تعیین ساختار پروتئین و ارزیابی سطح بیان ژن Tup1 قارچ Zymoseptoria tritici

Introduction

In the fungi kingdom, dimorphism refers to the capacity to change the morphology between yeast-like growth and filamentous states (1). Growth in the yeast state refers to the production of two independent cells by mitotic divisions, budding or fission. In the filamentous state, two different models were described. In the first model, the pseudohyphal model, cells following cytokinesis become elongated. They, however, fail to separate and remain joined as chains of cells. In the second model, true hyphae with long continuous tubes are produced, then nuclei separate in the tubes. Increasing evidence has shown that pathogenic fungi use this ability to manage growth between saprophytic and pathogenic forms (2). This morphological switch occurs in response to many environmental stimuli and the role of two cAMP and MAPK signaling pathways in regulating dimorphism has been verified (1, 3, 4).

Tup1, the global transcriptional repressor is a well-known regulator that controls dimorphism and has conserved status in evolutionarily distant organisms (5). In Saccharomyces cerevisiae, the model yeast, the mechanism of action, and the structure of Tup1p were studied comprehensively. In this yeast, Tup1p associates with Ssn6 (Cyc8) and forms the Ssn6-Tup1 transcriptional co-repressor complex consisting of four monomers of Tup1 and one monomer of Ssn6. Ssn6p is a protein with tetratricopeptide repeat (TPR) motifs that mediates protein-protein interactions (6). None of these proteins have a DNA binding region and their role was played by interaction with specific DNA-binding transcription factors (7). Tup1–Ssn6 is required for the transcriptional repression of more than 300 genes under standard growth conditions (8). A wide variety of gene families contain at least nine groups: glucose repressible genes (9), mating type regulated genes (10), Nucleo DNA damage-inducible genes (11), oxygen-regulated genes (12), Osmo stress-inducible genes (13), a fatty acid regulated gene (14), flocculation-related genes (15), sporulation-related genes (16), and a meiosis-related gene (17) which can be regulated by this complex.

Several studies showed that three functional domains with diverse impose on transcription regulation were defined for Tup1 protein in budding yeast. The first domain, N- terminal is a region with 72 amino acids (aa), the C-terminal has seven WD repeats (340–713 aa), and the middle region is between two parts. The N-terminal is involved in the interaction with Ssn6 and the formation of the co-repressor complex (18). It also mediates oligomerization (19). The middle region is involved in repression as well as interaction with the N-terminal of histones. This region is located at 73–385 and amino acids 120–316, and has a high affinity for binding to H3 and H4 histones (20) and provides molecular evidence for direct interaction between the Ssn6-Tup1 complex and chromatin organization. It was pointed out that in the middle region, the Tup1 histone-binding domain corresponded to the repression domain (21). According to another study (22), there are two independent repression domains, one in 73–200 aa and the other within 288–389 aa as suggested. Some studies suggested that Tup1p has two repression domains, one in the N-terminal which contains the Ssn6-association region, and the other in the C-terminal, in a region overlapping with the first WD repeats (7). Therefore, the histone binding region overlaps with two repression domains located in N- and C-terminals. The WD40 repeats domain, located in the C-terminal region, forms a highly conserved protein-protein interaction domain (20). Each WD repeat is a 40-amino-acid motif that possesses a highly-conserved tryptophan-aspartate or WD sequence. This motif was found in proteins and it often physically interacts with other proteins and functions as a scaffold in multimeric complexes construction (23). Proteins with WD repeats are involved in a wide variety of processes, including gene repression, signal transduction, secretion, RNA splicing, and progression through the cell cycle (24). According to this description, Lamas-Maceriras et al. (25) identified four functionally important domains in Tup1 protein in S. cerevisiae. In animal pathogenic fungi, the function of Tup1p as one of the best-known regulators of dimorphism was investigated but its role in most plant pathogenic fungi is unknown (26).

Zymoseptoria tritici is one of the most important fungal pathogens in wheat (27). This fungus causes septoria tritici blotch (STB), a destructive disease of wheat (28). Z. tritici is a dimorphic pathogen with two filamentous and yeast-like states. This switch in growth pattern plays an important role in the entry of the pathogen into the infection cycle (29). The identification and functional analysis of effectors can produce precise knowledge for disease management techniques (30). Therefore, various studies were dedicated to reveal molecular aspects of dimorphism and pathogenicity in Z. tritici with an emphasis on effectors discovery. Now, the critical role of different components of two cAMP and MAPK signaling pathways in this field has been verified (31-33) and these investigations are still ongoing (34, 35).

