A second-generation Bacillus cell factory for rare inositol production

Abstract

Some rare inositol stereoisomers are known to exert specific health-promoting effects, including scyllo-inositol (SI), which is a promising therapeutic agent for Alzheimer disease. We recently reported a Bacillus subtilis cell factory that performed the efficient production of SI from the cheapest and most abundant isomer myo-inositol (MI). In the cell factory all “useless” genes involved in MI and SI metabolism were deleted and overexpression of the key enzymes, IolG and IolW, was appended. It converted 10 g/L MI into the same amount of SI in 48 h of cultivation. In this addendum, we discuss further improvement in the cell factory and its possible applications.

Introduction

There are nine natural stereoisomers of inositol (1,2,3,4,5,6-cyclohexanehexol). One of the stereoisomers, scyllo-inositol (SI) is rare but has been regarded as a promising therapeutic agent for Alzheimer disease.¹ In fact, SI has received a fast-track designation from the US Food and Drug Administration for the treatment of mild to moderate Alzheimer disease.

Bacillus subtilis has the ability to metabolize at least three inositol stereoisomers, including SI, myo-inositol (MI), and D-chiro-inositol (DCI) (Fig. 1), and the gene set necessary for their utilization has been characterized.² The iolABCDEFGHIJ operon encodes enzymes involved in multiple steps of inositol metabolism, and transcription of the operon is regulated by the IolR transcriptional repressor.³ In the first step, MI and DCI are converted to scyllo-inosose and 1-keto-D-chiro-inositol, respectively, by the IolG enzyme with NAD⁺ reduction. Next, 1-keto-D-chiro-inositol is isomerized to scyllo-inosose by IolI. B. subtilis possesses two additional inositol dehydrogenases, IolX and IolW, which act specifically on SI with NAD⁺ and NADP⁺ reduction, respectively, to convert it to scyllo-inosose.⁴ In vivo and in vitro analyses have revealed that IolX plays the major physiological role in SI catabolism, whereas IolW efficiently reduces scyllo-inosose into SI with oxidation of NADPH.⁴scyllo-Inosose is metabolized sequentially in multiple steps involving the IolE, IolD, IolB, IolC, IolJ, and IolA enzymes to give common intermediates, dihydroxyacetone phosphate and acetyl-CoA, which enter glycolysis and the TCA cycle, respectively.⁵

Figure 1. Inositol metabolism in B. subtilis. Inositol metabolic pathway of B. subtilis is shown schematically to close up the reactions relevant to conversion of the three inositol stereoisomers including DCI, MI, and SI.

We developed a B. subtilis cell factory to produce SI, where MI was chosen as a starting material that is abundant in nature and cheaply available from rice bran. The first generation of the cell factory was realized in strain TM039, in which three genes including iolR, iolX, and iolI were disrupted and a missense mutation, iolE41, was introduced. Those modifications were introduced to enable the constitutive expression of the iolABCDEFGHIJ operon, including iolG, elimination of SI dehydrogenase activity and inactivation of both isomerization and dehydration of the key intermediate scyllo-inosose, and thus the accumulating scyllo-inosose was expected to be converted readily to SI. In fact, nearly half of the initial MI was converted to SI after 72 h of cultivation, but the strain was found to consume and waste the other half of MI.²

The Second-Generation B. subtilis Cell Factory

The first-generation B. subtilis cell factory wasted half of the MI starting material. To solve this problem, iolABCDEFHIJ, iolX, and iolR were deleted from the chromosome using a marker-free deletion technique⁶ to yield strain MYI04.⁷ As expected, MI was not wasted in MYI04, but the conversion rate of MI to SI was not improved. We accordingly attempted the overexpression of iolG and iolW, which encode the key enzymes for the conversion, under the control of strong and constitutively active promoters, even during stationary growth; the promoters were chosen from global B. subtilis transcriptome analysis data.⁸ Finally, strain KU106, in which iolG and iolW were simultaneously overexpressed under the control of the rpsO promoter, completely converted 10 g/L MI into the same amount of SI in 48 h of cultivation.⁷

Aiming at Further Improvement of the Cell Factory

In our preliminary attempt, the conversion was performed with 50 g/L MI; the resulting amount of SI produced in the medium was no more than 12 g/L. This amount of SI in the medium already exceeds its saturation point in water, but it does not have any cytotoxicity because KU106 was able to grow even in the presence of 20 g/L of SI. The results suggested that other unknown factors could limit conversion capacity. One such factor might be the transport of the substrate and product across the plasma membrane. IolT is responsible for uptake of MI^9,10 and is constitutively expressed in KU106. To test the possible effect of enhanced MI import, we gave it a shot to overexpress iolT but found no obvious elevation in the conversion so far. However, we do not know how SI is secreted. Identification of the SI exporter is currently under investigation, and once such a gene is found, its overexpression will be attempted.

