Positive selection and functional diversification of transcription factor Cmr1 homologs in Alternaria

Abstract Transcription factor Cmr1 (Colletotrichum melanin regulation 1) and its homologs in several plant fungal pathogens are the regulators of the 1,8-dihydroxynaphthalene (DHN)-melanin biosynthesis pathway and have evolved functional diversification in morphology and pathogenicity. The fungal genus Alternaria comprises the group of “black fungi” that are rich in DHN-melanin in the primary cell wall and septa of the conidia. Some Alternaria species cause many economically important plant diseases worldwide. However, the evolution and function of Cmr1 homologs in Alternaria remain poorly understood. Here, we identified a total of forty-two Cmr1 homologs from forty-two Alternaria spp. and all contained one additional diverse fungal specific transcription factor motif. Phylogenetic analysis indicated the division of these homologs into five major clades and three branches. Dated phylogeny showed the A and D clades diverged latest and earliest, respectively. Molecular evolutionary analyses revealed that three amino acid sites of Cmr1 homologs in Alternaria were the targets of positive selection. Asmr1, the homolog of Cmr1 in the potato early blight pathogen, Alternaria solani was amplified and displayed the sequence conservation at the amino acid level in different A. solani isolates. Asmr1 was further confirmed to have the transcriptional activation activity and was upregulated during the early stage of potato infection. Deletion of asmr1 led to the decreased melanin content and pathogenicity, deformed conidial morphology, and responses to cell wall and fungicide stresses in A. solani. These results suggest positive selection and functional divergence have played a role in the evolution of Cmr1 homologs in Alternaria. Key points • Cmr1 homologs were under positive selection in Alternaria species • Asmr1 is a functional transcription factor, involved in spore development, melanin biosynthesis, pathogenicity, and responses to cell wall and fungicide stresses in A. solani • Cmr1 might be used as a potential taxonomic marker of the genus Alternaria Supplementary Information The online version contains supplementary material available at 10.1007/s00253-023-12893-7.


Introduction
Melanin, a natural amorphous polymer of high molecular weight, is a ubiquitous pigment found in microorganisms, plants, and animals (Cordero and Casadevall 2020).Many plant pathogenic ascomycete and dothideomycete fungi synthesize melanin following the 1,8-dihydroxynaphthalene (DHN) pathway (Kimura and Tsuge 1993;Tsuji et al. 2000;Eisenman and Casadevall 2012;Schumacher 2016;Wang et al. 2021b;Li et al. 2022).In fungi, melanin confers protection against abiotic stresses, such as ultraviolet radiation, extreme temperatures, reactive oxygen species (ROS), high osmotic pressure, hypertonic and heavy metal exposure, and dehydration (Cordero and Casadevall 2017).Melanin is also involved in fungal cell development of conidia, black appressoria, and sclerotia (Money et al. 1998;Kheder et al. 2012;Yu et al. 2015;Liang et al. 2018;Wang et al. 2021a;Zhou et al. 2022).In addition, DHN-melanin has a function in virulence of some phytopathogenic fungi during their interactions with the host immune response.For those fungi that produce melanized appressoria for host penetration, such as the rice blast pathogen Magnaporthe oryzae and the cucumber anthracnose pathogen Colletotrichum lagenarium, DHN-melanin is required for appressorial melanogenesis for pathogenicity (Kubo et al. 1982;Howard and Valent 1996).However, similarly, in some fungi like the gray mold pathogen Botrytis cinerea that do not form appressoria, DHNmelanin also correlates with their virulence.Mutants of B. cinerea defective in three melanin biosynthesis-related genes fail to produce melanin, but they show increased virulence (Zhang et al. 2015;Zhou et al. 2022).
The worldwide-distributed fungal genus Alternaria is the typical representative of Dematiaceae and comprises the group of "black fungi", which are rich in DHN-melanin in the primary cell wall and septa of their conidia (Carzaniga et al. 2002;Thomma 2003).It is currently classified into 24 sections based on phylogenetic and morphological studies (Woudenberg et al. 2013).Some Alternaria species are fungal pathogens of many economically important plant diseases, which result in huge losses in agricultural production and postharvest agricultural products throughout the world (Thomma 2003).A. brassicicola is responsible for black spot disease on almost all cultivated Brassica species including cabbage, broccoli, canola, mustard, and rapeseed in the world (Neergaard 1945;Sigareva and Earle 1999;Westman et al. 1999).A. alternata is a pathogen on over 100 plant hosts including cereal crops, vegetables, and fruits (Rotem 1994).For example, A. alternata causes Alternaria brown spot of citrus (Gai et al. 2021), leaf blight on Ophiopogon japonicus (Wang et al. 2021a, b), and black spot of kiwifruit (Huang et al. 2021).Alternaria solani is the causal agent of early blight, one of the most destructive diseases of potato and tomato worldwide (Agrios 2005).In China, early blight ranks number 1 fungal disease on potato (Ma et al. 2020).A. solani also causes early blight on other members of the Solanaceae family including pepper, eggplant, black nightshade, horse nettle, and Jerusalem cherry (Farr and Rossman 2023).Although potato early blight is a devastating disease, only a few factors involved in pathogenicity have been identified in A. solani, such as toxins: alternaric acid and solanapyrones A, B, and C (Brian et al. 1949).
Currently, two Cmr1 homologs including Amr1 in A. brassicicola (Cho et al. 2012) and CmrA in A. alternata (Fetzner et al. 2014) have been identified and characterized in Alternaria.The function of Cmr1 homologs in regulating the DHN-melanin biosynthesis is highly conserved in A. brassicicola, A. alternata, and other fungi described above (Tsuji et al. 2000;Eliahu et al. 2007;Kihara et al. 2008;Cho et al. 2012;Fetzner et al. 2014;Zhou et al. 2017;Krishnan et al. 2018;Wang et al. 2021a, b;Zhang et al. 2022).Interestingly, based on the amino acid sequence comparison between Amr1 and CmrA, there was one amino acid insertion at site 934 in Amr1 and 31 polymorphic sites (data not shown), suggesting that the amino acid sequences have diverged in Amr1 and CmrA.Furthermore, Amr1 negatively regulates the pathogenicity of A. brassicicola (Cho et al. 2012).Cmr1 homologs in phytopathogenic fungi have evolved to differentiate in pathogenicity, which may relate to the interaction with their host plants.As a consequence, given the availability of the genome sequences of 44 Alternaria species to date and our laboratory stored 10 A. solani isolates from different geographic locations in China, we raised the following questions.How did Cmr1 homologs evolve in Alternaria and in A. solani?What kind of functions the Cmr1 homolog in A. solani may have?Of particular interest, whether there is functional difference in pathogenicity regulated by Cmr1 homologs in Alternaria?Here, we exploited Alternaria genomes for identifying Cmr1 homologs followed by investigating the diversity and evolutionary relationship of Cmr1 homologs in Alternaria as well as those in A. solani.We also engineered one A. solani isolate deficient in the production of a Cmr1 homolog using a gene knockout technology to test whether Cmr1 homolog acts as a virulence factor and whether DHN-melanin is important for pathogenesis of A. solani.

