Alcohols Inhibit Translation to Regulate Morphogenesis in C. Albicans 6 ⇑ 7 Nkechi E

YFGBI 2800 No. of Pages 11, Model 5G 6 April 2015

Fungal Genetics and Biology xxx (2015) xxx–xxx 1 Contents lists available at ScienceDirect

Fungal Genetics and Biology

journal homepage: www.elsevier.com/locate/yfgbi

2 Regular Articles

5 4 Alcohols inhibit translation to regulate morphogenesis in C. albicans 6 ⇑ 7 Nkechi E. Egbe, Caroline M. Paget, Hui Wang, Mark P. Ashe

8 Faculty of Life Sciences, University of Manchester, Manchester M13 9PT, United Kingdom

9 10 article info abstract 2412 13 Article history: Many molecules are secreted into the growth media by microorganisms to modulate the metabolic and 25 14 Received 4 December 2014 physiological processes of the organism. For instance, alcohols like butanol, ethanol and isoamyl alcohol 26 15 Revised 13 March 2015 are produced by the human pathogenic fungus, Candida albicans and induce morphological differentia- 27 16 Accepted 28 March 2015 tion. Here we show that these same alcohols cause a rapid inhibition of protein synthesis. More specifi- 28 17 Available online xxxx cally, the alcohols target translation initiation, a complex stage of the gene expression process. Using 29 molecular techniques, we have identified the likely translational target of these alcohols in C. albicans 30 18 Keywords: as the eukaryotic translation initiation factor 2B (eIF2B). eIF2B is the guanine nucleotide exchange factor 31 19 Candida albicans for eIF2, which supports the exchange reaction where eIF2.GDP is converted to eIF2.GTP. Even minimal 32 20 Protein synthesis 21 Filamentous growth regulation at this step will lead to alterations in the levels of specific proteins that may allow the exigen- 33 22 Eukaryotic initiation factor 2B (eIF2B) cies of the fungus to be realised. Indeed, similar to the effects of alcohols, a minimal inhibition of protein 34 23 synthesis with cycloheximide also causes an induction of filamentous growth. These results suggest a 35 molecular basis for the effect of various alcohols on morphological differentiation in C. albicans. 36 Ó 2015 Published by Elsevier Inc. 37 38

39 40 41 1. Introduction pseudohyphal growth. It has been suggested that these alcohols 64 might signal nitrogen scarcity to elicit these effects (Dickinson, 65 42 Candida albicans is a major fungal pathogen of humans, causing 1996). Consistent with this hypothesis, both non-pathogenic and 66 43 life threatening septicaemic infections, especially in immunocom- pathogenic yeasts switch from the budding yeast form to the fila- 67 44 promised individuals. In addition, C. albicans can cause frequent mentous form when starved for nitrogen (Csank and Haynes, 2000; 68 45 and recurrent infections that are difficult to treat in healthy Gimeno et al., 1992). 69 46 individuals. C. albicans has a variety of properties that have been The role of fusel alcohols and other alcohols in signalling to alter 70 47 implicated in pathogenicity, including the ability to switch growth the growth and physiology of Candida species is starting to become 71 48 between a variety of morphological forms, such as yeast, more widely appreciated. Martins et al. (2007) detected Isoamyl 72 49 pseudohyphal and hyphal forms. Such switches in morphology alcohol, 2-phenethylethanol, 1-dodecanol in supernatants of 73 50 occur as a response to environmental cues, where external stimuli planktonic and biofilm forms of C. albicans and C. dubliniensis, 74 51 are communicated via different signal transduction pathways where these alcohols inhibited the morphological transition from 75 52 (Biswas et al., 2007; Brown and Gow, 1999; Liu, 2001). yeast to filamentous form by over 50%. Aromatic alcohols i.e. phe- 76 53 Pseudohyphae are morphologically distinguishable from hyphae; nethyl alcohol, tyrosol and tryptophol are produced by C. albicans, 77 54 they also differ fundamentally in their cell cycle organisation and especially under nitrogen poor conditions (Chen and Fink, 2006; 78 55 in mechanisms of polarised growth (reviewed in Sudbery (2011)). Chen et al., 2004; Ghosh et al., 2008; Lingappa et al., 1969; 79 56 A variety of yeasts secrete alcohols such as butanol and isoamyl Martins et al., 2007). These aromatic amino alcohols induce 80 57 alcohol during growth (Dickinson et al., 1998, 2003; Ghosh et al., pseudohyphal formation in Saccharomyces cerevisiae, with tyrosol 81 58 2008; Hazelwood et al., 2008; Martins et al., 2007; Webb and also inducing germ-tube formation in C. albicans (Chen and Fink, 82 59 Ingraham, 1963). These alcohols are major components of fusel 2006). 83 60 oil; a by-product of yeast fermentation, and hence they have been In S. cerevisiae a number of examples exist where nutritional 84 61 collectively termed fusel alcohols (Webb and Ingraham, 1963). The alterations induce filamentous growth and also affect protein syn- 85 62 addition of fusel alcohols to a yeast culture has been shown to thesis. Hence, nitrogen limitation affects translation initiation and 86 63 induce a range of specific morphological effects such as also induces pseudohyphal growth in diploids (Gancedo, 2001; 87 Gimeno et al., 1992; Ibrahimo et al., 2006). Moreover, the effects 88 of various alcohols as well as glucose depletion have been shown 89 ⇑ Corresponding author. Tel.: +44 (0)161 306 4164. to modulate both protein synthesis and filamentous growth in S. 90 E-mail address: [email protected] (M.P. Ashe). cerevisiae (Ashe et al., 2000, 2001; Lorenz et al., 2000). 91

http://dx.doi.org/10.1016/j.fgb.2015.03.008 1087-1845/Ó 2015 Published by Elsevier Inc.

Please cite this article in press as: Egbe, N.E., et al. Alcohols inhibit translation to regulate morphogenesis in C. albicans. Fungal Genet. Biol. (2015), http:// dx.doi.org/10.1016/j.fgb.2015.03.008 YFGBI 2800 No. of Pages 11, Model 5G 6 April 2015

