Rapid screening of an ordered fosmid library to clone multiple polyketide synthase genes of the phytopathogenic fungus Cladosporium phlei

Journal of Microbiological Methods 91 (2012) 412–419

Contents lists available at SciVerse ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

Rapid screening of an ordered fosmid library to clone multiple polyketide synthase genes of the phytopathogenic fungus Cladosporium phlei Kum-Kang So a, 1, Jung-Mi Kim b, 1, Ngoc-Luong Nguyen a, Jin-Ah Park a, Beom-Tae Kim c, Seung-Moon Park a, Ki-Jun Hwang c, Dae-Hyuk Kim a,⁎ a b c

Institute for Molecular Biology and Genetics, Chonbuk National University, Jeonju, Chonbuk 561-756, Republic of Korea Department of Bio-Environmental Chemistry, Wonkwang University, Iksan, Chonbuk 570-749, Republic of Korea Research Center of Bioactive Materials, Chonbuk National University, Jeonju, Chonbuk 561-756, Republic of Korea

a r t i c l e

i n f o

Article history: Received 25 June 2012 Received in revised form 12 September 2012 Accepted 14 September 2012 Available online 25 September 2012 Keywords: Cladosporium phlei Perylenequinone Phleichrome Polyketide synthase

a b s t r a c t In previous studies, the biological characteristics of the fungus Cladosporium phlei and its genetic manipulation by transformation were assessed to improve production of the fungal pigment, phleichrome, which is a fungal perylenequinone that plays an important role in the production of a photodynamic therapeutic agent. However, the low production of this metabolite by the wild-type strain has limited its application. Thus, we attempted to clone and characterize the genes that encode polyketide synthases (PKS), which are responsible for the synthesis of fungal pigments such as perylenequinones including phleichrome, elsinochrome and cercosporin. Thus, we performed genomic DNA PCR using 11 different combinations of degenerate primers targeting conserved domains including β-ketoacyl synthase and acyltransferase domains. Sequence comparison of the PCR amplicons revealed a high homology to known PKSs, and four different PKS genes showing a high similarity to three representative types of PKS genes were amplified. To obtain full-length PKS genes, an ordered gene library of a phleichrome-producing C. phlei strain (ATCC 36193) was constructed in a fosmid vector and 4800 clones were analyzed using a simple pyramidal arrangement system. This hierarchical clustering method combines the efficiency of PCR with enhanced specificity. Among the three representative types of PKSs, two reducing, one partially reducing, and one non-reducing PKS were identified. These genes were subsequently cloned, sequenced, and characterized. Biological characterization of these genes to determine their roles in phleichrome production is underway, with the ultimate aim of engineering this pathway to overproduce the desired substance. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Photodynamic therapy (PDT) is a treatment that couples a photosensitizer or photo-sensitizing agent with a particular type of light to achieve an effect. When photosensitizers are exposed to light of specific wavelengths, they produce a form of radical oxygen that kills nearby cancer cells or virus-infected cells. The most common and best-characterized PDT agent is Photofrin II, the only commercially available photosensitizer (Reynolds, 1997). However, many new compounds have been studied in an attempt to increase the quantum yields of singlet oxygen ( 1O2), reduce toxicity, simplify their preparation and purification, reduce aggregation tendencies, and increase their metabolism in vivo (Hudson et al., 1997). Many phytopathogenic fungi produce perylenequinone pigments, which are light-activated and non-host-selective phytotoxins (Daub

⁎ Corresponding author. Tel.: +82 63 270 3440; fax: +82 63 270 3345. E-mail address: [email protected] (D.-H. Kim). 1 There was an equal contribution to the reported research by the first two authors. 0167-7012/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mimet.2012.09.014

et al., 2005). Perylenequinones are unique because they contain a core chromophore of phenolic quinone that can absorb light energy (photosensitizers) and produce reactive oxygen species (ROS) such as the hydroxyl radical (OH •), superoxide (O2•−), hydrogen peroxide (H2O2), and singlet oxygen ( 1O2) (Daub and Ehrenshaft, 2000; Daub et al., 2005; Liao and Chung, 2008). Recently, perylenequinones have gained attention because of their therapeutic potential (Hudson and Towers, 1991). Among them, phleichrome has been intensively studied with regard to its photodynamic activity (Olivo and Chin, 2006) and use as a pharmacophore to produce various derivatives. Phleichrome, a derivative of 4,9-dihydroxyperylene-3,10-quinone, is a member of a group of fungal perylenequinones and causes the deep red pigment in the mycelium of Cladosporium phlei de Vries (Yoshihara et al., 1975), a phytopathogenic hypomycetous fungus that causes the purple eyespot disease in timothy (Phleum pratense). Although phleichrome can be chemically synthesized and further transformed into more effective agents, the initial reaction is the rate-limiting step. Alternatively, C. phlei can biologically produce phleichrome, but strain improvement and optimization of culture conditions are required. In our previous studies, we established a