Plants, fungal isolates, and growth conditions
Potato (Solanum tuberosum cv.Zaodabai and cv.Favorita) plants were grown and maintained in a growth chamber (23 °C day/15 °C night temperatures with 14 h of light).A. solani isolates (Supplemental Table S1) and mutants were maintained as glycerol stocks.For conidial sporulation, the wild type (WT) A. solani isolate SH0806 and its mutants were grown on potato carrot agar (PCA) medium at 25 °C.

Database mining
The protein sequence of Amr1 was used as a query to BlastP searching against all annotated protein databases of Alternaria spp.available in the websites of GenBank, National Center for Biotechnology Information (http:// www.ncbi.nih.gov), JGI, Joint Genome Institute, US Department of Energy (http:// www.jgi.doe.gov), and AGD, Alternaria genome database (http:// alter naria.vbi.vt.edu) to identify Cmr1 homologs in Alternaria.For some Alternaria species without genome annotation, first, their corresponding genome sequences were predicted into genes through the AUGUSTUS server (Stanke and Morgenstern 2005).Second, the local protein databases were constructed using the Blast-2.12.0 program (Altschul et al. 1997).Third, BLASTP searching was performed against the generated local protein databases using Amr1 as a query.Finally, all achieved protein sequences were searching against the PFAM database in SMART server (Letunic et al. 2020) to confirm the existence of characteristic motifs of Cmr1.

Sequence analyses
The amino acid sequences corresponding to individual motifs of Cmr1 homologs were aligned using the ClustalX program (Thompson et al. 1997) and submitted to WebLogo server (Crooks et al. 2004) to calculate the consensus sequences.MEGA7.0.21 (Kumar et al. 2016) was used to make multiple sequence alignments of Cmr1 homologs followed by reconstructing a phylogenetic tree using the maximum likelihood (ML) method with 1000 bootstrapping replications.The Cmr1 (GenBank accession number ABI81496) in C. heterostrophus was used as an outgroup.The ML tree was used to estimate divergence times using the method of Reltime implemented in MEGA7.0.21.Two calibration points were used.That is, the node between Cmr1 in C. lagenarium and Pig1 in M. grisea was constrained to 299 million years ago (MYA) following the estimated divergence time (355 MYA) of Colletotrichum and Alternaria.Their divergence time estimates were derived from the TimeTree server (Kumar et al. 2022).
The selection pressures were performed using the CODEML program within the Phylogenetic Analysis by Maximum Likelihood (PAML) software package with the ML method (Yang 1997;Nielsen and Yang 1998;Yang et al. 2000;Yang and Bielawski 2000).A series of likelihood ratio tests (LRTs) were carried out to test for positive selection by comparing the null models (M0, M1a, M7) with their corresponding alternative models (M3, M2a, M8).The empirical Bayes theorem implemented in CODEML was used to calculate the posterior probability that a particular amino acid site belongs to a given selection class (neutral, deleterious or advantageous) (Yang 1997).Amino acid sites with a high probability coming from advantageous class of sites are more likely to be under positive selection.

DNA manipulations, gene cloning, and DNA sequencing
Genomic DNA was extracted from mycelia using the fungal genomic DNA extraction kit (Solarbio, Beijing, China).Based on the sequence of the Cmr1 homolog (AGD accession number ASL_PT00349) in A. solani, a pair of oligonucleotide primers Asmr1F/Asmr1R was designed for the amplification of the entire coding region of asmr1 from A. solani isolates.The amplified fragment was cloned into pTOPO-Blunt vector (Aidlab, Hong Kong, China) and confirmed by DNA sequencing.All primers used in this paper were listed in Supplemental Table S2.

Generation of the asmr1 deletion construct
The deletion construct was made by replacement of the gene with a hygromycin B (HygB) resistance cassette.The HygB phosphotransferase gene (hph) was used as a selective marker.The gene replacement construct was produced by a double-joint PCR method (Yu et al. 2004) and consisted of PCR fragments of 5′ and 3′ flanking region sequences of asmr1 overlapping within hph.The 5′ and 3′ flanking regions of asmr1 were amplified from A. solani using Asmr1-5F/ Asmr1-5R and Asmr1-3F/Asmr1-3R, respectively.The HygB resistance cassette was amplified from pAg1-H3 (Zhang et al. 2003) with hph1349F/hph1349R.The primer set Asmr1-nestF/Asmr1-nestR was used to amplify the fused knockout construct, which was confirmed by DNA sequencing.

Generation of the complemented construct of the Δasmr1-2
The Δasmr1-2 mutant was complemented with the wildtype (WT) allele under its native promoter.The gene fragment containing a 1497 bp 5′ flanking region and the coding region of asmr1 without the stop codon was amplified from isolate SH0806 using the primer set Asmr1-CF/Asmr1-CR and verified by DNA sequencing.The confirmed gene fragment and the linearized pYF11-gfp-Gen vector (Chen et al. 2018) were co-transformed into yeast strain XK1-25 using the Yeast Transformation Kit (Coolaber, Beijing, China) to achieve the fusion construct through the yeast gap repair approach.The resulting construct, pYF11:Asmr1 was verified by DNA sequencing.

