A 2016⎪Vol. 26⎪No. 0

J. Microbiol. Biotechnol. (2016), 26(3), 498–502http://dx.doi.org/10.4014/jmb.1509.09013 Research Article jmbReview

Improving 3’-Hydroxygenistein Production in Recombinant Pichiapastoris Using Periodic Hydrogen Peroxide-Shocking StrategyTzi-Yuan Wang1†, Yi-Hsuan Tsai2†, I-Zen Yu2, and Te-Sheng Chang2*

1Biodiversity Research Center, Academia Sinica, Taipei 115, Taiwan, Republic of China2Department of Biological Sciences and Technology, National University of Tainan, Tainan, 702, Taiwan, Republic of China

Orobol (3’-hydroxygenistein) was first purified from

Orobus tuberuosus in 1939 [2], and later from Calophyllum

polyanthum [9], Erycibe expansa [10], Sophora japonica (Sophorae

Flos) [3], and Eclipta prostrata [13]. This compound has also

been isolated from the fermentation broth of Aspergillus

niger [15] and from tempeh, a fermented soybean food [6].

Multiple bioactivities related to 3’-hydroxygenistein have

already been reported, including antiproliferative activity

toward T47D tumorigenic breast epithelial cells [11, 16],

HIV-1 integrase inhibitory activity [13], anti-inflammatory

activity [14], hepatoprotective activity [10], and anti-

melanogenesis activity [4], making 3’-hydroxygenistein

industrially important. However, 3’-hydroxygenistein rarely

occurs in nature [1]. Therefore, to overcome the natural

scarcity of 3’-hydroxygenistein, the recombinant Pichia

pastoris X-33 was constructed to produce detectable 3’-

hydroxygenistein from genistein (Figs. 1A and 1B) [4]. The

recombinant strain, harboring a fusion gene (CYP57B3-

sCPR) composed of the CYP57B3 gene from Aspergillus

oryzae and a cytochrome reductase gene (sCPR) from

Saccharomyces cerevisiae, is able to produce 3.5 mg/l of

3’-hydroxygenistein in a 5 L fermenter. However, since

this amount is still insufficient for industrial application,

improving the 3’-hydroxygenistein production by P. pastoris

has become an advanced research topic.

Based on the known antioxidant properties of carotenoids,

Reyes et al. [12] successfully obtained a 3-fold increase in

carotenoid production in an engineered S. cerevisiae strain

via periodic hydrogen peroxide treatment. Therefore,

this study initially assayed the antioxidant activity of 3’-

hydroxygenistein using a 2,2-diphenyl-1-picrylhydrazyl

(DPPH) radical-scavenging activity assay [5]. The results

showed that the IC50 value of 3’-hydroxygenistein (3.9 µM)

for DPPH radical-scavenging activity was lower than that

of ascorbic acid (8.9 µM) and over 400-fold lower than that

of its precursor, genistein (>1,600 µM) (Fig. 1C). Thus, since

the results revealed that 3’-hydroxygenistein is a potent

antioxidant, this encouraged application of the same

above-mentioned “hydrogen peroxide-shocking” strategy

[12] to improve the 3’-hydroxygenistein production in the

X-33 recombinant.

Three parallel cultures (P1 to P3) of the recombinant X-33

were initiated for 20 ml cultivations in YPD medium (1%

yeast extract, 2% peptone, and 2% dextrose), containing

Received: September 4, 2015

Revised: December 11, 2015

Accepted: December 14, 2015

First published online

December 23, 2015

*Corresponding author

Phone: +886-6-2602137;

Fax: +886-6-2606153;

E-mail: mozyme2001@gmail.com

†These authors contributed

equally to this work.

pISSN 1017-7825, eISSN 1738-8872

Copyright© 2016 by

The Korean Society for Microbiology

and Biotechnology

3’-Hydroxygenistein can be obtained from the biotransformation of genistein by the

engineered Pichia pastoris X-33 strain, which harbors a fusion gene composed of CYP57B3 from

Aspergillus oryzae and a cytochrome P450 oxidoreductase gene (sCPR) from Saccharomyces

cerevisiae. P. pastoris X-33 mutants with higher 3’-hydroxygenistein production were selected

using a periodic hydrogen peroxide-shocking strategy. One mutant (P2-D14-5) produced

23.0 mg/l of 3’-hydroxygenistein, representing 1.87-fold more than that produced by the

recombinant X-33. When using a 5 L fermenter, the P2-D14-5 mutant produced 20.3 mg/l of 3’-

hydroxygenistein, indicating a high potential for industrial-scale 3’-hydroxygenistein

production.