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

culturing system (Lee et al., 2007) and genetic manipulation protocol (Kim et al., 2009) for C. phlei. Phleichrome is a close analog of another fungal perylenequinone, cercosporin. The only structural difference between the two substances is the two methoxy groups in positions 6 and 7 of phleichrome instead of the methylenedioxy groups in cercosporin. Although no direct studies of phleichrome biosynthesis have been performed, the fact that cercosporin and other fungal photodynamic compounds such as hypericin and elsinochrome are synthesized via the polyketide pathway using acetate and malonate subunits (Kurobane et al., 1981; Kusari et al., 2009; Liao and Chung, 2008; Okubo et al., 1975) suggests that phleichrome is also synthesized via the polyketide pathway. It is known that the synthesis of all fungal polyketides is orchestrated by polyketide synthases (PKS), multimeric enzymes that function analogously to fatty acid synthase, in which carboxylic acid units are joined in a stepwise fashion (Crawford and Townsend, 2010). Therefore, it is of interest to clone the genes encoding the polyketide synthases, identify the specific PKS gene responsible for phleichrome synthesis, and explore the possibility of metabolically engineering phleichrome production by genetic manipulation. Construction of a genome library is the first step in study of non-sequenced genomes and isolation of putative genes involved in secondary metabolite production. Considering the large size of the gene product, the presence of multiple genes, and the low expression level of fungal PKS, genomic libraries are more desirable than expression libraries to obtain full-length genes (Kroken et al., 2003; Schumann and Hertweck, 2006). However, classical screening of genomic libraries by colony hybridization requires handling of a large number of membranes, which is highly time-consuming. Therefore, various strategies have been developed to account for this screening problem. The PCR screening technique is widely used since it allows for rapid detection of clones (Kim et al., 2003; Martinez-Castro et al., 2009). However, when nonspecific amplification occurs or degenerate primers are required, the use of PCR screening is problematic (Lee et al., 2005). Therefore, we constructed an ordered C. phlei genomic library in 96-well microtiter plates to allow for fast and efficient screening of PKS genes by successive PCR using gene-specific primers. In this study, screening of an ordered genomic library followed by successive PCR amplification of specific genes was used to clone at least one PKS of each subfamily (non-reducing, partially-reducing, and reducing). We identified four PKS genes from C. phlei by genomic library screening. Two of these were reducing PKSs, one was a partially reducing PKS, and the remaining was a non-reducing PKS. In this study, it was of a great interest to obtain the non-reducing polyketide synthase, which plays an important role in the metabolic pathways of most fungal pigments.

413

polymerase and T4 polynucleotide kinase (Epicentre, WI, USA), and ligated into the pCC1FOS™ fosmid vector using the CopyControl™ Fosmid Library Production kit (Epicentre). The ligated DNA was then packaged using MaxPlax™ Lambda Packaging Extracts (Epicentre). Escherichia coli EPI300 cells were infected using packaged DNA and plated on LB agar medium containing chloramphenicol (12.5 μg /ml). 2.3. Ordered library arrangement Approximately 200–300 colonies per chloramphenicol-containing LB plate were selected for library arrangement. An individual colony was inoculated into each well of 96-well microtiter plates containing 200 μl of LB broth supplemented with chloramphenicol, and incubated for 24 h at 37 °C. Next, sterile glycerol (40% final concentration) was added to each well. An ordered library consisting of 50 96-well microtiter plates was constructed using this method and stored at − 80 °C. To create the microtiter plate mix of fosmid clones, each fosmid clone was replica-inoculated into the 96-well microtiter plates containing 200 μl of LB broth supplemented with chloramphenicol, and incubated for 48 h at 37 °C. Then, the resulting 200 μl LB broth was used to inoculate 100 ml of LB broth supplemented with chloramphenicol, and incubated overnight at 37 °C. The 96 mixed fosmid DNAs were isolated using SV Miniprep kits (Promega, WI, USA). The isolated DNA was then used as a template for amplification of the gene of interest. 2.4. Identification of PKS genes

Cladosporium phlei (ATCC 36193) was maintained as frozen agar plugs containing actively growing young hyphae in 5% DMSO solution at − 70 °C, as described previously (Lee et al., 2007). The culture conditions and methods for preparation of the primary inoculum for liquid cultures have been described previously (Yi et al., 2011). The mycelia were collected and lyophilized as described previously, and stored until use (Yi et al., 2011). All chemicals were obtained from Sigma-Aldrich Co. (MO, USA), unless otherwise specified.

C. phlei genomic DNA was used as a template for amplification of polyketide synthase genes using different sets of degenerate primer pairs. The sequence information of each primer is provided in Table 1. A standard PCR reaction (50 μl) contained approximately 200–500 ng of genomic DNA template, 5 μl of 10 × buffer (iNtRON, Korea/Takara, Japan), 2 μl dNTP mix (0.1 mM each), 3 mM MgCl2, and 2.5 U of DNA Taq Polymerase (iNtRON i-StarTaqTM/Takara Ex Taq). The reactions were performed as follows: an initial 5 min denaturation step at 95 °C, followed by 35 cycles of 94 °C for 1 min, 50 °C or 55 °C for 1 min, and 72 °C for 1 min, with a final extension step of 5 min at 72 °C. The PCR products were electrophoresed on a 0.8% agarose gel, excised, and cloned into the pGEM-T Easy Vector (Promega, WI, USA). Sequences of cloned PCR amplicons were determined using the dideoxynucleotide method with universal and synthetic oligonucleotide primers. To obtain full-length sequences of the PKS genes, we screened the CopyControl™ pCC1FOS™ fosmid library for clones that contained genomic fragments encompassing putative PKS genes. The screening was performed using PCR in a hierarchical fashion (see Results section) using gene-specific primers identified from the corresponding putative PKS genes amplified using degenerate primers. After determination of the exact clones, the corresponding fosmids were subjected to chromosome walking in both directions until neighboring genes were reached or reasonable coverage of the genomic inserts had been achieved. For Southern blot analysis, genomic DNA (10 μg) was digested with BamHI, ClaI, EcoRI, HindIII, KpnI, NotI, and PstI, blotted onto a nylon membrane (GE healthcare, UK), and hybridized with radioactive probes. The probe was labeled with α-[ 32P]-dCTP using a random labeling kit (Amersham Pharmacia Biotech, NJ, USA).