Preparation of fungal protoplasts
Preparation of fungal protoplasts was conducted as previously described with a minor modification (Akamatsu et al. 1997).Details are as follows: Agar plugs containing WT A. solani (SH0806) or its mutants were cultured on potato dextrose agar (PDA) until spore sporulation.Spores were collected, transferred into potato dextrose broth (PDB), and shaken at 120 rpm at 28 °C for 1-2 days.Mycelia were harvested using miracloth (EMD Biosciences, Rockland, MA, USA), rinsed with sterilized ddH 2 O three times, and re-suspended in 50 ml of filter sterilized solution of enzymes containing 10 mg/ml lysing enzymes (Sigma, St. Louis, MO, USA) and 10 mg/ml Kitalase (Wako Pure Chemical Industries, Ltd., Osaka, Japan), which were dissolved in 0.7 M NaCl.The suspension was shaken at 70 rpm at 30 °C for 3-4 h, filtered by miracloth, and centrifuged at 3000 × g for 5 min.After the removal of the supernatant, the pellets of protoplasts were gently re-suspended with 0.7 M NaCl and centrifuged at 3000 × g for 5 min.Protoplasts were re-suspended with 1 ml of STC solution [1.2 M sorbitol, 10 mM Tris-HCl (pH 7.5), 10 mM CaCl 2 ] and centrifuged at 3000 × g for 5 min.Pelleted protoplasts were re-suspended with an appropriate volume of STC solution to obtain a final solution containing 1 × 10 8 protoplasts.The protoplasts were checked under the light microscope and the number of protoplasts was determined with a hemacytometer.

Transformation of A. solani and molecular confirmation of transformants
A. solani protoplasts were transformed following the method of Akamatsu et al. (1997) with some modification.One to 5 µg plasmid DNA and 5 µl of heparin sodium (5 mg/ml) were mixed with 1 ml of STC solution containing 1 × 10 8 protoplasts and incubated on ice for 30 min.One third amount of SPTC solution (STC solution containing 40% PEG 4000) was added into the mixture and incubated at room temperature for 20 min.Subsequently, 50 ml of RM (regeneration medium) (0.1% yeast extract, 0.1% casein enzymatic hydrolysate, 1 M sucrose) was added into the transformation mixture and incubated at 25 °C overnight followed by plating on PDA containing 100 µg/ml Hyg B. After incubating at 28 °C for 7-14 days, putative transformants were selected under two rounds of single-spore isolation.To obtain stable hph expressing putative transformants, single-spore isolated putative transformants were sub-cultured on PDA without HygB and PDA with HygB in turn for five generations.Selected transformants were verified by PCR with five primer pairs, H850F/H852R to check the presence of hph; Asmr1-InF/Asmr1-InR to check the presence of asmr1; Asmr1-nestF/Asmr1-nestR to check whether asmr1 was replaced by hph; Asmr1-nestF/H855R to check the correct integration of HygB resistance cassette into the upstream region of asmr1; H856F/nestR to check the correct integration of HygB resistance cassette into the downstream region of asmr1.Southern blot hybridization according to the manufacturer's instructions of the DIG-High Prime DNA Labeling and Detection Kit II (Roche, Indianapolis, IN, USA) followed by DNA sequencing using the primer pair HphCF/HphCR was performed to confirm the presence of one single copy of hph and the deletion of asmr1 in PCR verified transformants.Integration of the WT asmr1 allele under its native promoter into the Δasmr1-2 mutant genome was confirmed with PCR using the asmr1 internal region primer pair Asmr1-InF/Asmr1-InR.

Examination of colony growth and conidia
WT A. solani and its mutants from glycerol stocks were inoculated on PDA in the dark at 25 °C for 5 days.Agar plugs (5 mm in diameter) containing fresh mycelia on PDA were sub-cultured on PDA, PCA, or potato agar (PA) and PDA containing 0.2 g/L Congo red, 0.02% SDS (sodium dodecyl sulfate), 10 mM H 2 O 2 , 5 ppm mancozeb, 5 ppm chlorothalonil, 5 ppm azoxystrobin, 1 M KCl, 1 M NaCl, 1 M sorbitol, 10 mM FeSO 4 , 1.5 mM CuSO 4 or 250 mM CaCl 2 and incubated at 25 °C for 5 days.The diameter of each colony was measured using a Vernier caliper and the inhibitory rate of fungal mycelial growth was calculated.All experiments were conducted three times.Three replicates were used for each experiment.Significant differences in colony diameter and inhibitory rate were determined by oneway analysis of variance (ANOVA).Conidia were collected from PCA with sterilized ddH 2 O and examined for their morphology using a Nikon TE2000 inverted fluorescence microscope (Nikon, Miato, Japan).

Extraction and detection of fungal melanin content
Extraction of fungal melanin was conducted following the method described by Yang et al. (2012) with modification.A quarter of 1 g of fresh mycelia was collected, transferred into 10 ml of 1 M NaOH solution, incubated at 100 °C water bath for 5 h, and filtered using one layer of filter paper.The filtrate was adjusted to pH to 2.0 using HCl, incubated at 4 °C overnight, and centrifuged at 6000 × g for 15 min.The pellets collecting melanin were dissolved in 5 ml of 1 M NaOH and the absorbance of the solution at 459 nm was measured by UV spectrophotometer.The melanin content (Y, mg/g) of fungal mycelia was calculated as the following formula: Y = (X + 0.0114)/0.0495,where Y represents the yield of fungal melanin and X indicates the absorbance value at 459 nm.Three replicates were used for the experiment.Significant differences in melanin yield were determined by ANOVA.

Measurement of ROS
WT A. solani and its mutants from glycerol stocks were inoculated on PDA in the dark at 25 °C for 5 days.Agar plugs (5 mm in diameter) containing fresh mycelia were transferred to PDA and incubated at 25 °C for 3 days.Three days-old fungal cultures were stained with nitroblue tetrazolium (NBT) reagent (Beyotime Biotechnology, Shanghai, China) to detect the production of ROS.

Plant infection assays
A total of 20 µl of suspension containing 1 × 10 5 spores/ml from WT A. solani isolate SH0806 or its mutants was used to inoculate the detached leaves of potato.The inoculated leaves were placed in a plastic tray lined with wet paper towels, tightly fitted with a transparent plastic cover, and kept in a growth chamber at 25 °C under a photoperiod of 12 h.For quantitative real-time RT-PCR, leaves were harvested at 12 time points as described above, individually frozen in liquid nitrogen, and stored at -80 °C prior to RNA extraction.For pathogenicity test, the symptoms and necrotic lesion size were recorded at 14 days post-inoculation (dpi).The necrotic lesion size was calculated based on the square area (cm 2 ) of the lesion.The experiment was performed three times.Each time had three technical repeats.Significant differences in lesion size were determined by ANOVA.

