Improvement of pristinamycin production by genome shuffling and medium optimization for Streptomyces pristinaespiralis

To isolate an improved pristinamycin producing strain of Streptomyces pristinaespiralis, the technique of Genome shuffling was used which resulted in a high-yield recombinant G 3-56 strain. Strain G 3-56 yielded 322 ± 17 mg/L of pristinamycin which was 11.4-fold higher than that of the initial strain and 3.7-fold higher than strain UN-78 which previously had the highest yield of pristinamycin. The genetic characteristics of the recombinant G 3-56 strain was stable as revealed by our subculture experiments. The optimal production medium was determined using the orthogonal matrix method. Under the optimal medium conditions, the maximum yield of pristinamycin was 412 mg/L with about 1.24-fold higher than the original medium.     

Improved production of pristinamycin coupled with an adsorbent resin in fermentation by Streptomyces pristinaespiralis

Batch fermentation by Streptomyces pristinaespiralis with the addition of adsorbent resins was used to increase the production of pristinamycin. In consideration of the adsorption capacity and the desorption ability, a polymeric resin, JD-1, was finally selected. The maximum production of pristinamycin in Erlenmeyer flasks went up to 1.13 from 0.4 g l⁻¹, by adding 12% (w/v) resin JD-1 into the culture broth at 20 h after inoculation. In a 3 l bioreactor, pristinamycin fermentation with the addition of 12% (w/v) resin JD-1 at 20 h after inoculation reached 0.8 g l⁻¹, which was a 1.25-fold increase over fermentation without resin.     

Genomic variability among high pristinamycin-producing recombinants of Streptomyces pristinaespiralis revealed by amplified fragment length polymorphism

Amplified fragment length polymorphism (AFLP) was used to analyze genomic variability between high pristinamycin-producing recombinants of Streptomyces pristinaespiralis produced by genome shuffling and their ancestral strain. The AFLP fingerprints obtained with two restriction enzyme combinations of ApaI/TaqI and PstI/SacII showed together that there was no major polymorphism (less than 10%) between these high yield recombinants and their ancestor. However, the unique polymorphic bands, which might be related to the yield increasing of pristinamycin, could be distinguished from all the recombinants. Clustering analysis further indicated that the recombinants with similar ability of pristinamycin production had similar genomic variability.     

Improvement of pristinamycin-producing Streptomyces pristinaespiralis by rational screening

Summary According to the biosynthetic pathway of pristinamycin, a rational selection procedure with u.v. mutation was performed to obtain a high pristinamycin-producing strain. Aminoacetic acid-resistant mutants (AA^r), valine hydroxamate-resistant mutants (VH^r), kitasamycin-resistant mutants (KTM^r) and 2-deoxy-D-glucose-resistant mutants (DOG^r) were selected, successively. A strain Streptomyces pristinaespiralis 12–55 with AA^r, Val^r, KTM^r, and DOG^r was obtained, and its production of pristinamycin reached 3000 u/ml which is 100 times higher than that of the parent strain S. pristinaespiralis ATCC 25486. It is inferred that S. pristinaespiralis 12–55 can alleviate catabolite repression caused by carbon sources, provide more acetic acid and valine for pristinamycin biosynthesis and increase its resistance to pristinamycin produced by itself, all of which are favorable for pristinamycin production. The subculture experiments indicated that the hereditary character of high productivity of S. pristinaespiralis 12–55 is stable. The pristinamycin production of S. pristinaespiralis 12–55 in a 15-l fermentor could reach 3010 u/ml after a 56 h batch fermentation.     

Improved production of propionic acid using genome shuffling

Traditionally derived from fossil fuels, biological production of propionic acid has recently gained interest. Propionibacterium species produce propionic acid as their main fermentation product. Production of other organic acids reduces propionic acid yield and productivity, pointing to by‐products gene‐knockout strategies as a logical solution to increase yield. However, removing by‐product formation has seen limited success due to our inability to genetically engineer the best producing strains (i.e. Propionibacterium acidipropionici). To overcome this limitation, random mutagenesis continues to be the best path towards improving strains for biological propionic acid production. Recent advances in next generation sequencing opened new avenues to understand improved strains. In this work, we use genome shuffling on two wild type strains to generate a better propionic acid producing strain. Using next generation sequencing, we mapped the genomic changes leading to the improved phenotype. The best strain produced 25% more propionic acid than the wild type strain. Sequencing of the strains showed that genomic changes were restricted to single point mutations and gene duplications in well‐conserved regions in the genomes. Such results confirm the involvement of gene conversion in genome shuffling as opposed to long genomic insertions.     