2.2. Construction of a C. phlei library using a fosmid vector

2.5. Analysis of the PKS sequence

C. phlei genomic DNA was extracted based on the method of Kim et al. (2009). To construct the C. phlei genomic library, the sheared and size-separated DNA of approximately 40 kb was end-repaired to blunt the ends by using end-repair enzyme mix including T4 DNA

Sequencing results using the degenerate primer PCR were subjected to BLASTx searches against the NCBI non-redundant protein sequence database, restricted to fungal taxa. The structures of putative PKS genes and proteins were determined using FGENESH

2. Materials and methods 2.1. Fungal strains, culture media and growth conditions

414

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

Table 1 Primers used in this study. Primer

Orientation

Domain a specificity

Sequence

Reference

KAF1 KAF2 KAR1 KAR2 LC1 LC2 KS3 KS4 XKS1 XKS2 HR-F HR-R PR-F PR-R NR-F1 NR-R1 NR-F2 NR-R2 Cppks1-F Cppks1-R Cppks2-F Cppks2-R Cppks3-F Cppks3-R Cppks4-F Cppks4-R

Forward Forward Reverse Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

KS KS AT AT KS KS KS KS KS KS KS AT KS AT KS AT KS KS KS KS KS AT KS KS KS AT

GARKSICAYGGIACIGGIAC GARGCICAYGCIACITCIAC CCAYTGIGCICCRTGICCIGARAA CCAYTGIGCICCYTGICCIGTRAA GAYCCIMGITTYTTYAAYATG GTICCIGTICCRTGCATYTC TTYGAYGCIGCITTYTTYAA RTGRTTIGGCATIGTIATICC TTYGAYGCIBCITTYTTYRA CRTTIGYICCICYDAAICCAAA ACTTCGAGGCCCACGGNAANGGNAC ATCGCGGGCCACTGNRCNCCGTY CATCGCCGTCGACGCNGCNTGYGC CATGTCCGGCCACTGNGCNCCGTR GGCACCGGCACCCARGCNGGNGA GAACTCCTCCAGGATGGGNTCNACYTG CCCTCCTACACCGTCGAYACNGCNTG TCATCTCGACGGCGTCNCCNGCGYTG AGCTGATCTCGTGAGCATCG ATGGGTCGTCTTGCCCTTGTGAC ATTTCGAGCGTGTATGGCGA CGGAGAAGACCCAAGTCAAGC TCCTGATTGACGCCTGTCT ACATCAAACCACACGAGGC GCGATCCAGTTGAAGTTGCTGC ATCTTAGGGGGCGGACTGCTCT

Amnuaykanjanasin et al. (2005) " " " Bingle et al. (1999) " Nicholson et al. (2001) " Amnuaykanjanasin et al. (2005) " This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study

a

KS and AT represent the domains of the β-ketoacyl synthase and acyltransferase, respectively.

(http://linux1.softberry.com/berry.phtml?topic=fgenesh&group= programs&subgroup=gfind) and INTERPROSCAN (http://www.ebi. ac.uk/Tools/pfa/iprscan/), respectively. The deduced PKS protein sequences were aligned with those used to design novel degenerate primer sets, and phylogenetic trees were constructed using MEGA5 (Tamura et al., 2011). Multiple alignments were generated using the ClustalX program with the following parameters: Gap opening, 20; gap extension, 0.5 for both pairwise and multiple alignments. Phylograms were constructed using Neighbor Joining and p-distance with bootstrap analysis of 2000 replicates. 3. Results 3.1. Cloning and characterization of C. phlei PKS genes All PKS genes possess three principal domains; namely, the β-ketoacyl synthase (KS), acyltransferase (AT), and acyl carrier protein (ACP) domains (Kroken et al., 2003; Schumann and Hertweck, 2006). Three types of PKS have been identified. All fungal PKSs are type I (Crawford and Townsend, 2010; Hutchinson and Fujii, 1995), which repeatedly catalyze the condensation of subunits into PK backbones. Type I can be subdivided into two subclasses; namely, non-reducing (NR) and reducing (Crawford and Townsend, 2010; Kroken et al., 2003; Schumann and Hertweck, 2006). In addition, reducing-type PKSs can be grouped into PKSs for highly reduced PK (HR-PKS) and for partially reduced PK (PR-PKS), which are also classified as methylsalicylic acid synthase (MSAS)-type since they typically play a role in MSAS synthesis (Bingle et al., 1999; Nicholson et al., 2001; Schumann and Hertweck, 2006). In addition to the 10 degenerate primers consisting of four KA (ketosynthase-acyltransferase)-series primers (Amnuaykanjanasin et al., 2005), two LC (ketosynthase-specific)-series primers (Bingle et al., 1999), two XKS (modified KS-series primer)-series primers (Amnuaykanjanasin et al., 2005), and two KS (ketosynthase)-series primers (Nicholson et al., 2001), eight new degenerate primers were included in this study (Table 1). To design the new KA primers, multiple alignments of KS and AT domains from each fungal NR-, PR-, and HR-PKS groups were generated using ClustalW (Fig. 1). The NR-PKS

group contained 15 NR-PKSs from 14 fungal species. The PR-PKS group contained 8 PR-PKSs from 7 fungal species. The HR-PKS group contained 13 HR-PKSs from 11 fungal species. Degenerate primers were designed based on the conserved NR, PR, and HR sequences of KS and AT domains using the iCODEHOP (Boyce et al., 2009) online tool (http://dbmi-icode-01.dbmi.pitt.edu/i-codehop-context/Welcome). Of the possible new degenerate primers, the primer pairs NR-F1/ NR-R1, PR-F1/PR-R1, and HR-F1/HR-R1 with the least degeneracy were selected to cover the interdomain region between KS and AT for each NR-, PR-, and HR-PKSs, respectively. In addition to the NR-, PR-, and HR-PKS primer pairs, one further pair (NR-F2/NR-R2) was designed based on the conserved sequence of the NR-PKS KS domain. Therefore, a total of eleven primer pairs were used. Among these, seven primers pairs, including four pairs from previous studies and three new pairs, were used to amplify the interdomain region. Four primer pairs, including three pairs from previous studies and one new pair, were used to amplify the KS domain. All 11 primer pairs amplified one to several DNA fragments of the expected sizes, which were subsequently cloned for sequence analysis (Table 2). The PKS fragments amplified from the KS domain ranged from 468 to 731 bp. The PKS fragments amplified from the KS and AT domains ranged from of 470 to 1200 bp. Sequence analysis of the 30 cloned PCR fragments revealed that all but two primer pairs (XKS1/XKS2, NR-F1/NR-R1) produced PCR amplicons with homology to known PKS genes. Based on sequence comparisons of the cloned putative PKS PCR amplicons, four different PKS genes were amplified. Of the seven interdomain primer pairs, three (KAF1/KAR1, KAF1/ KAR2, HR-F1/HR-R1) amplified identical PCR amplicons with homology at the amino acid level to known fungal HR-PKSs from Peltigera membranacea (91%) and Botryotinia fuckeliana (82%). In addition three interdomain pairs (KAF2/KAR1, KAF2/KAR2, PR-F1/PR-R1) amplified identical DNA fragments with homology at the amino acid levels to known fungal PR-PKSs from Pertusaria subventosa (50%), Aspergillus terreus (49%), and Penicillium griseofulvum (50%). Among the four KS domain primers, two (LC1/LC2, NR-F2/NR-R2) amplified identical PCR fragments with homology at the amino acid levels to known fungal NR-PKSs from Epicoccum sp. (92%), Peyronellaea pinodella (93%), and Ascochyta pinodes (92%). The primer pair