Cmr1 homologs in Alternaria all contain one additional diverse fungal specific transcription factor motif
To identify Cmr1 homologs in the genus Alternaria, we mined publicly available Alternaria genomes representing 44 Alternaria species (Supplemental Table S3).After verification of the motifs, a total of 44 Cmr1 homologs (one per each species) including previously reported CmrA in A. alternata and Amr1 in A. brassicicola were identified (Supplemental Table S4 S4).All 42 Cmr1 homologs contained the expected motifs located at N-terminus: two Cys 2 His 2 zinc finger motifs (ZnF_C2H2, SM000355) at amino acid positions 2 to 24 and 30 to 52, respectively, as well as one GAL4-like Zn(II) 2 Cys 6 binuclear cluster DNA-binding motif (GAL4, SM00066) at amino acid positions 73 to 116 (Fig. 1a).In addition, they all carried one fungal specific transcription factor motif (fungal_trans, PF04082) at the C-terminus (amino acid positions 498 to 740 or 499 to 741) (Fig. 1a).Weblogo analysis of motifs of Cmr1 homologs revealed that the amino acid sequences within each of two ZnF_C2H2 motifs and the GAL4 motif were completely conserved (Fig. 1b-d).In contrast, the amino acid sequences within the fungal_trans motif were diverse at 17 sites in Cmr1 homologs (Fig. 1e).Our data indicates that Cmr1 homologs in Alternaria contain one extra diverse fungal_ trans motif except for three fully conserved characteristic motifs of Cmr1.

Five clades of Cmr1 homologs emerge in Alternaria
To determine the evolutionary relationship of Cmr1 homologs in Alternaria, a phylogeny (phylogenetic tree) was generated based on alignment of their amino acid sequences using the ML method implemented in the program MEGA7.0.21 (Kumar et al. 2016 and A. brassicicola, it appears that these Cmr1 homologs have been under selection for functional divergence during the interaction with their hosts or living environments.Most surprising, the phylogeny of Cmr1 homologs followed the authoritative classification system of the genus Alternaria established by Woudenberg et al. (2013).It implies that Cmr1 can be used as a potential taxonomic marker of the genus Alternaria, which can distinguish at least five groups of Alternaria species.

The A and D clades diverge latest and earliest, respectively
To investigate how divergent Cmr1 homologs appeared in Alternaria, we estimated the divergence times for all branching points in the phylogeny of 42 Cmr1 homologs using the Reltime ML method implemented in the Timetree Wizard window of the program MEGA7.0.21 (Kumar et al. 2016) and two calibration points (see "Materials and methods" for details).As seen in Fig. 2

Cmr1 homologs in Alternaria are under positive selection
The value of the ratios of nonsynonymous to synonymous nucleotide substitution rates in protein-coding DNA sequences (ω = d N /d S ) is the indicator of the type of selection pressures (Yang and Bielawski 2000).The ω values of 1, < 1, and > 1 represent neutral, positive, and negative selection, respectively.To examine the selection pressure underlying sequence diversification in Cmr1 homologs in Alternaria, we analyzed the 42 Cmr1 homologs using the ML method (a statistical distribution to model ω variations among sites along the protein) implemented in the program CODEML in the PAML software package (Yang 1997).First, LRTs were performed to test for selection.
In details, three pairs of ML models of codon substitution such as M3/M0, M8/M7, and M2a/M1a were used (Yang et al. 2000) (Table 1).Both the discrete model M3 and the selection model M2a did not identify any amino acid sites under positive selection.In contrast, the model M8  1).Thus, the null model M7 which does not allow for the presence of positive selection sites with ω > 1 was rejected, indicating that the detected positive selection under the discrete M8 was statistically significant at P value < 0.05.Second, the amino acid sites under positive selection when LRT tested the presence of sites with ω > 1 were identified.This was achieved by using the empirical Bayes theorem to calculate the (posterior) probability.We identified three sites including 171 S, 927 G, and 966 S implicated as being under positive selection with the confidence level of 98%, 78.6%, and 69.5%, respectively, under model M8.Interestingly, none of the three identified positive selection sites were located in any of the four identified protein motifs of Cmr1 homologs (Fig. 1a and Supplemental Fig. S3).The results suggest that positive selection has affected the evolution of Cmr1 homologs in Alternaria.

The Cmr1 homolog is fully conserved at the amino acid level in A. solani isolates
To study whether there is sequence polymorphism in Cmr1 homologs in A. solani, we amplified, cloned, and sequenced Cmr1 homologs from ten A. solani isolates from different potato or tomato producing areas in China including the northeast (Suihua, Keshan, Daqing, Jiagedaqi District, and Benxi), the northwest (Dingxi), the north (Bejing and Shandong), and the south (Jiansu) (Supplemental Table S1).
Multiple alignments of the nucleotide sequences of these ten Cmr1 homologs and two additional Cmr1 homologs from isolates BMP 0185 and altNL03003 (The Netherlands) revealed that polymorphisms were detected at 10 of 3277 nucleotides (data not shown).All polymorphic sites were located in the exon sequences," however, they could not cause amino acid changes.The results imply that Cmr1 homologs at amino acid levels are completely conserved and appear to undergo negative selection in A. solani isolates from different geographic locations.

Asmr1 exhibits transcriptional activity in yeast
To link the sequence analysis and the function of the Cmr1 homolog in A. solani, we validated the transcription factor activity of the Cmr1 homolog in A. solani using the transcription activation assay.The construct pGBKT7 carrying the open reading frame of Asmr1 (we designated the Cmr1 homolog amplified from isolate SH0806 A. solani melanin regulation transcription factor 1 or Asmr1) was first achieved.After transforming into the yeast strain Y2H GOLD, both empty pGBKT7 and pGBKT7:Asmr1 constructs enabled Y2H GOLD to grow on selective dropout medium without tryptophan (SD-T).However, only the strain Y2H GOLD carrying pGBKT7:Asmr1 construct could grow on SD medium without tryptophan, histidine, and adenine (SD-T-H-A).In addition, transformed colonies of pGBKT7:Asmr1 exhibited a blue color on SD-T-H-A medium supplemented with 5-bromo-4-chloro-3-indolylβ-D-galactopyranoside (X-α-Gal) (Fig. 3a).The results indicate that Asmr1 is able to activate the transcription of reporter genes including His3, Ade2, and Mel1, supporting that Asmr1 is a functional transcription factor.

