Analysis of differentially methylated regions of genomic DNA in maize (Zea mays L.) exposed to salt stress

ABSTRACT

Through the analysis of DNA methylation in a maize inbred line under salt stress, we investigated the molecular mechanism of salt tolerance in maize. The methylation sensitive amplification polymorphism (MSAP) combined with capillary electrophoresis was applied to detect DNA methylation, and some differentially methylated DNA sequences were cloned and analyzed to discover interesting genes which might contribute to maize salt tolerance. The maize inbred line LH196 experienced extensive DNA methylation variations, including hypo- and hyper-methylation and changes of DNA methylation types. Cloning, sequencing and real-time quantitative PCR tests found that a variety of related genes play important roles in maize salt tolerance.

KEYWORDS:

• Maize
• salt stress
• MSAP
• capillary electrophoresis

Introduction

Soil salinization is one of the important environmental factors affecting the normal growth of plants, and the poor physicochemical properties of the saline-alkali soils make it difficult for plants to grow, seriously affecting the ecological environment and the development of agricultural production. At present, more than 830 million hectares of land are affected worldwide (Deinlein et al. 2014), e.g. China has about 30 million hm² of saline-alkali land, mainly distributed in the northeast, north and northwest (Liu et al. 2009), which is still expanding, especially in coastal areas. The exploitation and utilization of salty soils can not only help solve the problems of grain, population and environment, but also improve the organic content in soil with the growth of salt-tolerant plants, which will further benefit soil reclamation (Bai et al. 2010). Among the technical measures of saline-alkali land development and utilization, cultivating and planting of salt-resistant crops is the most cost-effective way.

NaCl is one of the generators of saline-alkali soil, although Na⁺ plays an important role in plant growth. Lack of Na⁺ will lead to chlorosis or even necrosis of plants. Trace NaCl can benefit the plant growth and normal development (Jia and Zhao 1998), but too high salt concentration will cause damage such as ion penetration. Maize, a typical C4 plant, belongs to Gramineae, and is widely distributed in the world as one of the main food crops. In addition to its high yield per unit area, maize also has high nutritional value rich in calcium, glutamic acid, linoleic acid, dietary fiber, vitamins and other substances. Many researchers have studied many aspects of maize, due to the increasing demand for it (Yang et al. 2003; Dong et al. 2008; Ren et al. 2013), but little is known about the physiological responses of maize under abiotic stress, especially the methylation of DNA under salt stress (Ding 2008; Kimatu 2013). However, in response to the increasing salt damage, plants must evolve a response mechanism to adapt to it. Therefore, it is necessary to study the response mechanism of maize in high salt environment.

DNA methylation is a common DNA covalent modification method, which is one of the most important research contents of epigenetics (Nie and Wang 2007), and is common in higher plants. It plays an important role in the regulation of gene expression: the sequence of the active region of transcription is usually lower in DNA methylation than the specific coding region of the promoter and silencing genes (Freitag and Selker 2005). Although DNA methylation does not alter the DNA base sequence, it has significant epigenetic effects. The effects of DNA methylation on gene expression of plants are being studied extensively. Many studies have shown that DNA methylation plays an important regulatory role in gene expression, growth and development, resisting environmental stress, etc. (Richards 1997; Lukens and Zhan 2007; Tan 2010). In general, plant genome DNA will undergo methylation changes under adversity stress.

Methylation-sensitive amplification polymorphism (MSAP) technology (Xiong et al. 1999; Xu et al. 2000) is a further development of a DNA methylation assay based on amplified fragment length polymorphism (AFLP) (Vos et al. 1995). The method has been widely used in rice, Arabidopsis thaliana, cotton and other plants (Wang et al. 2016). In this study, MSAP technique integrated with capillary electrophoresis was used to detect genomic DNA methylation of maize under salt stress. The effect of salt stress on maize genomic DNA was analyzed. The important differential fragments were blasted in GenBank for genes related to salt tolerance, and then real-time quantitative PCR (qRT-PCR) was performed to verify the results of methylation detection. The mechanism of salt tolerance in maize was studied from the epigenetic level, which provided the basis for salt-tolerant maize breeding.

Materials and methods

Experimental materials

This study used LH196, a maize inbred line with the expiration of the right of protection, which was identified as salt-tolerant by Feng (2015).