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

415

Fig. 1. Multiple sequence alignment of the KS and AT domains of NR- (A), PR- (B), and HR-PKSs (C). Conserved regions are boxed and amino acid residues used to design novel primers are represented at the bottom line. Arrows on the top indicate primer directions. The accession numbers for all sequences are given at left. Penicillium citrinum MlcA (gi|23574643); Monascus pilosus MkA (gi|74275561); Aspergillus terreus LovB (gi|62510842); Aspergillus clavatus NRRL 1 PKS1 (gi|121714671); Magnaporthe oryzae 70–15 syn6 (gi|53850461); Talaromyces stipitatus ATCC 10500 PKS1 (gi|242816091); M. oryzae 70–15 syn7 (gi|53850465); Coccidioides posadasii C735 delta SOWgp (gi|303319795); Arthroderma otae CBS 113480 PKS1 (gi|296813984); Gibberella moniliformis PKS10 (gi|40806915); A. clavatus NRRL 1 PKS2 (gi|121704224); Aspergillus fumigatus Af293 PKS1 (gi| 146324546); Neosartorya fischeri NRRL 181 (gi|119485939); Arthroderma benhamiae CBS 112371 PKS1 (gi|302503845); Gibberella fujikuroi pks4 (gi|8216960); T. stipitatus ATCC 10500 PKS2 (gi|242792088); Penicillium marneffei ATCC 18224 (gi|212535122); A. fumigatus Af293 PKS2 (gi|71002828); Pyrenophora tritici-repentis Pt-1C-BFP PKS1 (gi| 189193405); A. otae CBS 113480 PKS2 (gi|296815510); Xylaria sp. BCC 1067 PKS12 (gi|22164068); Verticillium albo-atrum VaMs.102 (gi|302422136); Trichophyton verrucosum HKI 0517 PKS1 (gi|302659209); Ophiostoma piceae TOPA45 (gi|145279633); Cercospora nicotianae (gi|50080729); P. tritici-repentis Pt-1C-BFP PKS2 (gi|189194635); Elsinoe fawcettii PKS1 (gi|156446811); Cochliobolus miyabeanus PKS1 (gi|48675353); Penicillium griseofulvum pks2 (gi|1888549); Aspergillus ochraceus (gi|46452226); T. verrucosum HKI 0517 PKS2 (gi|302654618); Glarea lozoyensis pks2 (gi|60686921); A. benhamiae CBS 112371 PKS2 (gi|302507164); A. terreus pksM (gi|950203); Aspergillus parasiticus pksL2 (gi|1762235).

KS3/KS4 amplified a fragment with homology at the amino acid levels to known fungal HR-PKSs from Aspergillus clavatus (82%), Glomerella graminicola (83%), and Magnaporthe grisea (84%). Therefore, initial screening using 11 pairs of degenerate primers produced several

Table 2 Summary of PCR amplifications. Primer pair (Forward/Reverse)

No.a PCR clones

PKS type

KAF1/KAR1 KAF1/KAR2 KAF2/KAR1 KAF2/KAR2 LC1/LC2 KS3/KS4 XKS1/XKS2 HR-F/HR-R PR-F/PR-R NR-F1/NR-R1 NR-F2/NR-R2

3 2 2 2 3 3 4 2 2 3 4

HR-PKS HR-PKS PR-PKS PR-PKS NR-PKS HR-PKS N.D. HR-PKS PR-PKS NR-PKS N.D.

b

(Cppks4) (Cppks4)

c

partial DNA fragments; namely, one PR-PKS, one NR-PKS, and two HR-PKSs. Southern blot analysis was used to determine the PKS gene copy number using a cloned PCR amplicon probe representing four different PKSs. Genomic DNA was digested with various restriction enzymes including BamHI, ClaI, KpnI, NotI, and PstI, which did not cut the probe. Southern analysis using the PR-PKS specific probe showed only one hybridized band and no additional bands when the hybridization was performed under low stringency at 55 °C (Fig. 2). However, stronger intensity of the hybridizing band in the BamHI lane was of interest because it may suggest the presence of an additional copy. In addition, application of the NR-PKS specific probe resulted in generation of more than two hybridized bands (data not shown). These results indicate that more PKS genes are present in the C. phlei genome.

(Cppks3)

3.2. Hierarchical clustering methods to clone full-length PKS genes (Cppks4)

N.D. No significant homology to known genes. a Number of PCR amplicons of expected size that were cloned and sequenced. b Type of PKS genes based on sequence homology. c Indicates the specific HR-PKS gene out of two different HR-PKS genes.