Asmr1 is involved in spore development and melanin biosynthesis
To further investigate the function of Asmr1, a deletion mutant of the asmr1 gene was constructed by replacing the coding region with a HygB resistance cassette (Fig. 3b).Three HygB resistant transformants, namely, Δasmr1-1, Δasmr1-2, and Δasmr1-3 were obtained.The results of PCR analysis followed by Southern hybridization and DNA sequencing showed that Δasmr1-1 was an ectopic insertion mutant, whereas Δasmr1-2 and Δasmr1-3 were correct asmr1 replacement mutants (Supplemental Fig. S4a, Fig. 3b, c).We complemented the Δasmr1-2 mutant with the WT allele.PCR and RT-PCR analyses confirmed that Δasmr1-2C was a correctly asmr1 complemented transformant of mutant Δasmr1-2 (Supplemental Fig. S4b).
We determined the effect of asmr1-deficiency on conidia in A. solani.The WT and the Δasmr1-2 and Δasmr1-2C mutants all produced conidia on PCA and showed little difference in conidia production (data not shown).However, the conidia of the Δasmr1-2 mutant were pale, with nonblack color, and deformed, in comparison with the dark brown to black color and intact conidia of the WT and the Δasmr1-2C mutant (Fig. 4a and Supplemental Fig. S5).In certain individual conidia of the Δasmr1-2 mutant, some part was colorless or white while another part looked brownish.It is worth noting that the beak of the conidia of the Δasmr1-2 mutant became shorter compared with the WT and the Δasmr1-2C mutant (Fig. 4a and Supplemental Fig. S5).We subsequently measured the melanin content in A. solani.The color of the crude melanin extract from the Δasmr1-2 mutant looked very light and was close to transparent.In contrast, that from the WT and the Δasmr1-2C mutant displayed brown or dark brown color (Fig. 4b, c).Compared to WT, the melanin yield of the Δasmr1-2 mutant significantly decreased (Fig. 4c).Consequently, we quantified the transcript levels of A. solani orthologs of 13 DHN-melanin biosynthesis-related genes in fungi.Compared to WT,ten genes including pksA,aygA,aygB,brm1,brm2,brm3,lccB,lccC,lccE,and lccF were Page 10 of 20 all significantly less expressed in the Δasmr1-2 mutant, whereas the three laccase genes lccA, lccD, and lccG were all significantly upregulated in the Δasmr1-2 mutant (Fig. 4d), suggesting that the expression of the former genes was positively dependent on Asmr1 and that of the latter negatively controlled by Asmr1.Altogether, the results indicate that Asmr1 is involved in conidial morphogenesis and the regulation of melanin biosynthesis in A. solani.

Decreased pathogenicity of the Δasmr1-2 mutant in potato
To test whether the Asmr1 is involved in the infection process of A. solani in potato, we determined the expression of asmr1 during infection using quantitative real-time RT-PCR.The asmr1 gene showed elevated levels of expression throughout early infection stage from 0 to 12 hpi (Fig. 5a), suggesting that Asmr1 may involve in the infection process of A. solani in potato.This finding promoted us to perform a pathogenicity test.Two weeks after inoculation, the necrotic lesions caused by the Δasmr1-2 mutant were significantly smaller than those caused by the WT (Fig. 5b, c).The results suggest that the pathogenicity of the Δasmr1-2 mutant decreases in potato.