Salt tolerance test

We used the Hoagland nutrient solution to grow maize. When the seedlings grew to 4–5 cm, they were transplanted to a hydroponic device. Half nutrients were used on the first day and full nutrition was used later. Full and uniform seeds of the maize inbred line LH196 were selected. Seeds were treated with 2 h of sunlight exposure, 1 h rinsing in water, 4 min of immersion disinfection with 1:35 of 84% disinfectant, and then two immersion disinfections with 70% alcohol, for 4 min each. Seeds were then rinsed with sterile distilled water three times with 1–2 min per time after sterilization. After soaking in sterile water for 6 h, seeds were placed on wet sterile filter paper spread in sterilized Petri dish and then dark cultured at 28°C for two days in an incubator. Germinated seeds with 1–2 cm buds were transplanted to vermiculite for further light culture, under conditions of 28°C/14 h light and 22°C/10 h dark, and nutrient solution was replaced every two days. When maize seedlings grew to 4–5cm, they were transferred into the hydroponic device, and cultured with the semi-nutrient solution for one day and then replaced with the whole nutrient solution (Hoagland nutrient solution). At the three-leaf stage, the uniform seedlings were divided into two groups, one group was treated with 250 mM NaCl solution, and the other group was treated with fresh water as control. Leaves were sampled at the time points of 0, 1, 10, and 12 h and stored in a –80°C freezer.

DNA isolation

DNAs of the leaf samples of 0, 1, 10 and 12 h were extracted by liquid nitrogen milling, with hexadecyltrimethy ammonium bromide (CTAB) method performed according to Li et al. (2009a).

DNA methylation analysis

HpaII and MspI both recognize the 5ʹ-CCGG: HpaII cannot cleave sites containing double-stranded ^5mCCGG, C^5mCGG, ^5mC^5mCGG, but can cleave the hemimethylation sequence (single-stranded methylation); MspI can cleave C^5mCGG on single chain or double strand, but cannot cleave sites containing double stranded or single stranded ^5mCCGG and ^5mC^5mCGG sites (Reynalópez et al. 1997). For each single DNA sample, the MSAP patterns resulting from the digestions with the iso-schizomers were divided into the following four types: Type I bands, present only for EcoRI/MspI (0,1), which represent the case of full methylation of internal cytosine of 5ʹ-CCGG on the two strands; Type II bands, present only for EcoRI/HpaII (1,0), which represent the hemimethylated state of 5ʹ-CCGG sites due to methylation in one DNA strand (external cytosine or both external and internal cytosines of 5ʹ-CCGG) but not in its complementary strand; and Type III bands, absent from both enzyme combinations (0,0), which represent full methylation of the two strands where at least one external cytosine of 5ʹ-CCGG was methylated. Type IV bands, present for both enzyme combinations (1,1), which represent the case of no methylation, or the hemimethylated state of 5ʹ-CCGG sites due to methylation of internal cytosine of 5ʹ-CCGG on one DNA strand but not on its complementary strand. Type IV bands were usually treated as no methylation in analysis although it will underestimate actual methylation level. Type I, II, IV can be identified according to any single DNA sample; however, Type III (0, 0) can only be detected when comparing to other samples. The percentage of polymorphic MSAP bands can be calculated using the following formula: MSAP (%) = [(I+II+III)/(I+II+III+IV)]×100.

The DNA methylation status of maize leaves was analyzed with the MSAP analysis based on capillary electrophoresis as described in Wang et al. (2016). Differential sequences were recycled, cloned and sequenced. The adaptors and primers used in the experiments are shown in Table 1.

Table 1. Adaptors and primers used in MSAP analysis.

Adaptors/primers	Type	Sequences
Adaptors	EcoRI		5′-CTCGTAGACTGCGTACC-3 3′-CTGACGCATGGTTAA-5ʹ
	HpaII-MspI		5′-GACGATGAGTCTAGAA-3ʹ 3′-CTACTCAGATCTTGC-5ʹ
Pre-amplification primers	EcoRI		5′-GACTGCGTACCAATTCA-3ʹ
	HpaII-MspI		5′-GATGAGTCTAGAACGGT-3ʹ
Selective amplification primers	E1		5′-GACTGCGTACCAATTCAAC-3ʹ
	E2		5′-GACTGCGTACCAATTCAAG-3ʹ
	E3		5′-GACTGCGTACCAATTCACA-3ʹ
	E4		5′-GACTGCGTACCAATTCACT-3ʹ
	HM1		5′-GATGAGTCTAGAACGGTAG-3ʹ
	HM2		5′-GATGAGTCTAGAACGGTAC-3ʹ
	HM3		5′-GATGAGTCTAGAACGGTTG-3ʹ
	HM4		5′-GATGAGTCTAGAACGGTTC-3ʹ
	HM5		5′-GATGAGTCTAGAACGGTGT-3ʹ
	HM6		5′-GATGAGTCTAGAACGGTGC-3ʹ
	HM7		5′-GATGAGTCTAGAACGGTCT-3ʹ
	HM8		5′-GATGAGTCTAGAACGGTCG-3ʹ

