具有产生安莎类抗生素潜能的放线菌的分子筛选

                   作者：武临专 张会图 韩锋 高群杰 孙桂芝 王以光 

【摘要】  目的 建立一种高通量筛选具有产生安莎类抗生素潜能的放线菌的分子筛选方法。原理 3?氨基?5?羟基?苯甲酸(AHBA)是安莎类抗生素生物合成的前体。在通过氨基莽草酸途径生物合成AHBA的过程中，AHBA合酶是催化5?氨基?3?脱氢?5?脱氧莽草酸形成AHBA的一个特异性关键酶。AHBA合酶基因的存在可以作为放线菌菌株具有合成安莎类抗生素能力的一个必要而非充分的条件。AHBA合酶基因高度保守，通过PCR方法可以建立一种针对AHBA合酶基因存在与否的、具有安莎类抗生素产生潜能的放线菌菌株高通量筛选方法。方法 根据已知AHBA合酶序列的相似性，设计了针对AHBA合酶基因中高度保守的一段755bp片段大小的PCR筛选寡核苷酸引物，用于从土壤中分离的未知放线菌高通量筛选。结论 从1900株土壤分离的未知放线菌中筛选获得33株AHBA合酶基因阳性菌株，即具有安莎类抗生素产生潜能的放线菌菌株。

【关键词】  安莎霉素类； AHBA合酶基因； 聚合酶链反应； 分子筛选； 放线菌

Ansamycins are characterized by a benzenic or naphthalenic quinoid based chromophore with an ansa?bridge linking one end of the chromophore with the other end by an amide linkage (Fig.1). The chromophore is generated from a mC7N unit derived from 3?amino?5?hydroxybenzoic acid (AHBA) via aminoshikimate pathway; any remaining part of the chromophore, and the ansa? bridge are polyketide derived.Fig.1    All ansamycins use AHBA as the starter unit for their biosynthesis

    Ansamycins showed a variety of biological activities. Rifampicin, a semisynthetic derivative of rifamycin produced by Amycolatopsis mediterranei, is one of the most effective clinical treatments of mycobacterial infection. 17?allylamino?demethoxygeldanamycin (17?AAG), a semisynthetic derivative of geldanamycin produced by Strepromyces hygroscopicus, is currently under clinical trials for tumor treatment［1,2］. Ansamycins are also potential anti?viral agents. For example, rifamycin could block the RNA?dependent DNA polymerases of retroviruses, geldanamycin could inhibit the replication of Herpes Simplex Virus type 1 (HSV?1) in vitro［3］.

    Screening of novel ansamycins (and their producers) is of great importance in discovering promising drug?candidates or lead compounds with clinical potentials. During the past ten years, several ansamycins produced by streptomyces strains were discovered such as thiazinotrienomycins and cytotrienin A, both of them showed antitumor activities［4～7］. Some novel ansamycins were also obtained successfully by means of chemical and/or biological modifications of known ansamycins［8］.

    Up to now, most natural ansamycins are discovered by activity?based screening of microbial (especially, actinomycetes) fermentation and extraction of products. That is, a screening model was established first, based on detecting (assaying) a certain kind of biological activity, then the compound with such an activity was isolated and purified for further structural elucidation. In spite of the success of activity?based screening for ansamycins, their obvious disadvantages include: ① the limitations of the fermentation condition and the minimal bioactivity detection lead to missing the potential ansamycin in the initial screening; ② the structure determination complexity of numerous related chemistry resulted in blindfold laborious efforts for aimed screening of ansamycin; ③ the difficulties in setting?up the high?throughput screening program.

    We are interested in establishing a screening model that is ansamycin?targeted. From the biosynthetic viewpoint, AHBA, synthesized by so called aminoshikimate pathway, is the starter unit for all ansamycins. The aminoshikimate pathway, clarified by Floss HG et al［9,10］, involved a sequential biochemical reactions parallel (but different) in partial to the well?known shikimate pathway for aromatic amino acid biosynthesis. Comparison of the aminoshikimate pathway and the shikimate pathway revealed a terminal aromatization reaction unique to the aminoshikimate pathway that convert the 5?amino?3?dehydro?5?deoxyshikimate into AHBA, catalyzed by the AHBA synthase. Using AHBA synthase gene as a probe, the biosynthetic gene clusters of many AHBA?derived natural compounds have been identified［11～13］. With the accumulation of more and more sequence data of AHBA synthases, it is safe to conclude that all AHBA synthases are highly similar each other. We believe that the existence of AHBA within actinomycetes strain could be used as an indicator for ansamycin biosynthesis. As AHBA is a small aromatic molecule and an intermediate specifically involved in secondary metabolism, its simple and easy detection remains an unsolved problem. As an alternative, the highly conserved AHBA synthase gene could be easily detected by, e.g., PCR, and it could be used as a gene?marker of the biosynthetic potential of AHBA. So the AHBA synthase gene could be used in place of AHBA for the preliminary screening of potential ansamycin producers.

