Introduction

Tobacco is an industrial crop that is widely grown throughout the world. Tobacco leaf consists of lamina and midrib, with midrib accounting for about 25–30% of the leaf weight. Not only lamina but also midrib is widely used as raw materials for cigarette production through the making of reconstituted tobacco sheets or midrib cut. Therefore, tobacco midribs have great value when scientifically processed. Midrib has a lower tar level, so it plays a significant part in reducing the hazards of cigarettes1. The usage of midrib is also beneficial to cost cutting, thus improving the utilization efficiency of tobacco raw materials. However, the disadvantage of midrib is also obvious, among which is substantial higher levels of nitrate content and subsequent higher formation and accumulation of tobacco-specific nitrosamines (TSNAs)2,3 than that in the lamina. Nitrate content in midrib of cured tobacco leaf is usually more than 10 times higher than that in the lamina of the same cured leaf2.

TSNA is prone to induce malignant tumors in animals and was classified as the first class carcinogen by the International Agency for Research on Cancer4. It is well recognized that nitrate is an important precursor of tobacco-specific nitrosamines (TSNAs). Nitrate may easily be reduced to nitrite by microbial activity during leaf curing5 or produce gaseous NOx during leaf storage under warm or hot conditions6, and the subsequent nitrosation of tobacco alkaloids by these nitrosating agents may lead to much increased levels of TSNA formation and accumulation in midrib1. Therefore, the reduction of nitrate content is a key for reducing TSNA formation, and the investigation of the mechanisms of nitrate accumulation in midrib is essential, so as to lay the foundation for reducing nitrate content and TSNA formation in tobacco midribs and their products.

Nitrate (NO3−) is one of the main sources of nitrogen absorption by plants, which will accumulate to a large extent in plant cell vacuoles if not being reduced, reused, or transported into the cytoplasm7. Once absorbed by root cells, a larger proportion is transferred to the shoot, where it is rapidly turned into nitrite by nitrate reductase (NR) and nitrite reductase (NiR), and subsequently incorporated into glutamine by glutamine synthetase (GS), which is metabolized to glutamate (Glu) and glutamine (Gln) by Gln synthetase (GS) and Glu synthase (GOGAT), respectively8,9.

Carbon metabolism is highly correlated with nitrogen metabolism in plants. N assimilation requires both energy and organic carbon (C) which are provided by photosynthesis8,10. A previous study demonstrated that the lowering of pigment content, carbon fixation, and nitrogen assimilation were the main causes of nitrate accumulation in burley tobacco11. Moreover, some genes and transcription factors involved in nitrate transport, signaling, and use efficiency can affect the content of nitrate. Four protein families are known to be involved in nitrate uptake, distribution, or storage: the Nitrate Transporter 1/Peptide Transporter (NPF) family, the Nitrate Transporter 2 (NRT2) family, the Chloride Channel (CLC) family, and the Slow Anion Associated Channel Homolog (SLC/SLAH) family9. AtNPF6.3 (also known as CHLORATE RESISTANT 1, CHL1, or NRT1.1) was the first dual-affinity nitrate transporter and also founctions as a nitrate sensor12. OsNRT1.1B/OsNPF6.5 also operates as a dual-affinity nitrate transporter and mediates nitrate uptake and root-to-shoot transport13. AtNPF7.3/NRT1.5 modulates xylem loading of nitrate in root pericycle cells14,15. AtNPF7.2/NRT1.8 is chiefly expressed in xylem parenchyma cells and more nitrate is found in xylem sap in npf7.2 mutants14. Interestingly, the functions of NPF7.2 and NPF7.3 are antagonistic, and expressions of NPF7.2 and NPF7.3 are inversely regulated upon stress treatments13,16. AtNPF5.11, AtNPF5.12, and AtNPF5.16, localized in tonoplast, were proposed to mediate nitrate efflux from vacuoles and to regulate nitrate distribution between roots and shoots17. Moreover, AtNPF6.2/NRT1.4 is predominantly expressed in the petiole and midrib of leaves18. Compared with the wild type, less nitrate accumulates in the petiole, but more nitrate is detected in the leaf blade of npf6.2 mutants, indicating that NPF6.2 participates in nitrate storage of the petiole. NLPs have been suggested to be involved in mediating the early N response. For instance, transcription factor NIN-LIKE PROTEIN 7 (NLP7) was identified as a primary regulator in nitrate response in Arabidopsis and regulates the expression of several nitrate-responsive genes including NITRATE REDUCTASE 1 (NIA1), NIA2, NRT2.1, and NRT2.219,20. Furthermore, OsNLP4 transactivates the NRE motif at the promoter of OsNiR encoding nitrite reductase that is a key enzyme determining nitrogen assimilation in rice21. In addition to the genes above, other genes which mediate, for example, nitrate signalling and transcription factors play an essential role in nitrate metabolism22,23.