Primer design

According to the homologous gene/sequence of the cloned differential methylation bands, primers were designed using the Primer 3 software (http://www.simgene.com/Primer3), and synthesized by Shanghai Jierui Bioengineering Co., Ltd, Shanghai, China. The internal reference gene was GAPDH gene (X07156.1) (Table 2).

Table 2. Primers for real-time quantitative PCR.

Sequencing No./Gene name	GenBank Accession	Primer Sequence(5ʹ → 3ʹ)
HXC6-2	AC185486.5	-F-	GAGCGATTGAACACGAGGTC
		-R-	TACTTCTAAAATGCCGGCGC
HXC7-2	AC196472.3	-F-	AAGATGGAGTTGGTGGAGGG
		-R-	GAACTTCTTCGGGCGACATC
HXC8-1	AC226723.4	-F-	CATGTAAAGCTCCACGGTCG
		-R-	ATAGGTTCGGGCCACTTTGA
HXC27-2	BT062751.1	-F-	AGGAAACCGAGGAGCAATGA
		-R-	TGCTTGGCGTCATCACAAAA
HXC30-2	AC165178.2	-F-	CTTCTTCGGGCGACATCTTG
		-R-	AAGATGGAGTTGGTGGAGGG
HXC36-2	AC216070.4	-F-	TGATTGATGGGGCTGTGTCT
		-R-	CCCTCCACCAACTCCATCTT
HXC40-2	AC225222.3	-F-	CTCGGCTCGACATTCACAAG
		-R-	GTTGAGCTCCTCTTGCATCG
GAPDH	X07156.1	-F-	CCATCACTGCCACACAGAAAAC
		-R-	AGGAACACGGAAGGACATACCAG

RNA extraction

RNA from maize leaves was extracted with EASY spin RNA Rapid Extraction Kit (Aidlab Biotech, Beijing, China), and the RNA concentration of each material was measured by an ultraviolet spectrophotometer, and then reverse transcription was carried out.

RNA reverse transcription and real-time quantitative PCR analysis

RNA was reverse transcribed to cDNA using the M-MLV Reverse Transcription Kit (Invitrogen, Shanghai, China). qRT-PCR was carried out using the iCycler iQTM real-time PCR detection system (Bio-Rad, Jurong, Singapore). The reaction system (25 µl) was as follows: 12.5 µl of the FS Universal SYBR Green Master (Rox) reagent from Roche; 2 µl of the cDNA template; 8.5 µl of sterilized double distilled water and 2 µl of the primer (10µmol l^–1). qRT-PCR reaction was performed as follows: pre-denaturation at 95°C for 3 min; 50 cycles of denaturation at 95°C for 15 s, annealing at 60°C for 30 s. At the end of the cycle, we immediately carried out melting curve analysis. The melting curve was programmed at 65°C for 5 s, and then the temperature was raised from 65°C to 95°C gradually, and the temperature maintained for 5 s for every 0.5°C rise. Three parallel samples were made for each sample.

Results

Methylation level in maize inbred line under salt stress

In this study, DNA methylation of the maize inbred line LH196 under salt stress and control was detected. In the MSAP analysis, four EcoRI selective amplification primers and eight HpaII/MspI selective amplification primers were used, constituting 32 primer combinations.

DNA methylation status of maize inbred line LH196 under salt treatment and control at different time points were detected with 32 primer combinations, and in total 7683 CCGG sites were identified in each treatment. The number of methylation sites was 7268 at 0 h, and the numbers of methylation sites under salt treatment and under control were 6927 and 7431 at 1 h, 6740 and 6727 at 10 h, 7193 and 6981 at 12 h, accounting for 94.6%, 90.2%, 96.7%, 87.7%, 87.6%, 93.6%, and 90.9% of the total amplified sites respectively (Table 3).