From the CopyControl™ pCC1FOS™ fosmid library, a total of 4800 clones were cultured and preserved using 50 96-well microtiter plates. To validate the fosmid library, 20 colonies harboring the recombinant fosmid vector were randomly selected. Next, the vector DNA was extracted and subjected to restriction analysis. Restriction analysis using the BamHI and NotI enzymes, which are included in

416

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

3.3. Four PKS genes with complete ORFs

Fig. 2. Southern blot analysis of C. phlei using the 717 bp-PCR amplicon of the Cppks2 gene. The enzymes used to digest the DNA samples are indicated on the top of the lanes. Numbers on the left indicate the size in kb. B; BamHI, C; ClaI, E; EcoRI, K; KpnI, P; PstI.

the multiple cloning site and can be used to separate the vector and insert showed different banding patterns among the clones ranging from 36 to 42 kb in size, similar to the average insert size using the CopyControl™ pCC1FOS™ fosmid vector. These results indicated that our collection of clones from the fosmid genomic library represents random clones with the expected insert size. Considering the average genome size of filamentous fungi, and the insert size of our fosmid genomic library, our collection of fosmid clones represented a 4.8-fold coverage of the genome based on approximate genome size of 40.0 Mb (Deacon, 2006). Therefore, we used these collections to screen for PKS genes in C. phlei. The contents of each of the 50 96-well microtiter plates were inoculated into a 100 ml liquid culture, incubated overnight, and the mixed fosmid clones were extracted. The resulting fosmid mixture, which consisted of a fosmid from each clone in the 96-well microtiter plate, was labeled #1–#50, and then used as a template for amplification of the four PKSs using the corresponding gene-specific PCR primers (Cppks1-F/Cppks1-R, Cppks2-F/Cppks2-R, Cppks3-F/Cppks3-R, and Cppks4-F/Cppks4-R in Table 1). For NR-PKSs, six fosmid mixtures (#7, #12, #18, #26, #33, and #37) generated 722-bp PCR amplicons. For PR-PKSs, two (#36, and #49) generated 717-bp PCR amplicons. For one of the two HR-PKSs, two fosmid mixtures (#4, and #50) generated 731-bp PCR amplicons. In addition, two fosmid mixtures (#3, and #14) generated 803-bp PCR amplicons for the other HR-PKS. Each PCR amplicon was cloned into the pGEM-T Easy vector and sequenced. Sequence analysis revealed no differences among those PCR amplicons amplified using the same gene-specific PCR primers, suggesting that the gene-specific primers specifically amplified the corresponding gene and not homologous PKS genes. Moreover, sequence comparison revealed that the cloned PCR amplicon sequence obtained using gene-specific primers was identical to that obtained using degenerate primers, which suggests that a clone harboring the corresponding PKS gene was present in the mixture. The 96-well mixture was then further divided into twelve subgroups representative of the eight clones. These subgroups were screened by PCR to identify the fosmid mixture harboring the corresponding PKS genes. Thereafter, each positive subset was further screened by PCR to identify a specific clone harboring the corresponding PKS gene. After screening, the #7-B-3 and #36-B-10 clones were found to contain the NR-PKS and PR-PKS genes, respectively. In addition, the #4-E-6, and #3-F-8 clones contained HR-PKS genes. These four clones were then further evaluated by sequencing.

To obtain full-length sequence information for the corresponding PKS gene, a fosmid clone representing each PKS gene was selected and subjected to sequencing in both directions using synthetic primers until regions with significant homology to known genes or domains were reached. A total of 12,841, 11,239, 19,312, and 16,469 nt of the NR-PKS, PR-PKS, and two HR-PKS genes, respectively, were sequenced. BLAST searches of the sequenced NR-PKS, PR-PKS, and two HR-PKS genes revealed high similarity to the corresponding fungal PKS genes from Cochliobolus sativus (GenBank ID: HQ830033.1; 84% identity; E value 0), Penicillium patulum (GenBank ID: X55776.1; 72% identity; E value 2e−139), A. clavatus (GenBank ID: XM001274945.1; 68% identity, E value 9e−46), and Thielavia terrestris (GenBank ID: XM003649194.1; 79% identity, E value 4e−42), respectively. Therefore, we used Cppks1, Cppks2, Cppks3, and Cppks4 to refer to NR-PKS, PR-PKS, and the two HR-PKS genes in C. phlei, respectively. The genes neighboring the N-terminus of Cppks2, Cppks3, and Cppks4 are thought to be ones encoding a NADB-Rossmann superfamily gene, a putative enoyl reductase, and a FAD/FMN containing isoamyl alcohol oxidase, respectively. The genes neighboring the C-terminus of Cppks2, Cppks3, and Cppks4 are thought to be the ones encoding a carboxylesterase, a gene that is not determined due to short sequence information, and an MFS drug transporter, respectively. However, no neighboring gene with significant homology to Cppks1 was identified by BLAST search, even after analysis of 3.1 and 2.9 kb of sequence information from the putative start and stop codons, respectively. Next, we analyzed the lengths of Cppks1, Cppks2, Cppks3, and Cppks4 ORFs, as well as the number and size of the intron(s). Putative introns and exons were analyzed using the FGENESH eukaryotic gene prediction online software (http://linux1.softberry.com/berry.phtml? topic=fgenesh&group=programs&subgroup=gfind) using the model organism that showed the highest homology; i.e., the Pyrenophora, Penicillium, Aspergillus, and Botryotinia models for Cppks1, Cppks2, Cppks3, and Cppks4, respectively. In addition, the presence of introns was confirmed by frame analysis, the similarities of each frame to known PKS types, and the presence of signals similar to the canonical 5′ (GTRRGT) and 3′ (YAG) splicing signals (Bruchez et al., 1993). Sizes of the putative introns within these PKS genes ranged from 36 to 117 bp (Table 3). The sequences proximal to the putative start codons of Cppks1, Cppks2, Cppks3, and Cppks4 were (5′-CATCAUGTCT-3′), (5′-CACAAUGGCG-3′), (5′-GGTCAUGTCG-3′), and (5′-AAACAUGGGC-3′), respectively, which closely resemble the optimal eukaryotic initiation codons 5′-G/ANNAUGN-3′ (Kozak, 1983). ORF lengths and the putative Table 3 Characteristics of PKS gene-splicing regions of C. phlei. Gene

