DNA shuffling of methionine adenosyltransferase gene leads to improved S-adenosyl-l-methionine production in Pichia pastoris

Characterization and designing of an SAM riboswitch to establish a high-throughput screening platform for SAM overproduction in Saccharomyces cerevisiae

S-adenosyl- l-methionine (SAM) is a high-value compound widely used in the treatment of various diseases. SAM can be produced through fermentation, but further enhancing the microbial production of SAM requires novel high-throughput screening methods for rapid detection and screening of mutant libraries. In this work, an SAM-OFF riboswitch capable of responding to the SAM concentration was obtained and a high-throughput platform for screening SAM overproducers was established. SAM synthase was engineered by semirational design and directed evolution, which resulted in the SAM2S203F,W164R,T251S,Y285F,S365R mutant with almost twice higher catalytic activity than the parental enzyme. The best mutant was then introduced into Saccharomyces cerevisiae BY4741, and the resulting strain BSM8 produced a sevenfold higher SAM titer in shake-flask fermentation, reaching 1.25 g L-1 . This work provides a reference for designing biosensors to dynamically detect metabolite concentrations for high-throughput screening and the construction of effective microbial cell factories.

Genome-scale metabolic model analysis of Pichia pastoris for enhancing the production of S-adenosyl-L-methionine

Komagataella phaffii, formerly Pichia pastoris (P. pastoris), is a promising methylotrophic yeast used in industry to produce recombinant protein and valuable metabolites. In this study, a genome-scale metabolic model (GEMs) was reconstructed and used to assess P. pastoris' metabolic capabilities for the production of S-adenosyl-L-methionine (AdoMet or SAM or SAMe) from individual carbon sources along with the addition of L-methionine. In a model-driven P. pastoris strain, the well-established genome-scale metabolic model iAUKM can be implemented to predict high valuable metabolite production. The model, iAUKM, was created by merging the previously published iMT1026 model and the draught model generated using Raven toolbox from the KEGG database which covered 2309 enzymatic reactions associated with 1033 metabolic genes and 1750 metabolites. The highly curated model was successful in capturing P. pastoris growth on various carbon sources, as well as AdoMet production under various growth conditions. Many overexpression gene targets for increasing AdoMet accumulation in the cell have been predicted for various carbon sources. Inorganic phosphatase (IPP) was one of the predicted overexpression targets as revealed from simulations using iAUKM. When IPP gene was integrated into P. pastoris, we found that AdoMet accumulation increased by 16% and 14% using glucose and glycerol as carbon sources, respectively. Our in silico results shed light on the factors limiting AdoMet production, as well as key pathways for rationalized engineering to increase AdoMet yield.

Enhanced selenocysteine biosynthesis for seleno-methylselenocysteine production in Bacillus subtilis

Seleno-methylselenocysteine (SeMCys) is an effective component for selenium supplementation with anti-carcinogenic potential and can ameliorate neuropathology and cognitive deficits. In this study, we aimed to engineer Bacillus subtilis 168 for the microbial production of SeMCys. First, the accumulation of intracellular selenocysteine (SeCys) as the precursor of SeMCys was enhanced through overexpression of serine O-acetyltransferase, which was desensitized against feedback inhibition by cysteine. Next, the S-adenosylmethionine (SAM) synthetic pathway was optimized to improve methyl donor availability through expression of S-adenosylmethionine synthetase. Further, SeMCys was successfully produced through expression of the selenocysteine methyltransferase in SeCys and SAM-producing strain. The increased expression level of selenocysteine methyltransferase benefited the SeMCys production. Finally, all the heterologous genes were integrated into the genome of B. subtilis, and the strain produced SeMCys at a titer of 18.4 μg/L in fed-batch culture. This is the first report on the metabolic engineering of B. subtilis for microbial production of SeMCys and provides a good starting point for future pathway engineering to achieve the industrial-grade production of SeMCys. KEY POINTS: • Expression of the feedback-insensitive serine O-acetyltransferase provided B. subtilis the ability of accumulating SeCys. • SAM production was enhanced through expressing S-adenosylmethionine synthetase in B. subtilis. • Expression of selenocysteine methyltransferase in SeCys and SAM-accumulating strain facilitated SeMCys production.