Keywords: 3’-Hydroxygenistein, orobol, CYP57B3, Pichia pastoris

499 Wang et al.

J. Microbiol. Biotechnol.

100 µg/ml zeocin, 250 µM δ-aminolevulinic acid, and

500 µM genistein (Fig. 1). After 48 h of incubation, 0.1 ml of

each culture was mixed with 0.9 ml of glycerol as a stock

and stored at −80°C. A 0.5 ml sample from each culture was

also mixed with 0.5 ml of methanol for 3’-hydroxygenistein

analysis using ultra-performance liquid chromatography

(UPLC). Meanwhile, another 0.2 ml from each culture was

mixed with hydrogen peroxide and reacted at 25°C for

30 min, after which 0.01 ml of the hydrogen peroxide

mixture was inoculated into another 20 ml of fresh YPD

medium and cultivated following the method of Reyes et al.

[12]. This culture process was continued until the cultured

yeast stopped growing. Additionally, for comparison, two

cultures (P1 and P2) were initially treated with 250 mM

and then with 500 mM hydrogen peroxide after 20 days,

whereas P3 was initially treated with 500 mM hydrogen

peroxide. After the periodic hydrogen peroxide-shocking

treatment, six subcultures (as indicated by the arrows in

Fig. 2) that showed hyperproduction of 3’-hydroxygenistein

were selected to isolate the best mutant. The −80°C stocks

of the six selected subcultures were revived on YPD plates

by incubation at 28°C for 48 h. Five colonies were then

randomly selected from each subculture and cultured in

20 ml of a YPD medium as seed cultures. After 48 h of

incubation, each seed culture was transferred into three

20 ml batch cultures. Following a 72-h incubation, the 3’-

hydroxygenistein production was determined using UPLC.

The highest producer among the five colonies from each of

the six subcultures was then selected for a final comparison.

Fig. 2D shows that all the candidates produced significantly

higher concentrations of 3’-hydroxygenistein than the

original recombinant X-33, where P2-D14-5 showed the

highest 3’-hydroxygenistein production.

We then identified the genotype and phenotype differences

between the recombinant X-33 and P2-D14-5 (Fig. 3). In the

genotype analysis, the fusion gene CYP57B3-sCPR was

amplified with primers (forward, 5’-ATGATAGGGACG

GTCTTGGAC-3’; reverse, 5’-GACATCTTCTTGGTATCT

Fig. 1. 3’-Hydroxygenistein production from biotransformation of genistein by recombinant X-33.

(A) UPLC profile of fermentation broth after 72 h of cultivation of recombinant X-33 in a 5 L fermenter. The operational conditions for the UPLC

analysis were as previously described [4]. (B) Diagram of the biotransformation. (C) DPPH radical-scavenging activity of two soy isoflavones and

ascorbic acid. The DPPH scavenging activity was determined as previously described [5]. The IC50 values represent the concentrations required for

50% DPPH radical-scavenging activity. The mean (n = 3) is shown, and the SD is represented by error bars.

Improving 3'-Hydroxygenistein Production in P. pastoris 500

March 2016⎪Vol. 26⎪No. 3

ACCTG-3’) from the isolated genomic DNA and sequenced

(Genomics, Taipei, Taiwan). For the quantitative polymerase

chain reaction analysis, the oligonucleotide primers used

were for CYP57B3 (forward, 5’-GACACCATGACTGCT

GAAGG-3’; reverse, 5’-TTGAGGTACGGCATCTCTTG-3’)

and for the yeast actin ACT1 gene, as an internal control

(forward, 5’- GCGAATACTACGCGAGATGA-3’; reverse,

5’-CCTCAGTTTCCTGCTCCTTC-3’). The relative expression

of the CYP57B3 gene was normalized to that of the

ACT1 gene (∆Ct) and quantified using the ∆∆Ct relative

quantification method. The relative mRNA fold was

expressed as (mutant/parent) = 2[-∆∆Ct]. The genotype analysis

revealed that the CYP57B3-sCPR fusion gene sequences

were identical (data not shown), and no significant expression

differences were observed between the two strains after a

24-h incubation (Fig. 3A). Meanwhile, the phenotype

analysis revealed that the 3’-hydroxygenistein production

at 72 h of cultivation was 1.87-fold higher in P2-D14-5

(23.0 mg/l) than in the recombinant X-33 (12.3 mg/l);