In the present study, the sequence and structure of Tup1p of Z. tritici as an eligible candidate were studied. The survey was conducted by studying Tup1p from S. cerevisiae as the model organism and some dimorphic, plant pathogenic fungi. The presence of identified domains and the amount of their conservation were examined and the secondary structure of the protein was predicted. Finally, the best tertiary structure was drawn and presented and the expression profile of Tup1p in Z. tritici was examined in conidial and mycelial stages at different time points of the infection cycle in the greenhouse assay.

Materials and Methods

Species and Accession Numbers: Tup1p sequences of four plant pathogenic and dimorphic filamentous fungal species including Z. tritici IPO323 (Tup1, XP_003850038), Verticillium dahlia (rco-1, EGY14823), Ustilago maydis (Tup1, EAK84267), and Claviceps pupurea (rco-1, AAB63195) and three dimorphic yeast species including S. cerevisiae (Tup1p, AAA35182), Candida albicans (Tup1, AAB63195), and S. pombe (Tup1, AAB81475) were used for multiple alignment analysis. Tup1 sequence in U. maydis was provided from MIPS U. maydis Database (http://mips.gsf.de/genre/proj/ustilago/). V. dahlia Tup1 sequence was obtained from Verticillium group database (http://www.broadinstitute.org/). C. albicans and S. cerevisiae Tup1 sequences were provided from CGD (http://www.candidagenome) and SGD (http://www.yeastgenome.org/), respectively. The rest of the Tup1 sequences were obtained from the NCBI (http://www.ncbi.nlm.nih.gov/) (Table1).

Table 1- The Studied Microorganisms and the Accession Number of Their Tup1

Sequence Analysis: The physicochemical characteristics of Z. tritici Tup1p were defined using a primary sequence and with the help of the ProtParam tool (http://web.expasy.org/protparam/).

Sequence Alignment, Phylogenetic Analysis, and Domain Structure: The detailed study of the Tup1 protein sequences in these fungi was performed using ClustalW2 for multiple sequence alignments and the InterProScan sequence search tool from the European bioinformatics institute (http://www.ebi.ac.uk/) was applied for domain structure analysis. The schematic diagram of the obtained domains using Pfam was presented after the analysis by keeping proportions of each domain according to the total length of the proteins. The phylogenetic analysis of full-length protein sequences was performed using MEGA5 software with the neighbor-joining method using default setting (36).

Prediction of Secondary and Tertiary Structure: Chou and Fasman secondary structure prediction (CFFSSP) server at http://www.biogem.org/tool/chou-fasman/, SOPMA software at http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_sopma.html, GOR software at http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_gor4.html, and the program JPred 3 (http://www.compbio.dundee.ac.uk/wwwjpred/) were employed to predict the secondary structure.

I-TASSER software at http://zhanglab.ccmb.med.umich.edu/I-TASSER/, LOMETS software at http://zhanglab.ccmb.med.umich.edu/LOMETS/, MUSTER software at http://zhanglab.ccmb.med.umich.edu/MUSTER/, and PHYRE http://www.sbg.bio.ic.ac.uk/~phyre/ software were employed for tertiary structure prediction. The profile of energy minimization was computed by Swiss-PdbViewer software. The AMPAGE database at http://mordred.bioc.cam.ac.uk/~rapper/rampage.php and Swiss-PdbViewer software were used for the detection of the structural stability of the proteins by Ramachandran plot.

The Expression Level of Zt.Tup1: To study the expression profile of Zt.Tup1 in conidial and mycelium phases, IPO323 was inoculated on yeast glucose broth and maintained in shaker-incubator at 15 and 25 °^C, respectively, for five days at a speed of 125 rpm. Then, the samples were freeze-dried and used for RNA extraction. For greenhouse experiments, the susceptible wheat cultivar, Kavir, was used. Spore suspension of IPO323 was used for the inoculation of ten-day-old plants by using previously-used protocols (37). At 0, 4, 14, and 24 days after inoculation, three biological replications of the infected leaves were collected for RNA extraction. The total RNA isolation was done with RNX-Plus kit (Cinagene, Iran). Reverse transcription was performed using the RevertAid Reverse Transcriptase kit (Thermo Scientific, Germany) by the oligo-dT primer. For transcript quantification, the following primer pairs were designed, respectively:  for Tup1 and β-tubulin using the Primer3Plus online software (http://www. bioin forma tics.nl/prime rs3pl us): forward 5′-CACCTCCACAAGAGCAACAA -3′, reverse 5′- CTTGTCCTCCGCTCCTGTAG -3′ and Forward 5′-ACCATCTCCGGCGAACAT -3′, and reverse 5′-GGAATTCTCGACCAGCTGG -3′. The expression analyses of the Tup1 gene were carried out using SYBER Green qPCR Master mix (YTA, Iran) with Corbett RG-6000 Real-time PCR machine (Corbett Life Science, QIAGEN, Germany). Initially, the expression level of each sample was normalized with the β-tubulin gene and then the comparative Ct method was used to calculate the expression levels (38).