Improving ATP availability by sod1 deletion with a strategy of precursor feeding enhanced S-adenosyl-L-methionine accumulation in Saccharomyces cerevisiae

S-adenosyl-L-methionine (SAM), used in diverse pharmaceutical applications, was biosynthesized from L-methionine (L-met) and adenosine triphosphate (ATP). This study aims to increase the accumulation of SAM in Saccharomyces cerevisiae by promoting ATP availability. Strain ΔSOD1 was obtained from the parent strain WT15-33 (CCTCC M 2021915) by deleting gene sod1, which improved the supply of ATP. The SAM content in strain ΔSOD1 exhibited a 22.3% improvement compared to the parent strain, which reached 93.6 mg g^-1. The transformation of NADH (reduced nicotinamide adenine dinucleotide) and the relative expression of ATPase essential genes were investigated, respectively. The results showed that the lack of gene sod1 benefited the generation of ATP, which positively regulated the synthesis of SAM. Besides that, the production of SAM was further enhanced by improving substrate assimilation. With the infusion of 1.44 g L^-1L-met and 0.60 g L^-1 adenosine at 24 h (h) and 0 h following fermentation, the optimum medium could produce 1.54 g L^-1 SAM. Based on the regulations mentioned above, the SAM concentration of strain ΔSOD1 enhanced from 7.3 g L^-1 to 10.1 g L^-1 in a 5-L fermenter in 118 h. This work introduces a novel idea for the biosynthesis of ATP and SAM, and the strain ΔSOD1 has the potential for industrial production.

High-Throughput Screening and Directed Evolution of Methionine Adenosyltransferase from Escherichia coli

S-adenosyl-L-methionine (SAM) is the active form of methionine, which participates in various metabolic reactions and plays a vital role. It is mainly used as a precursor by three key metabolic pathways: trans-methylation, trans-sulfuration, and trans-aminopropylation. Methionine adenosyltransferase (MAT) is the only enzyme to produce SAM from methionine and ATP. However, there is no efficient and accurate method for high-throughput detection of SAM, which is the major obstacles of directed evolution campaigns for MAT. Herein, we established a colorimetric method for directed evolution of MAT based on detecting SAM by using glycine oxidase and glycine/sarcosine N-methyltransferase enzyme. Screening of MAT libraries revealed variant I303V/Q22R with 2.13-fold improved activity towards SAM in comparison to the wild type. Molecular dynamic simulation indicates that the loops more flexible and more conducive to SAM release.

The multiple effects of REG1 deletion and SNF1 overexpression improved the production of S-adenosyl-L-methionine in Saccharomyces cerevisiae

BACKGROUND

Saccharomyces cerevisiae is often used as a cell factory for the production of S-adenosyl-L-methionine (SAM) for diverse pharmaceutical applications. However, SAM production by S. cerevisiae is negatively influenced by glucose repression, which is regulated by a serine/threonine kinase SNF1 complex. Here, a strategy of alleviating glucose repression by deleting REG1 (encodes the regulatory subunit of protein phosphatase 1) and overexpressing SNF1 (encodes the catalytic subunit of the SNF1 complex) was applied to improve SAM production in S. cerevisiae. SAM production, growth conditions, glucose consumption, ethanol accumulation, lifespan, glycolysis and amino acid metabolism were analyzed in the mutant strains.

RESULTS

The results showed that the multiple effects of REG1 deletion and/or SNF1 overexpression exhibited a great potential for improving the SAM production in yeast. Enhanced the expression levels of genes involved in glucose transport and glycolysis, which improved the glucose utilization and then elevated the levels of glycolytic intermediates. The expression levels of ACS1 (encoding acetyl-CoA synthase I) and ALD6 (encoding aldehyde dehydrogenase), and the activity of alcohol dehydrogenase II (ADH2) were enhanced especially in the presence of excessive glucose levels, which probably promoted the conversion of ethanol in fermentation broth into acetyl-CoA. The gene expressions involved in sulfur-containing amino acids were also enhanced for the precursor amino acid biosynthesis. In addition, the lifespan of yeast was extended by REG1 deletion and/or SNF1 overexpression. As expected, the final SAM yield of the mutant YREG1ΔPSNF1 reached 8.28 g/L in a 10-L fermenter, which was 51.6% higher than the yield of the parent strain S. cerevisiae CGMCC 2842.

CONCLUSION

This study showed that the multiple effects of REG1 deletion and SNF1 overexpression improved SAM production in S. cerevisiae, providing new insight into the application of the SNF1 complex to abolish glucose repression and redirect carbon flux to nonethanol products in S. cerevisiae.