however, the growth rates of the two strains were the same

(0.19 gl-1h-1) (Fig. 3B). At present, there is no clear explanation

for the time delay observed between the mRNA expression

and 3’-hydroxygenistein production, although this may

reflect differences in protein synthesis or trafficking. In

addition, both strains showed higher hydrogen peroxide

resistance when cultivated in the presence of genistein for

Fig. 2. Periodic hydrogen peroxide-shocking of recombinant X-33.

(A to C) Hydrogen peroxide used (■ ) and 3’-hydroxygenistein production (●) profiles of three independent cultivations. (D) 3’-

Hydroxygenistein production by recombinant X-33 and isolated hyperproducing mutants after 72-h incubation in shaking flasks (arrows indicated

in panels A to C). The mean (n = 3) is shown, and the SD is represented by error bars.

501 Wang et al.

J. Microbiol. Biotechnol.

72 h than when cultured without genistein (Fig. 3C).

Therefore, our results indicated that the ability to produce

3’-hydroxygenistein made the strains more resistant to

hydrogen peroxide. Lee et al. [7, 8] also found that periodic

hydrogen peroxide-shocking up-regulated the expression

of genes involved in the aromatic compound-degrading

pathway and led to the accumulation of intracellular

antioxidant aromatic compounds, thereby increasing the

reactive oxygen species-scavenging activity and hydrogen

peroxide stress tolerance. Owing to the potent antioxidant

properties of 3’-hydroxygenistein (Fig. 1C), the accumulation

of 3’-hydroxygenistein in the present study also increased

the hydrogen peroxide stress tolerance (Figs. 3B and 3C).

Thus, the involvement of the P2-D14-5 mutations in oxidation-

reduction-related genes and adjusting the oxidation state to

increase 3’-hydroxygenistein production needs to be

investigated further. Therefore, future studies will focus on

the genome and/or transcriptome of P2-D14-5 to elucidate

the molecular mechanisms regulating 3’-hydroxygenistein

production in Pichia mutants.

In a scale-up experiment, P2-D14-5 was cultivated in a

5-L fermenter (Fig. 4). The resulting cell density in the

fermenter was 2-fold higher than that in flask culture

(Fig. 3B), but the maximal 3’-hydroxygenistein production

(20.3 mg/l) in the fermenter was slightly lower than that

observed in the flask (23.0 mg/l; Fig. 3B). Hence, the optimal

bioreactor operating parameters affecting productivity

remain to be established.

In summary, this study identified 3’-hydroxygenistein as

Fig. 3. Comparative analysis between recombinant X-33 and

P2-D14-5.

(A) Relative mRNA level of the CYP57B3-sCPR fusion gene. The

primer efficiency was 95% for the actin gene and 85% for the

CYP57B3-sCPR gene. The mean (n = 3) is shown, and the SD is

represented by error bars. (B) 3’-Hydroxygenistein production (bar)

and cell growth (curve) of recombinant X-33 (open) and P2-D14-5

(gray) cultured in shaking flasks. (C) Hydrogen peroxide resistance

determined by spot assays. The cells were cultivated with or without

genistein for 72 h. After hydrogen peroxide-shocking treatment, the

treated cells were series-diluted and spotted on YPD plates.

Fig. 4. 3’-Hydroxygenistein production (bar) and cell growth

(curve) profiles of P2-D14-5 in a 5 L fermenter.

The fermentation process was carried out as previously described [4],

with the YPD medium replacing the yeast nitrogen base medium. The

mean (n = 3) is shown, and the SD is represented by error bars.

Improving 3'-Hydroxygenistein Production in P. pastoris 502

March 2016⎪Vol. 26⎪No. 3

a powerful antioxidant with an IC50 value of 3.9 µM for

DPPH radical-scavenging activity, and demonstrated

that 3’-hydroxygenistein production could be successfully

increased 1.87-fold in a P. pastoris mutant when using a

periodic hydrogen peroxide-shocking strategy.

