Transcriptome analysis reveals the effects of alkali stress on root system architecture and endogenous hormones in apple rootstocks

LlU Xuan, LlANG Wei, Ll Yu-xing, Ll Ming-jun, MA Bai-quan, LlU Chang-hai, MA Feng-wang, Ll Cuiying

State Key Laboratory of Crop Stress Biology for Arid Areas/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling 712100, P.R.China

Abstract Soil alkalinity is a major factor that restricts the growth of apple roots. To analyze the response of apple roots to alkali stress,the root structure and endogenous hormones of two apple rootstocks, Malus prunifolia (alkali-tolerant) and Malus hupehensis(alkali-sensitive), were compared. To understand alkali tolerance of M. prunifolia at the molecular level, transcriptome analysis was performed. When plants were cultured in alkaline conditions for 15 d, the root growth of M. hupehensis with weak alkali tolerance decreased significantly. Analysis of endogenous hormone levels showed that the concentrations of indole-3-acetic acid (IAA) and zeatin riboside (ZR) in M. hupehensis under alkali stress were lower than those in the control.However, the trend for IAA and ZR in M. prunifolia was the opposite. The concentration of abscisic acid (ABA) in the roots of the two apple rootstocks under alkali stress increased, but the concentration of ABA in the roots of M. prunifolia was higher than that in M. hupehensis. The expression of IAA-related genes ARF5, GH3.6, SAUR36, and SAUR32 and the Cytokinin (CTK)-related gene IPT5 in M. prunifolia was higher than those in the control, but the expression of these genes in M. hupehensis was lower than those in the control. The expression of ABA-related genes CIPK1 and AHK1 increased in the two apple rootstocks under alkali stress, but the expression of CIPK1 and AHK1 in M. prunifolia was higher than in M. hupehensis. These results demonstrated that under alkali stress, the increase of IAA, ZR, and ABA in roots and the increase of the expression of related genes promoted the growth of roots and improved the alkali tolerance of apple rootstocks.

Keywords: alkali stress, apple rootstock, endogenous hormone, root architecture, transcriptome analysis

1. lntroduction

Soil alkalinity affects soil fertility and reduces agricultural productivity (Jin et al. 2006). Soil alkalinity affects 4.38×108ha of land in the world, which is detrimental to global crop production (Wang et al. 2016). Soil alkalization is induced mainly by excess NaHCO3, Na2CO3, and some neutral salts,but NaHCO3and Na2CO3led to more severe alkalization than neutral salts in soil (Shi and Sheng 2005). The stress to plants caused by soil alkalinity is due to osmotic stress, ion toxicity, and high pH (Rincon and Gonzales 1992). Among them, high pH ranked the first for its damage to plants, and osmotic stress ranked the last. Soil alkalinity is aggravating day by day, and this challenge has become a major factor that affects the quality of agricultural products (Guo et al.2015). Although the potential threat of alkali stress to plants has attracted more and more attention, the complex relationships caused by alkali stress are still unclear.

Apples, which are one of the most popular fruits, grow mainly in slightly acidic to neutral soil (Zhang et al. 2018).Soil alkalinity has a negative impact on apple yield and quality. For plants, the organ that first detects the pH value of the rhizosphere soil is the root. However, the morphology and vigor of roots directly affect the growth of above-ground parts of the plant (Davies and Zhang 1991). Therefore, a comprehensive study of the apple root system under alkali stress is worthwhile.

Endogenous hormones control physiological and metabolic responses of plants (Ljung et al. 2015).Indole-3-acetic acid (IAA) is a regulator of root growth,which is responsible for regulating cell identity and cell division of roots (Grieneisen et al. 2007). Cytokinin(CTK) regulates the formation and growth of lateral roots.Zwack and Rashotte (2015) found that CTK participated in various stress responses, and there was an extremely close relationship between CTK and plant abiotic stress tolerance. Zeatin riboside (ZR) is one of the most active CTKs and it is associated with abiotic stress resistance in plants (Bai et al. 2011; Wen et al. 2018). Plants that were deficient in abscisic acid (ABA) were extremely sensitive to stress. It has been reported that alkali stress causes accumulation of ABA in plant leaves and releases ABA from roots to the soil solution (Degenhardt et al. 2000).Spollen et al. (2000) have shown that under drought conditions, an increase in ABA promoted root growth.Previous studies in our laboratory showed differences in alkali tolerance among 17 apple rootstocks, and the inhibition on plant growth due to alkali stress was different among them (Zhang et al. 2016). Therefore,M. prunifolia with alkali tolerance and M. hupehensis with alkali sensitivity were used in this study.