Semi-rationally engineered variants of S-adenosylmethionine synthetase from Escherichia coli with reduced product inhibition and improved catalytic activity

S-adenosylmethionine synthetase (MAT) catalyzes the synthesis of S-adenosylmethionine (SAM) from ATP and L-methionine. SAM is the major methyl donor for more than 100 transmethylation reactions. It is also a common cosubstrate involved in transsulfuration and aminopropylation. However, product inhibition largely restrains the activity of MAT and limits the enzymatic synthesis of SAM. In this research, the product inhibition of MAT from Escherichia coli was reduced via semi-rational modification. A triple variant (Variant III, I303 V/I65 V/L186 V) showed a 42-fold increase in K_i,ATP and a 2.08-fold increase in specific activity when compared to wild-type MAT. Its K_i,ATP was 0.42 mM and specific acitivity was 3.78 ±0.19 U/mg. Increased K_i,ATP means reduced product inhibition which enhances SAM accumulation. The SAM produced by Variant III could reach to 3.27 mM while SAM produced by wild-type MAT was 1.62 mM in the presence of 10 mM substrates. When the residue in 104^th of Variant III was further optimized by site-saturated mutagenesis, the specific activity of Variant IV (I303 V/I65 V/L186 V/N104 K) reached to 6.02 ±0.22 U/mg at 37 °C, though the SAM concentration decreased to 2.68 mM with 10 mM substrates. Analysis of protein 3D structure suggests that changes in hydrogen bonds or other ligand interactions around active site may account for the variety of product inhibition and enzyme activity. The Variant III and Variant IV with reduced inhibition and improved enzyme activity in the study would be more suitable candidates for SAM production in the future.

Enhanced S-Adenosylmethionine Production by Increasing ATP Levels in Baker's Yeast ( Saccharomyces cerevisiae)

In the biosynthesis of S-adenosylmethionine (SAM) in baker's yeast ( Saccharomyces cerevisiae), ATP functions as both a precursor and a driving force. However, few published reports have dealt with the control of ATP concentration using genetic design. In this study we have adopted a new ATP regulation strategy in yeast for enhancing SAM biosynthesis, including altering NADH availability and regulating the oxygen supply. Different ATP regulation systems were designed based on the introduction of water-forming NADH oxidase, Vitreoscilla hemoglobin, and phosphite dehydrogenase in combination with overexpression of the gene SAM2. Via application of this strategy, after 28 h cultivation, the SAM titer in the yeast strain ABYSM-2 reached a maximum level close to 55 mg/L, an increase of 67% compared to the control strain. The results show that the ATP regulation strategy is a valuable tool for SAM production and might further enhance the synthesis of other ATP-driven metabolites in yeast.

Improving the production of S-adenosyl-L-methionine in Escherichia coli by overexpressing metk

S-adenosyl-L-methionine (SAM) has important applications in many fields including chemical therapy and pharmaceutical industry. In this study, the recombinant Escherichia coli strain was constructed for effective production of SAM by introducing the SAM synthase gene (metK). This strain produced 34.5 mg/L of SAM in basic medium in shake flask. Yeast extract, pH, and loaded volume had a significant positive effect on the yield of SAM. Their optimal values were 35 g/L, 7.5, and 30 mL, respectively. The final conditions optimized were as follows: glucose 20, g/L; peptone, 40 g/L; yeast extract, 35 g/L; NaCl, 10 g/L; MgSO₄, 1.2 g/L; L-methionine, 1 g/L; rotate speed, 220 rpm; loaded volume, 30 mL; inoculation, 1%; temperature, 37°C; and initial medium, pH 7.5. The recombinant strain produced 128.2 mg/L of SAM under the above conditions in shake flask. The production of SAM in a 5 L fermentor was also investigated. The maximal biomass of the recombinant strain was 60.4 g/L after the cells were cultured for 20 hr, and the highest yield of SAM was 300.9 mg/L after induction for 8 hr in a 5 L fermentor. This study provides a good foundation for the future production and use of SAM.