In recent year, the midrib is also widely used in cigarette production. However, the nitrate content of the midrib is markedly higher than that of the lamina. Shi et al.11 compared the carbohydrate and nitrate accumulation of flue-cured tobacco with that of burley tobacco, while inclusive studies about lamina and midrib have never been reported. This study aimed to identify the physiological and transcriptome differences between the lamina and midrib, so as to reveal the mechanism of nitrate accumulation in midrib. Significant findings were obtained that would provide insight into the difference in carbon and nitrogen metabolism and valuable gene resources that might explain the reason why midrib had higher nitrate content.

Results

Differences in enzymes activities and nitrogen compounds between lamina and midrib

The results showed that pigment content, enzyme activities, and nitrogen compounds were different between lamina and midrib (Fig. 1a–l). Chlorophyll a content, chlorophyll b, and carotenoid contents were significantly lower in midrib than those in the lamina. Also, SSA was always lower in midrib than that in the lamina. Lower pigment content may have an influence on carbon fixation and lead to low carbohydrate accumulation in midrib. Also the nitrate reductase activity (NRA) and glutamine synthetase activity (GSA) were lower in midrib than in the lamina. In addition, NH4-N, NO2-N, total nitrogen content (TN), and soluble protein content in midrib were dramatically lower than those in midrib while the NO3-N content and the ratio of NO3-N/total nitrogen content (TN) were significantly higher, indicating that the ability of nitrate reduction and assimilation in lamina was higher than midrib. It is noteworthy that the NO3-N content accumulated to 25.96 mg g−1 in midrib and was 3.1 times than that in the lamina, which might be due to the weak ability of nitrogen reutilization, leading to nitrate accumulation in midrib.

Figure 1
figure 1

(a) NO3-N content; (b) NO2-N content; (c) total nitrogen content; (d) NO3-N/TN; (e) nitrate reductase activity; (f) glutamine synthetase activity; (g) NH4-N content; (h) soluble protein content; (i) chlorophyll a; (j) chlorophyll b; (k) carotene content; (l) sucrose synthetase. Symbols ** and *indicates that the significant differences between lamina and midrib at 0.01 and 0.05.

Full size image

Quality control, gene expression, and correlation analysis between samples

After filtering the raw reads, a high rate of clean reads from each sample was achieved. In short, the mapping rates of all the samples to the reference genome were above 93%, the GC content of all samples was stable with the distribution ranging from 43.16 to 44.09% and the QC30 value of all samples was above 91% (Table 1), implying successful library construction and RNA sequencing. As shown in Fig. 2a, the FPKM expression levels for each sample were calculated. In addition, the range of correlation coefficients among intra-class was distributed between 0.98 and 1.00 (Fig. 2b). And principal component analysis (PCA) of the data profiles from all 6 samples revealed a high correlation among all samples (Fig. 2c). These results demonstrated that the sequencing data in the present study were adequately representative and valid.

Table 1 The primers used in real-time PCR.
Full size table
Figure 2
figure 2

(a) Gene expression levels in samples. (b) Heatmap of correlation between samples. (c) The results of principal component analysis.

Full size image

Differentially expressed gene (DEG) selection, Gene Ontology (Go) enrichment, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of DEGs

The fold change (FC) > 2 or FC < 0.5, and a P.adjust < 0.05, were used thresholds to determine the DEGs. A total of 7560 DEGs (3446 upregulated and 4114 downregulated) were identified between the lamina and the midrib groups (Fig. 3a). And the volcano of differentially expressed genes between the lamina and the midrib was achieved (Fig. 3b).