2 N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx

92 Translation initiation is the predominant phase where protein Table 1 93 synthesis is regulated. The process of translation initiation requires Yeast strains used in this study. 94 interaction of the initiator methionyl tRNA with eukaryotic initia- Strain Genotype Source 95 tion factor 2 (eIF2) bound to GTP to form a ternary complex. This is name 96 then recruited to the small ribosomal subunit (40S) to form the 43S CAI4 ura3::kimm434/ura3::kimm434 Fonzi and Irwin 97 complex. Following the binding of the 43S complex to the mRNA (1993) 98 and start codon recognition, the GTP on eIF2 is hydrolysed and gcn2D ura3::kimm434/ura3::kimm434 gcn2::hisG/ Tournu et al. gcn2::hisG (2005) 99 released as eIF2-GDP (Hershey and Merrick, 2000). The initial yMK2313 ura3::kimm434/ura3::kimm434 GCD1- This study 100 formation of the ternary complex represents a prominent point GFP::NAT/GCD1 101 of translational regulation across all eukaryotes. For instance, in efg1D ura3::kimm434/ura3::kimm434 efg1::hisG/ Lo et al. (1997) 102 S. cerevisiae activation of the eIF2a kinase, Gcn2p, during amino efg1::hisG-URA3-hisG cph1D ura3::kimm434/ura3::kimm434 cph1::hisG/ Liu et al. (1994) 103 acid starvation leads to the phosphorylation of the a subunit of cph1::hisG-URA3-hisG 104 eIF2. This phosphorylated form binds tightly and sequesters 105 eIF2B; the guanine nucleotide exchange factor that is responsible 106 for recycling eIF2-GDP to eIF2-GTP. Hence, the inhibition of eIF2B 107 results in less eIF2-GTP, less ternary complex and reduced levels growing strains in YPD to OD600 0.1. Butanol (0.5%, 1% and 2%), iso- 154 108 of global protein synthesis (Hinnebusch, 2005; Pavitt, 2005). amyl alcohol (0.1%, 0.25%, 0.5% and 1%) and ethanol (2%, 4%, 6% and 155 109 Even though regulation of eIF2B results in a global regulation of 8%) were added and growth was assessed hourly. 156 110 translation, specific mRNAs such as GCN4 are activated in a mecha- 111 nism requiring four upstream open reading frames (uORFs) 2.2. Morphogenesis assays 157 112 (Hinnebusch, 2005). GCN4 encodes a transcription factor that regu- 113 lates the expression of a host of genes involved in the regulation of Strains were grown at 30 °C, washed in water to remove trace 158 114 nitrogen resources in cells (Natarajan et al., 2001). Therefore Gcn4p media constituents and re-inoculated into fresh YPD medium sup- 159 115 up-regulation reverses the starvation by promoting amino acid plemented with alcohols to OD600 0.1 then incubated at 37 °C. As 160 116 biosynthesis as part of the general control pathway. C. albicans also positive and negative controls, cells were re-inoculated into 10% 161 117 harbours a single eIF2a kinase, Gcn2p (Tournu et al., 2005). As in S. serum medium or YPD lacking alcohol, respectively. At various 162 118 cerevisiae, amino acid starvation inhibits global translation via this times during incubation (1 h, 2 h, 3 h, 4 h), the morphology of the 163 119 kinase while activating translation of GCN4, yet in this case there cells was monitored using a Nikon Eclipse E600 and Axiocam 164 120 are just three uORFs involved in the regulation (Sundaram and MRm camera. Images were acquired using Axiovision 4.5 software. 165 121 Grant, 2014b). The proportion of cells forming germ tubes or pseudohyphae were 166 122 In this current study, we have assessed the translational and counted using a haemacytometer. Each count was repeated three 167 123 morphological responses of C. albicans to the effects of various times and the mean of the three counts recorded. Colony morphol- 168 124 alcohols. We confirm previous analyses showing that fusel alcohols ogy on solid media was assessed by plating on YPD agar or YPD 169 125 and ethanol induce a switch from vegetative growth to pseudohy- agar supplemented with the indicated alcohol, 10% serum or cyclo- 170 126 phal growth. We show that the alcohols also rapidly inhibit trans- heximide concentrations ranging from 10 lg/ml to 200 lg/ml. 171 127 lation initiation and induce the translation of the GCN4 mRNA in Micro-colonies were photographed using a Nikon Eclipse E600 172 128 both the wild type and the gcn2D strains. Therefore, Gcn2p is not and Axiocam MRm camera and the images were acquired using 173 129 involved in this pathway of translation regulation in C. albicans Axiovision 4.5 software. 174 130 and a pathway that is independent of the eIF2a kinases must regu- 131 late translation to induce GCN4 expression. Given the known 2.3. Polysome analysis 175 132 effects of fusel alcohols targeting eIF2B in S. cerevisiae, these results 133 are entirely consistent with the presence of a similar mechanism in C. albicans strains were grown to an OD600 of 0.7 and treated 176 134 C. albicans and opens up the possibility that such a regulation could with butanol, isoamyl alcohol and ethanol as described above. 177 135 be involved in morphogenetic alterations and pathogenicity. Extracts were prepared in 1 mg/ml cycloheximide and these were 178 136 Indeed we show that minimal inhibition of translation with cyclo- layered onto 15–50% sucrose gradients. The gradients were sedi- 179 137 heximide leads to a similar degree of filamentation as alcohol mented via centrifugation at 40,000 rpm using a SW41 Beckman 180 138 treatment. rotor for 2.5 h (Fullerton, CA) and the A254 was measured continu- 181 ously to give the traces shown in the figures (Ashe et al., 2000). 182

35 139 2. Materials and methods 2.4. [ S] methionine incorporation assay 183

140 2.1. Media and growth conditions C. albicans strains were grown to OD600 of 0.7 in synthetic com- 184 plete dextrose (SCD) medium lacking methionine (Guthrie and 185 141 The Candida strains used in this study are listed in Table 1. The Fink, 1991). The culture was split into two flasks and methionine 186 142 CAI4 background is used throughout (Fonzi and Irwin, 1993), how- was added to a final concentration of 60 ng/ml, of which 0.5 ng/ 187 35 143 ever experiments carried out comparing the CAI4 and CAF2-1 ml was [ S] methionine (cell-labelling grade 1175 Ci/mmol; New 188 144 strains showed that the two strains responded similarly to various England Nuclear, Boston, MA). Alcohol at the indicated concentra- 189 145 alcohol treatments in terms of growth and filament formation tion was added to one flask and samples (1 ml) were taken and 190 146 (data not shown). Therefore, as CAI4 is the parental strain for the processed as described previously (Ashe et al., 2000). For each 191 147 gcn2D mutant, this strain was selected for use in this study. The experiment, three biological replicates were assessed. 192 148 strains were grown and maintained on rich yeast extract- 149 Peptone-Dextrose (YPD) media (2% (w/v), glucose; 2% (w/v) bac- 2.5. Western blot analysis of phosphoserine 51 eIF2a 193 150 topeptone; and 1% (w/v), yeast extract) or YPD agar solid media

151 (YPD with 2% agar). Unless otherwise stated, alcohols (butanol, 50 ml OD600 0.7 of culture grown in YPD was harvested in a 194 152 isoamyl alcohol and ethanol) were added for 15 min at the particu- clinical centrifuge at 5000 rpm for 5 min. The cells were lysed, 195 153 lar concentrations stated. Alcohol tolerance was assessed by and protein samples were prepared as described previously 196