Table 3. MSAP-based cytosine methylation levels at different time-points under salt stress.

				Salt stress and control time-point
	Enzyme digestion^a			0 h	1 h		10 h		12h
Type	H	M	Methylation pattern^b	CK	Salt	CK	Salt	CK	Salt	CK
I	0	1	CCGG	1267	1488	1697	1723 *	1190	1249	1424
			GGCC
II	1	0	CCGG，CCGG	324	819 **	152	541 *	1023	764	520
			GGCC，GGCC
III	0	0	CCGG，CCGG，CCGG，CCGG，CCGG	5677	4620 *	5582	4476	4514	5180	5037
			GGCC，GGCC，GGCC，GGCC，GGCC
IV	1	1	CCGG，CCGG	415	756	252	943	956	490	702
			GGCC，GGCC
Total amplified loci ((I+II+III+IV）				7683	7683	7683	7683	7683	7683	7683
Total methylation loci(I+II+III)				7268	6927	7431	6740	6727	7193	6981
Type I loci no./primer combination				39.594	46.500	53.031	53.844*	37.188	39.031	44.500
Type II loci no./primer combination				10.125	25.594 **	4.750	16.906 *	31.969	23.875	16.250
Type III loci no./primer combination				177.406	144.375 *	174.438	139.875	141.063	161.875	157.406
Methylation loci no./primer combination				227.125	216.469 **	232.219	210.625	210.219	224.781	218.156
Total methylation loci (I+II+III)/ratio to total loci				94.6%	90.2%	96.7%	87.7%	87.6%	93.6%	90.9%
Non-methylation loci (IV)/ratio to total loci				5.4%	9.8%	3.3%	12.3%	12.4%	6.4%	9.1%

^aH and M represent fragments produced by digestion with HpaΙ Ι/EcoRΙ and Msp Ι/EcoRΙ respectively; 1: band present; 0: band absent; the same below.

^bC represents methylated cytosine; the 10 methylation patterns were divided into four types according to the enzyme digestion results.

*P < 0.05; ** P < 0.01.

In Table 3 is shown the DNA methylation levels in different samples. At 1 h, the average number of sites detected per primer combination for the salt treatment and the control were 216.469 and 232.219 respectively, the methylated site number under salt stress was significantly lower than that of the control (P = 0.003, P < 0.01). For the type III methylation, 144.375 and 174.438 methylated sites were detected under the salt treatment and control respectively, the methylation level under the salt treatment was significantly lower than that of the control (P = 0.024, P < 0.05). For the type II methylation, 819 and 152 methylated sites were detected under the salt treatment and control respectively, the methylation level under the salt treatment was significantly higher than that of the control (P = 0.0011, P < 0.01). At 10 h, 541 and 1023 methylated sites were detected under the salt treatment and control respectively in the type II methylation, and the methylation level under the salt treatment was significantly lower than that of the control (P = 0.031, P < 0.05). In total 1723 and 1190 methylated sites were detected under the salt treatment and control respectively in the type I methylation, and the methylation level under the salt treatment was significantly higher than that of the control (P = 0.034, P < 0.05). However, there was no significant difference between salt treatment and control at 12 h.

Variation of DNA methylation status of maize under salt stress

All the amplification modes in this experiment could be divided into 16 categories (Table 4). Types A, B, C and D indicated that the methylation status was the same between salt treatment and control without changes. These four types accounted for 57.82%, 63.91%, and 57.82% of total bands at 1, 10, and 12 h, respectively. Types E, F, G, H, and I indicated that DNA methylation level decreased under salt treatment, or hypomethylation happened under salt stress compared to the control. These five types accounted for 27.37%, 16.04%, and 14.08% of total bands at 1, 10, and 12 h respectively. Types J, K, L, M, and N indicated that DNA methylation level increased under salt treatment, or hypermethylation happened. These five types accounted for 11.22%, 16.54%, and 17.62% at 1, 10, and 12 h respectively. Types P and O indicated that DNA methylation status changed between salt treatment and control, whereas we cannot tell the difference of methylation level. These two types accounted for 3.59%, 3.51%, and 3.70% of total sites at 1, 10, and 12 h, respectively (Table 4).

Table 4. Analysis of DNA methylation patterns under salt stress with respect to control conditions in maize seedlings.