Biochemical characterisation of recombinant Streptomyces pristinaespiralis L-lysine cyclodeaminase

L-Lysine cyclodeaminase from Streptomyces pristinaespiralis was heterologously expressed in Escherichia coli, isolated to 90% purity after two purification steps and characterised. The size of the isolated recombinant enzyme was in agreement with the theoretical size calculated from the corresponding gene. We demonstrated that our preparation converts L-lysine to L-pipecolic acid (enantiomeric excess >95%) after isolating and identifying the conversion product by LC/MS, NMR and IR. This conversion followed Michaelis-Menten kinetics with a K(m) of 1.39+/-0.32 mM. The enzyme activity was maximal at pH 6.7. Reducing conditions, the presence of glycerol and in particular the presence of iron(II) significantly enhanced the L-lysine cyclodeaminase activity. Although the heat stability of the enzyme diminished significantly after 37 degrees C, the initial rate of reaction was maximal at 61 degrees C. We found no requirement for an external cofactor for full activity, although sequence data indicate NAD+ as cofactor. Upon enzyme denaturation, NAD+ release was observed, which indicates very tight binding of NAD+ to the enzyme. In parallel we developed selection and screening assays for lysine cyclodeaminase, which we adapted to microtitre plate format and validated. Among twenty-eight lysine analogues screened for turnover/binding to the enzyme, three were identified as substrates (L-ornithine, 5-hydroxylysine and L-4-thialysine), while another six (4-azalysine, L-2,4-diaminobutyric acid, 1,5-diaminopentane, N-epsilon-trifluoroacetyl-L-lysine, N-epsilon-Boc-L-lysine and N-epsilon-methyl-L-lysine) were shown to compete against L-lysine turnover without being converted by the enzyme. All substrates displayed Michaelis-Menten kinetics upon turnover by lysine cyclodeaminase. Our results indicate that the lysine cyclodeaminase from Streptomyces pristinaespiralis is a highly enantioselective enzyme at the substrate recognition and conversion levels, in both cases in favour of the l-isomer.     

Generation of high-yield rapamycin-producing strains through protoplasts-related techniques

Rapamycin is a 31-member ring macrolide produced by Streptomyces hygroscopicus and has many applications in clinical medicine. In the present work, several protoplasts-related techniques including protoplasts mutation, intraspecies and interspecies protoplasts fusion were tried to improve the rapamycin productivity in S. hygroscopicus. Although mutation and fusion of different protoplasts of S. hygroscopicus did not improve the productivity of rapamycin significantly, the interspecies fusion of protoplasts of S. hygroscopicus D7-804 and Streptomyces erythreus ZJU325 could have brought about one high-yield (345 mg/L) rapamycin producer with 23.6% higher than that of the parental strain. Then, with seven mutants of S. hygroscopicus with different features and rapamycin productivities as the parental strains, only one-round genome shuffling has generated a high-yield rapamycin-producing strain with an outstanding yield of 445 mg/L. The systematic research of protoplast-related techniques has established an applicable way to generate high-yield strains from original microorganisms which can only produce low amount of expected natural products, without information of target gene clusters and gene sequences.     