N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx 3

197 (Ashe et al., 2000; Taylor et al., 2010). Immunoblots were probed (YPD) and found that over a period of 2 days aerobic growth, levels 252 198 using phosphospecific eIF2a antibodies and antibodies raised of ethanol, isoamyl alcohol and isobutanol accumulated to 0.74% 253 199 against yeast elongation factor 1 (Tef1p) as a loading control. (5.88 g/l), 111 ppm (90 mg/l) and 34 ppm (27 mg/l) respectively. 254 Under anaerobic conditions, over a period of 6 days, the level of 255 200 2.6. Reporter assays ethanol and isobutanol accumulated to 1.2% (9.5 g/l) and 22 ppm 256 (18 mg/l) respectively, whereas isoamyl alcohol was only detected 257 201 Luciferase assay was performed essentially as explained pre- later after 14 days at 87 ppm (70 mg/l). 258 202 viously (Srikantha et al., 1996). Briefly, cells were harvested and In order to analyse the possible implications of alcohol accumu- 259 lation on the growth and morphology of the CAI-4 strain of C. albi- 260 203 extracted using RLUC buffer (0.5 M NaCl, 0.1 M K2HPO4 (pH), 204 1 mM Na2EDTA, 0.6 mM sodium azide, 1 mM phenylmethylsul- cans, the response to added n-butanol, isoamyl alcohol (IAA) or 261 205 fonyl fluoride, 0.02% bovine serum albumin). The assay was ethanol was studied. Firstly, in response to n-butanol, growth 262 206 initiated by addition of 1.25 lM coelentrazine h (Promega) to cell was weakly inhibited by concentrations of 0.5% (v/v), while at 2% 263 207 extracts, and activity was measured using GloMax 20/20 (v/v) the growth of the strain was completely inhibited (Fig. 1A). 264 208 luminometer (Promega). Luciferase activity (RLU) is expressed as A similar scenario was evident after the addition of IAA and etha- 265 209 relative luminescence per 10 s/mg protein. As a control, nol, only the concentrations causing weak and complete inhibition 266 210 Luciferase mRNA was evaluated relative to GAPDH mRNA by varied. For instance, 1% (v/v) IAA caused complete growth inhibi- 267 211 qRT-PCR and no significant changes were elicited by any of the tion (Fig. 1A), whereas higher concentrations of ethanol (6–8% (v/ 268 212 treatments used (data not shown). v)) were required to bring about the same effect (Fig. 1A). 269 Therefore, in terms of toxicity, the alcohols prevent growth at 270 concentrations of 2% (v/v) n-butanol, 1% (v/v) isoamyl alcohol 271 213 2.7. Microscopy and 6–8% (v/v) ethanol (Fig. 1A); levels that are significantly higher 272 than can be observed to accumulate in C. albicans. 273 214 Real-time 2D deconvolved projections from continuous The inhibition of growth of C. albicans by the alcohols used in 274 215 z-sweep acquisition were generated using a Delta Vision RT micro- this study, prompted us to examine the rate of protein synthesis 275 216 scope (Applied Precision, Isaaquah, WA) with an Olympus 100 following alcohol treatment. Cells were labelled with [35S] 276 217 1.40 NA DIC oil PlanApo objective (Melville, NY) and Roper methionine for 5 min prior to addition of a particular alcohol, after 277 218 CoolSnap HQ camera (Tucson, AZ) using Applied Precision which incorporation was measured for 10 min. In these experi- 278 219 Softworx 1.1 software and 2 2 binning at room temperature. ments, the addition of alcohols (2% (v/v) butanol, 1% (v/v) IAA or 279 220 Z-sweep acquisition allowed fast visualisation of all planes while 8% (v/v) ethanol) caused a 10-fold decrease in the rate of protein 280 221 minimising fluorescent bleaching. Time-course experiments were synthesis (Fig. 1B). Therefore, these alcohols inhibit both growth 281 222 performed by acquiring images every 5 s over 2 min. ImageJ and protein synthesis in C. albicans. 282 223 (http://rsb.info.nih.gov/ij/; NIH) was used to track the movement 224 of the 2B body on the images acquired over the time course. The 225 values calculated for the mean total distance moved by the 2B 3.2. Alcohols induce filamentation 283 226 body were subjected to statistical analysis via a two-sample t test 227 assessing significance after alcohol treatment relative to untreated On the basis of the growth considerations detailed above, an 284 228 cells. intermediate concentration of each alcohol (0.5% (v/v) n-butanol, 285 0.5% (v/v) isoamyl alcohol or 2% ethanol (v/v) was used to explore 286 229 2.8. Alcohol analysis the switch from vegetative to filamentous growth. It is well-estab- 287 lished that treatment of C. albicans with serum induces a robust 288 230 Strains were grown in YPD media in both aerobic flasks and switch to filamentous growth both in solid and liquid media 289 231 semi-anaerobic vials until steady state was reached. 2 ml samples (Feng et al., 1999; Liu, 2001). Indeed, microscopic examination of 290 232 were passed through a 0.22l filter into gas-chromatography (GC) cells early after serum treatment, suggest that the majority of cells 291 233 vials and analysed by GC-FID using an Agilent 6850A GC system were undergoing filamentous growth (Fig. 2C). All of the alcohols 292 234 with an automatic injector, sampler and controller (Agilent tested cause a switch to filamentous growth at some level 293 235 4513A). A DB-WAX capillary column (30 m 0.25 mm, 0.25 lM, J (Fig. 2A and B), but the response was significantly less robust than 294 236 & W Scientific) was used for separation. Samples were quantified for serum addition. For instance, roughly half the cells switch to a 295 237 relative to a standard of ethanol, isoamyl alcohol and isobutanol more pseudohyphal-like growth form early after alcohol addition 296 238 and butanol. (Fig. 2C), as judged by criteria such as relative cell shape, length 297 and the presence of constriction between cells (Sudbery et al., 298 2004; Sudbery, 2011). In addition, much less florid micro-colonies 299 239 3. Results are observed on alcohol containing relative to serum containing 300 solid agar (Fig. 2A). Overall these data show that the addition of 301 240 3.1. Growth and protein synthesis are inhibited in response to alcohols mildly inhibitory concentrations of n-butanol, isoamyl alcohol or 302 ethanol to the growth media of C. albicans slows the growth rate 303 241 The response of S. cerevisiae and C. albicans to alcohols varies and induces morphological differentiation both in solid and liquid 304 242 greatly: alcohols can induce morphological changes including growth media. 305 243 pseudohyphal and hyphal growth (Ashe et al., 2001; Chen and 244 Fink, 2006; Dickinson, 1996; Ghosh et al., 2008), or the alcohols 245 can prove toxic leading to cell death (Kitagaki et al., 2007). 3.3. Concentration dependent inhibition of translation initiation 306 246 Alcohols, under certain culture conditions, have also been reported 247 to inhibit filament formation in various Candida species (Chauhan To further investigate the stage of protein synthesis targeted by 307 248 et al., 2011, 2013; Martins et al., 2007). As a result it has been sug- the alcohols in C. albicans, and given that fusel alcohols bring about 308 249 gested that these alcohols are produced by yeast species to act as a rapid inhibition of translation at the initiation step in S. cerevisiae 309 250 concentration dependent signals. We investigated the level of such (Ashe et al., 2001), we set out to investigate if butanol and ethanol 310 251 alcohols in the CAI-4 strain of C. albicans grown on rich media exert a similar effect on translation initiation in C. albicans. 311

4 N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx

gcn2Δ A 0.6 0.6 0.6 WT

0.5 0.5 0.5 ) ) ) 1 1 1 0.4 0.4 0.4

0.3 0.3 0.3 Growth rate (h- Growth rate (h- Growth rate (h- 0.2 0.2 0.2

0.1 0.1 0.1

0 0 0 0 0.5 1 1.5 2 0 0.25 0.5 0.75 1 0 2 4 6 8 butanol (%) IAA (%) ethanol (%)

B 14 12 10 8 WT 6 gcn2Δ 4 translaon rate translaon

mean fold inhibion in inhibion fold mean 2 0 - 2% 8% 1% butanol ethanol IAA

Fig. 1. Concentration dependent inhibition of growth and protein synthesis by alcohols does not rely on GCN2. (A) Figure shows plots of growth rates at various

concentrations of ethanol, butanol and isoamyl alcohol (IAA) as indicated. Wild type CAI4 strain and the mutant gcn2D strain were grown to OD600 of 0.1, then treated and the

OD600 was taken hourly for 6 h to generate the growth rates plotted. (B) A bar chart depicting the fold inhibition in the rate of protein synthesis at the indicated concentrations of alcohol for wild-type and gcn2D cells. Translation rates were determined by measuring using 35S methionine incorporation over a 10 min period. Each treatment was analysed in triplicate; error bars are ±SEM.