		Banding pattern				Different treatment
		Control		Salt
Description of pattern	Class	HpaII	MspI	HpaII	MspI	1 h	10 h	12 h
	A	1	1	1	1	84	373	185
	B	1	0	1	0	23	180	114
No change	C	0	1	0	1	447	684	474
	D	0	0	0	0	4088	3673	4190
	Total					4442	4910	4963
						57.82%	63.91%	64.60%
	E	1	0	1	1	31	192	42
	F	0	1	1	1	378	199	193
Hypomethylation	G	0	0	1	1	263	179	70
	H	0	0	1	0	526	124	375
	I	0	0	0	1	905	538	402
	Total					2103	1232	1082
						27.37%**	16.04%	14.08%**
	J	1	1	1	0	31	168	70
	K	1	1	0	1	99	300	294
Hypermethylation	L	1	1	0	0	38	115	153
	M	1	0	0	0	61	450	285
	N	0	1	0	0	633	238	552
	Total					862	1271	1354
						11.22%	16.54%	17.62%
	O	0	1	1	0	239	69	205
Pattern changed	P	1	0	0	1	37	201	79
	Total					276	270	284
						3.59%	3.51%	3.70%

**P < 0.01 at the level of difference, the hypomethylation pattern was significantly higher than the hypermethylation band.

In particular, the analysis of methylation difference comparisons showed that hypomethylation level was significantly higher than that of hypermethylation at 1 h (P < 0.01) and significantly lower at 12 h (P < 0.01). Simultaneously, there was no significant difference under 10 h treatments (Table 4).

Differentially methylated DNA sequences and gene homology

The obtained 40 methylated differential sequences were all sequenced and analyzed by BLAST, and we found seven sequences that have high similarity with known genes. Seven sequences were encoded as HXC6-2, HXC7-2, HXC8–1, HXC30-2, HXC36-2, HXC27-2, and HXC40-2, which were homologous to genes related to zinc finger protein, NOD26-like membrane integral protein, non-specific lipid transporter, MYB-related protein, NRT1/PTR (NPF), protein kinase APK1B and receptor protein kinase ZmPK1, respectively (Table 5).

Table 5. The BLAST results of differentially methylated sequences.

Sequencing No.	Description	E value	Max ident	GenBank Accession
HXC6-2	Zea mays zinc finger protein CONSTANS-LIKE 16-like	2.00E-15	98%	AC185486.5
HXC7-2	Zea mays NOD26-like membrane integral protein ZmNIP2-1	2.00E-15	98%	AC196472.3
HXC8-1	Zea mays probable non-specific lipid-transfer protein 3	1.00E-18	90%	AC226723.4
HXC27-2	Zea mays protein kinase APK1B	5.00E-33	99%	BT062751.1
HXC30-2	Zea mays myb-related protein 306	5.00E-29	87%	AC165178.2
HXC36-2	Zea mays protein NRT1/PTR FAMILY 7.2-like	7.00E-48	95%	AC216070.4
HXC40-2	Zea mays putative receptor protein kinase ZmPK1	2.00E-10	96%	AC225222.3

Gene expression analysis of selected MSAP fragments

The expression of genes related to differentially methylated sequences was validated using real time quantitative PCR. The results showed that the expression of these seven related genes all had significant changes with continued salt treatment. The expression of two genes, namely AC226723.4 and AC216070.4, continued to increase with the increasing time of salt treatment. The expression of the other five genes, namely AC185486.5, AC196472.3, BT062751.1, AC165178.2, and AC225222.3 increased until 10 h, but decreased at 12 h (Figure 1).

Figure 1. The expression of selected genes detected by RT-qPCR in maize inbred line LH196 at different time points. The transcript levels were normalized to that of GAPDH, and the level of each gene in the control samples was set at 1.0. Error bars represent the SEM for three independent experiments. The small letters a, b, c and d represent differences at level of P < 0.05 between different processing times.

Discussion

DNA methylation and the resistance of plants

DNA methylation is one of the most important fields in epigenetics. It is a common process of DNA covalent modification found frequently in higher plants and it links genotype together with phenotype (Qing et al. 2011). With the growth and development of plants, the change in DNA methylation level plays a key role in regulation of gene expression, cell development and so on. Generally, the excessive methylation of plant genomes will inhibit the expression of genes, which leads to silence of genes; meanwhile hypomethylation is required for gene expression (Richards 1997; Lukens and Zhan 2007).