    In this paper, an ansamycin?targeted molecular screening model was successfully developed, which can be used for screening systemically of actimomycetes strains with ansamycin production potentials by PCR in a high?throughput manner. Positive actimomycetes strains from this screening model could be used for further exploring the real ansamycin producers.

    1  Materials and methods

    1.1  Strains

    Seven strains producing antibiotics other than ansamycins, Streptomyces griseolus CPHCC200279 for anisomycin, Actinoplanes luojiashanensis CPHCC200063 for dihydroxy anthraquinone, Streptomyces violaceus CPHCC200533 for violarine, Streptomyces viridochromogenes CPHCC200311 for xanthomycin, Streptomyces sp. CPHCC200014 for pluramycinA, Streptomyces spiramyceticus 200108 for spiramycin, and Saccharopolyspora erythraea CPHCC200101 for erythromycin. Four ansamycin?producing strains, Amycolatopsis mediterranei var. kanglensis CPHCC200092 for kanglemycin A［14］, Amycolatopsis mediterranei CPHCC200067 for rifamycin, Nocardia sp. CPHCC200096 for macbecin［15］, Streptomyces hygroscopicus CPHCC200178 for geldanamycin［11］. 1900 random?selected, unspecified actinomycetes strains from soil isolates, each as liquid fermentation culture. All the strains were provided kindly by Center for Culture and Collection of Pharmaceutical Microorganisms (CPHCC, China).

    1.2  Reagents

    TaKaRa LA TaqTM with GC Buffer kit (for all PCR amplifications), agarose gel DNA fragment recovery kit ver. 2.0, restriction enzymes, and pMD19?T vector were all purchased from TaKaRa Biotechnology (Dalian) Co., Ltd.. Polyvinylpyrrolidone?4000 (PVP?4000), a product of Amresco purchased from Shanghai Sangon Biological Engineering Technology and Service Co., Ltd.. DNA molecular weight marker III (200bp, 500bp,800bp,1200bp, 2000bp,3000bp, and 4000bp) was purchased from Beijing Dingguo Biotechnology Co. Ltd..

    1.3  Extraction of genomic DNA from actinomycetes strains

    A modified procedure for genomic DNA extraction from liquid culture of actinomycetes strains in eppendorf tubes was developed from reference［16］. Briefly, for each actinomycetes strain, 70μl of liquid culture was drawn into an 1.5ml eppendorf tube, centrifuged at 10,000r/min for 1min, the supernant was discarded. Then 1ml Washing Solution (Tris?HCl 50mmol/L, EDTA 25mmol/L, PVP?4000 0.1%, pH7.7) was added into the tube, vortexed for 5s, centrifuged at 10,000r/min for 2min, discarded the supernant. 150μl Lysis Buffer (Tris?HCl 50mmol/L, EDTA 25mmol/L, SDS 3%, PVP?4000 1.2%, pH8.0) was added into the pellet and mixed well by vortexing for 5s. It was then put into a microwave oven and heated to just nearly boiling. 450μl Extraction Buffer (Tris?HCl 10mmol/L, EDTA 1mmol/L, NaAc 0.3mol/L, PVP?4,000 1.2%, pH8.0) was added and mixed up and down by pipeting for a few times, then an equal volume of phenol∶chloroform∶isoamyl alcohol (25∶24∶1) was added and mixed up and down for 3～5min. After centrifuging at 10,000r/min for 2min, the supernant was pipeted into a new 1.5ml eppendorf tube and precipitated with an equal volume of isopropyl alcohol. The pellet was washed with 70% alcohol, dried, and dissolved in 20μl TE buffer with RNaseA (40μg/ml), then it can be used as template for PCR. For pooled genomic DNA extraction of 10 actinomycetes strains, 40μl of the liquid culture of each strain were mixed in an eppendorf tube, then following the extraction procedure above.

