Whole-transcriptome analysis of oxidative stress response genes in carotenoid-producing Enterococcus gilvus

Abstract

Whole-transcriptome analysis of aerobic stress response gene in Enterococcus gilvus was performed using RNA-sequencing to identify carotenoid-based stress response genes in lactic acid bacteria. The expression of gene responsible for pyruvate dehydrogenase complex synthesis was highly upregulated after aerobic treatment. In addition, the expression of transcriptional regulator spx and genes encoding UvrABC system protein was also upregulated.

Keywords:

• Transcriptome
• carotenoid
• oxidative stress
• lactic acid bacteria

Oxidative stress can damage lactic acid bacteria (LAB) cells [1]. It is important to investigate oxidative stress mechanism in LAB because antioxidants produced by LAB could attenuate colitis [2], and stress tolerance mechanisms improve the activity of LAB under stress conditions, such as oxygen, low-pH, and the gut environment [3–5]. With advancements in genomic techniques over the past few years, transcriptome analyses using RNA-sequencing and microarray have been performed to investigate various stress response networks in LAB [6,7].

Our previous study has demonstrated that diaponeurosporene, a yellow-pigmented carotenoid, could enhance oxidative stress tolerance in LAB [8], suggesting that carotenoid production is one of the stress response mechanisms in LAB. Diaponeurosporene is a triterpenoid produced by various microorganisms, including Lactobacillus, Enterococcus, and Staphylococcus [8–10]. In LAB, diaponeurosporene is produced via mevalonate pathway [11]. The resultant five-carbon precursor isopentenyl diphosphate (IPP) or its isomer dimethylallyl diphosphate (DMAPP) is subsequently converted to farnesyl diphosphate (FPP). Finally, FPP is converted to diaponeurosporene by crtM and crtN, encoding dehydrosqualene (diapophytoene) synthase and dehydrosqualene desaturase, respectively [12].

Carotenoids are antioxidants, and their production is linked to photo-oxidative stress [13]. In eukaryotic microorganisms, such as Phaffia rhodozyma (Xanthophyllomyces dendrorhous), photo-oxidative stress generates singlet oxygen, which leads to a subsequent increase in carotenoid levels [14]. In prokaryotic microorganism Thermus thermophiles, light induces carotenoid production [15]. In E. gilvus, diaponeurosporene production is not induced by light; however, diaponeurosporene production can be increased under aerobic conditions rather than anaerobic conditions [12,16]. Our previous gene expression analysis targeting the isoprenoid biosynthesis pathway has shown that gene expression levels of crtM and crtN were induced under aerobic conditions. In addition, the expression levels of almost all genes associated with conversion from acetyl-CoA to FPP (hmgr, mvk, pmvk, mpd, ipi, and ispA) significantly increased under aerobic conditions [12]. However, changes in upstream gene expression of acetyl-CoA production genes, such as pyruvate metabolism genes linked to carotenogenesis in LAB, remain unclear.

Regarding stress response factors linked to carotenogenesis, Takano et al. [15] have reported that light induced carotenoid production and gene expression of the DNA repair system in T. thermophiles. Another previous study has indicated that the accumulation of global regulator Spx led to a decrease in carotenoid production in S. aureus [17]. In LAB, the factors responsible for regulating carotenoid production and other concomitant damage repair systems remain unknown. Thus, in the present study, whole-transcriptome analysis using RNA-sequencing was performed to identify other upregulated genes (particularly, targeting the upstream genes of Acetyl-CoA responsible for biosynthesis, regulation, and the DNA repair system) with the same pattern of upregulation of carotenoid biosynthesis genes under aerobic condition, gene expression levels of E. gilvus under aerobic were compared with those under anaerobic conditions.

E. gilvus CR1 was incubated at 30 °C for 24 h in M17 medium (Difco Laboratories, Detroit, MI, USA) supplemented with 0.5% glucose (GM17 medium) under anaerobic conditions [8]. Aerobic conditions were maintained according to a method in a previous report [12]. Briefly, CR1 cells from preculture were inoculated in 200 mL GM17 medium (0.5%, v/v) in an Erlenmeyer flask (500 mL). After static incubation for 4 h in an AnaeroPack system (AnaeroPack, Mitsubishi Gas Chemical, Tokyo, Japan), the culture medium was shaken at 110 rpm at 30 °C for 30 min. For anaerobic conditions, a culture was statically incubated at 30 °C for 4.5 h in an AnaeroPack system.