312 Polysome analysis can be used to investigate both the level of pro- (Dever, 2002; Dever et al., 1992; Hinnebusch, 2005). Like S. cere- 343 313 tein synthesis occurring in cells and, under stress conditions, what visiae, C. albicans possesses a single eIF2a kinase gene GCN2, and 344 314 step of the translation pathway is inhibited (Ashe et al., 2001). An the encoded Gcn2p kinase has been shown to mediate a number 345 315 analysis of the distribution of polysomes across a sucrose gradient of responses to cellular stress (Tournu et al., 2005). Previous 346 316 revealed that at increasing concentration of the alcohols, a change studies have highlighted the similarities between C. albicans 347 317 in the polysome profiles was observed. The 80S peak increased and S. cerevisiae in terms of the Gcn2p-dependent regulation of 348 318 dramatically and the polysome peaks were reduced (Fig. 3A). translation initiation (Sundaram and Grant, 2014a; Tournu 349 319 This change in profile is termed polysome run-off and is character- et al., 2005). In both organisms, either amino acid starvation or 350 320 istic of an inhibition of initiation step of translation (Ashe et al., oxidative stress causes the activation of Gcn2p leading to 351 321 2001; Hartwell and McLaughlin, 1969). Once again, the concentra- increased phosphorylation of eIF2a and a resultant inhibition 352 322 tion required to induce these effects varied across the alcohols of translation initiation. Even though in S. cerevisiae it has been 353 323 tested (Fig. 3B). Butanol and IAA effectively inhibit translation at shown that fusel alcohols target translation in a Gcn2p-indepen- 354 324 2% (v/v) and 1% (v/v) respectively, whereas much higher concen- dent manner (Ashe et al., 2001), it was important in this current 355 325 trations of ethanol are required. study to independently assess the role of Gcn2p in the response 356 326 The difference in concentration dependence for the inhibition of to alcohols in C. albicans. 357 327 protein synthesis could reflect differences in the precise mecha- Therefore, a C. albicans homozygous GCN2 deletion mutant 358 328 nism by which translation initiation is inhibited; especially for (gcn2D) was used to assess the requirement for the Gcn2p kinase 359 329 ethanol relative to the other alcohols. However, these differences in the various alcohol-dependent effects described above. 360 330 are analogous to variations observed in the concentration depen- Consistent with studies on butanol in S. cerevisiae (Ashe et al., 361 331 dent inhibition of growth (Figs. 1 and 2), with the order of alcohol 2001), the inhibition of either growth or translation initiation by 362 332 potency being ethanol < n-butanol < isoamyl alcohol. On the basis butanol, IAA or ethanol in C. albicans is insensitive to the deletion 363 333 of this, it seems plausible that the inhibition of protein synthesis of the GCN2 gene (Figs. 1 and 3). In particular, a sensitive compar- 364 334 could be the cause of the growth inhibition observed for cultures ison of the polysome profiles shows that there is no obvious differ- 365 335 treated with these alcohols. However, it is unclear how the difference between the response of a gcn2D mutant and a wild type 366 336 ent alcohols impact upon protein synthesis and whether the strain in terms of inhibition of translation by either fusel alcohols 367 337 reduced rate of growth observed is important in the morpho- or ethanol (Fig. 3A). Furthermore, quantitation of the change in 368 338 genetic switching of colonies. polysome/monosome ratios caused by the alcohols suggests that 369 GCN2 is not required for the alcohol-dependent polysomal 370 339 3.4. Inhibition of translation initiation by alcohols in C. albicans does run-off (Fig. 3B). Therefore, it would appear that the Gcn2p kinase 371 340 not rely upon eIF2a kinase activation plays little or no role in the inhibition of translation initiation by 372 alcohols in C. albicans. These results also support a conclusion that 373 341 A key mechanism by which global translation initiation is treatment with alcohols elicits translational inhibition by a differ- 374 342 controlled in eukaryotes is via the activation of eIF2a kinases ent mechanism to amino acid starvation and oxidative stress. 375

N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx 5

YPDA + YPDA + YPDA + YPDA + A YPD Agar 10% serum 2% ethanol 0.5% butanol 0.5% IAA

Day 1

Day 2

YPD + YPD + YPD + YPD + B YPD 10% serum 2% ethanol 0.5% butanol 0.5% IAA

pseudohyphal C yeast hyphal-like 100 -like

% cells 40

0 serum 0.5% butanol 2% ethanol 0.5% IAA

Fig. 2. Alcohols cause morphological differentiation in C. albicans. (A) Serial dilutions of overnight cultures of the wild type CAI4 strain were spread onto YPD agar plates containing 0.5% butanol, 2% ethanol, 0.5% IAA or 10% serum. Representative microcolonies were photographed (10 magniﬁcation) after 1 and 2 days of growth as indicated. (B) Images of cells from cultures that were grown at 37 °C in the indicated media for 3 h. (C) Cells from the analysis in B were scored for yeast, hyphal and pseudohyphal growth in a cell counting chamber. Percentages given are an average (±SEM) calculated from three biological replicates.

376 3.5. Alcohols do not cause eIF2a phosphorylation, but do induce GCN4 was seen in the precise level of induction for both reporters 400 377 expression depending on the alcohol used, however, a significant induction 401 was always observed (Fig. 4A and B). In S. cerevisiae, the GCN4 402 378 Given that fusel alcohols inhibit translation initiation in S. cere- reporter system is often used as a means to indirectly assess the 403 379 visiae by targeting the guanine nucleotide exchange factor, eIF2B levels of ternary complex; where the activity of the guanine 404 380 (Ashe et al., 2001), we set out to investigate if the alcohols exert nucleotide exchange factor eIF2B is a key determinant. As similar 405 381 a similar effect in C. albicans. A key observation that pointed observations have been made using the GCN4 system in C. albicans 406 382 towards eIF2B as a target for butanol in S. cerevisiae was the (Sundaram and Grant, 2014), the induction of GCN4 in response to 407 383 demonstration that translation of a GCN4 reporter mRNA is up- alcohols is at least consistent with eIF2B being important in this 408 384 regulated after butanol addition (Ashe et al., 2001). A variety of cel- regulation. 409 385 lular stresses that inhibit eIF2B to reduce ternary complex and Similar to the results for global protein synthesis above, dele- 410 386 inhibit translation initiation, also induce GCN4 mRNA translation tion of the GCN2 gene did not impact significantly upon the level 411 387 (Hinnebusch, 2005; Tripathi et al., 2002). of luciferase induction observed from either reporter for any of 412 388 In order to test whether alcohol treatment in C. albicans induces the alcohols. Once again, this suggests that neither Gcn2p nor 413 389 Gcn4p translation leading to a transcriptional up-regulation of eIF2a phosphorylation is required for the translational effects of 414 390 genes harbouring GCRE elements in their promoters, we used butanol, IAA or ethanol in C. albicans. 415 391 strains bearing two different reporter genes. The first reporter con- However, even though Gcn2p is not required for the inhibition 416 392 tains five copies of the GCRE cloned in a basal promoter upstream of translation initiation, the induction of Gcn4p prompted us to 417 393 of a luciferase reporter gene (Srikantha et al. (1996). Following the investigate any changes in the levels of phosphorylated eIF2a fol- 418 394 addition of 1% butanol, 0.5% IAA or 6% ethanol, there was an lowing alcohol treatment. Initially, an attempt was made to detect 419 395 increase in GCRE-Luc expression relative to untreated controls phosphorylated eIF2a in C. albicans cells using phosphospecific 420 396 (Fig. 4A). Similarly, using strains bearing the second reporter, a antibodies to eIF2a phosphoserine 51, but a very low basal signal 421 397 GCN4-Luc reporter construct where the 50-leader sequence of the for phosphorylated eIF2a is only just detectable after prolonged 422 398 GCN4 gene was cloned upstream of the luciferase gene, all three exposure for an untreated sample (Fig. 4C, long vs. short). The level 423 399 alcohols used in this study caused induction (Fig. 4B). Variation of this signal was reduced by butanol treatment, suggesting that 424

6 N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx

A wild type 80S gcn2∆ mutant sedimentation

80S

polysomes 60S 40S 60S 40S polysomes

untreated 0.5% 1% 2% Butanol untreated 0.5% 1% 2%

untreated 2% 4% 6% 8% Ethanol untreated 2%4% 6% 8%

untreated 0.1% 0.25% 0.5% 1% IAA untreated 0.1% 0.25% 0.5% 1%

B 1 1 1

0.5 0.5 0.5 Polysome/monosome ratio Polysome/monosome Polysome/monosome ratio Polysome/monosome Polysome/monosome ratio Polysome/monosome

0 0 0 012 048 00.51 % butanol % ethanol % IAA

Fig. 3. Translation initiation is inhibited by alcohols in CAI4 and gcn2D strains. (A) Figure shows polysome analyses assessing the effect of alcohols on translation initiation in the CAI4 wild type and gcn2D strains of C. albicans. Strains were grown in YPD and the indicated concentrations of the alcohols were added for 15 min prior to extract preparation. Extracts were sedimented on 15–50% sucrose gradients and the absorbance at 254 nm was continuously measured to generate the traces depicted. The position of 40S, 60S, 80S and polysome peaks are labelled and the direction of sedimentation is also depicted. (B) Quantitation of the polysome: monosome ratio across the polysome proﬁles. Solid lines represent the wild type, dashed lines represent gcn2D strain.