In this study, to understand how alkali stress affected the growth and development of root systems of two different apple rootstocks, the endogenous hormone levels of roots were measured, and the transcriptome analysis of the roots of M. prunifolia was performed to identify the candidate genes associated with the changes in endogenous hormones. This is significant for the apple breeding and improvement of apple yield and quality of alkali-resistant apple rootstock.

2. Materials and methods

2.1. Plant materials and experimental design

Fuping Qiuzi (M. prunifolia, alkali-tolerant) and Pingyi Tiancha (M. hupehensis, alkali-sensitive) were used in this study. The seedlings of Fuping Qiuzi were obtained by tissue culture. Because Pingyi Tiancha is apomictic, the seedlings were obtained from seeds provided by Shandong Agricultural University, China.

When the seedlings grew to 6-8 leaves, 400 plants were selected from each apple variety and transferred into half-strength Hoagland nutrient solution (pH=6.0) with 50 plants per pot (47 cm×35 cm×17 cm); the nutrient solution was replenished every 5 d. Each basin was considered as one replicate, and four replicates were maintained in total. The method proposed by Bai et al. (2008) was used for hydroponic culture at a light intensity of 8 000-9 000 lx,artificial light for 14 h, and temperature at (24±1)°C. During the experiment, an air compressor pump was used to aerate the nutrient solution (ventilating for 40 min, stopping for 1 h). After 10 d of pre-culture, the seedlings were divided equally into two groups: one group was the control that was cultured in standard 1/2-strength Hoagland solution at a pH of 6.0 that was attained with the addition of concentrated H2SO4. The treatment group was treated with 1 mol L-1NaHCO3and 1 mol L-1Na2CO3(1:1) to adjust the pH of the nutrient solution to 9.0. The pH of nutrient solutions in the control group and the treatment group were adjusted daily to maintain a pH of 6.0 and 9.0, respectively. However, the remaining conditions were not changed.

2.2. Analysis of root system architecture

After 15 d of alkali stress, 10 plants were selected from each treatment for root scanning. The morphological features of roots were analyzed by using an Epson Perfection V700 Photo Scanner (SEIKO EPSON CORP, Japan) (n=10).

2.3. Analysis of endogenous hormone levels

The tender white root tips were sampled at 0, 6, 24, and 48 h of alkali stress treatment and subsequently stored at-80°C for the measurement of concentration of endogenous hormones, such as ABA, IAA, and ZR. An indirect ELISA method proposed by Bai et al. (2008) was used to extract and purify the contents of ABA, IAA, and ZR.

2.4. Transcriptome analysis

After 0, 6, 24, and 48 h of alkali stress, the tender white root tips of M. prunifolia were collected and stored at -80°C for transcriptome analysis. A TianGen?plant RNA isolation kit was used to extract total RNA. Oligo (dT) beads were used to enrich eukaryotic mRNA, and aRibo-ZeroTM Magnetic Kit (Epicentre) was used to enrich prokaryotic mRNA by removing rRNA. Then, the enriched RNA fragments were transformed into short fragments using fragment buffer and retrieved into cDNAs by random primers. cDNA fragments were purified by a ChiaQuickPCR extraction kit, the end was repaired, and poly(A) was added and connected to the Illumina sequencing adapter. The ligation products were screened by agarose gel electrophoresis and amplified by PCR, and IlluminaHiSeqTM 2500 was used for the sequencing.

Reads obtained from the sequencing machines include raw reads containing adapters or low quality bases which will affect the following assembly and analysis. Thus, to get high quality clean reads, reads will be further filtered. Short reads alignment tool Bowtie2 (Langmead and Salzberg 2012) was used for mapping reads to ribosome RNA (rRNA)database. The rRNA mapped reads will be removed. The remaining reads were further used in assembly and analysis of transcriptome. The reads removed from each sample by rRNA were then mapped to a reference genome by TopHat2(Kim et al. 2013). Differentially expressed genes with a fold change ≥2 and a false discovery rate (FDR) ≤0.05 were identified by using the software edgeR package.