Acknowledgments

This study was funded by the Ministry of Science and

Technology of Taiwan (MOST 103-2221-E-024-009-MY2).

References

1. Chang TS. 2014. Isolation, bioactivity, and production of

ortho-hydroxydaidzein and ortho-hydroxygenistein. Int. J.

Mol. Sci. 15: 5699-5716.

2. Charaux C, Rabate J. 1939. Constitution chimique de l'orobol.

Bull. Soc. Chim. Biol. (Paris) 21: 1330-1333.

3. Chen QC, Zhang WY, Jin W, Lee IS, Min BS, Jung HJ, et al.

2010. Flavonoids and isoflavonoids from Sophorae Flos

improve glucose uptake in vitro. Planta Med. 76: 79-81.

4. Ding HY, Chiang CM, Chen YC, Chang TS. 2015. Identifying

3'-hydroxygenistein as potent melanogenesis inhibitor from

biotransformation of genistein by recombinant Pichia pastoris.

Process Biochem. 50: 1614-1617.

5. Ding HY, Lin PW, Wang HW, Chang TS. 2014. Methanol

partition fraction of ethanol extract of Dioscorea nipponica

Makino inhibits melanogenesis. Trop. J. Pharm. Res. 13: 719-

726.

6. Kiriakidis S, Hogemeier O, Starcke S, Dombrowski F, Hahne JC,

Pepper M, et al. 2005. Novel tempeh (fermented soyabean)

isoflavones inhibit in vivo angiogenesis in the chicken

chorioallantoic membrane assay. Br. J. Nutr. 93: 317-323.

7. Lee JY, Lee HJ, Seo J, Kim ES, Lee HS, Kim P. 2014. Artifical

oxidative stress-tolerant Corynebacterium glutamicum. AMB

Express 4: 15.

8. Lee JY, Seo J, Kim ES, Lee HS, Kim P. 2013. Adaptive

evolutin of Corynebacterium glutamicum resistant to oxidative

stress and its global gene expression profiling. Biotechnol.

Lett. 35: 709-717.

9. Ma CH, Chen B, Qi HY, Li BG, Zhang GL. 2004. Two

pyranocoumarins from the seeds of Calophyllum polyanthum.

J. Nat. Prod. 67: 1598-1600.

10. Matsuda H, Morikawa T, Xu F, Ninomiya K, Yoshikawa M.

2004. New isoflavones and pterocarpane with hepatoprotective

activity from the stems of Erycibe expansa. Planta Med. 70:

1201-1209.

11. Nguyen DT, Hernandez-Montes E, Vauzour D, Schonthal AH,

Rice-Evans C, Cadenas E, Spencer JP. 2006. The intracellular

genistein metabolite 5,7,3',4'-tetrahydroxyisoflavone mediates

G2-M cell cycle arrest in cancer cells via modulation of the

p38 signaling pathway. Free Radic. Biol. Med. 41: 1225-1239.

12. Reyes LH, Gomez JM, Kao KC. 2014. Improving carotenoids

production in yeast via adaptive laboratory evolution.

Metab. Eng. 21: 26-33.

13. Tewtrakul S, Subhadhirasakul S, Cheenpracha S, Karalai C.

2007. HIV-1 protease and HIV-1 integrase inhibitory substances

from Eclipta prostrata. Phytother. Res. 21: 1092-1095.

14. Tewtrakul S, Subhadhirasakul S, Tansakul P, Cheenpracha S,

Karalai C. 2011. Antiinflammatory constituents from Eclipta

prostrata using RAW264.7 macrophage cells. Phytother. Res.

25: 1313-1316.

15. Umezawa H, Tobe H, Shibamoto N, Nakamura F,

Nakamura K. 1975. Isolation of isoflavones inhibiting DOPA

decarboxylase from fungi and streptomyces. J. Antibiot.

(Tokyo) 28: 947-952.

16. Vauzour D, Vafeiadou K, Rice-Evans C, Cadenas E, Spencer

JP. 2007. Inhibition of cellular proliferation by the genistein

metabolite 5,7,3',4'-tetrahydroxyisoflavone is mediated by

DNA damage and activation of the ATR signalling pathway.

Arch. Biochem. Biophys. 468: 159-166.