In research into plant cold stress, Wang et al. (2013) analyzed changes of DNA methylation in four kinds of potato before and after ultralow temperature preservation utilizing MSAP technology, and found that hypomethylation and hypermethylation were produced after cryopreservation, and hypomethylation was the main trend of methylation changes. In the study of changes of DNA methylation level and patterns in tea plant (Camellia sinensis) during cold acclimation, the researchers found that the overall trend of methylation changes increased (Zhou et al. 2015). Pan et al. (2011) identified an increase in DNA methylation in the leaves and hypomethylation in the panicles in IR64 at the booting stage under cold stress. Wang et al. (2011b) found that under the conditions of cold stress, DNA methylation was decreased in rice. In the study of salt stress, Baek et al. (2011) found that the putative small RNA target region and the tandem repeat were essential for maintaining AtHKT1 expression patterns crucial for salt tolerance. In other studies, Li (2014) analyzed the total level of DNA methylation and dynamics of the whole genomes in different stages of development in immature embryo of pecan with MSAP and found that the methylation level reached 32.48% at the 12th week after the natural pollination of the pecan. The lowest was 26.34% at the ninth week and the 10th week. Liu et al. (2015) found that the hypomethylation rate of leaf and xylem of dwarf ornamental Cunninghamia lanceolata was significantly higher than that of wild type in the analysis of its DNA methylation levels and patterns. Gao et al. (2011) investigated DNA methylation in Brassica campestris L. during heat stress. They found that both DNA methylation and demethylation were detected, and more DNA demethylation bands were recorded.

The MSAP method based on capillary electrophoresis was used in this study, and DNA methylations of the salt-tolerant maize inbred line LH196 at different time points of salt treatment were detected. Our results showed that after 1 h of treatment, the level of hypomethylation was obviously higher than that of hypermethylation; after 10 h of treatment, the level of hypomethylation was not significantly different from that of hypermethylation; after 12 h of treatment, the level of hypomethylation was significantly lower (P < 0.01; Table 4). The results suggested that maize may respond to salt stress by reducing the level of whole-genome methylation in a certain period; however, with the salt treatment continued, the level of whole-genome methylation may increase, meaning that gene expression level will decrease, which might be beneficial to protect plants from salt stress. Some previous studies had similar conclusions to our results at 1 h, whereas others had similar conclusions to our results at 12 h. Li et al. (2009b) found that after NaCl treatment, cotton root genomic DNA methylation rate was lower than the control group. Under salt stress conditions, Wang et al. (2011a) found that demethylation of genes was an active response to salt stress in roots at the rice seeding stage. Zhao et al. (2010) found that DNA demethylation of cotton played a positive role in salt tolerance, whereas hypermethylation negatively affected salt tolerance. In studies related to maize, Shan et al. (2013) also found results which were mainly based on hypomethylation. However, there are some studies with contrary conclusions. For example, in the study of Labra et al. (2002), they concluded that water stress induces cytosine hypermethylation in the pea genome. Bilichak et al. (2012) found that most of the promoters of Arabidopsis progenies exhibited hypermethylation under salt stress. Zhong and Wang (2007) found that after 10 days treatment with 150 mmol l^–1 NaCl, salt-tolerant wheat expressed a higher level of methylation than salt-sensitive wheat under salt stress.

DNA methylation is tissue-specific, and many researchers have found that the methylation is variable in different plants or different organs of the same species (Ruiz-García et al. 2005; Sha et al. 2005; Gehring and Henikoff 2007; Lu et al. 2008).The different results from these previous studies might be due to the adoption of different materials or different tissues, or different time points. The advantage of our current research is that different time points after salt treatment were analyzed, so we can draw a comprehensive conclusion by detecting the dynamic variations of DNA methylation at different time points.

The different expression of genes related to DNA methylation under salt stress

In order to explore the molecular mechanism of salt tolerance in maize, we recycled, cloned and sequenced partial differently methylated fragments and found the related genes with BLAST analysis. Seven sequences highly similar to known genes were found, including the zinc finger protein, non-specific lipid-transfer protein 3, protein kinase APK1B, MYB-related protein 306, NRT1/PTR(NPF), NOD26-like membrane integral protein and receptor protein kinase ZmPK1 (Table 5).