Results

Based on the results of the ProtParam tool, the atomic composition of Z. tritici Tup1p contains 9021 atoms comprising carbon (2860), hydrogen (4422), nitrogen (846), oxygen (871), and sulfur (22). Also, C₂₈₆₀H₄₄₂₂N₈₄₆O₈₇₁S₂₂ has been retrieved as Zt.Tup1 molecular formula. The retrieved primary sequence of Z. tritici Tup1p has 603 amino acids, 65.299 kD molecular weight, and 6.44 theoretical pI. A majority of amino acids are present in the sequence including glycine (11.3%), proline (9.6%), alanine (7.6%), and glutamine (7.6%). Cysteine exhibited a minimum frequency (1.2%) of residue. The sequence included about 60 negatively-charged residues (Asp + Glu) and 54 positively-charged residues (Arg + Lys). The aliphatic index and the instability index were calculated as 65.52 and 43.98, respectively. This tool graded the protein as unstable and the grand average of hydropathicity (GRAVY) was -0.579.

The result of multiple sequence alignments and domain structure analyses of Tup1 proteins from these fungi are shown in Fig. 1 (and Sup 1). The three main and functionally-identified domains of Tup1p structure in S. cerevisiae as a model yeast (18) are shown in Fig. 1A (and Sup 1): (i) N-terminal region contains 72 amino acids and is involved in complex formation with Ssn6 and formation of the oligomeric state of Tup1p; (ii) middle region (residues 73–385) contains different parts: two repression domains (R1 and R2) that have the ability to interact with other factors, two glutamine regions (Q1 and Q2) and one ST-rich domain; and (iii) C-terminal region with seven WD repeats that are involved in the interaction with other proteins and it is shown that the first WD repeat partially overlaps with R2 repression domain.

Based on S. cerevisiae, these domains include the Tup_N domain in the amino-terminal region, seven WD40 domain repeats in the carboxyl-terminal region, and a poorly-conserved middle region.

Tup1p sequence similarity in Z. tritici and S. cerevisiae was 39.66 % (Sup 2). Among the sequences examined, Tup1p sequence of C. purpurea has the highest similarity with Z. tritici Tup1p (60.76%).

The phylogenetic analysis of full-length protein sequences showed that the sequences sort into two clades, one of them includes plant pathogenic species and the other includes yeast species (Fig. 2). The topology of the clades showed the closest relationship between the V. dahlia and C. purpurea sequences in the first clade and S. cerevisiae and C. albicans in the second clade.

The secondary structural analysis of the protein was done.  The random coil is the most frequent (62.02%), followed by an extended strand (21.56%) and finally Alpha helix as being the least frequent (16.42%) (Table 2). The dominance of the coiled regions indicates the high level of conservation and stability of the protein structure (39).

Since the factor exerts its numerous effects through complex and multiple interactions, it was necessary to perform further investigations to show the presence of the characterized domains and motifs. The1–72 residues in the N-terminal region of Tup1p are required for the organization of both Tup1-Ssn6 complex and oligomeric status of Tup1 (22). In Zt.Tup1p, a partly-conserved helix structure is formed in the secondary structure at positions 1-72 (these residues are indicated with h in Sup 3.). Two glutamine-rich elements (Q1 and Q2) described in Sc.Tup1 p (25) were not present in Zt.Tup1p. Therefore, the region 1–72 of Zt.Tup1p did not conserve in the major part of the sequence. Also, the sequence covering positions 96–116 lacked the α-helical structure and the Q1-rich element. This indicated the structural difference in this part of the sequence which included a repression-associated region that has not been previously reported. Considering that Zt.Tup1 N-terminus did not have a high similarity with Sc.Tup1p, we tried to compare the sequence among filamentous fungi species used in this study (Sup 3).