425 eIF2a may have become dephosphorylated. On account of this very Therefore, it appears that in both C. albicans and S. cerevisiae the 439 426 low basal level, we decided to investigate the impact of alcohol alcohols are activating some phosphatase activity to dephos- 440 427 treatment after a preinduction of eIF2a phosphorylation. In order phorylate eIF2. Indeed previous work in S. cerevisiae suggests that 441 428 to achieve a high level of eIF2a phosphorylation prior to alcohol a type 2A-related phosphatase (Sit4p) might be somehow involved 442 429 treatment, the strains were ﬁrst pretreated with 1 mM Cadmium in the dephosphorylation of eIF2a after exposure to fusel alcohols 443 430 Sulphate which signiﬁcantly increases eIF2a phosphorylation (Taylor et al., 2010). In sharp contrast to the effects of the fusel 444 431 (Fig. 4C). Subsequent treatment with butanol and IAA cause a dose alcohols, ethanol had little or no observable effect on the level of 445 432 dependent de-phosphorylation of eIF2a (Fig. 4C). In contrast, no phospho-eIF2a (Fig. 4C). If, as suggested by the GCRE/GCN4 repor- 446 433 phosphorylation of eIF2a was detected in a gcn2D mutant indicat- ter assays, the alcohols all inhibit translation initiation by a similar 447 434 ing that phosphorylation is entirely dependent on the Gcn2p mechanism, the fact that they vary in their capacity to induce 448 435 protein kinase (Fig. 4C). This result is similar to the effects of both eIF2a dephosphorylation once again suggests that the dephospho- 449 436 fusel alcohols and volatile anaesthetics in S. cerevisiae, where rylation has no bearing on the control of protein synthesis. 450 437 inhibition of translation initiation also leads to dephosphorylation Therefore, our interpretation of these data is that fusel alcohols 451 438 of eIF2a (Palmer et al., 2005). induce eIF2a dephosphorylation by activating a phosphatase 452

N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx 7

Δ gcn2Δ A WT gcn2 B WT 3 4

3 2

2 GCN4::Luc GCRE::Luc 1 1

0 0 6% 1% 0.5% - - 6% 1% 0.5% ethanol butanol IAA ethanol butanol IAA

C Wild type gcn2Δ

Pretreatment - - + + + + Pretreatment - - + + + + + butanol (%) - 2 - 0.5 1 2 butanol (%) - 2 - 0.5 1 2 WT short P-eIF2α P-eIF2α long P-eIF2α Tef1p Tef1p

Pretreatment - - + + + + + Pretreatment - - + + + + + IAA (%) - 2 - 0.1 0.5 1 2 IAA (%) - 2 - 0.25 1 2 WT P-eIF2α short P-eIF2α Tef1p long P-eIF2α Tef1p

Pretreatment - - + + + + + Pretreatment - + + + + + ethanol (%) - 8 - 2 4 6 8 ethanol (%) - 2 4 6 8 WT short P-eIF2α P-eIF2α long P-eIF2α Tef1p Tef1p

Fig. 4. Alcohols do not induce eIF2a phosphorylation, but does induce GCN4 activation. (A and B) Luciferase assays from strains bearing either GCRE::Luc or GCN4::Luc reporters treated with the alcohols and concentrations indicated for 2 h. Values are plotted relative to untreated and represent a mean of three biological replicates. Error bars are ±SEM and all the alcohol dependent inductions are statistically signiﬁcant (p < 0.05). (C) Western blots on extracts from wild type and gcn2D strains probed with antibodies to Tef1p and phospho-eIF2a. Strains were pretreated with 1 mM Cadmium then treated with various concentrations of alcohols (% v/v) for 15 min as indicated. Long and short exposures to ﬁlm are indicated.

453 pathway that is not induced by ethanol, but this pathway is not cerevisiae that butanol impedes the movement of the 2B body in 473 454 required for the translational regulation. a manner that correlates with the sensitivity/resistance of strains 474 (Taylor et al., 2010). To investigate whether butanol, IAA or ethanol 475 impact on the dynamics of the 2B body in C. albicans, we performed 476 455 3.6. eIF2B is the likely translational target of alcohol treatment in C. time lapse microscopy. In all, 25 images were collected over a 2- 477 456 albicans min period and the 2B bodies were tracked to follow their move- 478 ment over the time course. Quantitation of the total distance 479 457 eIF2B is a guanine nucleotide exchange factor that has a critical moved over the time course reveals that 1% butanol causes a 480 458 role in the formation of ternary complex. Previous work has impli- decrease of 50%, while 6% ethanol reduces the movement by about 481 459 cated eIF2B as the target for the alcohol regulation of protein syn- 70% (Fig. 5B). IAA caused a less pronounced yet still significant 482 460 thesis in S. cerevisiae. A key observation concerns the dynamic reduction in the movement of the eIF2B body. The observation that 483 461 localisation of eIF2B in response to alcohols (Taylor et al., 2010). the 2B body moves less after exposure to concentrations of these 484 462 In S. cerevisiae eIF2B is localised to a single large cytoplasmic body alcohols, which also inhibit translation initiation, correlates with 485 463 termed the ‘2B body’, where guanine nucleotide exchange has been the possibility that at least part of the inhibition of translation 486 464 suggested to occur (Campbell et al., 2005). The 2B body was also initiation is linked to increased tethering of the 2B body. 487 465 found to be dynamic; transiting all areas of the cytoplasm, and 466 alcohol treatment limits this movement (Taylor et al., 2010). 467 We generated heterozygous strains of C. albicans CAI4 strain 3.7. Alcohols still regulate protein synthesis and promote filamentation 488 468 where one of the genomic copies of the GCD1 gene (encoding the in Candida signalling pathway mutants 489 469 c subunit of eIF2B) is C-terminally tagged with GFP. Using epifluo- 470 rescence microscopy, we observed that, in C. albicans, eIF2B loca- One of the key virulence factors of C. albicans is its ability to 490 471 lises to a large cytoplasmic body in a similar manner to what has undergo reversible morphological transitions (Lo et al., 1997; 491 472 been observed in S. cerevisiae (Fig. 5A). It has been shown in S. Odds, 1994). Several signalling pathways control morphogenesis 492

8 N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx

Individual slls Time lapse merge A

B UT 30

* 1% 20 butanol *

10 Distance moved moved (μ) Distance * 6% ethanol

0.5% IAA

Fig. 5. Fusel alcohols and ethanol impede the movement of eIF2B bodies. (A) Images from time-lapse microscopy studies using an eIF2Bc-GFP expressing C. albicans strain. The strains were incubated in media with 1% butanol, 0.5% IAA or 6% ethanol, or they were left untreated (UT) for 15 min as indicated. Each row contains two stills from a series of 25 images over a period of 2 min, as well as a merged image of all 25 stills, which serves to depict the total extent of 2B body movement. (B) Bar chart depicting the mean distance moved in l over a 2-min period from 24 time-lapse experiments. Error bars, ±1 SEM; ⁄ p < 0.01.

