Para-Aminobenzoic Acid (PABA) Synthase Enhances Thermotolerance of Mushroom Agaricus bisporus

Most mushrooms are thermo-sensitive to temperatures over 23°C, which greatly restricts their agricultural cultivation. Understanding mushroom’s innate heat-tolerance mechanisms may facilitate genetic improvements of their thermotolerance. Agaricus bisporus strain 02 is a relatively thermotolerant mushroom strain, while strain 8213 is quite thermo-sensitive. Here, we compared their responses at proteomic level to heat treatment at 33°C. We identified 73 proteins that are differentially expressed between 02 and 8213 or induced upon heat stress in strain 02 itself, 48 of which with a known identity. Among them, 4 proteins are constitutively more highly expressed in 02 than 8213; and they can be further upregulated in response to heat stress in 02, but not in 8213. One protein is encoded by the para-aminobenzoic acid (PABA) synthase gene Pabs, which has been shown to scavenge the reactive oxygen species in vitro. Pabs mRNA and its chemical product PABA show similar heat stress induction pattern as PABA synthase protein and are more abundant in 02, indicating transcriptional level upregulation of Pabs upon heat stress. A specific inhibitor of PABA synthesis impaired thermotolerance of 02, while exogenous PABA or transgenic overexpression of 02 derived PABA synthase enhanced thermotolerance of 8213. Furthermore, compared to 8213, 02 accumulated less H2O2 but more defense-related proteins (e.g., HSPs and Chitinase) under heat stress. Together, these results demonstrate a role of PABA in enhancing mushroom thermotolerance by removing H2O2 and elevating defense-related proteins.


Introduction
Agaricus bisporus, also called button mushroom, is one of the most popular edible basidiomycete fungus worldwide and well-known for its low content of fat and high content of dietary cellulose and pharmacological components [1,2]. So far, A. bisporus has been agriculturally cultivated in more than 100 countries, especially in China, and has already become one of the most prevailing species of dietary mushroom [3]. Like other mushrooms, A. bisporus is thermo-sensitive and vulnerable to high temperatures. Once the ambient temperature exceeds 23uC, A. bisporus fruit body grows slowly and its sporophores become smaller and brown, which greatly affects the quality and yield in cultivation. In order to ensure a large scale off-season cultivation of this commercially important species, special cooling instruments and more energy expense are required to maintain the ambient temperature below 20uC, which has become the major restrictive factor in mushroom industry worldwide [4]. Therefore, understanding the intrinsic mechanisms of acquired thermotolerance of mushrooms under heat stress is of both theoretical and practical significance, and eventually contributes to the mushroom cultivation industry.
Organisms use different ways to respond to environmental stress such as high temperature. One of the best characterized factors is the heat shock protein (HSP) family, which plays a central role in thermotolerance of plants and animals [5]. Three major classes of HSPs are present in organisms including the small HSPs (ranging in molecular weight from 15 to 28 kD), the moderate HSPs (such as HSP60 and HSP70) and high HSPs (such as HSP90 and HSP101). Most HSPs can ameliorate protein misfolding and aggregation during heat stress, but each major HSP family uses a distinct and rigorous mechanism to prevent heat stress induced impairment. For example, Hsp60 and HSP70 can prevent protein aggregation by binding to protein intermediates, while HSP101 can reactivate a protein that has already aggregated. In addition to HSPs, many other factors, such as plant hormone ABA, reactive oxygen and salicylic acid, are also involved in acquired thermotolerance [6,7,8,9]. However, very little is known about the thermotolerant signaling in higher basidiomycetes.
PABA (para-aminobenzoic acid), also known as 4-aminobenzoic acid, is a precursor for the synthesis of folic acid (also known as vitamin B 9 or folacin). As an enzyme cofactor, folic acid is involved in numerous basic biological reactions, including nucleotide biosynthesis, DNA repair and DNA methylation. For example, children and adults both require abundant folic acid to produce healthy erythrocytes and avoid anemia [10]. As the primary source of folic acid [11], plants firstly synthesize the intermediate PABA in chloroplasts via PABA synthase and then use PABA for folic acid synthesis in mitochondria [12,13,14]. Most previous studies focused on the role of the PABA metabolite in folic acid synthesis. Interestingly, salicylic acid (SA), an analog of PABA, is well known for its wide range of functions from human pain relief to plant system defenses [15]. In recent years, SA has been found to protect Arabidopsis, tomato, bean [16], potato, mustard and tobacco against heat stress [9,17]. Due to the structural similarities between PABA and SA, it is highly possible that PABA may also play a role mediating plant thermotolerance response, but its roles in plants responding to environmental stress remain elusive.
In this study, we set out to identify novel signaling pathways and novel secondary messengers that are involved in thermotolerance of mushroom against heat stress. We started by performing comparative proteomic profiling on two A. bisporus strains, 8213 and 02, which are phylogenetically closely related, but otherwise completely different in thermotolerance. For mycelium growth, strain 8213 (will be called 8213 herein) is thermo-sensitive and vulnerable to high temperatures over 33uC, while strain 02 (will be called 02 herein) is thermotolerant to such temperature. By employing 2D-electrophoresis, a set of proteins differentially expressed between 8213 and 02, with or without heat stress, were identified. Among them, one protein encoding mushroom PABA synthase was constitutively more abundant in 02 than in 8213 at both mRNA and protein level, and was further upregulated under heat stress in 02. Further physiological and transgenic experiments demonstrated that PABA may be a messenger that mediates mushroom thermotolerance response. Our findings provide insights into the role of PABA in mushrooms thermotolerance and implicate a novel target for transgenic improvement of mushroom thermotolerance.