Total RNA was extracted from harvested E. gilvus cells grown under aerobic and anaerobic conditions using RNeasy Protect Bacteria Mini Kit (QIAGEN Valencia, CA, USA), as described previously [12]. Total RNA was extracted from three independent E. gilvus cell cultures for each culture condition. Equal amounts of total RNA from each culture condition were pooled prior to RNA-seq analysis. The RNA samples were sequenced and analyzed at Hokkaido System Science (Sapporo, Hokkaido). After the elimination of rRNA using Ribo-Zero Magnetic kit (bacteria) (Illumina CA, USA), a cDNA library was prepared using TruSeq RNA Sample Prep kit (Illumina). The cDNA was sequenced using paired-end methods (100 bp) by Illumina Hiseq. Pair-end sequences were mapped to E. gilvus ATCC BAA-350 genomes. Gene expression level of CR1 grown under aerobic conditions was compared with that of CR1 grown under anaerobic condition. To confirm RNA-seq results, real-time quantitative reverse transcription PCR (qRT-PCR) was performed, as described previously [12]. Primer sequences are listed in Supplemental Table S1. Primers were designed using sequence information for E. gilvus (ASWH01000001). Gene expression ratios were normalized to 16S rRNA and calculated using the 2^−ΔΔCt method. Relative gene expression levels in E. gilvus cells without aerobic treatment (anaerobic conditions) were set at 1. Student’s t-test was performed for statistical comparisons, and p < 0.05 was considered significant.

Results of transcriptome analysis by RNA-seq revealed that 439 genes were upregulated (>two-fold increase) and 222 genes were downregulated (0.5 > fold decrease) following aerobic treatment (Supplemental Table S2). First, the RNA-seq results were validated by comparing the observed changes in gene expression levels of isoprenoid biosynthesis pathway with results of our previous study using qRT-PCR [12]. Transcriptome analysis indicated the upregulation of six genes (crtN, crtM, mvk, pmvk, mpd, and ipi) following aerobic treatment (Table 1). Regarding isoprenoid biosynthesis pathway, our results of transcriptome analysis using RNA-seq were similar to the results of a qPCR analysis in our previous report [12].

Table 1. Upregulated genes in isoprenoid biosynthesis pathway.

No.	Gene name	Products	Locus-tag	Ratio (Aerobic/Anaerobic)^a
				RNA-seq (This study)	qPCR^b (Ref. [12])
1	crtN	Dehydrosqualene desaturase	I592_01445	4.8	5.9 ± 0.40
2	crtM	Dehydrosqualene synthase	I592_01446	4.6	5.3 ± 0.37
3	mvk	Mevalonate kinase	I592_00331	1.7	2.3 ± 0.14
4	pmvk	Phosphomevalonate kinase	I592_00329	1.6	1.9 ± 0.13
5	mpd	Diphosphomevalonate decarboxylase	I592_00330	1.5	2.0 ± 0.21
6	ipi	Isopentenyl-diphosphate delta-isomerase	I592_00328	1.4	2.3 ± 0.15
7	ispA	Geranyltranstransferase	I592_02163	1.0	1.3 ± 0.05
8	hmgs	Hydroxymethylglutaryl-CoA synthase	I592_01009	0.9	1.2 ± 0.09
9	hmgr	Acetyl-CoA C-acetyltransferase	I592_01010	0.9	1.2 ± 0.08

^a Fold increase in expression when treated under aerobic condition compared to under anaerobic condition.

^b Results are average of three independent experiments (Mean ± standard deviation). These qPCR results have been represented in our previous reports [12]. The expression levels of these genes excepting hmgs are significantly greater when CR1 was treated under aerobic than under anaerobic conditions (p < 0.05).