493 in C. albicans. These include the MAPK pathway which is dependent 3.8. Delicate inhibition of protein synthesis impacts upon 522 494 upon Cph1p, a homologue of S. cerevisiae Ste12p (Liu et al., 1994), morphogenesis 523 495 and the Ras-cAMP signalling pathway which requires functional 496 Efg1p (Bockmühl and Ernst, 2001). Both pathways are thought to In order to further characterise any link between the regulation 524 497 activate filamentous growth in response to starvation or serum. of protein synthesis and the switch to filamentous growth, we 525 498 Transitions from yeast to filamentous growth are therefore com- studied the effects of cycloheximide (CHX) on morphological dif- 526 499 promised in cph1/cph1 or efg1/efg1 mutants (Lo et al., 1997). ferentiation. CHX is a widely used inhibitor of eukaryotic protein 527 500 In order to assess whether there is any correlation between the synthesis that blocks the elongation phase of eukaryotic transla- 528 501 activities of these transcription factors and the translational tion by interfering with deacylated tRNA release from the ribosome 529 502 response of C. albicans to fusel alcohols or ethanol, we analysed to block eEF2-mediated translocation (Schneider-Poetsch et al., 530 503 the sensitivity of the efg1D and the cph1D mutant strains to vary- 2010). In S. cerevisiae, CHX inhibits translation across many strains, 531 504 ing concentrations of the alcohols. In terms of protein synthesis, however resistant mutations are possible that can mean greater 532 505 assessed via polysome analysis, the cph1D mutant responds to concentrations are required for inhibition (Fried and Warner, 533 506 the alcohols in a similar manner to the wild type (Fig. 6A). For 1982). Indeed, in the Candida or Kluyveromyces genera, a polymor- 534 507 the efg1D mutant, while the response to ethanol is very similar phism in the RPL41A gene has been implicated in their CHX resis- 535 508 to wild type, this mutant exhibits slight resistance to IAA and buta- tance (Dehoux et al., 1993; Kawai et al., 1992). As a result high 536 509 nol (Fig. 6A). Equally, in terms of morphogenesis, especially in the concentrations 1 mg/ml CHX are necessary in C. albicans to inhibit 537 510 efg1D mutant, the response to butanol and IAA is marginally less protein synthesis (data not shown). 538 511 robust early after treatment (Fig. 6B). However, overall it is clear In order to study the effects of CHX on morphological dif- 539 512 that in the two mutants not only is translation still inhibited by ferentiation in C. albicans, we tested various dilutions of CHX. 540 513 all three alcohols but also the alcohols still cause changes in mor- Interestingly, lower sub-lethal concentrations of CHX, 10–100 lg/ 541 514 phogenesis (Fig. 6B). This result suggests that alcohols might elicit ml induce some filamentation, but as the concentration increase 542 515 a switch from vegetative to filamentous growth in a pathway that beyond 200 lg/ml, cells grow in the yeast form as regular colonies 543 516 does not rely solely on either of the transcription factors mutated (Fig. 7A). All of the concentrations used are insufficient to com- 544 517 in these strains. Furthermore, given the lack of connection between pletely inhibit growth. These observations are similar to the impact 545 518 these pathways, it remains entirely plausible that the regulation of of fusel alcohols or ethanol on C. albicans, where lower concentra- 546 519 protein synthesis by alcohols could represent a key determinant in tions that result in slight alterations to translation, induce 547 520 the induction of filamentation that primes cells for changes in gene pseudohyphal formation, while higher concentrations induce yeast 548 521 expression at the transcriptional level. form growth (Fig. 7B and C). On the basis of these results, it appears 549

N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx 9

A efg1Δ

- 0.5% 1%IAA 1% 2%butanol 6% 8% ethanol

cph1Δ

- 0.5% 1%IAA 1% 2%butanol 6% 8% ethanol

10% 0.5% 2% 0.5% B serum butanol ethanol IAA

efg1Δ

cph1Δ

Fig. 6. Separate alcohols affect morphological transcriptional regulator mutants. (A) Figure shows polysome analyses assessing the effect of alcohols on translation initiation in efg1D and cph1D mutant strains of C. albicans. Yeast strains were grown in YPD and various concentrations of alcohols were added as indicated for 15 min prior to extract preparation. Extracts were sedimented on 15–50% sucrose gradients and the absorbance at 254 nm was continuously measured. (B) Figure shows images of the mutant strains relative to wild type after treatment in liquid culture for 3 h with various alcohols, as indicated.

550 that low concentrations of these chemicals have a dual effect; they concentrations of other agents that target protein synthesis, such 567 551 mildly inhibiting protein synthesis and induce morphological as cycloheximide, also lead to the same effect. These results are 568 552 differentiation. suggestive that weak regulation of protein synthesis, where the 569 presumed focus is proteome alteration, appears to contribute to 570 the morphogenetic switch from vegetative to filamentous growth 571 553 4. Discussion in response to alcohols. 572 The fact that alcohols promote morphogenetic change is not 573 554 Microorganisms are known to produce a range of molecules unprecedented. In our study, we show that addition of ethanol 574 555 that regulate their growth, morphology, metabolism and virulence. (1–3%), butanol (0.5–1%) or isoamyl alcohol (0.1–0.5%) induces 575 556 In the genera Candida and Saccharomyces, various secreted alcohols pseudohyphal formation in C. albicans both in liquid and solid 576 557 have been characterised from culture supernatants, these include: media. Previous work has also shown that the addition of fusel 577 558 fusel alcohols, aromatic alcohols and ethanol (Martins et al., 2007; alcohols (butanol or isoamyl alcohol) to both C. albicans and S. cere- 578 559 Webb and Ingraham, 1963). In this study, we show that C. albicans visiae can induce a range of morphological effects, including fila- 579 560 produces both ethanol and fusel alcohols (isoamyl alcohol and ment formation (Dickinson, 1996; Lorenz et al., 2000). In terms 580 561 isobutanol) under both aerobic and semi-anaerobic culture condi- of the impact of ethanol on C. albicans, previous studies are some- 581 562 tions. We demonstrate that such alcohols inhibit growth and pro- what perplexing: ethanol has been described to induce germ tube 582 563 tein synthesis at high concentrations via a mechanism that likely formation (Pollack and Hashimoto, 1985; Zeuthen et al., 1988) yet 583 564 results from a targeted regulation of eIF2B. At sub-inhibitory con- at 4%, ethanol inhibits this process (Chauhan et al., 2011; Chiew 584 565 centrations these alcohols lead to changes in morphogenesis, in et al., 1982) Furthermore, concentrations of ethanol ranging from 585 566 that they promote filamentation. Indeed sub-inhibitory

10 N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx

10μg/ml 50μg/ml 200μg/ml A UT cycloheximide cycloheximide cycloheximide

3% 6% B UT Serum ethanol ethanol

0.5% 2% C UT butanol butanol

Fig. 7. Inhibition of protein synthesis by alcohols also impacts on ﬁlamentation in C. albicans. Figure shows micro-colonies formed after exponentially grown cells were plated on solid media containing the indicated supplements: (A) cycloheximide, (B) serum or ethanol, and (C) butanol, then grown for 1 day.