Mushroom Strains and their Growth Conditions
Mushroom (A. bisporus) strains 02 and 8213 were provided by the Mushroom Research and Development Station, Fujian Academy of Agricultural Sciences, China. They were cultured at 23uC on a potato dextrose agar (PDA) plate for two weeks to obtain mycelium growth. The fruit body cultivation was performed at 18uC as previously described [18].

Thermotolerance Assay
The heat sensitivity of mushroom strains was determined primarily by the mycelium elongation assay. The well-grown mycelia were inoculated into a test tube that contained a cottonseed hull medium (78% cottonseed hull, 15% cow dung, 5% wheat bran, 2% plaster and lime mixture). The test tubes were kept vertical throughout this test. After 5 days of normal growth at temperatures at 23uC, the mycelium elongation position of each test-tube was marked as the starting point. Subsequently, the experimental groups were subjected to 33uC heat stress conditions. Mycelium elongation of each test tube was marked every 7 days, and the elongation length was recorded. In this assay, we used the mycelium elongation length to quantify heat sensitivity of strains 02, 8213 and each transgenic line. Furthermore, we used the hypha damage observation and the fruit body formation as supplements to thermotolerance determination that provided similar conclusions [19,20].

Thermotolerance Induction and Sample Collecting
After growing in 100 ml liquid PDA medium in 250-ml Erlenmeyer flasks for 2 weeks at 23uC on a rotary shaker (160 rpm), the mycelia culture of strains 02 and 8213 were divided into heat treatment and control groups. For the heat treatment groups, both strains were exposed to the stress temperature of 33uC for 24 h. Controls strains were cultivated for the same period of time at 23uC. After heat treatments, the mycelia were collected, frozen in liquid nitrogen immediately and stored at 270uC. The analysis of proteomes was carried out for the two comparative groups.

Two-DE Proteome Profiling and Image Analysis
Mushroom mycelium samples collected from each experimental group were ground to a powder in liquid nitrogen. Total protein extraction and quantification were conducted according to previously described methods [21]. In brief, the samples were extracted with a dehydration buffer (2 M thiourea, 7 M urea, 2% CHAPS, 40 mM DTT, 0.002% bromophenol blue, 2% IPG buffer), and stored at 270uC. Each protein sample (500 mg) was loaded on an IPG (immobilized pH gradient) gel for 12 h at 20uC for in-gel rehydration. The first-dimension IEF was carried out with 18 cm-long, pH 3-10, non-linear Immobiline DryStrips (GE Biosciences) and performed on an IPGphor (Amersham) system using the following program: 500 V for 1 h, 1000 V for 1 h, and 8000 V for 8 h 20 min (total 68,000 Vh). The focused IPG strips were placed in equilibrium buffer (75 mM of Tris-HCl, pH 8.8, 6 M urea, 29.3% v/v glycerol, 2% w/v SDS, and 0.002% bromophenol blue) containing 10 mg/ml of DTT for 15 min followed by another 15 min treatment in the same buffer containing 25 mg/ml iodoacetamide. The second dimension electrophoresis was carried out using a 12% SDS-PAGE to separate proteins by molecular weight.
All gels were stained with colloidal Coomassie Brilliant blue G-250 and scanned by an image scanner with a resolution of 300 dpi. Image analysis was carried out with Image-Master 2D Platinum version 5 software (GE Biosciences). Each spot was assigned a value as the volume of differential expression (V DE ), which is defined as the sum of the gray-level values of all the pixels in the spot. Three biological replicates were used to calculate the average values. Candidate spots exhibiting 2-fold changes or greater were selected for identification, and the Student's t-test was performed to calculate statistical significance.