Among the highly upregulated genes (Table 2), the following genes encoding pyruvate dehydrogenase complex were highly upregulated under aerobic conditions: pdhA, encoding pyruvate dehydrogenase E1 component subunit α (26.6-fold); pdhB, encoding E1 component subunit β (29.0-fold); dlat, encoding E2 component dihydrolipoamide acetyltransferase (28.3-fold); and dld, encoding E3 component dihydrolipoyl dehydrogenase (29.6-fold). Genes encoding a hypothetical protein were also highly upregulated (Table 2; 41.8-fold and 20.7-fold). The former gene (I592_02660) has rhodanese homology domain associated with sulfur metabolism, whereas the latter gene (I592_02659) has no putative conserved domains according to the NCBI database. In addition, aldB encoding alpha-acetolactate decarboxyrase and als encoding acetolactate synthase, which convert pyruvate to acetoin, were upregulated (29.1-fold and 28.0-fold increase, respectively). Further, the validation of the RNA-seq analysis in Table 2 was confirmed by qPCR analysis. Other studies have also demonstrated an increase in the expression of the pyruvate dehydrogenase complex and als with aeration [18,19]. Our results were partly similar to those of previous studies. Pyruvate dehydrogenase complex converts pyruvate to acetyl-CoA which is the first compound in the isoprenoid (carotenoid) biosynthesis [20]. Although it has been established that pyruvate can be converted to lactate, acetoin, ethanol, and acetate [18], our results showed that pyruvate can also be converted to carotenoid via acetyl-CoA, and genes of pyruvate dehydrogenase complex undergo upregulation with the same pattern of upregulation of isoprenoid biosynthesis genes under aerobic condition .

Table 2. Highly upregulated genes (Top10).

No.	Gene name	Products	Locus-tag	Ratio (Aerobic/Anaerobic)^a
				RNA-seq	qPCR^b
1		Hypothetical protein	I592_02660	41.8	59.4 ± 8.4
2	glo	Glyoxalase	I592_03589	32.8	24.9 ± 1.8
3	dld	Dihydrolipoyl dehydrogenase	I592_01649	29.6	32.8 ± 4.1
4	aldB	Alpha-acetolactate decarboxylase	I592_01868	29.1	26.5 ± 2.2
5	pdhB	Pyruvate dehydrogenase E1 component subunit β	I592_01651	29.0	34.7 ± 6.6
6	dlat	Dihydrolipoamide acetyltransferase	I592_01650	28.3	35.4 ± 6.8
7	als	Acetolactate synthase	I592_01869	28.0	39.7 ± 2.5
8	pdhA	Pyruvate dehydrogenase E1 component subunit α	I592_01652	26.6	32.6 ± 3.7
9		Small integral membrane protein	I592_03691	20.9	25.1 ± 1.8
10		Hypothetical protein	I592_02659	20.7	11.9 ± 2.6

^a Fold increase in expression when treated under aerobic condition compared to under anaerobic condition.

^b Results are averages of three independent experiments (Mean ± standard deviation). The expression levels of these genes are significantly greater when CR1 was treated under aerobic than under anaerobic conditions (p < 0.05).

Regarding other stress response genes (Table 3), the expression of spx encoding spx/MgsR family transcriptional regulator was highly upregulated, which coincided with the upregulation of general oxidative stress response genes (nox and sod). One of the spx genes is located on a plasmid and another is located on the chromosome in the reference data (ASWH01000001). The two spx genes showed an 11.6- and 3.4-fold increase according to RNA-seq, and a 9.8- and 4.8-fold increase according to qPCR results, respectively. In S. aureus, an yjbH-null mutant led to the accumulation of Spx and decrease in carotenoid production [17]. Although our results showed an increase in the expression of spx and an increase in carotenoid production following aerobic treatment, Spx has been considered to regulate carotenoid production. On the other hand, the catabolite control protein A (ccpA, I592_00780), which has been shown to be a major transcriptional regulator of carbon catabolism under aerobic and respiratory condition [21], was not highly upregulated (1.3-fold increase compared to anaerobic conditions in transcriptome analysis, data not shown). To clarify the relationship between regulators and the expression of carotenoid biosynthesis genes, further studies using the deletion or overexpression of spx and other genes in E. gilvus are required.

