ C-WT complemented strains after incubation for 24h in copper-sufficient (YPD + 10 µM CuSO 4 ) 235 and copper-deficient (YPD + 250 µM BCS) conditions. No significant changes were detected in total cell 236 wall β-glucan between the WT and cbi1∆ cells in either growth condition (Supp Fig 2A). However, the 237 cbi1∆ mutant exposed to copper-deficiency displayed a greater than 50% reduction in total cell wall chitin 238 and in chitosan compared to WT and complemented strains (Fig 3 A-B). These results are consistent with 239 reduced transcript levels for the CDA2 and FPD1 chitin deacetylase genes in the copper-starved cbi1∆ 240 strain. Therefore, the reduction of cell wall thickness and altered cell wall integrity in the cbi1∆ strain 241 during copper starvation is, in part, likely due to a reduction in the inner cell wall chito-oligomer layer.

242
To examine detailed changes in patterns of chitin and chitosan deposition, we performed 243 microscopy using chitin-and chitosan-specific fluorescent stains. We double-stained Cu-sufficient and Cu-244 deficient WT, cbi1∆ and cbi1∆ C-WT cells with Calcofluor white (CFW), a small molecule globally staining 245 chitin, and AlexaFluor488-conjugated wheat germ agglutin (WGA-Alexa 488), a lectin that binds exposed 246 chito-oligomers. To a similar extent as in the biochemical chitin assays, we observed a reduction in CFW 247 staining intensity of Cu-deficient cbi1∆ cells (Fig 3 C, Supp Fig 2B). WGA staining of WT and cbi1∆ C-WT 248 complemented cells only revealed exposed chito-oligomers, primarily at regions of cell separation, 249 budding sites, and bud scars (

254
We also performed flow cytometry on CFW and WGA-Alexa 488 stained cells to more precisely 255 quantify the altered chito-oligomer staining pattern in these strains (

264
To explore the role of chitin and chitosan for modulating the resistance to copper stress, we 265 assessed the growth effects during low and high copper stress for strains with mutations in either copper 266 homeostasis or chitosan synthesis (Fig 4). As previously described, we observed poor growth of the cbi1∆ 267 and ctr1∆ strains during copper deficiency ( Fig 4B) To assess the role of the chitin and chitosan inner layer on resistance to copper toxicity, we tested 277 the chs3∆ and cda2∆ mutant strains for growth phenotypes in the presence of increasing copper 278 concentations. The chitosan-deficient chs3∆ strain was more sensitive to high copper stress than the wild-279 type, and similar in its copper sensivity to strains with mutations in the Cn metallothionein genes CMT1 280 and CMT2 that mediate scavenging of excess copper ( Fig 4C). In contrast to chs3∆, the cda2∆ strain, with 281 a mutation in a single chitin deacetylase gene but relatively preserved cell wall chitosan levels, 282 demonstrated resistance to toxic copper levels compared to WT. The cbi1∆ mutant displayed a similar 283 copper resistance profile as the cda2∆ strain. This increased copper resistance was not shared with the 284 ctr1 or ctr4 copper transporter mutants, suggesting that altered copper transport was not responsible 285 for this phenotype. This finding suggests an unexpected new role for Cbi1 in modulating growth during 286 high copper stress, potentially by its known role in modulating CDA2 function.

288
Cell wall changes in response to low copper stress lead to increased macrophage activation and altered 289 caspofungin tolerance 290 We tested the physiological relevance of altered copper homestasis to other infection-related 291 processes involving the cell wall. C. neoformans strains with enhanced exposure of cell wall chito-292 oligomers often display increased activation of host innate immune cell activity [31,32]. To investigate 293 the physiological consequences of the aberrant cbi1 cell wall structure in host cell interactions, we co-294 incubated C. neoformans strains with murine bone marrow-derived macrophages (BMM), assessing TNF-α 295 production as a marker of host immune cell activation. Macrophages exposed to copper-deficient cbi1∆ 296 cells showed a statistically significant increase in TNF-α secretion compared to macrophages co-incubated 297 with similarly treated WT or complemented strains ( Fig 5A). No differences in TNF- production were 298 noted for macrophages incubated with any of these strains grown in the presence of copper. production are more susceptible to this drug

569
Once the dehydration was complete, ethanol was removed and the dehydrated sample was 570 embedded into resin (53,5% (w/v) resin, 20.5% (w/v) DDSA, 26% (w/v) NMA, 1.4% (v/v) DMP-30). The 571 sample were incubated in resin mix at RT overnight. The following day, samples were incubated at 50-572 60C for 10 minutes, the old resin mix was replaced by freshly made resin mix and incubated for 10 mins 573 at RT, followed by 10 minutes at at 50-60C. This resin wash step was repeated one more time, followed 574 by a 48h incubation at 50-60C. to an OD 600 of 0.05 in 50 mL YPD + 10 μM CuSO 4 (=Cu sufficiency) or 250 μM BCS (=Cu deficiency) and 598 cultivated for 24h at 30°C. The following day, 10 to 25 mL of the cells were harvested and washed twice 599 with dH 2 O. In the last wash step, cells were counted, spun down and lyophilized. Chitin and chitosan levels 600 were quantified from lyophilized yeast using a modified MBTH (3-methyl-benzothiazolinone hydrazine 601 hydrochloride) method as previously described [13]. β-glucan was quantified using the megazyme yeast