Cofactors Required for Inositol Conversion

The cofactors required for inositol dehydrogenases possibly play important roles in efficient conversion. In the SI-producing cell factory, MI is once converted to scyllo-inosose with NAD⁺ reduction, and then scyllo-inosose is reduced to SI with NADPH oxidation (Fig. 1). NAD⁺ is supposed to be easily regenerated by respiration, given that B. subtilis is a typical aerobic bacterium. However, given that B. subtilis is believed to have no transhydrogenase, NADPH must be regenerated by direct reduction of NADP⁺.¹¹ In the previous study, we showed that the reduction of scyllo-inosose catalyzed by IolW requiring NADPH could be the most important step in SI conversion and also that the reduced amounts of the major nutrient Bacto soytone drastically reduced SI conversion.⁷ The results suggested that higher concentrations of Bacto soytone might stimulate NADPH regeneration. In addition, our preliminary observation suggested that inactivation of some of NADPH regenerating enzymes could result in decreased SI conversion even in the presence of higher concentrations of Bacto soytone. We are currently investigating the mechanism underlying this phenomenon. The regeneration of NADPH would be artificially and efficiently controlled, for instance, by enhancing the expression levels of NADPH regenerating enzymes. This might thus be applied not only to SI conversion but also to the production of other valuable chemicals, such as alcohols and fatty acids, which usually requires reaction steps driven by reducing power of NADPH and/or NADH.

Possible Application to Producing Other Inositols

Not only SI but also some of other rare inositol stereoisomers are known to exert specific and useful biological activity. For example, DCI is claimed for its functional role in insulin action and its deficit in insulin resistance.¹² In addition, epi-inositol was reported to be a possible antidepressant.¹³ As described above, B. subtilis IolG can act on MI and DCI but not on SI, whereas IolW and IolX act on SI, indicating the differential substrate specificities of inositol dehydrogenases. Within the B. subtilis genome are found additional iolG homologs, namely yfiI, ntdC, yrbE, yteT, and yulF,³ and some of their gene products may exhibit inositol dehydrogenase activity. We also characterized three inositol dehydrogenases of Geobacillus kaustophilus, which exhibited mutually differing substrate specificities.¹⁴ This differential substrate specificity indicates that every inositol dehydrogenase has its own preference for dehydrogenation of the secondary alcohol at a specific position of an inositol isomer to generate a ketone group at the same position. Furthermore, there are inosose isomerases that can move the ketone group to the next position, similar to the IolI enzyme.⁵ We are currently collecting possible inositol dehydrogenases and inosose isomerases from databases and natural isolates. Those enzyme genes are to be cloned in combination into the B. subtilis cell factory platform similarly to iolG and iolW, with the aim of identifying combinations that enable the conversion of MI to other rare inositol isomers.

SI Production from Raw Materials

Phytic acid (MI-1,2,3,4,5,6-hexaphosphate) is the principal storage form of phosphorus in many plant tissues, in particular bran and seeds. Phytase is a special class of phosphatases that catalyze the sequential hydrolysis of phytic acid to liberate MI and phosphate.^15-17 Because B. subtilis has high ability to secrete enzymes,^18,19 the B. subtilis cell factory could be modified to secrete phytase, possibly enabling SI production from waste materials such as rice bran rich in phytic acid.

It is known that MI is synthesized from glucose-6-phosphate in two steps. Glucose-6-phosphate is the starting compound of glycolysis, appearing when glucose is incorporated into the cell via the phosphotransferase system.²⁰ Glucose-6-phosphate is converted by inositol-3-phosphate synthase to MI 1-phosphate, which is then dephosphorylated by inositol monophosphatase to yield MI. We have already cloned a heterogeneous inositol-3-phosphate synthase gene from Mycobacterium tuberculosis²¹ and have also identified yktC encoding endogenous inositol monophosphatase in B. subtilis (details of these results will be reported elsewhere). If these two enzyme genes could be efficiently expressed in the future, it might be possible to develop a novel cell factory producing SI from glucose, sucrose, or starch.

Conclusion

SI is a promising therapeutic agent for Alzheimer disease. Here we discussed that the B. subtilis cell factory converting MI into SI could have a potential to be improved further. Our previous study suggested that SI conversion might depend on NADPH-regeneration capacity in the cell, which seemed to be stimulated in the presence of higher concentration of Bacto soytone, and we are currently investigating the mechanism underlying this phenomenon. Applying the B. subtilis cell factory concept, we might be able to produce not only SI but also other rare inositol stereoisomers from raw materials in the future.

Glossary

Abbreviations:

SI

scyllo-inositol

MI

myo-inositol

DCI

D-chiro-inositol

10.4161/bioe.29897

Tanaka K, Tajima S, Takenaka S, Yoshida K. An improved Bacillus subtilis cell factory for producing scyllo-inositol, a promising therapeutic agent for Alzheimer’s disease. Microb Cell Fact. 2013;12:124. doi: 10.1186/1475-2859-12-124.