    1.4  PCR amplification of AHBA synthase genes

    Two pairs of degenerative oligonucleotides, 755a (5′?agaggatccTTCGAGCRSGAGTTCGC?3′, small letters are those added for easy cloning, same below) and 755b (gcaggatccGGAMCATSGCCATGTAG?3′), 235a (5′?GAGGTSATCGTSCCSGCSTTC?3′) and 235b (5′?TGSGCGTGSGCSGCGTCCTG?3′) were designed to amplifiy, respectively, a 755bp portion and a 235bp portion of the highly conserved AHBA synthase genes. For each PCR amplification reaction, the following procedure was followed. In an 15μl PCR system, adding: Taq LA 2×GC Buffer I 7.5μl; ddH2O 5μl; dNTP (2.5mmol/L) 0.5μl; primers (755a/755b; or 235a/235b), each 0.5μl (to a final concentration of 20μmol/L); genomic DNA 1μl; Taq LA (TaKaRa) 0.5U. The PCR parameters were denaturing at 96℃ for 25s, annealing at 58℃ for 25s, extension at 72℃ for 1min, 30 cycles. 4～5μl of PCR amplification product was used for agarose gel (1%) electrophoresis analysis.

    1.5  Screening of actinomycetes strains with AHBA synthase gene by PCR

    Pooled genomic DNAs of 10 different actinomycetes strains were used as template for PCR amplification/detection of the 755bp portion of AHBA synthase gene. The overall strategy for high?throughput screening of actinomycetes strains with AHBA synthase gene is illustrated in Fig.2.

    1.6  Sequencing and blast analysis of PCR product

    DNA amplification products of 755bp were recovered and purified from agarose gel. Most DNA recovery samples could be sent for direct sequencing, but some samples needed to be cloned into vectors (e.g., pMD19?T) before sequencing. DNA sequencing services were accomplished by TaKaRa Biotechnology (Dalian) Co.,

    Fig.2    Schematic diagram of the PCR screening procedure for AHBA synthase gene

    positive strains. Each number of 1～10 and each letter of A?J represented a

    pool of 10 strains genomic DNA. The cross of two positive pools, like 5E,

    was chosen for subsequent single strain PCR screeningLtd.. The DNA sequencing data were then blast?analyzed on NCBI′s nucleotide data bank (http://www.ncbi.nlm.nih.gov/BLAST) to obtain similarity parameters with published AHBA synthase genes.

    1.7  Nucleotide sequence accession numbers

    Sequence data have been deposited with GenBank and assigned the accession number from EU178762 to EU178794, which can be retrieved by Entrez PopSet as a group.

    2  Results

    2.1  Oligonucleotides 755a/755b are suitable for targeted screening of potential ansamycin producers

    Five known AHBA synthases for ansamitocin (AAC13997), mitomycin C (AAD32723), ansatrienin (AAD31835), rifamycin (AAC01720), and naphthomycin (AAD31828) from NCBI were chosen for multiple alignment (http://bioinfo.genopole?toulouse.prd.fr/multalin/multalin.html). Two pairs of oligonucleotides (see 1.4) were designed and synthesized, then tested for their specificity and effectiveness for PCR amplification of AHBA synthase genes using four strains (see 1.1) as positive controls for ansamycin producers, and seven carefully chosen strains (which can be divided into two groups, one group of first five strains producing antibiotics with aromatic moiety ring, the other group of last two strains producing macrolide antibiotics. See 1.1) as negative controls for non?ansamycin producers. Both pair of oligonucleotides produced expected DNA amplification fragments from the four positive controls, and both pairs of oligonucleotides failed to generate expected DNA amplification fragments from the seven negative controls. These results suggested that PCR screening for potential ansamycin producers was practicable. The oligonucleotides, 755a/755b showed a better specificity for PCR amplification of AHBA synthase gene in the positive controls. Sequencing and blast analysis of the four 755bp DNA fragments amplified by the pair of oligonucleotides 755a/755b, suggested that they all belonged to AHBA synthase genes. Therefore, the pair of oligonucleotides, 755a/755b, was chosen for large?scale PCR screening of actinomycetes strains with ansamycin producing potentials.

    2.2  A total of 33 potential ansamycin?producing actinomycetes strains obtained

    To facilitate the screening process, a high?throughput screening strategy (see Fig.2) was developed by pooled?amplification of a mixture of genomic DNAs of 10 actinomycetes strains, which saved both labor and time, and therefore made screening more convenient and cost?effective.