Fig. 1- A) The characterized functional domains of Tup1p in Saccharomyces cerevisiae are shown in this diagram (adapted by Lamas-Maceiras et al.(25)). B) The comparison of the conserved domains of Tup1 proteins among different fungi species. The conserved structure of Tup1 proteins in S. cerevisiae (Sc.Tup1), Zymoseptoria tritici (Zt.Tup1), Candia albicans (Ca.Tup1), Saccharomyces pombe (Sp.Tup1), and Ustilago maydis (Um.Tup1), Verticillium dahlia (Vd. rco-1), and Claviceps purpurea (Cp. rco-1). The domains were depicted with the comparison to functionally characterized-domains of S. cerevisiae using InterPro (Pfam) software. All domains described for Sc.Tup1 are available in the rest and include the N-terminal Tup_N domain (green square), seven WD40 domains in the C-terminal region (red squares), and a less-conserved middle region.

Fig. 2- Dendrogram of Tup1p protein in different species, based on ClustalW alignment of full-length translated protein sequences

Table 2- Secondary Structure Elements of the Zt.Tup1p

The results indicated that the secondary structure of Um.Tup1p had a significant difference with the rest of the sequences. On the other hand, Zt.Tup1p, Cp. Rcor-1 and Vd. Rcor-1 showed a high similarity in this region. The multi-alignment analysis Tup1p homologous in these fungi revealed that they did not have R1 and R2 regions with the same sequences which were observed in Sc.Tup1p. ST region and also Q1 and Q2 sequences did not exist but Vd. Rcor-1 had one Q-rich region in 168-183 (Sup. 1).

The carboxy-terminal region of Tup1 protein is an important fragment in protein-protein interaction. Tup1p appertains to the WD family of eukaryotic proteins specified by having a conserved region with seven repeats named WD40 motifs. This repeat is a degenerate one known as a beta-transducin repeat. WD40 is a variable region in both length and composition with approximately 40 amino acids and contains a core region with a uniform length. It was suggested that proteins containing WD40-repeats have a variety of functions such as signal transduction, transcription regulation, cell cycle control, autophagy, and apoptosis (40). The common function of all WD40-repeat units is to provide a rigid scaffold for protein interactions and to form multi-protein complexes. The results showed high conservation in the amino-acid sequences of this region. Also Zt.Tup1p, Cp.Rcor-1 and Vd.Rcor-1 C-terminus comparison showed that all three sequences had a high similarity and conservation in seven WD repeats (Fig. 1 and Sup. 1). 

All the generated tertiary structures by I-TASSER, LOMETS, MUSTER of the Zt.Tup1 protein models were subjected to a series of tests for assaying their internal consistency and reliability. The result of the structural stability test of the obtained best tertiary structure by Ramachandran plot is shown in Fig. 3.

According to this test, 94.7% of the residues are located in a favored region. Also, 4.3% and 1% of the residues are located in allowed and outlier regions, respectively. Zt.Tup1p tertiary structures were depicted by the Accelrys Ds visualize software and is shown in Fig. 4.

Fig. 3- Ramachandran plot of the best tertiary structure of Zt.Tup1p

Evaluating Zt.Tup1 transcriptional level, higher expression was observed in the mycelia stage in comparison with the conidial stage (Fig. 5). In planta, the expression profile corresponded remarkably well with the early stage of infection or biotrophic phase (4 dpi) and also the necrotrophic phase (>8 dpi, Fig. 5).

Fig. 4- Zt.Tup1p tertiary structure depicted by the Accelrys DS visualizer software

Fig. 5- Transcript levels of Zt.Tup1 in conidial and mycelium phases and during infection or biotrophic phases as detected by RT-qPCR.  Error bars represent the standard error (N = 3). Normalization of reads was done with respect to the β-tubulin as a reference gene.

Discussion and Conclusion

In Z. tritici as a phytopathogenic ascomycetea, the switch from yeast-like growth to a filament phenotype plays a critical role in the infection cycle of the pathogen. This process occurs in response to different environmental stimuli and is highly controlled by complex genetic pathways (41, 42).

The Zt.Tup1p and Tup1p of other filamentous fungi studied in this paper contained most of the identified functional domains of Sc.Tup1p. Also, the complete sequence comparison showed that they had a different degree of sequence similarity. In other words, they include public structures available in the other Tup1-like eukaryotic proteins, with an N-terminal domain, a poorly-conserved middle region, and a C-terminal domain with seven WD repeats. Overall, despite the differences in the sequence, the structure of functional domains was preserved in all studied proteins. These results indicate that these functional domains have been preserved during the evolution due to their critical importance (18).