Genome shuffling of <Emphasis Type="Italic">Streptomyces</Emphasis> sp. U121 for improved production of hydroxycitric acid

(2S, 3R)-Hydroxycitric acid (HCA) from Hibiscus subdariffa inhibits pancreatic α-amylase and intestine α-glucosidase, leading to reduction of carbohydrate metabolism. In our previous study, Streptomyces sp. U121 was identified as a producer of (2S, 3R)-HCA [Hida et al. (2005) Bioscience, Biotechnology, and Biochemistry 69:1555–1561]. Here, we applied genome shuffling of Streptomyces sp. U121 to achieve rapid improvement of HCA production. The initial mutant population was generated by nitrosoguanidine treatment of the spores, and an improved population producing fivefold more HCA over wild type was obtained by three rounds of genome shuffling. For efficient screening of the mutant library, trans-epoxyaconitic acid (EAA), an antibiotic analog of HCA, was utilized. EAA inhibited the regeneration of nonfused protoplasts, resulting in selective screening of shuffled strains. Mutant strains with enhanced EAA resistance exhibited significantly higher HCA production in liquid media. Furthermore, the best mutant showed increased cell growth in flask culture, as well as increased HCA production.     

Genome shuffling: Progress and applications for phenotype improvement

Although rational method and global technique have been successfully applied in strain improvement respectively, the demand for engineering complex phenotypes required combinatorial approach. The technology of genome shuffling has been presented as a novel whole genome engineering approach for the rapid improvement of cellular phenotypes. This approach using recursive protoplast fusion with multi-parental strains offers the advantage of recombination throughout the entire genome without the necessity for genome sequence data or network information. Genome shuffling has been demonstrated as an effective method, which is not only for producing improved strain but also for providing information on complex phenotype. In this review we attempt to present the advantage of genome shuffling, introduce the procedure of this technology, summarize the applications of this approach for phenotype improvement and then give perspective on the development of this method in the future.     

Novel methods of genome shuffling in <Emphasis Type="Italic">Saccharomyces cerevisiae</Emphasis>

Genome shuffling can improve complex phenotypes; however, there are several obstacles towards its broader applicability due to increased complexity of eukaryotic cells. Here, we describe novel, efficient and reliable methods for genome shuffling to increase ethanol production of Saccharomyces cerevisiae. Using yeast sexual and asexual reproduction by itself, mutant diploid cells were shuffled through highly efficient sporulation and adequate cross among the haploid cells, followed by selection on the special plates. The selected strain obtained after three round genome shuffling not only distinctly improved the resistance to ethanol, but also, increased ethanol yield by up to 13% compared with the control.     

Ethanol production from selected lignocellulosic hydrolysates by genome shuffled strains of Scheffersomyces stipitis

Two genome-shuffled Scheffersomyces stipitis strains, GS301 and GS302, exhibiting improved tolerance to hardwood spent sulphite liquor, were tested for growth and fermentation performance on three wood hydrolysates: (a) steam-pretreated enzymatically hydrolyzed poplar hydrolysate from Mascoma Canada, (b) steam pretreated poplar hydrolysate from University of British Columbia Forest Products Biotechnology Laboratory, and (c) mixed hardwoods pre-hydrolysate from FPInnovations (FPI). In the FPI hydrolysate, the wild type (WT) died off within 25 h, while GS301 and GS302 survived beyond 100 h. In fermentation tests, GS301 and GS302 completely utilized glucose and xylose in each hydrolysate and produced 0.39–1.4% (w/v) ethanol. In contrast, the WT did not utilize or poorly utilized glucose and xylose and produced non-detectable to trace amounts of ethanol. The results demonstrated cross tolerance of the mutants to inhibitors in three different wood hydrolysates and reinforced the utility of mating-based genome shuffling approach in industrial yeast strain improvement.     

Genome shuffling amplifies the carbon source spectrum and improves arachidonic acid production in Diasporangium sp.

The aim of this study was to develop a new fungal strain that simultaneously amplifies the carbon source spectrum and increases arachidonic acid (AA) productivity using genome shuffling between Diasporangium sp. and inactive Aspergillus niger. Through three rounds of genome shuffling, one of the stable daughter strains (F₁) acquired the ability to produce arachidonic acid and utilize various carbon sources. Compared to the parental Diasporangium sp., which could only use four out of eight carbon sources tested, F₁ could utilize all eight carbon sources. During fermentation with CMC-Na as the carbon source, F₁ was able to obtain 30.16% of lipid effectively whereas the parental Diasporangium sp. was not able to grow at all. When glucose was used as the carbon source, the CMCase activity of F₁ was 879.36 U, 298.23% higher than that of the parental Diasporangium sp. Under optimized fermentation conditions in a 5-L fermentation container, the AA yield of F₁ reached 0.81 g/l, 94.78% higher than that of the parental generation. These results indicate that inter-kingdom genome shuffling can be used successfully in eukaryotic microorganisms and that it can effectively improve the production of desired metabolites within a short period of time. The findings of this study may be useful for extending the application of genome shuffling in eukaryotic microbial breeding.     