In-gel Digestion
Protein spots were excised from the gel above and washed 3 times with ultrapure water, and then destained twice with 50% acetone in 25 mM NH 4 HCO 3 . The gels were dehydrated with 50 mL 100% acetonitrile, lyophilized and further digested overnight with trypsin (Promega, Madison, WI, US) in 25 mM NH4HCO3 containing 10% CAN at 37uC. For each spot, the supernatant was transferred to a separate Eppendorf tube and lyophilized for MS analysis.

MALDI-TOF (Matrix-Assisted Laser Desorption/Ionization Time-of-Flight)/TOF Analysis and Database Search
Tryptic-digested peptides were extracted according to a previous protocol [21], and MS analysis was performed with a MALDI-TOF/TOF mass spectrometer (Bruker-Franzen Analytik) in reflectron mode with an accelerating voltage of 25 kV and a delayed extraction of 50 ns. Mass spectra were obtained in an automatic mode with the AutoXecute module of Flexcontrol (Bruker-Franzen Analytik) (laser power ranged from 40 to 60%, 1000 shots). Flex-Analysis software (Bruker-Franzen Analytik) was employed to analyze the spectra with the auto proteolysis peptides of trypsin (m/z 842.51, 1045.56, 2211.10) as the internal calibration. Peptides were selected in the mass range of 800-3500 Da. For each sample, 4-5 ion peaks with signal-to-noise ratios over 100 were selected as precursors for secondary mass spectrum analysis, and the TOF/TOF signal for each precursor was accumulated with 2000 laser points. The peptide MS/MS spectra were compared against the domestic database download from the public mushroom web site (http://genome.jgi-psf.org/ week before fruit body formation, each was then divided into two groups. One half was maintained at 18uC (control group); the other half was maintained at 30uC (high-temperature group) for 7 days. At Day 7, the fruit bodies were photographed (A), and the pileus diameter and biomass weight were recorded (B). Three independent replicates were performed. Data are expressed as average 6 SEM. Unpaired t-tests were performed, ns: P.0.05, **: P,0.01. doi:10.1371/journal.pone.0091298.g001 Agabi_varbisH97_2/Agabi_varbisH97_2.home.html) using MAS-COT software (http://www.matrixscience.com). Fixed modification was set for the carbamidomethylation of cysteine, while the oxidation of methionine was set as a viable modification. The peptide tolerance was 60 ppm, and the MS/MS tolerance was 0.25 Da. One incomplete cleavage was allowed for each protein.
Keratin contamination was removed, and a MASCOT score over 60 (p,0.05) was considered to be a positive hit.

Transgenic Manipulation
For mushroom transgenic manipulation, the ORF of A. bisporus PABA synthase gene (Pabs) was PCR-amplified from the cDNA of strain 02 and subsequently cloned into the binary vector pBHg provided by the Mushroom Research and Development Station, Fujian Academy of Agricultural Sciences, China. The construct was then transformed into strain 8213 using the Agrobacterium mediated gill-soaking method as described previously [18]. At least 10 independent lines were selectively grown on a PDA plate containing 50 mg l 21 hygromycin for further analysis.

Quantitative Real-time PCR Analysis
Total RNA was extracted, DNaseI digested and reverse transcribed for use as a template. SYBR Green dye and genespecific primers were added to perform the real-time PCR quantification with a LightCycler real-time PCR system (ROCHE). A melting curve analysis of each gene was performed to check the specificity of amplification, and Gapdh was used as a reference control and corrected for PCR efficiency differences between target and reference samples with efficiency correction Relative Quantification Software (LightCycler Software 4.05). Each of the amplifications was performed in duplicate, and the mean value was calculated as the final result.

Total PABA Extraction and HPLC Quantification
Total PABA extraction was performed as described by Eoin P. Quinlivan [22] with some modification. Briefly, mycelium (1 g of fresh weight) was ground in liquid nitrogen and extracted twice with 3 ml of methanol. The two parts of the supernatant were then pooled and evaporated. The dried sample was dissolved in 1 ml of 0.1 M sodium acetate buffer, pH 5.5, and digested with 0.04 U/ml glycosidase (Sigma) at 30uC overnight to change the conjugated PABA into free type. Subsequently, the product was evaporated, re-dissolved in 0.8 ml of 0.1 M sodium citrate, pH 3.7, and partitioned against 2.4 ml of ethyl acetate (EA). PABA was recovered from the organic phase by back extraction with 0.7 ml of 50 mM NaOH and neutralized. After final evaporation, the sample was dissolved in methanol solvent.
The extracted sample was then injected onto a 5 mm C18 (250 64.6 mm, Agilent) column and detected by fluorescence (270 nm excitation, 350 nm emission). The column was eluted (1 ml/min) with 0.5% acetic acid/methanol (80:20, v/v) [23]. The total PABA (free PABA and conjugated PABA) was quantified according to the standards. The spiked PABA sample was also quantified for the recovery rate correction.