2.5. Quantitative analysis of genes

Premier 6 Software was used to design specific primers,and qRT-PCR analysis was performed using an ABI StepOnePlus real-time PCR system. The 10 μL reaction volume consisted of 5 μL SYBR, 3.5 μL ddH2O, 0.4 μL forward and reverse primers, 0.2 μL rox reference dye, and 0.5 μL cDNA template. Each qRT-PCR reaction was tested in triplicate. The SPSS19 statistical analysis software was used to calculate the correlation coefficient (R) of each gene between qRT-PCR and RNA-seq.

3. Results

3.1. Effect of alkali stress on root system architecture

After plants were under alkaline conditions for 15 d, the root growth of M. hupehensis was inhibited significantly compared with the control (Fig. 1). The analysis of root system architecture showed that under alkali stress, the root surface area and forks number in both M. hupehensis and M. prunifolia seedlings decreased, but the reduction was different (Table 1). M. prunifolia showed no significant difference in root length, diameter, volume, and tips number between the control and alkali stress treatment,but M. hupehensis was the opposite except for root volume.The reduction of tips number in M. hupehensis was about 1.7 times as much as in M. prunifolia. The inhibition of root growth in M. hupehensis with alkali sensitivity was more serious than that in M. prunifolia with alkali tolerance.

Fig. 1 Effects of alkali stress on root systems of Malus prunifolia (AT) and Malus hupehensis (AS) seedings after 15 d of treatment. A, AS-CK. B, AS-T. C, AT-CK. D, AT-T. AT,alkali-tolerant; AS, alkali-sensitive; CK, control; T, alkali stress.

3.2. Effect of alkali stress on the concentrations of endogenous hormones in roots

To quantify the difference in root system architecture between the two apple rootstocks under alkali stress, the concentrations of IAA, ZR, and ABA were measured after alkali stress for 0, 6, 24, and 48 h. IAA concentration in the roots of M. hupehensis in the control was always higher than that in alkali treatment, but it was the opposite in M. prunifolia. After 6 h, IAA concentrations of the two apple rootstocks were different between the control and the treatment (Fig. 2-A). Compared with the control,IAA concentration increased by 12.4% in the roots of M. prunifolia with alkali tolerance. In contrast, it decreased by 10.67% in the roots of M. hupehensis with alkali sensitivity.

ZR concentration in the roots of M. hupehensis in the control was always higher than that in alkali treatment,but it was the opposite in M. prunifolia (Fig. 2-B). After 6 h of stress, ZR concentration increased slightly (6%)in M. prunifolia. However, ZR concentration in roots of M. hupehensis decreased.

ABA concentration of two apple rootstock seedlings increased under alkali stress (Fig. 2-C). ABA concentration ofM. prunifolia did not change significantly after 24 h of alkali stress, but at 48 h, it suddenly increased to the maximum value. After 48 h of treatment, ABA concentrations of the roots of both M. prunifolia and M. hupehensis increased by 15.33 and 8.3%, respectively, compared with the control.ABA concentrations of two rootstocks showed a trend of decreasing first and then increasing.

Table 1 Effect of high pH on root system architecture of two apple rootstock seedlings

Fig. 2 Changes of endogenous hormones in the roots of Malus prunifolia (pru) and Malus hupehensis (hup) sampled at 0, 6, 24,and 48 h. A, B, and C represent the concentrations of indole-3-acetic acid (IAA), zeatin riboside (ZR), and abscisic acid (ABA),respectively. CK, control; T, alkali stress. Values are mean±SE (n=3). ＊, P＜0.05 (t-test).

3.3. RNA-seq analysis of differentially expressed genes (DEGs)

The root tips of M. prunifolia were collected at 0, 6, 24,and 48 h, and the morphological changes in the root tips were analyzed to understand the molecular mechanism of different root system structures. Heatmap analysis of hormone-related genes at different time points of alkali stress was shown in Fig. 3. Venn diagrams with upregulated and down-regulated DEGs in roots of M. prunifolia are shown in Fig. 4.