Intron

Size

5′ splice site

3′ splice site

Cppks2 Cppks3

1 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8

53 59 54 50 117 51 55 83 39 36 51 56 51 68 56 49 51 53 52

GTGAGT GTAAGC GTACGT GTATGT GTAAGT GTGAGT GTTCGT GTAAGT GTGAGT GTAGGT GTATGT GTAAGA GTACGA GTGAGC GTAAGT GTCAGT GTGAGA GTATGA GTATGT

CAG CAG TAG TAG CAG CAG CAG CAG CAG TAG TAG CAG TAG CAG TAG CAG AAG CAG CAG

Cppks4

Note that the Cppks1 gene is predicted to be intronless.

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

number of introns were determined for Cppks1 (6525 bp and 0), Cppks2 (5432 bp and 1), Cppks3 (12,337 bp and 10), and Cppks4 (7606 bp and 8). The deduced Cppks1 sequence contained no introns, and the protein product CpPKS1 consisted of 2174 amino acids, with an estimated molecular mass of 235.1 kDa and a pI of 6.02. (GenBank ID: JX129223). Alignment of the deduced amino acid sequence showed the cloned CpPKS1 to be related to known fungal NR-PKS proteins from Mycosphaerella graminicola IPO323 (75% identity), Elsinoe fawcettii (67%) and C. sativus (65%). The deduced Cppks2 protein product CpPKS2 consisted of 1792 amino acids, with an estimated molecular mass of 192.4 kDa and a pI of 5.26 (GenBank ID: JX129224). Alignment of the deduced amino acid sequence showed that CpPKS2 is related to known fungal PR-PKS proteins from A. terreus (57%), Byssochlamys nivea (57%) and A. clavatus (56%). The deduced Cppks3 protein product CpPKS3 consisted of 3913 amino acids, with an estimated molecular mass of 427.8 kDa and a pI of 5.30. (GenBank ID: JX129225). Alignment of the deduced amino acid sequence showed that CpPKS3 is related to known fungal HR-PKS proteins from Coccidioides posadasii str. Silveira (40%), Arthroderma gypseum CBS 118893 (38%) and A. clavatus NRRL 1 (38%). The deduced Cppks4 protein product CpPKS4 consisted of 2389 amino acids, with an estimated molecular mass of 258.3 kDa and a pI of 5.70 (GenBank ID: JX129226). Alignment of the deduced amino acid sequence showed that CpPKS4 is related to known fungal HR-PKS proteins from B. fuckeliana (49%), P. membranacea (48%) and Aspergillus oryzae RIB40 (44%). In addition, the deduced amino acid sequence of CpPKS4 showed the 31% identity to the sequence of CpPKS3, the other HR-PKS protein from C. phlei. Domain scanning using full-length deduced amino acid sequences revealed that each PKS gene had a domain structure characteristic of its corresponding PKS type (Fig. 3). For example, CpPKS1 had duplicated ACP domains followed by thioesterase (TE) domain at the C-terminal region. In addition, CpPKS3 and CpPKS4 had additional reducing domains such as DH, KR or ER domains, characteristic of HR-PKSs. 3.4. Comparison of PKS sequences The deduced amino acid sequences of four PKS genes from C. phlei were compared with other known PKS genes using ClustalX (Tamura et al., 2011). A total of 39 amino acid sequences, including 15 NR-PKS, 8 PR-PKS, and 13 HR-PKS genes from other fungi, were included in

417

this multiple alignment. In addition, three reference fatty acid synthase (FAS) genes were included as an out-group (Kroken et al., 2003). The sequence alignment was then phylogenetically analyzed using the Neighbor Joining method. Phylograms using full-length sequences, as well as AT and KS domains, of aligned PKS genes revealed similar clustering (Fig. 4). We identified three major clades for fungal PKSs depending on the reducing conditions of their polyketide product. The NR-PKS (CpPKS1) from C. phlei clustered with other NR-PKSs, within which CpPKS1 formed a separate subclade including the NR-PKS genes for elsinochrome, another member of the perylenequinones, from E. fawcettii, melanin from Cochliobolus miyaneanus, and a yellow pigment from Pyrenophora tritici-repentis. The PR-PKS (CpPKS2) from C. phlei grouped with other PR-PKSs that play roles in MSA synthesis, such as patulin. Two HR-PKSs (CpPKS3 and CpPKS4) from C. phlei clustered with known fungal HR-PKSs. Domain analysis indicated that CpPKS3 had lost the ER domain. This type of PKS is responsible for the synthesis of PKs, which are predicted to either lack reduced alkyl groups or to contain an alkyl group that is reduced by the product of an external ER domain-containing gene (Kroken et al., 2003). CpPKS4 was unique and formed a separate subclade from CpPKS3. Bootstrap analysis supported all the major clades containing PKS genes from C. phlei. 4. Discussion C. phlei can be applied to sustainably produce phleichrome, a derivative of 4,9-dihydroxyperylene-3,10-quinone, which is proven to be chemically modified for an improved PDT agent (Lee et al., 2007). In addition, genetic manipulation techniques such as transformation have been established for this fungus (Kim et al., 2009). Therefore, cloning of the novel PKS genes in this fungus is important for identification of the gene responsible for phleichrome production and construction of a genetically engineered strain for high-level phleichrome expression. PCR amplification using degenerate primers has been applied to clone PKS genes from fungi (Amnuaykanjanasin et al., 2005; Blanco et al., 1993). Our screening using domain specific primers, such as KS-, XKS-, and LC-series, amplified two different types of PKS genes (Cppks1, Cppks4); namely, NR-PKS and HR-PKS genes. In addition, the previous interdomain specific primers, such as KA-series, amplified two types of PKS genes (Cppks2, Cppks3); namely, a PR-PKS and an HR-PKS gene. Our new type-specific PCR primers amplified three different types of PKSs (Cppks1, Cppks2, and Cppks3). However, one of the two NR-PKS primer pairs did not amplify the PKS gene, and

Fig. 3. Schematic organization of four PKS genes in C. phlei. Thick lines represent the ORF region of each PKS, in which the position and appropriate size of putative introns are represented by short vertical lines. All PKS-conserved domains were detected using the INTERPROSCAN program and are shown at their corresponding positions in the ORF. Flanking regions other than the ORF of PKS and neighboring genes are shown as thin lines and open boxes, respectively. The size of the DNA sequence is shown at the top line.