H 2 O 2 Concentration and Antioxidant Enzymes Activities Measurement
The mushroom mycelium (0.5 g) were quickly frozen in liquid Nitrogen and ground to a fine powder. The frozen powdered samples were then homogenized in 5 mL of extraction buffer (10 mM Tris-Cl, pH 8.0, 1 mM phenylmethylsulfonyl fluoride, 10 mM MgSO 4 , 5 mM KCl, 5 mM NaCl, 10 mM oxyhemoglobin and 10 units ml 21 of catalase). The extracts were then centrifuged (15,0006g, 10 min) at 4uC, and the supernatants were used for H 2 O 2 determination; H 2 O 2 content was measured using a previously described method [24]. The antioxidant enzymes including Catalase and SOD enzyme activities were measured as previous method [24].

Western Blotting Analysis
With liquid nitrogen grounding, total protein from mycelium was extracted and dissolved in SDS sample buffer as described.
Then the protein concentration of each sample was determined using Coomassie Blue assay, and diluted to equal. After 15% (w/v) SDS-PAGE separation, the proteins were blotted onto PVDF membrane and immunoblotted against the anti-HSP27 (cell signaling) and anti-HSP70 (Stressgen) respectively with a dilution of 1:1,000 (v/v). The antibodies against mushroom PABA synthase and Chitinase were prepared by immunizing the rabbit with the purified PABA synthase protein or synthesized peptides from mushroom Chitinase (VTFNGHLWQNK).