Screening and breeding of high taxol producing fungi by genome shuffling

To apply the fundamental principles of genome shuffling in breeding of taxol-producing fungi, Nodulisporium sylviform was used as starting strain in this work. The procedures of protoplast fusion and genome shuffling were studied. Three hereditarily stable strains with high taxol production were obtained by four cycles of genome shuffling. The qualitative and quantitative analysis of taxol produced was confirmed using thin-layer chromatography (TLC), high performance liquid chromatography (HPLC) and LC-MS. A high taxol producing fungus, Nodulisporium sylviform F4-26, was obtained, which produced 516.37 μg/L taxol. This value is 64.41% higher than that of the starting strain NCEU-1 and 31.52%―44.72% higher than that of the parent strains.     

Electron microscopic analysis of protoplast fusion in Streptomyces hygroscopicus and consideration on structural alterations in fusing membranes

Protoplasts of Streptomyces hygroscopicus were treated with polyethylene glycol and prepared for electron microscopic investigation as ultrathin sections. About 5% binary fusion products and 0.9% multicellular fusion products have been obtained in the sections. Three main types may be differentiated among binary fusion products, characterized by a successive loss of the bispherical shape and of continuous membrane structures in fusion zones.Analysing the membrane alterations a contact zone characterized by intact cytoplasmic membranes in both protoplasts, a fusion zone with a trilaminar fusion membrane of about 13–17 nm in thickness, and a fusion zone without continuous membrane structure can be distinguished. The different fusion areas are considered as stages in the fusion process. The data will be discussed in conjunction with a model for membrane alterations during fusion at the molecular level.     

Restriction analysis of the Streptomyces glaucescens genome by agarose gel electrophoresis

A method for the analysis of total DNA of Streptomyces glaucescens is described. The relevant steps are (a) extraction and purification of DNA, (b) restriction of DNA samples with type II restriction enzymes, (c) one dimensional separation of restriction fragments by agarose gel electrophoresis. A typical banding pattern was obtained for each wild type strain, independant of growth conditions or age of the culture. Mutant strains exhibited in most cases the same banding pattern as the parent wild type strain. Only in one specific mutant class a fragment of about 9 megadalton was missing.     

Strain improvement of the pentose-fermenting yeast Pichia stipitis by genome shuffling

Genome shuffling based on cross mating was used to improve the tolerance of the pentose-fermenting yeast Pichia stipitis towards hardwood spent sulphite liquor (HW SSL). Six UV-induced mutants of P. stipitis were used as the starting strains, and they were subjected to 4 rounds of genome shuffling. After each round, improved strains were selected based on their growth on HW SSL gradient plates. Mutant libraries were established after each round and these improved mutant strains served as the starting pool for the next round of shuffling. Apparent tolerance to HW SSL on the gradient plate increased progressively with each round of shuffling up to 4 rounds. Selected improved mutants were further tested for tolerance to liquid HW SSL. After 4 rounds of shuffling, 4 mutants, two from the third round (designated as GS301 and GS302) and two from the fourth round (designated as GS401 and GS402), were selected that could grow in 80% (v/v) HW SSL. GS301 and GS302 grew also in 85% (v/v) HW SSL. GS301 was viable in 90% (v/v) HW SSL, although no increase in cell number was seen. The P. stipitis wild type strain (WT) could not grow on HW SSL unless it was diluted to 65% (v/v) or lower. Genome-shuffled strains with improved tolerance to HW SSL retained their fermentation ability. Fermentation performance of GS301 and GS302, the 2 strains that exhibited the best tolerance to liquid HW SSL, was assessed in defined media and in HW SSL. Both strains utilized 4% (w/v) of xylose or glucose more efficiently and produced more ethanol than the WT. They also utilized 4% (w/v) of mannose or galactose and produced ethanol to the same extent as the WT. GS301 and GS302 were able to produce low levels of ethanol in undiluted HW SSL.     