Figure 3
figure 3

(a) The number of differentially expressed genes between lamina and midrib. (b) The volcano of differentially expressed genes between lamina and midrib. (c,d) Go enrichment of differentially expressed genes between lamina and midrib. (c,d) KEGG enrichment of differentially expressed genes between lamina and midrib.

Full size image

The DEGs in lamina vs midrib were further analyzed using Gene Ontology (Go) enrichment, and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses (Fig. 3c–f). In detail, the down-regulated DEGs in lamina vs midrib were significantly enriched in photosynthesis-antenna proteins (ko00196), photosynthesis (ko00195), porphyrin and chlorophyll metabolism (ko00860), carbon fixation in photosynthetic organisms (ko00710), carotenoid biosynthesis (ko00906), photosystem II (GO:0009523), and photosystem I (GO:0009522). The up-regulated genes were mostly involved in cell wall organization (GO:0071555), phenylpropanoid biosynthesis (ko00940), steroid biosynthesis (ko00100), xyloglucan metabolic process (GO:0010411), pentose and glucuronate interconversions (ko00040), and transporter activity (GO:0005215).

Comparative analysis of DEGs correlated with carbon and nitrogen metabolism

Transcriptome sequencing technology provides a large amount of information regarding the DEGs that are involved in specific biological responses. Figure 4 showed that porphyrin and chlorophyll metabolism, carotenoid biosynthesis, photosynthesis-antenna proteins, photosynthesis, carbon fixation in photosynthetic organisms, starch and sucrose metabolism, nitrogen metabolism, and biosynthesis of amino acids were lower in midrib than in lamina. In addition, we searched the genes involved in porphyrin and chlorophyll metabolism (LOC107777980, LOC107786828, and LOC107788874) (Fig. 4a), carotenoid biosynthesis (LOC107772713, LOC107763949, LOC107763628, and (LOC107797654) (Fig. 4b), photosynthesis-antenna proteins (LOC107773808, LOC107776229, LOC107778264, LOC107782430, LOC107772663, LOC107773232, and LOC107764358) (Fig. 4c), photosynthesis (LOC107763149, LOC107810205, LOC107784985, LOC107766588, and LOC107768924) (Fig. 4d), carbon fixation in photosynthetic organisms (LOC107780142, LOC107777241, LOC107771723, and LOC107766567) (Fig. 4e), starch and sucrose metabolism (LOC107761864, LOC107825407, and LOC107771409) (Fig. 4f), nitrogen metabolism (LOC107768773, and LOC107766022) (Fig. 4g) and biosynthesis of amino acids (LOC107785928, LOC107784332, LOC107766022, and LOC107794948) (Fig. 4h) were greatly suppressed in midrib. To explore the reason why midrib holds higher nitrate than lamina, we analyzed the differences in gene expression levels of nitrate response, transport, and assimilition. The results showed that genes of NLP4 (LOC107782967), NLP7 (LOC107806749), NPF2.13 (LOC107775674), NPF3.1 (LOC107829632), NPF6.3 (LOC107799198), NPF7.3 (LOC107768465), NIA (LOC107794079), GS (LOC107802035), and GOGAT (LOC107781744) were down-regulated in midrib while genes of NPF1.2 (LOC107789301) and NPF7.2 (LOC107770138) were up-regulated in (Fig. 4i), which might also be the cause for higher nitrate content in the midrib.

Figure 4
figure 4

(a) Porphyrin and chlorophyll metabolism; (b) carotenoid biosynthesis; (c) photosynthesis-antenna proteins; (d) photosynthesis; (e) carbon fixation in photosynthetic organisms; (f) starch and sucrose metabolism; (g) nitrogen metabolism; (h) biosynthesis of amino acids; (i) expression of genes involved in nitrate response, transport and assimilation. Box-whisker plot represents dispersity of minimum, first quartile, median, third quartile in genes expression level of treatments. Y-axis represents expression level. The columns represent six samples. The name of gene is on the right side, the up-or down-regulated proteins are indicated in red and green, respectively. The intensity of the colors increases with increasing expression level as noted on the color bar on the right side.