Agaricus bisporus Strain 02 is more Thermotolerant than Strain 8213
Agaricus bisporus strain 02 is one of the common mushroom strains that are used in agriculture and research. Strain 02 is characterized as a thermotolerant strain. Mycelium growth of 02 is resistant to an ambient temperature over 33uC, which is a lethal temperature for most of the other common strains. In contrast, during our breeding practice, we screened out strain 8213 that was significantly more sensitive to heat stress than regular A. bisporus strains. As shown in Figure 1A & 1B, at 18uC, a regular ambient temperature for mushroom cultivation, the fruit bodies of 02 and 8213 grew similarly well. When the ambient temperature rose to 30uC, the fruit body growth of 8213 was severely reduced in both pileus diameter and biomass, while the 02 still had a comparable growth of biomass as at 18uC. Phylogenetic analysis showed that 8213 and 02 both belong to the same species Agaricus bisporus, and 02 is evolutionally closer to 8213 than other strains of A. bisporus ( Figure S1). Thus, 02 and 8213 provide a pair of experimental models for studying the mechanisms of intrinsic thermotolerance in mushroom.
Proteomic Profiling of 02 and 8213 Reveals a Potential Role of PABA-synthase in Enhancing Thermotolerance To investigate the mechanisms underlying higher thermotolerance of 02, we employed comparative 2-dimensional electrophoresis (2-DE) to profile the proteomics difference between 02 and 8213 with or without heat stress.
To avoid inducing drought stress, we choose PDA liquid culture for mushroom mycelia growth. We first determined optimal heat stress condition for our studies. The optimum temperature for mushroom mycelium growth in liquid culture is 23uC; we tested the thermo-sensitive strain 8213 in a set of temperatures (23uC, 33uC, and 43uC) for 48 h, and then inoculated them separately into fresh PDA culture for another week of growth at 23uC. We found that mycelia treated with 43uC failed to re-grow in fresh medium, while mycelia treated with 33uC grew to reduced biomass of about 70% of the 23uC treated group, suggesting that 33uC may be a appropriate temperature for heat stress. We then treated strain 02 with 33uC for a time course (12 h, 24 h, and 48 h) and used the heat shock proteins (HSP20 and HSP70) as markers to determine the best time point for proteomic profiling. As a result, we found that both Hsp20 and Hsp70 began to show accumulation at 48 h time point, without any noticeable accumulation at 24 h time point, which suggested that, under 33uC heat stress condition, the upregulation of those heat stress terminal effectors happens in the time gap between 24 h to 48 h (data not shown). In order to better profile proteins that govern the expression of HSPs, we used 33uC for 24 h as the heat stress condition for proteomic profiling.
Strains 02 and 8213 were treated with or without the heat stress condition determined above. Proteins were isolated from their mycelia and resolved by 2-DE. For brevity, we will use 02-NS and 02-HS to indicate non-stressed strain 02 and heat-stressed 02, respectively; and use 8213-NS and 8213-HS to indicate nonstressed strain 8213 and heat-stressed 8213, respectively. Figure 2 and Figure S2 show the results for strain 02 and 8213 under normal condition (23uC) and heat stress condition (33uC), respectively. The gels were stained by CBB R-50, scanned and analyzed with Image-Master 2D Platinum version.5 software (GE Biosciences). We found a total of 340 protein spots showing differential expression in one of the three pairs of comparisons (02-NS vs. 02-HS, 8213-NS vs. 02-NS and 8213-HS vs. 02-HS). Among them, 73 spots reach the criterion of 2-fold change with statistically significance (P,0.05). Those 73 spots were then subjected to MS/MS analysis and peptide mass fingerprinting (PMF) identification in the NCBI database. The identities of 48 proteins were successfully determined (Table S1).
We set the abundance of each protein spot in 02-NS as the reference value of 1.0, and then converted the protein abundance of that spot from 02-HS, 8213-NS and 8213-HS gels into relative value. We generated heatmap for all the 48 proteins between all 4 samples and found that 25 out of 48 spots shows changed expression level (.2.0 fold) in thermotolerant strain 02 upon heat stress ( Figure 3A top). Based on their predicted metabolic and functional features (using the COGNITOR tool in Cluster of Orthologous Groups (COG) database), those 25 proteins could be classified into six groups as shown in Figure.3A. Among the 25 proteins, 20 spots show up-regulation, while the other 5 spots show down-regulation, indicating that upregulated proteins are more likely to be responsible for thermotolerance enhancement in 02. We decided to focus on those 20 up-regulated proteins for further studies. Most of the up-regulated proteins in 02 upon heat stress belong to the defense response group (26.1%) or the antioxidant protein family (26.1%) (Figure 3A bottom). Since strain 02 is a thermotolerant strain and is also the center of this study, we next screened out candidate genes that are responsive to heat-stress in 02 but not in 8213, and are expressed higher in 02 than in 8213 in non-stressed conditions. Severn out of 20 spots were also upregulated (.2.0 fold) in 8213 after heat stress, and were excluded (since their up-regulation did not confer thermotolerance to 8213) ( Figure 3B, middle). For the remaining 13 proteins, we excluded those that are not expressed higher in strain 02 over 8213 either in non-stress condition or in heat stress conditions, leaving 4 proteins as our candidates ( Figure 3B, bottom).
Among these four proteins, protein spot 36 ( Figure 2B-a) was indentified by MS/MS spectra peak analysis to encode a mushroom para-aminobenzoic acid (PABA) synthase ( Figure 4A). Consistently, the EST fragment of PABA synthase gene (Pabs) was found to be more highly expressed in strain 02 than in 8213 in our suppressive subtractive hybridization (SSH) (EST Genbank accession#: GH159019.1). Collectively, these studies identify PABA synthase as likely playing a role in enhancing thermotolerance of strain 02.

PABA Synthesis is Stimulated in Thermotolerant Strain 02 under Heat Stress
By performing the RACE technique, we cloned the Pabs gene ORF sequence in mushroom (GenBank accession#: FJ617437), and determined that its full-length open reading frame (ORF) is 1800 bp and encodes a 600-amino acid PABA synthase.
To determine whether the accumulation of PABA synthase is due to transcriptional level induction upon heat stress or due to changes in protein stability, we perform qRT-PCR experiment with mRNA samples of the corresponding sets of mushroom mycelia. Consistent with the pattern of protein level change, the PABA synthase gene (Pabs) constitutively expresses higher in 02 than in 8213 and upregulates dramatically in strain 02, but only slightly changes in strain 8213, under heat stress ( Figure 4B). Thus, transcriptional regulation of the Pabs gene likely accounts for its protein level changes in 02 and 8213 in non-stress and heat stress conditions.
To study the functions of Pabs in thermotolerance, we subcloned its ORF into the pGEX-4T1, expressed the protein in E.coli BL-21 strain ( Figure S3), and purified the protein for enzymatic analysis. Our results showed that purified PABA synthase protein exhibited a considerable PABA synthesizing activity in vitro (data not shown). We then generated antibody against PABA-synthase by immunizing the rabbit with the bacterially expressed PABA synthase protein. With this antibody, we confirmed that the endogenous PABA synthase significantly increased over time under heat stress in 02, but not in 8213 ( Figure 4C).
Since PABA synthase is expected to synthesize PABA in mushroom, we also measured the PABA content in 02 and 8213 with or without high-temperature treatment. Results showed that    PABA level was less than 2 fold higher in 02 than in 8213 at the onset of high-temperature treatment. After 72 h of high-temperature treatment, PABA content increased dramatically in 02, reaching to around 10 mg/g FW, while to around 3 mg/g FW in 8213 treated in parallel ( Figure 4D).