The amino acid residues of 72 N-terminal are involved in a complex formation with Ssn6. It was suggested that Ssn6 may play the role of an adapter between Tup1p and DNA-bound proteins (43). The formation of two α-helical segments in the first 72 amino acids of Zt.Tup1p was confirmed showing that the Ssn6 interaction domain is conserved. Considering that in the Zt.Tup1p sequence, Q1 region (residues 96–116) is not conserved, Q2 (residues 173–194) is not present, the α-helix predicted in this segment is not formed in the Zt.Tup1p structure. Furthermore, contrary to what was identified in S. cerevisiae, the N-terminal domain comparison between Zt.Tup1p and the other filamentous fungi showed that most of them do not have two glutamine-rich elements (Q1 and Q2). Probably, in pathogenic filamentous fungi, this sequence does not have virulence effects, but confirmation of this hypothesis needs further research.

Although the R1 repression domain was not conserved among these filamentous fungi, R2 was partly conserved. Lamas-Maceiras et al. suggested that this difference could have several effects on the structure and function of the protein since the amino acid sequence (residues 96–116) may have other roles or such changes may create a new capability in repression responses and ultimately may be ineffective in protein function (25).

The crystal structure of the Tup1p WD domain has been characterized. This structure confirmed that the seven repeated units create a propeller-like figure with seven blades in which each blade is made up of four β strands (43, 44). The study showed that WD repeats in C-terminus are available and highly conserved in Zt.Tup1p and other filamentous fungi. WD domain is responsible for transcriptional repression and protein-protein interactions. The ST sequences of the middle region (residues 365–385) between WD1 and WD2 are not present in the filamentous fungi studied in this research. Zhang et al. found that the deletion of the ST region in S. cerevisiae Tup1p did not show loss of repression. They suggested that this region may be unique to yeasts Tup1p and is not present in other higher eukaryotic homologs nor is important for repression (18).

The phylogenetic analysis of full-length protein sequences showed that the sequences are divided into two clades:  one includes plant pathogenic species and the other one includes yeast species. The topology of the clades showed the closest relationship between the V. dahlia and C. purpurea sequences in the first clade as well as S. cerevisiae and C. albicans in the second clade. Considering these results, it seems that the filamentous fungal and yeast species in evolutionary terms have two separate paths. During the course of evolution, mutations occurred for better fitness and these changes led to the separation of filamentous fungi and yeasts in phylogenetic studies.

As mentioned earlier, Tup1 protein is known for its role in dimorphism. In the present study, Zt.Tup1 expression was examined in two stages of the fungal life cycle that are associated with transformation.

The expression of the gene was evaluated in both conidial and mycelia stages as well as different post-inoculation time points during the pathogenic interaction with wheat. According to the results of the present study, the up-regulation during the mycelial stage suggests that Zt.Tup1p may play a major role during mycelium development. During the stage of infection, low expression was found in the first-time point. Subsequently, the expression was strongly induced (about 20-fold) during biotrophy at 4 dpi and then constantly decreased until 24 dpi. In addition, Zt.Tup1 was partially up-regulated during necrotrophy at 14 and 24 dpi. These findings show that Zt.Tup1p may play a key role in the pycnidium development.

Overall, the expression profiling suggested that Tup1 may play a fundamental role in Z. tritici growth and development. Contrary to Chen et al.’s study on Magnaporthe oryzae, the authors of the present study observed a higher transcript level for Tup1 in the mycelia stage (45). However, it was proved that Mo.Tup1 is essential for vegetative growth and correct hyphal branching. Even though the role of Tup1p in fungal dimorphism seems conserved, the control pathways associated with this process frequently differed among fungi (26). Roldán et al. stated that the role of Tup1p in the regulation of morphological changes in many plant pathogenic fungi is partly unknown (46).

In summary, the present study provided basic information on the structure and function of the Zt.Tupl p. It is expected that these results could be useful in the development of new pathogen management strategies and could be an initial step for further research on the role of this protein in the mechanisms of pathogenicity in Z. tritici.

Acknowledgments

This work was supported by the National Institute of Genetic Engineering and Biotechnology and Tarbiat Modares University, Tehran, Iran.