Compared with the control, the number of up-regulated genes of M. prunifolia at three time points were 751, 362,and 693, and the number of down-regulated genes were 548, 147, and 499. The number of up-regulated genes and down-regulated genes that overlapped at three different periods were 134 and 27, respectively.

3.4. Quantitative verification of candidate genes

We selected seven DEGs related to ABA, IAA, and CTK to carry out qRT-PCR to verify further the reliability of the RNAseq data and to compare them in the two apple rootstocks.Interestingly, the expressed trends in M. prunifolia between qRT-PCR and RNA-seq analyses were the same. In M. prunifolia, the expression of IAA-related genes increased suddenly after 6 h, but the magnitude of increase was different in these genes. The expression of ARF5, GH3.6,SAUR36, and SAUR32 was 4.69, 5.89, 3.06, and 7.49 folds higher, respectively, than that of the control at 6 h (Fig. 5-AD). The expression of GH3.6 and SAUR36 in M. hupehensis did not change significantly after 6 h of stress compared with the control, but it decreased after 48 h of treatment. In contrast, the expression of ARF5 and SAUR32 decreased continuously after stress (Fig. 6-A-D). The expression level of IPT5, which is related to CTK synthesis, increased at first and then decreased, and it reached the highest level at 6 h in M. prunifolia (Fig. 5-E). However, the expression level of IPT5 was lower than that of the control in M. hupehensis(Fig. 6-E). The expression levels of ABA-related genes CIPK1 and AHK1 in M. prunifolia reached the maximum at 48 h, which were 3.81 and 3.53 times of that in the control, respectively (Fig. 5-F and G). In M. hupehensis,the expression of CIPK1 was down-regulated. The lowest value was at 48 h, which was 0.88 times lower than that of the control. The expression of AHK1 was up-regulated slightly at 6 and 48 h, but there was no significant differences compared with the control (Fig. 6-F and G).

Fig. 3 Heatmap that showed the expression patterns of hormone-related genes in Malus prunifolia. CK, control; T1, T2, and T3,alkali stress for 6, 24, and 48 h, respectively.

Fig. 4 Three pairs of differentially expressed genes in Malus prunifolia shown in Venn diagrams. The intersection of Venn diagram indicated that the differentially expressed genes (DEGs) were (A) up- and (B) down-regulated at three different time points under alkali stress. CK, control; T1, T2, and T3, alkali stress for 6, 24, and 48 h, respectively.

4. Discussion

Alkali stress has negative effects on plant growth. It can significantly affect the growth and development of plants to different degrees (Wang et al. 2015). Our study showed that alkali stress suppressed the growth of apple roots(Table 1). When plants were subjected to alkali stress for 15 d, the root growth of two apple rootstocks was inhibited.However, the degree of inhibition in M. prunifolia was lower than that in M. hupehensis, which was consistent with what we identified previously (Wen et al. 2018). In addition, the high pH environment around the root system inhibited plant growth by affecting the normal physiological function of the root and destroying the cellular structure.

Fig. 5 qRT-PCR validation of candidate differentially expressed genes (DEGs) in Malus prunifolia at three different time points after alkali treatment. Data from qRT-PCR are mean±SE (n=3). Data from RNA-seq are the mean of three replicates and were log2 transformed.

Fig. 6 qRT-PCR validation of candidate differentially expressed genes (DEGs) at three different time points of Malus hupehensis.Data from qRT-PCR are mean±SE (n=3).