418

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

Fig. 4. Phylogram of fungal PKSs inferred based on Neighbor Joining analysis of the full-length deduced amino acid sequences. Vertical bars indicate major clades, each of which represents the reducing state of polyketides. Numbers below the branches indicate the percentage bootstrap support for each clade. The accession numbers for all sequences are given in the legend of Fig. 1. Three FAS genes were included as an out-group: Homo sapiens FAS (gi|1049053); Gallus gallus FAS (gi|1345958); Bombyx mori FAS (gi|2058460). Bar represents 0.05 substitutions per amino acid position.

one of the HR-PKS primer pairs amplified only a single HR-PKS gene (Cppks4) under our PCR conditions. Screening of an ordered library using gene specific primers can be used to obtain the sequence of a full-length unknown gene. Compared to whole genomic DNA, use of a fosmid mixture as PCR template, prepared from a mixed culture of a few different fosmids (in this case 96 different recombinant fosmids), generated specific target DNA amplification products. In addition, the fosmid mixture can be reused for PCR amplification to screen for other unknown genes. Genome coverage determines the quality of the genomic library which is supposed to represent the whole genome of the target organism. The 4800 fosmid clone library is likely to sufficiently represent the C. phlei genome because more than one specific fosmid clone was obtained

for each target. Thus, using PCR amplification targeted towards a specific clone, it is possible to identify more PKS genes by PCR using the fosmid mixtures as a template combined with degenerate primers. This method can also be applied to cloning of unidentified genes of interest from C. phlei using a similar approach. We identified four PKS genes; namely, Cppks1, Cppks2, Cppks3, and Cppks4, which represent the three different types of PKS genes in C. phlei. Among these, Cppks1 is a non-reducing PKS-type that does not contain introns, while Cppks2 contained only one intron. However, the two HR-PKS genes (Cppks3, and Cppks4) contained multiple (10 and 8) introns. Considering the comparable size of the intronless Cppks1 gene to the HR-PKS genes, it is possible that these genes split from a common HR-PKS-type ancestor.

K.-K. So et al. / Journal of Microbiological Methods 91 (2012) 412–419

On average, four reducing PKS genes per fungus were detected using KA series primers (Amnuaykanjanasin et al., 2005). However, we amplified only two reducing PKS genes. In addition, Southern blot analysis suggested that additional PKS genes were present in the genome. Therefore, given that a fungal species has as many as 25 PKS genes (Kroken et al., 2003), it is possible that there are more unidentified PKS genes in this fungus. Considering the various functions of polyketides as toxins, pigments, and signaling molecules for growth, hyphal fusion, aerial hyphae formation, conidiation, and sexual reproduction (Chiang et al., 2010), the presence of multiple PKS genes in C. phlei suggests that they play an important role in the biology of this fungus. Therefore, cloning and characterization of the PKS genes will increase our understanding of the diversity of polyketides and their functions. Phylogenic analysis confirmed that the Cppks1 gene is most closely related to the NR-PKS genes involved in elsinochromes biosynthesis, as well as red or orange perylenequinone pigments from many phytopathogenic Elsinoe spp. (Liao and Chung, 2008), which are structurally similar to the phleichrome. Accordingly, the Cppks1 gene in this fungus might be responsible for phleichrome synthesis. However, production of various pigments (other than the reddish-purple phleichrome) has been reported in this fungus, such as a dark-gray melanin-type pigment (Lee et al., 2007). Therefore, the Cppks1 gene (or other PKS genes) may play a role in the synthesis of these fungal pigments. To our knowledge, this is the first report of cloning of C. phlei PKS genes. We are in the process of identifying the PKS gene responsible for phleichrome synthesis. Acknowledgments This research was supported by the Bio-industry Technology Development Program, Ministry for Food, Agriculture, Forestry and Fisheries, Republic of Korea and in part by the KOSEF grant (R11-2008-062-02002-0). This work was also supported by research funds of Chonbuk National University in 2012. We thank the Research Center of Bioactive Materials and the Institute for Molecular Biology and Genetics at Chonbuk National University for kindly providing the facilities for this research. References Amnuaykanjanasin, A., Punya, J., Paungmoung, P., Rungrod, A., Tachaleat, A., Pongpattanakitshote, S., Cheevadhanarak, S., Tanticharoen, M., 2005. Diversity of type I polyketide synthase genes in the wood-decay fungus Xylaria sp. BCC 1067. FEMS Microbiol. Lett. 251, 125–136. Bingle, L.E., Simpson, T.J., Lazarus, C.M., 1999. Ketosynthase domain probes identify two subclasses of fungal polyketide synthase genes. Fungal Genet. Biol. 26, 209–223. Blanco, G., Brian, P., Pereda, A., Mendez, C., Salas, J.A., Chater, K.F., 1993. Hybridization and DNA-sequence analyses suggest an early evolutionary divergence of related biosynthetic gene sets encoding polyketide antibiotics and spore pigments in Streptomyces spp. Gene 130, 107–116. Boyce, R., Chilana, P., Rose, T.M., 2009. iCODEHOP: a new interactive program for designing COnsensus-DEgenerate Hybrid Oligonucleotide Primers from multiply aligned protein sequences. Nucleic Acids Res. 37, W222–W228. Bruchez, J.J.P., Eberle, J., Russo, V.E.A., 1993. Regulatory sequences in the transcription of Neurospora crassa genes: CAAT box, TATA box, Introns, Poly(A) tail formation sequences. Fungal Genet. Newsl. 40, 89–97.