586 0.06% to 8% reduce the level of germ tube formation observed in yeast the response to ethanol is very similar to the response to 618 587 response to a range of inducing conditions (Chauhan et al., 2011). other alcohols in that translation is inhibited at relatively low con- 619 588 It is entirely possible that many of the differences observed in centrations. In contrast, S. cerevisiae has evolved mechanisms to 620 589 the response to alcohols could stem from the different culture con- maintain translational function even at higher concentrations of 621 590 ditions employed in the various filamentation assays. Our results ethanol. Taken altogether, these results suggest that generally alco- 622 591 equate well with the fact that a range of alcohols have been found hols could be targeting eIF2B to exert a specific effect on transla- 623 592 to impact upon filamentation at physiological concentrations and tion initiation in C. albicans. 624 593 they show that conditions/concentrations of alcohol that cause a A variety of connections between the regulation of protein syn- 625 594 very weak regulation of protein synthesis favour filamentation, thesis and alterations in cell fate such as morphogenesis, dif- 626 595 whereas more severe regulatory conditions prevent the response. ferentiation or the response to cellular stress have recently 627 596 The fact that alcohols inhibit protein synthesis in C. albicans is emerged. For instance, low concentrations of cycloheximide alter 628 597 directly comparable to our work in S. cerevisiae (Ashe et al., translation to promote cAMP-dependent morphological dif- 629 598 2001). We have shown previously that fusel alcohols (butanol in ferentiation of rat C6 glioma cells (Liu et al., 2010). Equally alter- 630 599 particular) cause a rapid inhibition of protein synthesis via inhibi- ations in ribosome biogenesis impact on Drosophila female 631 600 tion of eIF2B and reduced ternary complex levels. As a consequence germline stem cell differentiation, where reduced rRNA production 632 601 translation of the GCN4 mRNA is induced resulting in activation of and presumably reduced protein synthetic capacity induces dif- 633 602 the general control pathway. Here we show that fusel alcohols ferentiation (Zhang et al., 2014). Furthermore, studies on the rela- 634 603 cause very similar effects in C. albicans. Indeed, as in S. cerevisiae, tive contributions of different stages in gene expression during 635 604 eIF2B resides in a large cytoplasmic granule in C. albicans and in monocyte to macrophage differentiation identify the rate of pro- 636 605 response to butanol, the movement of this body is impeded. One tein synthesis as a principal regulator (Kristensen et al., 2013). 637 606 of the key differences between this study investigating the Even in single-celled eukaryotes such as S. cerevisiae minimal 638 607 response of C. albicans to alcohols and our previous studies on S. cycloheximide treatment results in a lower proportion of 639 608 cerevisiae is the response to ethanol. In S. cerevisiae, ethanol can pseudohyphal cells (Kron et al., 1994). In addition, the disruption 640 609 inhibit translation initiation but at much higher levels and does of established mechanisms of translation repression via the dele- 641 610 not appear to target eIF2B to the same extent, whereas here the tion of the genes for either the yeast eIF4E binding proteins 642 611 response to ethanol is almost identical to that of the two fusel alco- (Eap1p or Caf20p) or GCN2 inhibits the capacity of S. cerevisiae to 643 612 hols. This makes sense in terms of the precise metabolic and undergo filamentous growth in response to nitrogen limitation 644 613 physiological make-up of the two yeasts. For instance, unlike S. (Ibrahimo et al., 2006). More recent work in C. albicans suggests 645 614 cerevisiae, C. albicans is a Crabtree negative yeast (Rozpeßdowska that translational regulation of the UME6 mRNA specifically may 646 615 et al., 2011) and hence has not evolved to primarily ferment glu- be involved in the morphogenetic switch (Childers et al., 2014). 647 616 cose to ethanol, but instead relies largely upon the aerobic metabo- Therefore, it is entirely plausible that alcohol accumulation 648 617 lism of glucose. Therefore, it perhaps not surprising that in this represents a stress condition where at the concentrations that 649