    Each 755bp DNA fragment amplified by PCR were sequenced and blast?analyzed with NCBI GenBank (Blastn and Blastp). It was very easy to identify each of the 755bp DNA sequence as being the AHBA synthase gene or not, because all AHBA synthases shared very high similarity/identity percentages. Amino acid sequences of each 755bp DNA fragment was then multi?aligned with known AHBA synthases to check if it fit well with the 6～7 nearly?unvariable sequence blocks. A total of 33 AHBA synthase genes were obtained.

    3  Discussions

    The AHBA synthase gene had been used as a probe, to clone successfully the biosynthetic gene clusters of several ansamycins such as rifamycin, geldanamycin, etc. It is based on these successes that a high?throughput screening method, which detects the existence of AHBA synthase gene, for potential ansamycins producers was developed in this paper.

    Because of the varieties of biological activities ansamycins displayed, it is nearly impossible to establish an activity?based assaying screening model suitable for screening all ansamycins. There were some novel ansamycins disclosed during the past ten years, but they were all discovered by certain kinds of activity?based screening models［4～7］.

    There are several reports about PCR screening of actinomycetes strains with potentials to biosynthesize, for examples, aromatic macrolides and macrolactone macrolides［17,18］. All these PCR screening methods are based on the amplification of a specific fragment/portion of a highly conserved gene that is indispensable for the biosynthesis of antibiotics with certain structures. As an extension of the PCR screening strategy, Zazopoulos et al.［19］ established a genomics?guided approach to search for actinomycetes strains with the potential to biosynthesize antibiotics with enediyne structure, which probed the existence of a cassette of 5 conserved genes that encoded the 5 essential enzymes for enediynes biosynthesis.

    Of the 1900 actinomycetes strains screened in this study, 1.74% were AHBA synthase gene positive. If this is a typical case of all the actinomycetes strains in nature, then ansamycins may not be a class of scarce secondary metabolites of actinomycetes.

    AHBA, the starter unit for all kinds of ansamycins, may also be employed for the biosynthesis of other secondary metablites of microorganisms, although this is a vary rare case. As an example, Toshikazu Komoda reported recently the biosynthesis of tetrapetalones, which utilized AHBA as one of its building units (possible, the starting unit) of biosynthsis［20］. We believe that most of the 33 AHBA synthase gene positive strains obtained in this study are potential ansamycins producing strains, but one or a few of them may be potential other than ansamycins producing strains that utilize AHBA as their biosynthetic building blocks.

    Although the oligonucleotides 755a/755b were designed for amplification of AHBA synthase genes, their specificity was not absolute. Bioinformatic analysis of the 755bp DNA were necessary. One 755bp DNA fragment (No.22～157) was eliminated after sequencing and blast analysis (part of this 755bp DNA showed high similarities to genes encoding the nitrile hydratase activator or cobalamin synthesis protein from some microorganisms).

    This paper provided a simple and easy way for discovering, systemically, potential ansamycins producing strains. For the discoveries of novel ansamycins from the fermentation cultures of these AHBA synthase gene positive strains, chemical data such as TLC and MS?MS, with known ansamycins (many commerially available) as references for comparison, are needed to eliminate first those which belong to known ansamycins. Sequence comparison/alignment of AHBA synthase genes could not provide convincible evidences for novel ansamycins production, as all the AHBA synthase genes are highly conserved.

    According to our incomplete collection and analysis of published articles, about 40～50 different kinds of natural ansamycins have been discovered, with the majorities of them being produced by actinomycetes. It is believed there are still a lot of novel ansamycins to be discovered, because the combinations of the ansa?bridge variations in length and reduction level, the different substituting groups on the ansa? bridge and the chromophore, and the two kinds of chromophores (the benzenic and the naphthalenic based aromatic rings) should result in far more than 50 different kinds of structures of ansamycins. At present, the AHBA synthase gene positive strains obtained in this study are under intensive investigation of their ansamycin producting abilities.

    It is worthy to note that the screening strategy employed in this paper should aslo be applicable to the systemic screening of potential producing strains of aminoglycosides, anthracyclines, β?lactams, macrolides, non?ribosomal peptides, etc, if some essential biosynthetic gene(s) is/are selected and then targeted?screened by PCR.

【】
  ［1］ Weigel B J, Blaney S M, Reid J M, et al. A phase I study of 17?allylaminogeldanamycin in relapsed/refractory pediatric patients with solid tumors: a Children′s Oncology Group study ［J］. Clin Cancer Res,2007,13(6):1789～1793.