Table 3. Other stress response genes which was simultaneously upregulated with carotenoid biosynthesis genes.

No.	Gene name	Products	Locus-tag	Ratio (Aerobic/Anaerobic)^a
				RNA-seq	qPCR^b
1	spx	Spx/MgsR family transcriptional regulator (plasmid)	I592_03537	11.6	9.8 ± 1.4
2	nox	NADH oxidase	I592_01535	9.1	7.2 ± 0.8^c
3	sod	Superoxide dismutase [Fe]	I592_02089	6.1	8.0 ± 0.6^c
4	spx	Spx/MgsR family transcriptional regulator (chromosome)	I592_02651	3.4	4.8 ± 0.8
5	uvrB	UvrABC system protein B	I592_01743	3.3	3.7 ± 0.7
6	uvrA	UvrABC system protein A	I592_01742	3.3	3.7 ± 0.7

^a Fold increase in expression when treated under aerobic condition compared to under anaerobic condition.

^b Results are average of three independent experiments (Mean ± standard deviation). The expression levels of these genes are significantly greater when CR1 was treated under aerobic than under anaerobic conditions (p < 0.05).

^c qPCR results of nox and sod have been represented in our previous reports [12].

Carotenoid production is strongly linked to photo-oxidative stress response mechanism [13]. In Myxococcus xanthas, photo-stress has been shown to generate singlet oxygen (¹O₂), which in turn increases carotenoid production [22]. In T. thermophiles, photo-stress has been shown to activate carotenoid production and the DNA repair system [15]. In our study, the expression levels of uvrA and uvrB, which encode the UvrABC complex and plays a central role in locating and excising DNA lesions [23], were, respectively, 3.3- (3.7-fold in qPCR results) and 3.3-fold (3.7 in qPCR results) higher under aerobic conditions than under anaerobic condition. In addition, transcriptome analysis revealed that the expression of recA, known to be involved in DNA repair [24], was 2.5-fold higher under aerobic conditions than under anaerobic conditions (Supplemental Table S2). In L. helveticus, the expression of uvrA has been shown to be induced by both UV and oxidative stresses [25]. In addition, Maraccini et al. [26] have reported that enterococcal carotenoids can increase photo-stress tolerance. Taken together, these results imply that UvrABC system and carotenoid production function in coordination for the attenuation of photo-oxidative or oxidative stress.

Based on the results of our previous reports [12] and of the present study, we can conclude that aerobic treatment increases gene expression of the pyruvate dehydrogenase complex, which converts pyruvate to acetyl-CoA. Further, this upregulation coincides with upregulation pattern observed in isoprenoid biosynthesis genes in E. gilvus (Supplemental Figure 1). Concurrently, genes encoding the transcriptional regulators Spx, UvrABC repair system, and the acetoin biosynthesis enzyme were upregulated. Regarding other carotenoid biosynthesis genes, diaponeurosporene is converted to staphyloxanthin by oxidase CrtP, glycosyltransferase CrtQ, and acyltransferase CrtO in S. aureus [27]. In E. gilvus, the homolog of these genes was not detected (data not shown). Thus, diaponeurosporene has been considered to be the final compound in isoprenoid biosynthesis pathway of E. gilvus. This is the first report of transcriptome analysis of oxidative stress response genes coinciding with the upregulation of isoprenoid biosynthesis genes. These results provide a novel understanding of carotenoid-based stress response mechanism in LAB.

Author contributions

We declare that all three authors contributed to the design of this study critical and revision of manuscript. Acquisition of the data and drafting this manuscript were mostly done by the corresponding author (Tatsuro HAGI).

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was partially supported by JSPS KAKENHI [grant number 26850176] and [grant number 16K08012].

Supplemental data

Supplemental data for this article can be accessed at https://doi.org/10.1080/09168451.2017.1399790.

Acknowledgments

The authors would like to thank Enago (www.enago.jp) for the English language review.