Responses of the Δasmr1-2 mutant to cell wall and fungicide stresses
Melanin plays a role in protecting fungi against a diverse array of stresses (Cordero and Casadevall 2017).To this end, we examined the effect of stressors including cell wall inhibitors, oxygen stressor, fungicides, osmotic stressors, and metal ions on the Δasmr1-2 mutant.The mycelial growth of the Δasmr1-2 mutant was significantly inhibited in the presence of Congo red or SDS (Fig. 6a), suggesting that Asmr1 may be involved in the regulation of the cell-wallrelated pathway in A. solani.Similarly, the mycelial growth of the Δasmr1-2 mutant was also significantly reduced on the medium containing 10 mM H 2 O 2 (Fig. 6a) .This is likely due to the increase of ROS in the Δasmr1-2 mutant, which results in a counterbalance between ROS and H 2 O 2 .Thus, we decided to examine the ROS content in the Δasmr1-2 mutant through staining 3-day-old fungal cultures with NBT reagent.NBT reacts with superoxide anion to form blue or blue purple formazan.The higher the content of superoxide anion, the darker the color (Halliwell and Gutteridge 1985).As seen in Fig. 6b, the entire colony and hyphae of the Δasmr1-2 mutant exhibited a blue purple color, whereas both WT and the Δasmr1-2C mutant changed not in color.
The results indicate that the ROS content in the Δasmr1-2 mutant is much higher than that of WT.Perhaps, without protection from melanin, asmr1-deficient A. solani increases the ROS content to combat against external adverse environmental conditions.Three economic and practical fungicides including mancozeb, chlorothalonil, and azoxystrobin for the prevention and control of potato early blight were used in this study.Compared to WT, the Δasmr1-2 mutant showed a significant decrease in the mycelial growth on the medium supplemented with 5 ppm of mancozeb, chlorothalonil, or azoxystrobin (Fig. 7).These results indicate that the deletion of asmr1 results in the reduced tolerance of A. solani to these three fungicides.It also indirectly suggests that the effect of mancozeb on preventing and controlling potato early blight is better than that of chlorothalonil and azoxystrobin.
The mycelial growth of the Δasmr1-2 mutant was comparable to that of the WT and Δasmr1-2C mutant on the medium containing KCl, NaCl or sorbitol (Supplemental Fig. S6a).Since KCl, NaCl, and sorbitol are related to fungal osmotic stress, we speculate that Asmr1 may not be involved in the regulation of the osmotic stress pathway in A. solani.In addition, the mycelial growth of the WT and of the Δasmr1-2 and Δasmr1-2C mutants were completely inhibited on the medium supplemented with Fe 2+ (Supplemental Fig. S6b).The colony size was not significantly different between the WT and the Δasmr1-2 mutant under Cu 2+ stress.However, the Δasmr1-2 mutant showed a significant reduction in the mycelial growth on the medium containing Ca 2+ .Altogether, it suggests that Asmr1 may be involved in the regulation of the A. solani response to Ca 2+ , but not to Fe 2+ and Cu 2+ .
We also determined the effect of asmr1-deficiency on vegetative growth of the WT and the Δasmr1-2, and Δasmr1-2C mutants on PDA, PCA, or PA (Supplemental Fig. S7).The Δasmr1-2 mutant displayed orange or near pink and light orange color (with less dense aerial hyphae) on PDA and PCA, respectively, compared to the blackcolored WT.On PA lacking sugar, WT changed from black on the medium containing sugar to gray, while the Δasmr1-2 Fig. 7 Effects of the asmr1 deletion on A. solani responses to three fungicides.Comparison of mycelial growth inhibitory rates between mutants and wildtype on PDA in the presence of indicated fungicides.Values represent means from three independent experiments with three replicates each.Significant differences in growth inhibitory rates were determined by ANOVA.***P < 0.001 (comparison with wild-type).Abbreviations: WT, wild-type A. solani; Δasmr1-2, asmr1 deletion mutant; Δasmr1-2C, complementation mutant Δasmr1-2:asmr1 Fig. 8 The structures of CmrA, Asmr1, and Amr1.The protein structures of Asmr1 and Amr1 were predicted by SWISS-MODEL server at http:// swiss model.expasy.org using CmrA (UniProt accession num-ber A0A075QC79) as a template.The structures were visualized and aligned using the PyMOL software mutant remained little changed.Compared to the WT, the colony size in the Δasmr1-2 mutant was significantly smaller on PDA, PCA, or PA.The phenotype was comparable for both the Δasmr1-2C mutant and the WT.The results suggest that Asmr1 affects the vegetative growth in A. solani.
Both Asmr1 and Amr1 share highly similar structures with CmrA Through multiple sequence alignment of Asmr1, Amr1, and CmrA, a total of 39 polymorphic amino acid sites were identified (data not shown).Therefore, we predicted the protein structures of Asmr1 and Amr1 through the SWISS-MODEL server (Waterhouse et al. 2018) using the public available protein structure of CmrA as a template.As seen in Fig. 8, both Asmr1 and Amr1 folded into a structure highly similar to CmrA.The alpha-helix, beta-sheet, and random coil were accounted for 34.4%, 3%, and 53.1% of the entire protein structure, respectively.Six and two out of 39 polymorphic sites were located in an alpha-helix or a beta-sheet while the remaining 31 sites were in random coils.Furthermore, two of six polymorphic sites in alpha-helixes were located in the fungal_trans motif.Taken together, the results suggest that structural conservation occurs in Asmr1, Amr1, and CmrA.