419

Chiang, Y.M., Oakley, B.R., Keller, N.P., Wang, C.C., 2010. Unraveling polyketide synthesis in members of the genus Aspergillus. Appl. Microbiol. Biotechnol. 86, 1719–1736. Crawford, J.M., Townsend, C.A., 2010. New insights into the formation of fungal aromatic polyketides. Nat. Rev. Microbiol. 8, 879–889. Daub, M.E., Ehrenshaft, M., 2000. The photoactivated cercospora toxin cercosporin: Contributions to plant disease and fundamental biology. Annu. Rev. Phytopathol. 38, 461–490. Daub, M.E., Herrero, S., Chung, K.R., 2005. Photoactivated perylenequinone toxins in fungal pathogenesis of plants. FEMS Microbiol. Lett. 252, 197–206. Deacon, J., 2006. Fungal Biology, fourth ed. Blackwell Publishing, Malden. Hudson, J.B., Towers, G.H., 1991. Therapeutic potential of plant photosensitizers. Pharmacol. Ther. 49, 181–222. Hudson, J.B., Imperial, V., Haugland, R.P., Diwu, Z., 1997. Antiviral activities of photoactive perylenequinones. Photochem. Photobiol. 65, 352–354. Hutchinson, C.R., Fujii, I., 1995. Polyketide synthase gene manipulation — a structure– function approach in engineering novel antibiotics. Annu. Rev. Microbiol. 49, 201–238. Kim, C.G., Fujiyama, A., Saitou, N., 2003. Construction of a gorilla fosmid library and its PCR screening system. Genomics 82, 571–574. Kim, J.A., Kim, J.M., Kim, H.G., Kim, B.T., Hwang, K.J., Park, S.M., Yang, M.S., Kim, D.H., 2009. Protoplast-mediated transformation of the filamentous fungus Cladosporium phlei: evidence of tandem repeats of the integrated transforming vector. Plant Pathol. J. 25, 179–183. Kozak, M., 1983. Comparison of initiation of protein synthesis in procaryotes, eucaryotes, and organelles. Microbiol. Rev. 47, 1–45. Kroken, S., Glass, N.L., Taylor, J.W., Yoder, O.C., Turgeon, B.G., 2003. Phylogenomic analysis of type I polyketide synthase genes in pathogenic and saprobic ascomycetes. Proc. Natl. Acad. Sci. U. S. A. 100, 15670–15675. Kurobane, I., Vining, L.C., McInnes, A.G., Smith, D.G., Walter, J.A., 1981. Biosynthesis of elsinochromes C and D. Pattern of acetate incorporation determined by 13C and 2 H nmr1. Can. J. Chem. 59, 422–430. Kusari, S., Zuhlke, S., Kosuth, J., Cellarova, E., Spiteller, M., 2009. Light-independent metabolomics of endophytic Thielavia subthermophila provides insight into microbial hypericin biosynthesis. J. Nat. Prod. 72, 1825–1835. Lee, K.M., Lee, C.K., Choi, S.U., Park, H.R., Kitani, S., Nihira, T., Hwang, Y.I., 2005. Cloning and in vivo functional analysis by disruption of a gene encoding the gammabutyrolactone autoregulator receptor from Streptomyces natalensis. Arch. Microbiol. 184, 249–257. Lee, J.K., Kim, B.T., Kim, J.A., Chung, H.J., Park, S.M., Yang, M.S., Hwang, K.J., Kim, D.H., 2007. Cultural characteristics and extraction of the fungal pigment phleichrome from the phytopathogenic fungus Cladosporium phlei. Biotechnol. Bioprocess Eng. 12, 508–515. Liao, H.L., Chung, K.R., 2008. Genetic dissection defines the roles of elsinochrome phytotoxin for fungal pathogenesis and conidiation of the citrus pathogen Elsinoe fawcettii. Mol. Plant Microbe Interact. 21, 469–479. Martinez-Castro, M., Solera, E., Martin, J.F., Barreiro, C., 2009. Efficient pyramidal arrangement of an ordered cosmid library: rapid screening of genes of the tacrolimusproducer Streptomyces sp. ATCC 55098. J. Microbiol. Methods 78, 150–154. Nicholson, T.P., Rudd, B.A., Dawson, M., Lazarus, C.M., Simpson, T.J., Cox, R.J., 2001. Design and utility of oligonucleotide gene probes for fungal polyketide synthases. Chem. Biol. 8, 157–178. Okubo, A., Yamazaki, S., Fuwa, K., 1975. Biosynthesis of cercosporin. Agric. Biol. Chem. 39, 1173–1175. Olivo, M., Chin, W., 2006. Perylenequinones in photodynamic therapy: cellular versus vascular response. J. Environ. Pathol. Toxicol. Oncol. 25, 223–237. Reynolds, T., 1997. Photodynamic therapy expands its horizons. J. Natl. Cancer Inst. 89, 112–114. Schumann, J., Hertweck, C., 2006. Advances in cloning, functional analysis and heterologous expression of fungal polyketide synthase genes. J. Biotechnol. 124, 690–703. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. Yi, M.H., Kim, J.A., Kim, J.M., Park, J.A., Kim, B.T., Park, S.M., Yang, M.S., Hwang, K.J., Kim, D.H., 2011. Characterization of a mutant strain of a filamentous fungus Cladosporium phlei for the mass production of the secondary metabolite phleichrome. J. Microbiol. 49, 680–683. Yoshihara, T., Shimanuki, T., Araki, T., Sakamura, S., 1975. Phleichrome, a new phytotoxic compound produced by Cladosporium phlei. Agric. Biol. Chem. 39, 1683–1684.