Penicillin-binding proteins of protoplast and sporoplast membranes of Streptomyces griseus strains

Membrane-bound penicillin-binding proteins (PBPs) of two Streptomyces griseus strains that sporulate well in liquid and solid medium have been investigated during the course of their life-cycle. The PBP patterns were analyzed by sodium dodecylsulphate polyacrylamide-gel electrophoresis and fluorography. One strain (No. 45 H) has only a single band (mol wt: 27,000) in early log phase, and two additional PBPs of higher mol wt (69,000 and 80,000) in the late log phase. The other strain (No. 2682) possessed two bands with mol wts 27,000 and 38,000 which did not change during its vegetative phase. In strain No. 2682, a new PBP with a mol wt of 58,000 appeared in spore membranes while one of those (mol wt 38,000) present in mycelial membranes disappeared. Our results suggest that appearance of the new PBP in the spore may be associated with the sporulation process. The major PBP band (mol wt: 27,000) present in all stages of the life cycle of these strains, may be characteristic of S. griseus while the other PBPs reflect certain stages of the life cycle. A new method was developed for the production of spore protoplasts by consecutive enzymatic treatments.Abbreviation PBP penicillin-binding protein     

Optimization of glycerol feeding for clavulanic acid production by Streptomyces clavuligerus with glycerol feeding

Glycerol at 10–20 g l^–1 increased clavulanic acid production by Streptomyces clavuligerus in shake-flask culture. The biosynthesis of clavulanic acid continued for longer by feeding glycerol and production increased to 250 mg l^–1 compared with 115 mg l^–1 without feeding. In fermenter batch culture, degradation of clavulanic acid began after 72 h. With glycerol feeding in fed-batch culture, clavulanic acid production was not only increased further to about 280 mg l^–1 but also remained stable up to 130 h.     

Mechanism of natural rifampin resistance of Streptomyces spp

In a previous phylogenetic study of the genus Streptomyces using the rpoB gene, N531, which stands for an aspargine residue in position 531 of RpoB instead of serine (S531), known to be associated with natural rifampin resistance in several organisms, was also observed in the RpoB of several Streptomyces species. To determine whether N531 is associated with the rifampin resistance of Streptomyces strains, we analyzed the rifampin minimum inhibitory concentrations (MICs) of 11 strains of the N531 RpoB type (putative rifampin resistant strains) and of 12 strains of the S531 RpoB type. (putative rifampin susceptible strains). In general, the N531 RpoB types showed higher MIC levels (16-128 microg/ml) than the S531 RpoB types (0-8 microg/ml). To determine the isolation frequencies of N531 RpoB types versus rifampin concentration, we applied screening methods involving different rifampin concentrations (0, 20 and 100 microg/ml) to Korean soils. Higher isolation frequencies of the N531 RpoB types were observed at the higher rifampin concentrations. In addition, during the course of this study we developed an allele specific PCR method to detect rifampin resistant Streptomyces strains. Our results strongly suggested that N531 might be involved in a major mechanism of natural rifampin resistance in strains of the genus Streptomyces.     

Identification and functional characterization of phenylglycine biosynthetic genes involved in pristinamycin biosynthesis in Streptomyces pristinaespiralis

Pristinamycin I (PI), a streptogramin type B antibiotic produced by Streptomyces pristinaespiralis, contains the aproteinogenic amino acid l-phenylglycine. Recent sequence analysis led to the identification of a set of putative phenylglycine biosynthetic genes. Successive inactivation of the individual genes resulted in a loss of PI production. Production was restored by supplementation with externally added l-phenylglycine, which demonstrates that these genes are involved in phenylglycine biosynthesis and thus probably disclosing the last essential pristinamycin biosynthetic genes. Finally, a putative pathway for phenylglycine synthesis is proposed.