Increasing PABA Enhances Thermotolerance of the Thermo-sensitive Strain 8213
To determine the role of PABA in mushrooms undergoing hightemperature stress, we compared the responses of 02 and 8213 to exogenous addition of PABA to the culture medium. We used the mycelium breakage as the criteria for sensitivity of mushrooms to high temperatures because mushroom mycelium easily breaks and cannot grow well under high temperatures. As shown in Figure 5A, for strain 8213, most mycelia broke under 33uC after 48 hours, as compared to control sets which were cultured under regular 23uC. Addition of PABA to medium alleviated the mycelium breakage caused by high temperatures, and this alleviation effect was even more obvious at 10 mg/L PABA concentration in the medium. For strain 02, addition of exogenous PABA did not further improve its thermotolerance ( Figure 5A&5B), suggesting that the endogenous levels of PABA in 02 might approach saturating point.
Since thermotolerant strain 02 has higher level of endogenous PABA synthase, and addition of PABA in medium helped to enhance thermotolerance of the thermo-sensitive strain 8213, we further tested whether transgenic over-expression of PABA synthase would confer 8213 with higher thermotolerance. We subcloned the 02-derived Pabs ORF into the intermediate vector pBHg, which uses a constitutive mushroom Gapdh promoter to drive the transcription of target gene in vivo. Then we introduced the Pabs expressing plasmid into thermo-sensitive strain 8213 using agobacteria mediated transformation, and obtained several independent transgenic lines by antibiotic selection. By using real time qPCR to examine the transcriptional level of Pabs in these transgenic strains, we obtained and confirmed two transgenic lines, named TB-2 and TB-3, that constitutively express higher levels of Pabs (23uC), which is 5 fold higher than the constitutive Pabs level in strain 02 ( Figure 6A). As shown, although the transcription of Pabs in TB-2 and TB-3 do not upregulate upon heat stress, their abundance is still higher than the induced Pabs level of strain 02 under heat stress (33uC). We further measured the endogenous PABA content of 02 and 8213 as well as the two transgenic strains. Indeed, TB-2 and TB-3 contained higher level of PABA in both normal (23uC) and heat stress (33uC) conditions compared to strain 8213 ( Figure 6B). After validating the effectiveness of the transgenic manipulation, we determined the effects of overexpression of Pabs on thermotolerance. In Figure 6C and 6D, we used the mycelia elongation essay to quantitatively compare thermotolerance capability between those mushroom strains. All strains grew well under normal conditions (23uC), as measured by their similar mycelia elongation length. Under heat stress condition, the mycelia growth of 8213 was almost completely abolished, while 02 continued to grow although at a slightly reduced rate. Interestingly, TB-2 and TB-3, which originated from 8213 but constitutively overexpress 02-derived Pabs gene, showed similar mycelia growth as 02 ( Figure 6D).
Collectively, these studies demonstrate that gain-of-function of 02-derived Pabs gene in thermo-sensitive strain 8213 enhanced its thermotolerance capability.

Suppressing PABA Synthesis Compromises
Thermotolerance of the Thermotolerant strain 02 PABA synthesis and utilization can be suppressed by sulfanilamide, a structural analog of PABA [25]; and strain 02 has higher basal and induced level of Pabs gene and PABA production. To verify the relationship between PABA function and thermotolerance, we treated 02 with sulfanilamide to carry out loss-of-function studies. We found that addition of sulfanilamide reduced PABA production in 02 under high temperature ( Figure 7A). Without the inhibitor, strain 02 contained PABA content around 9 mg/g FW under heat stress; 0.1 mM sulfanilamide treatment abolished about 70% of its PABA production. This effect reached saturation at concentrations over 0.25 mM. After validating the inhibitory effect of sulfanilamide on PABA production, we investigated whether reduction of PABA content in strain 02 would compromise its thermotolerance. Indeed, the percentage of intact mycelia of 02 dramatically decreased following treatment of 0.25 mM sulfanilamide for 48 hours ( Figure 7B). The biomass of 02 was also reduced by sulfanilamide treatment (data not shown). The same sulfanilamide treatment did not further aggravate the damage seen in strain 8213, likely because 8213 natively express less PABA and has less intact mycelium under heat stress. Thus, inhibition of PABA synthase dramatically compromised thermotolerance capability of thermotolerant strain 02, providing further evidence for the relationship between PABA synthase function and thermotolerance.