Progress in the microbial production of S-adenosyl-L-methionine

S-Adenosyl-L-methionine (SAM), which exists in all living organisms, serves as an activated group donor in a range of metabolic reactions, including trans-methylation, trans-sulfuration and trans-propylamine. Compared with its chemical synthesis and enzyme catalysis production, the microbial production of SAM is feasible for industrial applications. The current clinical demand for SAM is constantly increasing. Therefore, vast interest exists in engineering the SAM metabolism in cells for increasing product titers. Here, we provided an overview of updates on SAM microbial productivity improvements with an emphasis on various strategies that have been used to enhance SAM production based on increasing the precursor and co-factor availabilities in microbes. These strategies included the sections of SAM-producing microbes and their mutant screening, optimization of the fermentation process, and the metabolic engineering. The SAM-producing strains that were used extensively were Saccharomyces cerevisiae, Pichia pastoris, Candida utilis, Scheffersomyces stipitis, Kluyveromyces lactis, Kluyveromyces marxianus, Corynebacterium glutamicum, and Escherichia coli, in addition to others. The optimization of the fermentation process mainly focused on the enhancement of the methionine, ATP, and other co-factor levels through pulsed feeding as well as the optimization of nitrogen and carbon sources. Various metabolic engineering strategies using precise control of gene expression in engineered strains were also highlighted in the present review. In addition, some prospects on SAM microbial production were discussed.

Rapid and selective enzymatic assay for L-methionine based on a pyrophosphate detection system

An enzymatic assay for L-methionine was developed by coupling adenosylmethionine synthetase (AdoMetS) to a pyrophosphate (PP(i)) detection system, which was constructed using pyruvate, phosphate dikinase. To expand the use of this assay, the PP(i) detection system was embodied as three different forms, which allowed PP(i) to be measured by UV, visible, and fluorescent light detectors. The assay system was robust and could tolerate the addition of inorganic phosphate and ATP to the assay mixtures. L-Methionine could be accurately determined by coupling the PP(i) detection system and AdoMetS. This AdoMetS coupling assay was highly selective to L-methionine and exhibited no significant activity to other proteinaceous amino acids, ammonia, or urea, unlike conventional enzymatic assays for L-methionine. Spike and recovery tests showed that the AdoMetS assay could accurately and reproducibly determine increases in L-methionine in human plasma samples without any pretreatment to remove proteins and potentially interfering low-molecular-weight molecules. The high selectivity and robustness of the AdoMetS assay provide rapid and high-throughput analysis of L-methionine in various kinds of analytes.

Expression, purification, and characterization of a recombinant methionine adenosyltransferase pDS16 in Pichia pastoris

Methionine adenosyltransferase (MAT, EC2.5.1.6) catalyzes the synthesis of S-adenosylmethionine (SAM) using L-methionine and adenosine triphosphate (ATP) as substrates. The mutant MAT pDS16 was obtained through DNA shuffling previously in our lab. Overexpression of pDS16 in Pichia pastoris led to about 65 % increase of MAT activity and SAM accumulation, compared with the strain overexpressing Saccharomyces cerevisiae MAT gene SAM2. Different strategies were tested to facilitate the expression and purification of pDS16. However, addition of the hexahistidine tag to pDS16 was shown to decrease the enzyme activity, and the yeast α-factor signal sequence could not effectivley direct the secretion of pDS16. The intracellular pDS16 was purified by a simple two-step procedure combining an ion exchange and hydrophobic interaction chromatography. Protein purity was verified by sodium dodecyl sulfate polyacrylamide gel electrophoresis to be 93%, with the specific activity of 1.828 U/mg. Two-dimensional electrophoresis revealed pI of ∼5.5. The purified enzyme followed Michaelis kinetics with a Km of 1.72 and 0.85 mM, and Vmax of 1.54 and 1.15 μmol/min/mg for ATP and L-methionine, respectively. pDS16 exhibited optimal activity at pH 8.5 and 45 °C with the requirement of divalent cation Mg(2+) and was slightly stimulated by the monovalent cation K(+). It showed an improved thermostability, about 50% of the enzyme activity was retained even after preincubation at 50 °C for 2 h.

Enhanced S-adenosyl-l-methionine production in Saccharomyces cerevisiae by spaceflight culture, overexpressing methionine adenosyltransferase and optimizing cultivation

AIMS

S-adenosyl-l-methionine (SAM) is an important biochemical molecule with great potential in the pharmacological and chemotherapeutic fields. In this study, our aims were to enhance SAM production in Saccharomyces cerevisiae.