Discussion
In this study, we found that Cmr1 homologs were present in all examined Alternaria species and defined their gene structures with regard to consensus sequences at three intron junctions.This will eventually contribute to future prediction, identification, and annotation of Cmr1 homologs in other Alternaria species.Furthermore, identification of three completely conserved motifs suggests that the functional conservation of these motifs occurs in Cmr1 homologs.Most notably, one extra diverse fungal_trans motif was identified in Cmr1 homologs in Alternaria.However, until now, not all identified Cmr1s and their homologs in other fungi carry this motif.The motif is present in Cmr1 of C. heterostrophus, Bmr1, and StMR1, but not in Cmr1 of C. lagenaria, Pig1, BcSMR1, Zmr1, and CgCmr1 (Supplemental Fig. S8).It is not clear whether the fungal_trans motif originally existed in Cmr1 or was obtained during evolution.The phylogeny of eleven described Cmr1 homologs including those from other fungi and Asmr1, Amr1, and CmrA revealed that the six Cmr1 homologs with fungal_trans motif from six different fungal species belonging to four distinct genera clustered together to form one clade (Supplemental Fig. S8b).Given the fungal_trans motif sequences were polymorphic, therefore, we compared the motif structures of the six Cmr1 homologs.All fungal_trans motifs of the six Cmr1 homologs folded into a similar structure (Supplemental Fig. S9a, b, c).Despite the different morphologies of these six species, as well as the taxonomic status of four different genera, Cmr1 homologs still share similarities among the six species, such as the close phylogenetic relationship, the similar gene structure, that all carry the four identified protein motifs, and similar folded structures of the fungal_trans motif.We speculate that Cmr1 homologs may have undergone divergent evolution in different fungi.This is likely due to the long period of selection pressure from their ecological niches and hosts.
Due to the large differences in the morphology, identification of Alternaria spp. is usually not correct only based on one single gene.However, the reliability in Alternaria spp.identification using four to five genes is high (Woudenberg et al. 2013).We found here that Cmr1 could distinguish different groups of Alternaria spp., indicating that Cmr1 has potential application values in taxonomy.It therefore not only augments the known function of melanin, but also expands the identification methods of Alternaria spp., that is, Cmr1 might be used as a novel taxonomic marker, just like the secondary metabolite toxin (Andersen et al. 2008;Patriarca et al. 2019).
Here, we show that positive selection has promoted divergence of Cmr1 homologs in Alternaria at three amino acid sites.These three positive selection sites were found to be located in random coils, which may alter the secondary structure of Cmr1.To rapidly adapt to novel and changing environments for increasing the survival probability, biological populations need to maintain new beneficial mutations or a standing genetic variation of adaptive traits (Messer and Petrov 2013).Melanization is such an adaptive trait.It is known that transcription factors that regulate expression of effectors in plant-pathogen interactions, secreted enzymes, and secondary metabolites, are required for the adaption of plant pathogenic fungi to the host environment (van der Does and Rep 2017).Thus, adaptive changes in the coding region of Cmr1 homologs in Alternaria might be responsible for the regulation of the variation in melanin accumulation.It would allow fungal populations to adapt and specialize on new hosts, or to adapt/tolerate to stress conditions.Future functional analyses, such as site-directed mutagenesis and identification of host targets of Cmr1 could be explored to help understanding the nature of adaptive changes in Cmr1 homologs in Alternaria.
Asmr1 was confirmed to have transcriptional activation activity and be involved in the melanin biosynthesis pathway.Noticeably, the deletion of asmr1 did not completely abolish melanin production, suggesting the presence of a crosstalk between melanin biosynthesis and other signaling pathways regulating melanin.The MAP kinase (MAPK) pathways are critical for fungi to regulate responses to environmental stimuli (Zhang et al. 2021).Three conserved MAPK pathways including the cell wall integrity (CWI), invasive growth (IG), and high osmolarity growth (HOG) pathways have been well characterized in plant fungal pathogens (Zhang et al. 2021).Indeed, one IG pathway gene, AbSte7 in A. brassicicola and one CWI pathway gene, AaSLT2 in A. alternata are required for melanin biosynthesis (Yago et al. 2021;Xu et al. 2016).In addition, AaSet2 encoding a histone methyltransferase is involved in melanin accumulation (Meng et al. 2022).Histone methylation is known to be vital for transcriptional regulation and diverse biological processes.We predict that the incomplete decrease of melanin accumulation in the Δasmr1-2 mutant could be due to the regulation role from other melanin-related signaling pathways, such as the MAPK pathway.
Asmr1 also affected other biological processes of A. solani.After the deletion of asmr1, the structure of certain conidia of A. solani became abnormal and the numbers of transverse and longitudinal septa of conidia also changed.This finding is similar to results presented in previous reports about the involvement of DHN-melanin in regulating longitudinal septal number and cell wall integrity of A. alternata conidia (Yago et al. 2021;Kheder et al. 2012) and that of CmrA in controlling the spore morphology and spore formation of A. alternata (Fetzner et al. 2014).However, the deletion of amr1 in A. brassicicola has no effect on the conidial morphology (Cho et al. 2012).It suggests that the functional divergence of spore morphology occurs in Cmr1 homologs in Alternaria.In addition, we found that there were some pigments remaining in conidia, although the melanized degree of conidia and hyphae decreased after the deletion of asmr1.A recent study reported that many proteins including the CmrA transcription factor, which are required for the DHN-melanin biosynthesis pathway, are also involved in fungal toxin production via the perylene quinones biosynthesis pathway in A. alternata (Gao et al. 2022).Therefore, we speculate that Asmr1 may be a global transcription factor participating in multiple pathways.
Except for controlling melanin biosynthesis and morphogenesis, Asmr1 was required for the resistance to Congo red or SDS.It is critical for plant fungal pathogens to maintain the cell wall integrity and stability for disease establishment within the host (Zhang et al. 2021).Melanin is located within the fungal cell wall (Thomma 2003).The absence of Asmr1 might affect the cell wall integrity and stability of A. solani, further influencing the tolerance of A. solani to these cell wall stressors.Plant pathogenic fungi secrete a series of cell wall-degrading enzymes (CWDEs) during infection (Kubicek et al. 2014).Some researches indicate that the type and activity of fungal CWDEs is associated with metal ions, such as Ca 2+ , Zn 2+ , and Mg 2+ etc. (Piero and Pascholati 2000;Martino et al. 2000;Sucheta et al. 2001).For instance, Ca 2+ inhibits the activity of polygalacturonase in Fusarium oxysporum f. sp.vasinfectum infecting cotton (Piero and Pascholati 2000).One possible explanation for the observed phenotype of the decreased tolerance to Ca 2+ in the Δasmr1-2 mutant could be that the deletion of asmr1 might affect the cell wall integrity of A. solani, leading to a change of cell wall permeability, facilitating the easier attack of certain CWDEs by Ca 2+ .Therefore, this would influence the tolerance of A. solani to Ca 2+ .
Similar to the silenced mutant of CmrA in A. alternata (Fetzner et al. 2014), the Δasmr1-2 mutant of A. solani reduced tolerance to H 2 O 2 .In contrast, a deletion mutant of Amr1 in A. brassicicola remained unchanged in the sensitivity against H 2 O 2 (Cho et al. 2012).This implies the diversification of responding to oxidative stress in Cmr1 homologs in Alternaria.Plants generate toxic ROS, primarily H 2 O 2 and superoxide as a defense response against pathogens.For a plant fungal pathogen, its sensitivity to ROS likely depends on the effectiveness of its own ROS detoxification ability.In A. alternata, the detoxification of ROS is regulated by the redox-responsive transcription regulators YAP1 and SKN7 or by a nonribosomal peptide synthetase (NPS) mediated pathway (Lin et al. 2009;Yang et al. 2009;Chen et al. 2012Chen et al. , 2013)).The ability of A. alternata to detoxify ROS is also required for virulence on citrus (Lin et al. 2009;Yang et al. 2009).Hence, the lack of Asmr1 might likely affect the ability of A. solani to effectively detoxify H 2 O 2 through impairing ROS detoxification pathway, further influencing the tolerance of A. solani to external H 2 O 2 stress.
One of the interesting findings of this study is the observation on the involvement of a melanin-regulating gene in fungicide sensitivity.Asmr1 was found to be required for cellular resistance to mancozeb, chlorothalonil, and azoxystrobin.Mancozeb is shown to be an inhibitor of metal-and/or sulfhydryl-containing enzymes related to lipid metabolism, respiration, and ATP production, resulting in the inhibition of fungal spore germination (Dias et al. 2010;Gullino et al. 2010).Chlorothalonil has a multi-site mode of action that inhibits specific NAD thiol-dependent glycolytic and respiratory enzymes (Tillman et al. 1973).Azoxystrobin is a quinone outside inhibitor (QoI) fungicide with a site-specific mode of action.It inhibits mitochondrial respiration by binding to Qo site of the cytochrome b, blocking electron transfer, and thus halting ATP synthesis (Bartlett et al. 2002).Based on the mode of action of these tested fungicides, we predict that the reduced level of melanin due to the deletion of asmr1 might lead to the decreased detoxification ability of A. solani on substances affecting fungal spore germination (mancozeb/chlorothalonil), glycolysis (chlorothalonil), and respiration (azoxystrobin/chlorothalonil).Genomewide expression analysis between the Δasmr1-2 mutant and the WT under appropriate conditions are worthy to do to further characterize the function of Asmr1, and to understand its role in abiotic stress response.
Asmr1 was found to be related to the pathogenicity in A. solani.Given the involvement of the Δasmr1-2 mutant in the sensitivity to cell wall stressors, it seems likely that the ability of the Δasmr1-2 mutant to respond to cell wall stress was associated with its decreased pathogenicity.Asmr1 might regulate the expression of a set of genes encoding CWDEs that are important for decomposing and utilizing plant material during infection.In this context, it is worth noting that many glycoside hydrolases 61 family genes are highly expressed in the deletion mutant of Amr1 in A. brassicicola during pathogenesis (Cho et al 2012).Thus, A. brassicicola can quickly and efficiently utilize pectin and other cell wall components of green cabbage plant cells, causing the host cells to be more easily damaged.Another explanation for the decreased pathogenicity may be a connection between melanin and spore germination or the ROS detoxification ability.Although the deletion of asmr1 did not affect the yield of conidia, the deformed conidia in the Δasmr1-2 mutant might influence spore germination.The possibility of a positive relationship between Asmr1 and CWDEs/spore germination/ROS detoxification could be explored further through experimental investigation of expression profiles between the Δasmr1-2 mutant and WT during the infection on potato using the RNA-seq approach followed by functional assays.
While in the same genus, Amr1 can negatively regulate the pathogenicity in A. brassicicola (Cho et al. 2012).Asmr1 positively regulated the pathogenicity in A. solani.It suggests there is the pathogenicity diversification regulated by Cmr1 homologs in Alternaria.Significantly, their predicted secondary structures were almost identical, despite the sequence variation between Asmr1, Amr1, and CmrA.This is possibly due to the fact that the amino acid substitutions among them include a high proportion of the same physico-chemical groups, such as nonpolar and non-charged amino acids.Considering the conservation of the protein structure and diversification in pathogenicity, we speculate that the structural architecture is under high selective constraint in the function of Asmr1 and Amr1, which is driven by the co-evolution between pathogens A. solani and A. brassicicola and their corresponding hosts.With the increase in the genomic sequences available of Alternaria spp., this will contribute to improve our understanding the evolutionary relationships among Cmr1 homologs in Alternaria.This study provides an experimental framework to identify Cmr1 homologs in Alternaria spp.and explore their evolution and function.In combination with the identification of targets of Cmr1 homologs, this will ultimately help to make a link between the evolution of Cmr1 homologs and their functional roles in the biology of Alternaria spp.