PABA Alleviates Oxidative Damage by Heat Stress through Increasing the Antioxidases Activity
Oxidative stress usually accompanies heat stress, and could directly induce injury following heat stress in Arabidopsis [26] and yeast [27]. On the other hand, PABA could scavenge reactive oxygen species and protect DNA against free radical damage in vitro [28]. It is unknown whether PABA can also alleviate heat stress or oxidative stress induced damage in plants or higher fungus (e.g., mushroom). Indeed, heat stress induced fast accumulation of H 2 O 2 in thermo-sensitive strain 8213, while such effect was much more moderate in thermotolerant strain 02 ( Figure 8A). Pabs over-expressing transgenic strains TB-2 and TB-3 showed significantly lower H 2 O 2 as compared to their parent strain 8213 under heat stress. To directly investigate whether H 2 O 2 could affect mushroom viability under heat stress, we treated the mushroom with 5 mM H 2 O 2 and found that it inhibited mycelia elongation of both 02 and 8213 under heat stress. On the other hand, treatment with BHT (Butylated hydroxytoluene), an artificial scavenger of H 2 O 2 , alleviated the inhibition of mycelia elongation by heat stress ( Figure S4). We further found that treatment with exogenous PABA assuaged the H 2 O 2 accumulation in 8213 under heat stress, while treatment with the PABA synthase inhibitor sulfanilamide aggravated it in both 02 and 8213 under heat stress ( Figure 8B). These data reveal a negative correlation between PABA abundance and the H 2 O 2 accumulation in mushroom.
Furthermore, we found that Catalase (CAT) and Superoxide dismutase (SOD), two key antioxidants responsible for clearing oxidative free radical in vivo, both exhibited higher enzymatic activity in strain 02 as compared to 8213 after 2 days of heat stress ( Figure 9). As a control set, addition of exogenous PABA further enhanced their enzyme activities, while treatment of sulfanilamide compromised them. As expected, the transgenic strain TB-2, which has higher level of basal PABA, also possesses high activity of CAT and SOD. Together, these data suggest that PABA may reduce the H 2 O 2 accumulation through increasing CAT and SOD activity in mushroom under heat stress, thus eventually helps to alleviate the direct oxidative damaged incurred by heat stress.

PABA Elevates HSPs and Chitinase Proteins to Enhance Thermotolerance of Mushroom under Heat Stress
Heat shock proteins (HSPs) play essential roles in organisms against high-temperature stress; and Chitinase was upregulated and contributed to the resistance against the biotic and abiotic stress in tobacco [29]. Here, we hypothesized that under heat stress, thermotolerant strain 02 should express higher level of defense-related proteins (e.g., HSPs and Chitinase), and that the induction of these proteins is mediated by PABA.
We first measured the time course of expression of HSPs in strains 02 and 8213 under heat stress. As shown in Figure 10A, HSP70, HSP90 and Chitinase protein showed faster and more robust accumulation in strain 02 than in strain 8213, with abundance peaked at 48-hour time point post-treatment. Further, we investigated whether the upregulation of those stress-resistant proteins was mediated by PABA induction or just parallel event under stress. To address this, we treated the mushrooms with either exogenous PABA or PABA synthase inhibitor sulfanilamide under heat stress to determine whether expressions of HSPs and Chitinase would change accordingly. Our results showed that application of exogenous PABA increased HSP20, HSP70 and Chitinase protein levels in thermo-sensitive strain 8213 under heat stress ( Figure 10B), and inhibiting PABA production by sulfanilamide reduced the accumulations of HSPs and Chitinase in thermotolerant strain 02 ( Figure 10C). Moreover, the transgenic line TB-2 also showed higher protein accumulation of HSPs and Chitinase than strain 8213 under heat stress ( Figure 10D). Together, those results demonstrate that PABA synthase and its product PABA can also induce the upregulation of HSPs and Chitinase protein upon heat stress. Thus, PABA may enhance thermotolerance of mushroom with multiple mechanisms.