METHODS AND RESULTS

Through spaceflight culture, a SAM-accumulating strain, S. cerevisiae H5M147, was isolated and found to produce 86·89% more SAM than its ground control strain H5. Amplified fragment length polymorphism (AFLP) analysis demonstrated that there were genetic variations between strain H5M147 and its ground control. Through recombinant DNA technology, the heterologous gene encoding methionine adenosyltransferase was integrated into the genome of strain H5M147. The recombinant strain H5MR83 was selected because its SAM production was increased by 42·98% when compared to strain H5M147. Furthermore, cultivation conditions were optimized using the one-factor-at-a-time and Taguchi methods. Under optimal conditions, strain H5MR83 yielded 7·76 g l(-1) of SAM in shake flask, an increase of 536·07% when compared to the strain H5. Furthermore, 9·64 g l(-1) of SAM was produced in fermenter cultivation.

CONCLUSIONS

A new SAM-accumulating strain, S. cerevisiae H5MR83, was obtained through spaceflight culture and genetic modification. Under optimal conditions, SAM production was increased to a relative high level in our study.

SIGNIFICANCE AND IMPACT OF THE STUDY

Through comprehensive application of multiple methods including spaceflight culture, genetic modification and optimizing cultivation, the yield of SAM could be increased by 6·4 times compared to that in the control strain H5. The obtained S. cerevisiae H5MR83 produced 7·76 g l(-1) of SAM in the flask cultures, a significant improvement on previously reported results. The SAM production period with S. cerevisiae H5MR83 was 84 h, which is shorter than previously reported results. Saccharomyces cerevisiae H5MR83 has considerable potential for use in industrial applications.

Genetic modification and bioprocess optimization for S-Adenosyl-L-methionine biosynthesis

S-Adenosyl-L-methionine is an important bioactive sulfur-containing amino acid. Large scale preparation of the amino acid is of great significance. S-Adenosyl-L-methionine can be synthesized from L-methionine and adenosine triphosphate in a reaction catalyzed by methionine adenosyltransferase. In order to enhance S-adenosyl-L-methionine biosynthesis by industrial microbial strains, various strategies have been employed to optimize the process. Genetic manipulation has largely focused on enhancement of expression and activity of methionine adenosyltransferase. This has included its overexpression in Pichia pastoris, Saccharomyces cerevisiae and Escherichia coli, molecular evolution, and fine-tuning of expression by promoter engineering. Furthermore, knocking in of Vitreoscilla hemoglobin and knocking out of cystathionine-β-synthase have also been effective strategies. Besides genetic modification, novel bioprocess strategies have also been conducted to improve S-adenosyl-L-methionine synthesis and inhibit its conversion. This has involved the optimization of feeding modes of methanol, glycerol and L-methionine substrates. Taken together considerable improvements have been achieved in S-adenosyl-L-methionine accumulation at both flask and fermenter scales. This review provides a contemporary account of these developments and identifies potential methods for further improvements in the efficiency of S-adenosyl-L-methionine biosynthesis.

Structural and functional characterization of S-adenosylmethionine (SAM) synthetase from Pichia ciferrii

S-adenosylmethionine synthetase (SAM-s) catalyzes the synthesis of S-adenosylmethionine (SAM), which is essential for methylation, transcription, proliferation, and production of secondary metabolites. Here SAM-s from Pichia ciferrii were selectively cloned using RNA CapFishing and rapid amplification of cDNA ends (RACE). The putative full-length cDNA of SAM-s encoded a 383 amino acid protein (42.6 kDa), which has highly conserved metal binding sites, a phosphate-binding site, and functionally important motifs. The corresponding enzyme was over-expressed in a heterologous host of Pichia pastoris, and then purified to a homogenous form. Enzyme kinetics, immunoblotting, circular dichroism (CD), high performance liquid chromatography (HPLC), and molecular modeling were conducted to characterize the SAM-s from P. ciferrii. Structural and functional studies of SAM-s will provide important insights for industrial applications.