Full size image

Expression levels of genes related to nitrate transport

Nitrate transporters play an essential role in nitrogen metabolism. The expression level of genes involved in nitrate transporting (LOC107782967, LOC107806749, LOC107775674, LOC107829632, LOC107799198, LOC107768465) was down-regulated while LOC107789301 and LOC107770138 were up-regulated in midrib compared to that in the lamina (Fig. 5). And the qRT-PCR results showed that the expression patterns of the eight genes were identical to those detected by transcriptome sequencing, which confirmed the reliability of RNA-seq data and explained the reason why nitrate content was higher in midrib than that in the lamina.

Figure 5
figure 5

Expression levels of genes related to nitrate transport by qRT-PCR. The x-axis indicates the two samples. YP: lamina of flue-cured tobacco; ZM: midrib of flue-cured tobacco. The left y-axis indicated relative expression level of qRT-PCR. Error bars represent standard error of mean.

Full size image

Discussion

In recent year, the midrib has been widely used in cigarette production in the form of tobacco sheets. However, our study showed that the midrib had higher NO3-N content of more than 3 times than lamina (Fig. 1a), which is not beneficial to tobacco safety and harm reduction. One strategy to decrease the content of nitrate is to identify the physiological and molecular mechanisms contributing to nitrate accumulation in the midrib. In the studies presented here, the pots experiment was employed to study the physiological and transcriptome differences between lamina and midrib. Overall, the present study demonstrated that the expression of genes involved in porphyrin and chlorophyll metabolism, carotenoid biosynthesis, photosynthesis-antenna proteins, photosynthesis, carbon fixation in photosynthetic organisms, starch and sucrose metabolism, nitrogen metabolism, and biosynthesis of amino acids were significantly lower in midrib than in the lamina (Fig. 4a–h), which might be the cause for higher nitrate accumulation in the midrib.

It has long been recognized that chlorophyll content is used as an indicator of photosynthetic capacity and photosynthesis and C metabolism functions to provide both energy and C skeletons for plant growth and N assimilation8,10. Our results showed that the midrib had lower chlophyll a chlophyll b, carotenoid, and SSA than that of the lamina (Fig. 1i–l). The previous study has shown that the midribs tend to have fewer chloroplasts in C3 and CAM plants, which might be the reason for lower chlophyll content in the midrib24. More than 30 genes are involved in the chlophyll biosynthesis pathway and any genetic mutation may affect the synthesis of chlorophyll25. HEMA1 is considered to play the major role in tetrapyrrole biosynthesis and antisense HEMA1 Arabidopsis plants showed decreased levels of chlophyll26. In Arabidopsis thaliana, Alexey et al.27 showed that the chlorophyll biosynthesis pathway was suppressed in this ChlI mutant. In accordance with this, our results found that some key genes related to pigment biosynthetic process and C metabolism were significantly down-regulated in the midrib, including LOC107777980 (MgPME)28, LOC107783891 (CHLI)29, LOC107763283 (hemA), and LOC107783257 (CHLP)30, which play crucial roles in chlophyll biosynthesis, LOC107772713 (PSY2)31,32, which encodes phytoene synthase and controls the carbon flux through the carotenoid biosynthetic pathway, LOC107785687 (SPS2) edcoding sucrose-phosphate synthase that plays the role of rate-limiting steps in sucrose synthesis in higher plant33. The down-regulation of these genes might decrease the chlorophyll formation and photosynthesis efficiency in the midrib. Besides, the lower chlorophyll content resulted in a decrease of the chlorophyll a/b binding proteins in midrib. LOC107772663 (LHCb1), which is one of the most abundant chloroplast proteins in plants and mainly functions to collect and transfer light energy to photosynthetic reaction centers34, was significantly repressed in midrib. Previous studies showed that that in CAM plants the photochemical parameters describing the performance of PSII were significantly lower in the midribs than in the interveinal leaf area, which reduced the photosynthesis24. In the present study, PSI, PSII, and photosynthetic electron transport are key components in the photosynthetic pathway. While LOC107810205 (PsbR), LOC107784985 (PsaO) and LOC107803171 (petF), which were involved in PSI, PSII, and photosynthetic electron transport were down regulated in midrib. LOC107771723 (rbcS), which encodes a key enzyme in the calvin cycle and assimilates atmospheric CO2 into the biosphere35, was also down regulated in midrib. This is consistent with the physiological differences between lamina and midrib. Carbon metabolism is closely related to nitrogen metabolism. The lower capacity of photosynthesis and carbon fixation might influence the nitrogen metabolism and resulted in higher level nitrate in the midrib.