Discussion
Comparative Proteomics Studies Unveils a Protein Network Supporting Thermotolerance of Strain 02 under Heat Stress Mushroom A. bisporus strain 02 and strain 8213 have a very close phylogenetic relationship ( Figure S1), and share various biological and agricultural characteristics. However, their capability to tolerate high-temperature stress is dramatically different from each other: strain 02 is thermotolerant to 33uC, which is lethal to strain 8213 (Figure 1).
To investigate the global protein changes of these two strains under heat stress, we performed a comparative proteomic analysis to identify differentially expressed proteins. First, we observed more induction of the HSPs and Chitinase protein families in strain 02 compared to strain 8213, which may contribute to thermotolerance of stain 02. Antioxidant enzyme proteins were also increased in strain 02 under high-temperature stress suggesting their function in clearing ROS (reactive oxygen species) in mushrooms. Accordingly, strain 02 showed lower level of H 2 O 2 accumulation compared to 8213 under heat stress. Similarly, increasing antioxidant enzyme activities such as SOD and ascorbate peroxidase (APX) can enhance thermotolerance in potato and wheat [30,31,32,33]. Meanwhile, we found that many proteins belonging to energy and metabolism groups showed higher accumulation in strain 02 than in 8213 under high temperature, which is consistent with previous studies that the proteins in the class of energy and metabolism greatly contribute to plant tolerance against high-temperature stress [34,35,36,37]. The proteins from the group of cell structures and division were also induced to higher accumulation in strain 02, which are likely responsible for the quick increase of biomass in strain 02 under high temperature. Some proteins involved in RNA folding, such as ATP-dependent RNA helicase DRS1 and ATP-dependent RNA helicase DBP8 were also differentially regulated, which indicates that they possibly modulate the RNA structure to adapt to hightemperature stress, which confirmed previous reports that RNA helicase takes part in plants' response to abiotic stress such as cold and salt stress [38,39]. Further, we found that Chitinase was highly induced under high-temperature stress, suggesting a function of this enzyme in mushroom thermotolerance.
Differential regulation of many proteins involved in calcium and G-protein signaling were observed in mushrooms responding to high temperature, suggesting their involvement in thermoregulation. The mitogen-activated protein kinase (MAPK) pathway is involved in plants' response to multiple stresses [40,41]. Consistently, we identified several protein spots that encode MEKK and related serine/threonine protein kinases (e.g., spot 17) with upregulation in strain 02 in response to high-temperature stress.
Collectively, we propose that mushrooms use multiple strategies to enhance their tolerance against high-temperature stress.

PABA Enhances Mushroom Thermotolerance by Removing H 2 O 2 and Elevating Defense-related Proteins
Among the differentially expressed proteins related to heat stress, we found that a PABA synthase is constitutively more abundant in 02 and heat stress further increased its expression. We further demonstrate that mRNA level of Pabs gene and its synthesis product PABA correlated with the PABA synthase protein level in response to heat stress. Since PABA is structurally similar to SA, and SA is known to play a role in plants' resistance to abiotic stress [16,17], it's possible that PABA may play an important role in enhancing mushroom thermotolerance. In the ensuing physiological studies, we found that exogenous PABA application could greatly improves thermotolerance of sensitive strain 8213, as measured by reduced mycelia breaks. On the other hand, treatment with a PABA synthase inhibitor reduced the PABA accumulation and reduced thermotolerance of strain 02. To carry our studies at a potentially applicable transgenic level, we cloned the Pabs gene of strain 02, confirmed its PABA synthase activity in vitro, and transgenically overexpressed it in thermosensitive strain 8213. These studies established that Pabs gene of strain 02 can substantially improve thermotolerance of strain 8213. In addition, we found that reducing PABA synthase transcription using a transgenic antisense sequence against Pabs gene in strain 02 impaired its thermotolerance (data not shown). We therefore conclude that PABA modulates the mushroom's thermotolerance.
The higher accumulation of H 2 O 2 in strain 8213 compared to strain 02 under heat stress may explain the more severe damage in strain 8213. We found that PABA plays a role in clearing heat stress induced H 2 O 2 accumulation (Figure 8) likely through increasing antioxidant enzyme SOD and CAT activities (Figure 9), thus alleviating the injury. Moreover, we showed that the abundance of thermotolerance-related proteins, such as HSPs and Chitinase, was higher in strain 02 as compared to strain 8213 under heat stress. Importantly, the induction of HSPs and Chitinase can be further strengthened either through in vitro application of PABA or through transgenic overexpression of Pabs gene. PABA synthase inhibitor sulfanilamide, on the other hand, had the opposite effects. We therefore propose that PABA functions as a mediator to stimulate signaling leading to systematic acquired resistance (SAR). Future experiments using ''plus and minus PABA'' in proteomic profiling combined with RNA-Seq will likely identify more PABA targets and downstream pathways.
To summarize, we propose a model for the role of PABA in enhancing mushroom thermotolerance through mediating the expression of defense-related effecters and clearing H 2 O 2 accumulation ( Figure 11). Mushrooms have adopted multiple strategies to adapt to heat stress; important among these is the upregulation of PABA. Experimentally increasing PABA level enhances, while reducing PABA level impairs, mushroom's tolerance against heat stress. Our findings provide new insight into the role of PABA in enhancing thermotolerance of mushroom.