N.E. Egbe et al. / Fungal Genetics and Biology xxx (2015) xxx–xxx 11

650 accumulate in cells, the impact on global rates of protein synthesis Ghosh, S. et al., 2008. Regulation of aromatic alcohol production in Candida 724 albicans. Appl. Environ. Microbiol. 74, 7211–7218. 725 651 is quite minimal. However, such effects are sufficient to alter the Gimeno, C.J. et al., 1992. Unipolar cell divisions in the yeast S. cerevisiae lead to 726 652 precise selection of specific mRNAs for translation. Such a mecha- filamentous growth: regulation by starvation and RAS. Cell 68, 1077–1090. 727 653 nism would lead to a rapid switch in the cell’s proteome to max- Guthrie, C., Fink, G.R., 1991. Guide to Yeast Genetics and Molecular Biology. 728 729 654 imise the timeliness of a cell’s response to specific external cues. Academic Press, San Diego, California. Hartwell, L.H., McLaughlin, C.S., 1969. A mutant of yeast apparently defective in the 730 655 More extreme conditions where higher concentrations of alcohol initiation of protein synthesis. Proc. Natl. Acad. Sci. U.S.A. 62, 468–474. 731 656 are encountered would alter global protein synthesis rates more Hazelwood, L.A. et al., 2008. The Ehrlich pathway for fusel alcohol production: a 732 733 657 dramatically and as such, cells would be targeted for destruction century of research on Saccharomyces cerevisiae metabolism. Appl. Environ. Microbiol. 74, 2259–2266. 734 658 pathways. Hershey, J.W., Merrick, W.C., 2000. Pathway and mechanism of initiation of protein 735 synthesis. Cold Spring Harbor Monogr. Ser. 39, 33–88. 736 659 Acknowledgments Hinnebusch, A.G., 2005. Translational regulation of GCN4 and the general amino 737 acid control of yeast. Annu. Rev. Microbiol. 59, 407–450. 738 Ibrahimo, S. et al., 2006. Regulation of translation initiation by the yeast eIF4E 739 660 We are grateful to Prof. Alistair Brown (Aberdeen University) binding proteins is required for the pseudohyphal response. Yeast 23, 1075– 740 661 and Prof. Chris Grant (The University of Manchester) for providing 1088. 741 Kawai, S. et al., 1992. Drastic alteration of cycloheximide sensitivity by substitution 742 662 strains. We thank Lydia Castelli for critical reading of the manu- of one amino acid in the L41 ribosomal protein of yeasts. J. Bacteriol. 174, 254– 743 663 script and Arunkumar Sundaram for his advice and technical help. 262. 744 664 The Bioimaging Facility microscopes used in this study were pur- Kitagaki, H. et al., 2007. Ethanol-induced death in yeast exhibits features of 745 apoptosis mediated by mitochondrial fission pathway. FEBS Lett. 581, 2935– 746 665 chased with grants from BBSRC, Wellcome Trust and the 2942. 747 666 University of Manchester Strategic Fund. Special thanks goes to Kristensen, A.R. et al., 2013. Protein synthesis rate is the predominant regulator of 748 667 Peter March and Steve Mardsen for their help with this micro- protein expression during differentiation. Mol. Syst. Biol. 9. 749 750 668 scopy. NEE is funded by the Tertiary education trust fund Kron, S.J. et al., 1994. Symmetric cell division in pseudohyphae of the yeast Saccharomyces cerevisiae. Mol. Biol. Cell 5, 1003–1022. 751 669 Nigeria, and CMP is funded by a BBSRC project grant (BB/ Lingappa, B.T. et al., 1969. Phenethyl alcohol and tryptophol: autoantibiotics 752 670 K002767/1). produced by the fungus Candida albicans. Science 163, 192–194. 753 Liu, H., 2001. Transcriptional control of dimorphism in Candida albicans. Curr. Opin. 754 Microbiol. 4, 728–735. 755 671 References Liu, H. et al., 1994. Suppression of hyphal formation in Candida albicans by 756 mutation of a STE12 homolog. Science 266, 1723–1726. 757 672 Ashe, M.P. et al., 2000. Glucose depletion rapidly inhibits translation initiation in Liu, X. et al., 2010. Induction of cell cycle arrest at G1 and S phases and cAMP- 758 673 yeast. Mol. Biol. Cell 11, 833–848. dependent differentiation in C6 glioma by low concentration of cycloheximide. 759 674 Ashe, M.P. et al., 2001. A novel eIF2B-dependent mechanism of translational control BMC Cancer 10, 684. 760 675 in yeast as a response to fusel alcohols. EMBO J. 20, 6464–6474. Lo, H.-J. et al., 1997. Nonfilamentous C. albicans mutants are avirulent. Cell 90, 939– 761 676 Biswas, S. et al., 2007. Environmental sensing and signal transduction pathways 949. 762 677 regulating morphopathogenic determinants of Candida albicans. Microbiol. Lorenz, M.C. et al., 2000. Characterization of alcohol-induced filamentous growth in 763 678 Mol. Biol. Rev. 71, 348–376. Saccharomyces cerevisiae. Mol. Biol. Cell 11, 183–199. 764 679 Bockmühl, D.P., Ernst, J.F., 2001. A potential phosphorylation site for an A-type Martins, M. et al., 2007. Morphogenesis control in Candida albicans and Candida 765 680 kinase in the Efg1 regulator protein contributes to hyphal morphogenesis of dubliniensis through signaling molecules produced by planktonic and biofilm 766 681 Candida albicans. Genetics 157, 1523–1530. cells. Eukaryot. Cell 6, 2429–2436. 767 682 Brown, A.J., Gow, N.A., 1999. Regulatory networks controlling Candida albicans Natarajan, K. et al., 2001. Transcriptional profiling shows that Gcn4p is a master 768 683 morphogenesis. Trends Microbiol. 7, 333–338. regulator of gene expression during amino acid starvation in yeast. Mol. Cell. 769 684 Campbell, S.G. et al., 2005. Dynamic cycling of eIF2 through a large eIF2B-containing Biol. 21, 4347–4368. 770 685 cytoplasmic body: implications for translation control. J. Cell Biol. 170, 925– Odds, F.C., 1994. Pathogenesis of Candida infections. J. Am. Acad. Dermatol. 31, S2– 771 686 934. S5. 772 687 Chauhan, N.M. et al., 2011. A morphogenetic regulatory role for ethyl alcohol in Palmer, L.K. et al., 2005. Inhibition of translation initiation by volatile anesthetics 773 688 Candida albicans. Mycoses 54, e697–e703. involves nutrient-sensitive GCN-independent and -dependent processes in 774 689 Chauhan, N.M. et al., 2013. Effect of alcohols on filamentation, growth, viability and yeast. Mol. Biol. Cell 16, 3727–3739. 775 690 biofilm development in Candida albicans. Braz. J. Microbiol. 44, 1315–1320. Pavitt, G.D., 2005. EIF2B, a mediator of general and gene-specific translational 776 691 Chen, H., Fink, G.R., 2006. Feedback control of morphogenesis in fungi by aromatic control. Biochem. Soc. Trans. 33, 1487–1492. 777 692 alcohols. Gene Dev. 20, 1150–1161. Pollack, J.H., Hashimoto, T., 1985. Ethanol-induced germ tube formation in Candida 778 693 Chen, H. et al., 2004. Tyrosol is a quorum-sensing molecule in Candida albicans. albicans. J. Gen. Microbiol. 131, 3303–3310. 779 694 Proc. Natl. Acad. Sci. U.S.A. 101, 5048–5052. Rozpeßdowska, E. et al., 2011. Candida albicans–a pre-whole genome duplication 780 695 Chiew, Y.Y. et al., 1982. The effects of ergosterol and alcohols on germ-tube yeast–is predominantly aerobic and a poor ethanol producer. FEMS Yeast Res. 781 696 formation and chitin synthase in Candida albicans. Can. J. Biochem. 60, 15–20. 11, 285–291. 782 697 Childers, D.S. et al., 2014. A 50 UTR-mediated translational efficiency mechanism Schneider-Poetsch, T. et al., 2010. Inhibition of eukaryotic translation elongation by 783 698 inhibits the Candida albicans morphological transition. Mol. Microbiol. 92, 570– cycloheximide and lactimidomycin. Nat. Chem. Biol. 6, 209–217. 784 699 585. Srikantha, T. et al., 1996. The sea pansy Renilla reniformis luciferase serves as a 785 700 Csank, C., Haynes, K., 2000. Candida glabrata displays pseudohyphal growth. FEMS sensitive bioluminescent reporter for differential gene expression in Candida 786 701 Microbiol. Lett. 189, 115–120. albicans. J. Bacteriol. 178, 121–129. 787 702 Dehoux, P. et al., 1993. Natural cycloheximide resistance in yeast. Eur. J. Biochem. Sudbery, P.E., 2011. Growth of Candida albicans hyphae. Nat. Rev. Microbiol. 9, 737– 788 703 213, 841–848. 748. 789 704 Dever, T.E., 2002. Gene-specific regulation by general translation factors. Cell 108, Sudbery, P. et al., 2004. The distinct morphogenic states of Candida albicans. Trends 790 705 545–556. Microbiol. 12, 317–324. 791 706 Dever, T.E. et al., 1992. Phosphorylation of initiation factor 2 alpha by protein kinase Sundaram, A., Grant, C.M., 2014a. Oxidant-specific regulation of protein synthesis in 792 707 GCN2 mediates gene-specific translational control of GCN4 in yeast. Cell 68, Candida albicans. Fungal Genet. Biol. 67, 15–23. 793 708 585–596. Sundaram, A., Grant, C.M., 2014b. A single inhibitory upstream open reading frame 794 709 Dickinson, J.R., 1996. ‘Fusel’ alcohols induce hyphal-like extensions and (uORF) is sufficient to regulate Candida albicans GCN4 translation in response 795 710 pseudohyphal formation in yeasts. Microbiology 142 (Pt 6), 1391–1397. to amino acid starvation conditions. RNA. 796 711 Dickinson, J.R. et al., 1998. An investigation of the metabolism of valine to isobutyl Taylor, E.J. et al., 2010. Fusel alcohols regulate translation initiation by inhibiting 797 712 alcohol in Saccharomyces cerevisiae. J. Biol. Chem. 273, 25751–25756. eIF2B to reduce ternary complex in a mechanism that may involve altering the 798 713 Dickinson, J.R. et al., 2003. The catabolism of amino acids to long chain and complex integrity and dynamics of the eIF2B body. Mol. Biol. Cell 21, 2202–2216. 799 714 alcohols in Saccharomyces cerevisiae. J. Biol. Chem. 278, 8028–8034. Tournu, H. et al., 2005. Global role of the protein kinase Gcn2 in the human 800 715 Feng, Q. et al., 1999. Ras signaling is required for serum-induced hyphal pathogen Candida albicans. Eukaryot. Cell 4, 1687–1696. 801 716 differentiation in Candida albicans. J. Bacteriol. 181, 6339–6346. Tripathi, G. et al., 2002. Gcn4 co-ordinates morphogenetic and metabolic responses 802 717 Fonzi, W.A., Irwin, M.Y., 1993. Isogenic strain construction and gene mapping in to amino acid starvation in Candida albicans. EMBO J. 21, 5448–5456. 803 718 Candida albicans. Genetics 134, 717–728. Webb, A.D., Ingraham, J.L., 1963. Fusel oil. Adv. Appl. Microbiol. 5, 317–353. 804 719 Fried, H.M., Warner, J.R., 1982. Molecular cloning and analysis of yeast gene for Zeuthen, M.L. et al., 1988. Ethanol tolerance and the induction of stress proteins by 805 720 cycloheximide resistance and ribosomal protein L29. Nucleic Acids Res. 10, ethanol in Candida albicans. J. Gen. Microbiol. 134, 1375–1384. 806 721 3133–3148. Zhang, Q. et al., 2014. Changes in rRNA transcription influence proliferation and cell 807 722 Gancedo, J.M., 2001. Control of pseudohyphae formation in Saccharomyces fate within a stem cell lineage. Science 343, 298–301. 808 723 cerevisiae. FEMS Microbiol. Rev. 25, 107–123. 809