NR and GS are two of the most important enzymes in N assimilation9. The ammonium taken up by AMTs or derived from nitrate is used to produce a variety of amino acids via the GS/GOGAT cycle8. Lu et al.36 showed that expression of a constitutively activated nitrate reductase (NR) enzyme dramatically decreases leaf nitrate levels in burley tobacco. Meanwhile, recent literature also suggests that the overexpression of GS is able to increase the activity of GS and promote N assimilation efficiency37. NLP7 is a primary regulator in nitrate response and regulates the expression of several nitrate-responsive genes including NIA1, NIA2, NRT2.1, and NRT2.219,20. And OsNLP4 transactivats the NRE motif at the promoter of OsNiR encoding nitrite reductase in rice21. Xiang et al.38 has demonstrated that NLP7-overexpressing plants showed lower nitrate accumulation. In this study, NLP7 and NLP4 were down-regulated in the midrib, which was inconducive to the decrease of nitrate accumulation in the midrib. Further investigation of the expression of genes encoding nitrate response, transport, and assimilation led to the discovery of nitrate response genes (NPF6.3, NLP4, and NLP7), nitrate transporters (NPF2.13, NPF3.1, NPF7.3, NPF1.2, and NPF7.2), and nitrate assimilation genes (NIA, GS and GOGAT) with contrasting transcriptional responses in lamina and midrib. And our results showed that midrib was lower in NR activity, GS activity, NH4-N, and soluble protein content while higher in NO3-N and NO3-N/TN than midrib, suggesting that midrib might retain a weaker capacity of nitrate assimilation. In plants, NO3− accumulation depends on its absorption, transport, and metabolism, among which there is a close interdependency that facilitates the coordinated regulation of NO3− accumulation in plants. NPF7.3/NRT1.5 mediates efflux of NO3– to the xylem vessels, whereas NPF7.2/NRT1.8 performs the opposite function and retrieves NO3–from the xylem sap into xylem parenchyma cells15,39. NPF2.13 can facilitate outward nitrate transport by phloem loading. Moreover, NPF1.2 is expressed in the companion cells of the major veins in expanded leaves and involved in diverting root-derived nitrate into phloem in the major vein of mature and expanded leaves40. qRT-PCR results showed that the nitrate transporter genes LOC107782967 (NLP4), LOC107806749 (NLP7), LOC107775674 (NPF2.13), LOC107829632 (NPF3.1), LOC107799198 (NPF6.3), LOC107768465 (NPF7.3) were down-regulated while LOC107789301 (NPF1.2) and LOC107770138 (NPF7.2) were up-regulated in midrib but not in the lamina, indicating that midrib had poor ability in reallocation nitrate transported by roots.

In Conclusion, significant differences were observed in nitrate accumulation between lamina and midrib of flue-cured tobacco. Pigment content and SSA in midrib were significantly lower than that in the lamina, which resulted in insufficient C skeleton for nitrogen metabolism. Meanwhile, the greater nitrate accumulation was probably conferred by more disadvantageous aspects such as weak nitrogen reduction, weak nitrogen assimilation, poor ability in reallocation, and high capacity of accumulating nitrate in midrib than in the lamina. The above insights to the physiological and molecular basis of carbon and nitrogen differences in lamina and midrib would be helpful for providing direction for decreasing nitrate accumulation in the midrib.

Materials and methods

Plant material and study design

The flue-cured tobacco variety K326 was used in this study. Seeds were sterilized with 2% (v/v) sodium hypochlorite for 5 min twice and then were sown in a floating system. Forty days after sowing, seedlings were transplanted in 7.1 cm × 7.8 cm (diameter × depth) plastic pots and cultivated with Hoagland solution. Pot experiments were conducted on substrate culture in the greenhouse that maintained a temperature of 25 ± 2 °C, an average photosynthetic photon flux density of 400 μmol m−2 s−1, and relative humidity of 80%. Laminas and midribs were collected separately 15 days after seedlings being transplanted. Fully expanded leaves (length > 5 cm, up to down, the fourth leaf from top) from the same position in three pots of each treatment was sampled in an ice box. Half of the samples were frozen in liquid nitrogen and stored in a freezer at − 80 °C, while the other half were deactivated at 105 °C for 20 min and then dried at 60 °C for 48 h. Frozen samples were used for transcriptome analysis, enzyme activity determination, soluble protein and NH4-N content investigation. Dried samples were used for determination of nitrate content. Every treatment had three biological replicates. The K326 seeds used in this study were provided by Yunnan Tobacco Company and the collection of the plant material complied with relevant institutional, national and international guidelines and legislation. In preliminary tests, laminas and midribs of seedlings were collected on the 7th, 15th, and 21st days after seedlings being transplanted to determine the difference in nitrate content. The results showed that the nitrate content of midrib was significantly higher than that of the lamina on the 15th day. So laminas and midribs were collected separately 15 days after seedlings being transplanted.