Plant hormones are capable of enhancing plant resistance to adversity (Vanstraelen and Benková 2012). The results of our study indicated that, compared with the control, the IAA concentration in the roots of M. prunifolia increased at 6 h,but the level of IAA declined in M. hupehensis. Meanwhile,IAA concentration in the roots of M. hupehensis was always higher in the control than in the treatment (Fig. 2-A). Similar results on the effect of alkali stress on IAA concentration was reported previously for Arabidopsis thaliana (Li et al. 2015).Some studies showed that auxin biosynthetic genes were expressed in the niche of root stem cells, which increased auxin concentration (Brady et al. 2007; Stepanova et al.2008). ARF5 belongs to the ARF gene family that includes 23 ARF genes. It encodes the auxin-induced tfs gene,which can promote root formation. Our results showed that the highest expression of ARF5 was after 6 h of treatment in M. prunifolia (Fig. 5-A), which exhibited the same trend as the change in IAA concentration in roots. However, the expression of ARF5 in M. hupehensis showed a decreasing trend (Fig. 6-A). This may imply that under alkali stress,plants could increase their alkaline tolerance by inducing the expression of ARF5. Auxin homeostasis is maintained by the Gretchen Hagen 3-like (GH3) protein family, and auxin can induce GH3.6 expression (Costa et al. 2018).Huang et al. (2016) found that the small auxin-up RNA(SAUR) family participated in auxin signal transduction, and genes belonging to this family are often used as markers for early auxin reactions. In Arabidopsis, SAUR32 regulated cell division by interacting with type-A ARR and PP2C.A.Consistent with those findings, we discovered that GH3.6,SAUR32, and SAUR36 were up-regulated in M. prunifolia(Fig. 5-B-D) but were down-regulated in M. hupehensis(Fig. 6-B-D). The results showed that all of these genes promoted the formation of IAA and improved the alkali tolerance of plants.

Although auxin is essential for regulating root growth,other hormones, such as ABA and CTK, are also essential for root system development. CTK can delay the senescence of plants and enhance their resistance to stress(Albacete et al. 2014). The biosynthesis of CTK begins with isoamylenyltransferase (IPT), and the gene IPT5 is involved(Sakamoto et al. 2006). Drought tolerance in rice (Oryza sativa) was increased by increasing the overexpression of the IPT5 gene (Reguera et al. 2013). We found that the concentration of ZR in M. prunifolia peaked at 6 h, although the concentration of ZR in M. hupehensis decreased gradually (Fig. 2-B). Meanwhile, the CTK biosynthesisrelated gene IPT5 in M. prunifolia with alkali tolerance was up-regulated (Fig. 5-E), but it was down-regulated in M. hupehensis with alkali sensitivity (Fig. 6-E). These observations demonstrated that alkali stress increased the concentration of ZR in roots by inducing the expression of IPT5, thereby enhancing the alkali tolerance of plants.

ABA is a stress hormone, which promotes the survival of plants under adversity due to its rapid accumulation and its regulation of stress responses (Zhang et al. 2006). AHK1 is a monoethylene receptor histidine kinase and functions in salt stress response and ABA signal transduction. Earlier studies proved that kinase CIPK1 could be a focal point of ABA-dependent and ABA-independent stress responses by alternately forming complexes with CBL1 or CBL9 (Dangelo et al. 2006). In our study, the concentration of ABA in the two species only increased after 48 h, but the concentration of ABA in M. hupehensis was lower than that in M. prunifolia(Fig. 2-C). The ABA-related genes AHK1 and CIPK1 also showed the same trend in M. prunifolia (Fig. 5-F and G).In M. hupehensis, CIPK1 gene expression was downregulated, and AHK1 was slightly up-regulated at 6 and 24 h after alkali treatment (Fig. 6-F and G). Therefore, these findings revealed that alkali tolerance of plants could be improved by increasing the concentration of ABA.

5. Conclusion

The root growth of two apple rootstock species with different alkali tolerance was inhibited under alkali stress.The analysis of endogenous hormones showed that the concentrations of IAA, ZR, and ABA increased greatly in M. prunifolia. Subsequently, we selected seven hormonerelated genes from transcriptome analysis. Quantitative results showed that the expression of these seven hormonerelated genes was up-regulated in M. prunifolia with alkali tolerance, but down-regulated in M. hupehensis with alkali sensitivity. These results indicated that alkali stress increased the concentration of hormones in roots by inducing the expression of hormone-related genes, which ultimately enhanced the alkali tolerance of plants.

Acknowledgements

This study was supported by the earmarked fund for the China Agriculture Research System (CARS-27). The authors sincerely thank Thomas A. Gavin, Professor Emeritus, Cornell University, for help with editing this paper.