Fig. 3 Fig. 4
Fig.3Transcriptional activation analysis of Asmr1 in yeast and verification of asmr1 deletion and complementation mutants.a Yeast Y2H GOLD strains carrying the pGBKT7 vector (negative control) and pGBKT7:Asmr1 construct grew on SD-T and SD-T-H-A + X-α-Gal medium.b Schematic diagram of replacement of the asmr1 coding region with a HygB resistance cassette.The lowest panel shows the sequencing results of the Δasmr1-2 mutant with the primer pair, HphF/HphR.c Southern blot analysis of the asmr1 deletion mutants with the hph probe amplified with the primer set Hph1349F/Hph1349R and the asmr1 probe amplified with the primer pair Asmr1-InF/Asmr1-InR. Abbreviations: WT = wildtype A. solani; Δasmr1-1, Δasmr1-2, Δasmr1-3, asmr1 deletion mutants; Δasmr1-2C, complementation mutant Δasmr1-2:asmr1

Fig. 5
Fig. 5 Time course expression of asmr1 during infection of potato by A. solani and pathogenicity assay of the asmr1 deletion mutant on potato.a Expression levels of asmr1 at 0, 2, 4, 6, 8, 10, 12, 24, 36, 48, 72, and 96 hpi of susceptible potato cv.Zaodabai infection with A. solani isolate SH0806 and normalized to A. solani actin1 expression levels.Error bars present standard deviations of three biological replicates.b Inoculated leaves of susceptible potato cv.Favorita with wild-

Fig. 6
Fig. 6 Effects of asmr1 deletion on A. solani responses to cell wall and oxygen stressors and determination of ROS contents.a Comparison of mycelial growth inhibitory rate between mutants and wildtype on PDA in the presence of indicated chemicals.Values represent means from three independent experiments with three replicates each.The significant difference in growth inhibitory rate was determined by ANOVA.*P < 0.05; ***P < 0.001 (comparison with wild-type).b Colony (upper panel) and hyphae (lower panel) of 3-day-old WT, asmr1 deletion mutant, and complementation mutant on PDA stained with NBT reagent.Abbreviations: WT, wild-type A. solani; Δasmr1-2, asmr1 deletion mutant; Δasmr1-2C, complementation mutant Δasmr1-2:asmr1 The clade A comprised 13 species including members of the Alternaria section Alternata and 2 species belonging to an unclassified Alternaria section, Alternaria alstroemeriae and Alternaria fragariae.The B clade was composed of 8 species, which were included in the Alternaria section Porri, such as A. solani.The C clade consisted of Alternaria atra and Alternaria consortialis, belonging to the Alternaria section Ulocladioides.The D clade also contained 2 species, A. brasssicicola from the Alternaria section Brassicicola and Alternaria brassicae from an unclassified Alternaria section.The E clade included 12 species, all from the Alternaria section Infectoriae.The results showed that the A, B, C, and D clades were more closely related to each other than to the E clade.In addition, the A clade was the most divergent one.Three species including Alternaria panax, Alternaria gansuensis, and Alternaria rosae formed individually three branches.Note that, phylogenetic analysis placed CmrA in A. alternata, the Cmr1 homolog in A. solani, and Amr1 in A. brassicicola in A, B, and D distinct clades, respectively.When coupled with the divergence in host range ofA.alternata, A. solani, ). Forty-two Cmr1 homologs were classified into five closely related genetic groups, clades A, B, C, D, and E supported with high bootstrap values of 95, 99, 99, 75, and 88, respectively, and three individual branches (Fig.1f).
, the last common ancestor of group I and group II of the 42 Cmr1 homologs had an estimated age of 135.77MYA (million years ago).The origin of group I including the A, B, C, and D clades as well as two individual branches, A. panax and A. gansuensis, was estimated to be 97.14MYA.The origin of group II was inferred to be 18.28 MYA, suggesting that group II emerged later than group I.In addition, among the five clades, the D clade diverged earliest (59.93 MYA), followed by the B clade (18.5 MYA), the E clade (11.2 MYA), and the C clade (10.79 MYA), and the A clade diverged latest (8.71MYA).