［2］ Solit D B, Ivy S P, Kopil C, et al. Phase I trial of 17?allylamino?17?demethoxygeldanamycin in patients with advanced cancer ［J］. Clin Cancer Res,2007,13(6):1775～1782.

［3］ Li Y H, Tao P Z, Liu Y Z, et al. Geldanamycin, a ligand of heat shock protein 90, inhibits the replication of herpes simplex virus type 1 in vitro ［J］. Antimicrob Agents Chemother,2004,48(3):867～872.

［4］ Stead P, Latif S, Blackaby A P, et al. Discovery of novel ansamycins possessing potent inhibitory activity in a cell?based oncostatin M signalling assay ［J］. J Antibiot (Tokyo),2000,53(7):657～663.

［5］ Hosokawa N, Naganawa H, Hamada M, et al. New triene?ansamycins, thiazinotrienomycins F and G and a diene?ansamycin, benzoxazomycin ［J］. J Antibiot (Tokyo),2000,53(9):886～894.

［6］ Hosokawa N, Naganawa H, Inuma H, et al. Thiazinotrienomycins, new ansamycin group antibiotics ［J］. J Antibiot (Tokyo),1995,48(6):471～478.

［7］ Kakeya H, Zhang H P, Kobinata K, et al. Cytotrienin A, a novel apoptosis inducer in human leukemia HL?60 cells ［J］. J Antibiot (Tokyo),1997,50(4):370～372.

［8］ Hu Z H, Liu Y, Tian Z Q, et al. Isolation and characterization of novel geldanamycin analogues ［J］. J Antibiot (Tokyo),2004,57(7):421～428.

［9］ Arakawa K, Muller R, Mahmud T, et al. Characterization of the early stage aminoshikimate pathway in the formation of 3?amino?5?hydroxybenzoic acid: the RifN protein specifically converts kanosamine into kanosamine 6?phosphate ［J］. J Am Chem Soc,2002,124(36):10644～10645.

［10］ Kim C G, Yu T W, Fryhle C B, et al. 3?Amino?5?hydroxybenzoic acid synthase, the terminal enzyme in the formation of the precursor of mC7N units in rifamycin and related antibiotics ［J］. J Biol Chem,1998,273(11):6030～6040.

［11］ He W Q, Wu L Z, Gao Q J, et al. Identification of AHBA biosynthetic genes related to geldanamycin biosynthesis in Streptomyces hygroscopicus 17997 ［J］. Curr Microbiol,2006,52(3):197～203.

［12］ Yu T W, Bai L, Clade D, et al. The biosynthetic gene cluster of the maytansinoid antitumor agent ansamitocin from Actinosynnema pretiosum ［J］. Proc Natl Acad Sci USA,2002,99(12):7968～7973.

［13］ Chen S, von Bamberg D, Hale V, et al. Biosynthesis of ansatrienin (mycotrienin) and naphthomycin. Identification and analysis of two separate biosynthetic gene clusters in Streptomyces collinus Tü 1892 ［J］. Eur J Biochem,1999,261(1): 98～107.

［14］ Wang N J, Fu Y, Yan G H, et al. Isolation and structure of a new ansamycin antibiotic. Kanglemycin A from a Nocadia ［J］. J Antibiot (Tokyo),1988,41(2):264～267.

［15］ Muroi M, Izawa M, Kosai Y, et al. Macbecins I and II, new antitumor antibiotics. II. Isolation and characterization ［J］. J Antibiot (Tokyo),1980,33:205～212.

［16］ Xu P, Li W J, Xu L H, et al. A microwave?based method for genomic DNA extraction from actinomycetes ［J］. Microbiology (China),2003,30(4):73～75.

［17］ Xu P, Li W J, Gao H Y, et al. Molecular screening and distribution of aromatic polyketide antibiotics producers from actinomycetes ［J］. Chin J Antibiot,2005,30(3):134～137.

［18］ Xu P, Li W J, Zhang Y G, et al. Molecular screening and distribution of polyketide antibiotics producers from actinomycetes ［J］. Chin J Antibiot,2003,28(3):321～324.

［19］ Zazopoulos E, Huang K, Staffa A, et al. A genomics?guided approach for discovering and expressing cryptic metabolic pathways ［J］. Nat Biotechnol,2003,21(2):187～190.