Assays of nitrate reductase activity (NRA), sucrose synthetase activity (SSA), and glutamine synthetase activity (GSA)

Frozen samples were powdered with liquid N2. The activities of SS, NR, and GS were determined using SS, NR, and GS microdetermination kits (Suzhou Comin Biotechnology Co., Ltd, Jiangsu, China), respectively.

Measurement of pigment content, nitrate, soluble protein, and NH4-N content

Nitrate content was determined by the method described in Cataldo41. Samples were frozen in liquid N2 and used to investigate the pigment content and soluble protein content according to Zou42. About 0.5 g of each sample were frozen in liquid N2 and used to investigate the NH4-N content according to Fan43.

RNA extraction, preparation of cDNA library, and sequencing

Total RNA was extracted using the mirVana miRNA Isolation Kit (Ambion, Waltham, MA, USA) following the manufacturer’s protocol. RNA integrity was evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The samples with RNA integrity (RIN) ≥ 7 were used for the subsequent analysis. The libraries were constructed using TruSeq Stranded mRNA LTSample Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. These libraries were then sequenced on the Illumina sequencing platform (HiSeqTM 2500) and 125 bp/150 bp paired-end reads were generated. Quality control was assessed on the remaining reads using the NGS QC Toolkit44. After removing low quality date, the clean reads were mapped to the reference genome of N. tabacum (assembly Ntab-K326) (ftp://ftp.solgenomics.net/genomes/Nicotiana_tabacum/assembly/Ntab-K326_AWOJ-SS.fa.gz) using tophat software45 (v2.1.0).

Enrichment analysis of differentially expressed genes (DEGs)

Transcript profiles of RNA-seq data were analyzed by calculating the read fragments per kilobase per million mapped reads (FPKM). The FPKM value of each gene was calculated using cufflinks, and the read counts of each gene were obtained using htseq-count46,47. DEGs were identified using the DESeq (2012) functions to estimate size factors and using nbinomTest48. A P.adjust < 0.05 and |logFC|> 2 were set as the thresholds for significantly differential expression. Gene function was annotated based on databases of NR (NCBI non-redundant protein sequences), KOG (Clusters of Orthologous Groups of proteins)49, Swiss-Prot (A manually annotated and reviewed protein sequence database)50, KO (KEGG Ortholog database)51, GO (Gene Ontology)52. GO enrichment and KEGG pathway enrichment analyses of the DEGs were conducted using R package GOstats (version: 2.40.0, http://bioconductor.org/packages/release/bioc/html/GOstats.html)53.

Gene expression analysis by qRT-PCR

Expression of eight genes related to nitrogen metabolism was observed. qRT-PCR was performed using Light Real-time PCR Instrument (7900HT FAST, ABI). Reactions were incubated in a 384-well optical plate (Roche, Basel, Swiss) at 50.0 °C for 2 min, 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s, 60 °C for 60 s. TKF and TKR were used as the endogenous control (Table 2). The expression levels of mRNAs were normalized and calculated using the 2−ΔΔCt method54.

Table 2 Statistics of sequencing data quality.
Full size table

Statistical analysis

The figures were processed using GraphPad Prism (v. 8.0.1, GraphPad Software Inc., CA, USA) and correlation analysis and variance between treatments were all processed using SPSS 20.0 (IBM, Palo Alto, CA, USA). For comparison between two data sets, a Student’s t test was used. *P < 0.05, **P < 0.01 were considered statistically significant. All presented data is the mean of three biological replicates (n = 3).