GAP promoter library for fine-tuning of gene expression in Pichia pastoris

A library of engineered promoters of various strengths is a useful genetic tool that enables the fine-tuning and precise control of gene expression across a continuum of broad expression levels. The methylotrophic yeast Pichia pastoris is a well-established expression host with a large academic and industrial user base. To facilitate manipulation of gene expression spanning a wide dynamic range in P. pastoris, we created a functional promoter library through mutagenesis of the constitutive GAP promoter. Using yeast-enhanced green fluorescent protein (yEGFP) as the reporter, 33 mutants were chosen to form the functional promoter library. The 33 mutants spanned an activity range between ∼0.6% and 19.6-fold of the wild-type promoter activity with an almost linear fluorescence intensity distribution. After an extensive characterization of the library, the broader applicability of the results obtained with the yEGFP reporter was confirmed using two additional reporters (β-galactosidase and methionine adenosyltransferase [MAT]) at the transcription and enzyme activity levels. Furthermore, the utility of the promoter library was tested by investigating the influence of heterologous MAT gene expression levels on cell growth and S-adenosylmethionine (SAM) production. The extensive characterization of the promoter strength enabled identification of the optimal MAT activity (around 1.05 U/mg of protein) to obtain maximal volumetric SAM production. The promoter library permits precise control of gene expression and quantitative assessment that correlates gene expression level with physiologic parameters. Thus, it is a useful toolbox for both basic and applied research in P. pastoris.

Enzymatic and whole cell catalysis: finding new strategies for old processes

The use of enzymes and whole bacterial cells has allowed the production of a plethora of compounds that have been used for centuries in foods and beverages. However, only recently we have been able to master techniques that allow the design and development of new biocatalysts with high stability and productivity. Rational redesign and directed evolution have lead to engineered enzymes with new characteristics whilst the understanding of adaptation mechanisms in bacterial cells has allowed their use under new operational conditions. Bacteria able to thrive under the most extreme conditions have also provided new and extraordinary catalytic processes. In this review, the new tools available for the improvement of biocatalysts are presented and discussed.

Generation of mutant of Rhodococcus rhodochrous PA-34 through chemical mutagenesis for hyperproduction of nitrile hydratase

Rhodococcus rhodochrous PA-34 has been reported to produce nitrile hydratase enzyme that converts 3-cyanopyridine to nicotinamide. A mutant of R. rhodochrous PA-34 was generated through chemical mutagenesis using N-methyl-N-nitro-N-nitrosoguanidine (MNNG) that exhibited 2 times higher nitrile hydratase activity as compared to wild strain. The reaction conditions using resting cells of this mutant strain for the conversion of nicotinamide were optimized. Under the optimized reaction conditions the mutant strain exhibited maximum nitrile hydratase activity [7.8 U/mgdcm (milligram dry cell mass)] at 55 degrees C in 0.3 M potassium phosphate buffer (pH 5.5).

Open access to sequence: browsing the Pichia pastoris genome

The first genome sequences of the important yeast protein production host Pichia pastoris have been released into the public domain this spring. In order to provide the scientific community easy and versatile access to the sequence, two web-sites have been installed as a resource for genomic sequence, gene and protein information for P. pastoris: A GBrowse based genome browser was set up at and a genome portal with gene annotation and browsing functionality at . Both websites are offering information on gene annotation and function, regulation and structure.

In addition, a WiKi based platform allows all users to create additional information on genes, proteins, physiology and other items of P. pastoris research, so that the Pichia community can benefit from exchange of knowledge, data and materials.

Optimization of L: -methionine feeding strategy for improving S-adenosyl-L: -methionine production by methionine adenosyltransferase overexpressed Pichia pastoris

The recombinant Pichia pastoris harboring an improved methionine adenosyltransferase (MAT) shuffled gene was employed to biosynthesize S-adenosyl-L: -methionine (SAM). Two L: -methionine (L: -Met) addition strategies were used to supply the precursor: the batch addition strategy (L: -Met was added separately at three time points) and the continuous feeding strategies (L: -Met was fed continuously at the rate of 0.1, 0.2, and 0.5 g l(-1) h(-1), respectively). SAM accumulation, L: -Met conversion rate, and SAM productivity with the continuous feeding strategies were all improved over the batch addition strategy, which reached 8.46 +/- 0.31 g l(-1), 41.7 +/- 1.4%, and 0.18 +/- 0.01 g l(-1) h(-1) with the best continuous feeding strategy (0.2 g l(-1) h(-1)), respectively. The bottleneck for SAM production with the low L: -Met feeding rate (0.1 g L(-1) h(-1)) was the insufficient L: -Met supply. The analysis of the key enzyme activities indicated that the tricarboxylic acid cycle and glycolytic pathway were reduced with the increasing L: -Met feeding rate, which decreased the adenosine triphosphate (ATP) synthesis. The MAT activity also decreased as the L: -Met feeding rate rose. The reduced ATP synthesis and MAT activity were probably the reason for the low SAM accumulation when the L: -Met feeding rate reached 0.5 g l(-1) h(-1).