Through the qRT-PCR test, we found that the expression of the zinc finger protein CONSTANS-LIKE 16-like (GenBank Accession AC185486.5) increased as salt treatment continued, whereas its expression decreased after 10 h (Figure 1). Zinc finger protein is a kind of transcription factor first discovered in the transcription factor TF IIIA of Xenopus oocytes in 1985 (Miller et al. 1985) and it exists widely in animals, plants and microorganisms. It regulates gene expression by combining with DNA, RNA or interacts with other proteins, and plays an important role in plant growth and stress resistance. In many studies, researchers have found that the zinc finger protein is closely related to stress resistance under abiotic stress, such as drought, salt, cold and heat (Xiang et al. 2012). Jain et al. (2008) found that Arabidopsis with transferred zinc finger protein genes OsTOP6A1 had improved salt tolerance and the expression of the response gene under the adversity stress increased. Studies have shown that the expression of GhDi19-1 and GhDi19-2 genes in the zinc finger protein gene of cotton can be strongly induced by salt and drought, and results indicated that excessive expression in Arabidopsis thaliana led to excessive sensitivity to high salt and abscisic acid (ABA) (Li et al. 2010a). In the transgenosis test, potato zinc finger protein gene StZFP1 (Tian et al. 2010) can improve salt tolerance in plants. In this research, the expression of zinc finger protein gene increases with the increase of salt treatment time until 10 h, suggesting that its expression is essential to maize salt-resistance; whereas longer time of salt stress may hurt plants and lead to decrease of the gene expression at 12h.

In this study, we also found that the expression of non-specific lipid transfer protein 3 (GenBank Accession AC226723.4) had clearly increased with continuing salt treatment (Figure 1), which suggested that it was very important in the maize salt resistance response. Plant non-specific lipid-transfer protein is an alkalinity gene family protein and exists widely in plants (Kader 1996). Previous studies found that nsLTP was overwhelmingly expressed by the effect of the plant abiotic stress. Dani et al. (2005) found that nsLTP was overwhelmingly expressed in tobacco leaf. Wu et al. (2005) and Gao et al. (2008) came to the same conclusion. These results indicated that maize might improve its salt tolerance through increasing its nsLTP expression.

In previous studies, Arabidopsis protein kinase 1B (APK1B) was mainly expressed in the stomata guard cell (Leonhardt et al. 2004). Elhaddad et al. (2014) concluded that APK1b was required for full stomatal opening in the light but not for stomatal closing. In this research, we also found this gene in maize (BT062751.1, Table 5), and with time going by, the expression significantly increased, although the expression decreased a little after 10 h (Figure 1), suggesting that under salt stress conditions maize may respond to stress by regulating the cell stomata.

Recent studies have found that MYB protein is important in plant abiotic stress regulations including salt stress (Jung et al. 2008). Xiong et al. (2014) and Cheng et al. (2013) found induced MYB-related protein genes included OsMYB48 and LcMYBI by salt stress. Some peanut (Chen et al. 2014) and wheat (Zhang et al. 2012) MYB-related genes were also confirmed to be induced by salt stress. In this study, the expression of MYB-related protein gene (AC165178.2, Table 5, Figure 1) showed a trend of increasing with the time of salt treatment, which was validated by a qRT-PCR expression test. It was possible to conclude that maize responded to salt stress with the increase of MYB-related gene expression.

Plant NRT1/PTR (NPF) (GenBank Accession AC216070.4) is a translocator that transports a variety of substrates, such as nitrate, peptide, amino acid, ester, glucosinolate, IAA and ABA. Plant NRT1/PTR proteins display homology with proteins from other kingdoms (Léran et al. 2014). Bielach (2010) found that NRT1.1 not only transported nitrate but also facilitated uptake of the phytohormone auxin. Li et al. (2010b) found that the nrt1.8 mutant showed a nitrate-dependent cadmium-sensitive phenotype, and an increased amount of cadmium was transported to the wildtype roots compared with the nrt1.8-1 mutant. They speculated that the function of NRT1.8 in removing nitrate from xylem sap also allowed Cd²⁺ to stay in the roots, and consequently enhanced Cd²⁺ tolerance. Our research showed the NPF (AC216070.4) expression increased significantly with salt treatment continued (Figure 1), and it suggested that maize would increase its NPF expression under salt stress, which played an important role under maize salt tolerance.

Aquaporins (AQPs) or major intrinsic proteins (MIPs) including complete membrane protein can form the transmembrane ion channel of water and other small molecules. The MIP is divided into five subfamilies in the higher plants: plasma membrane (PIPs), tonoplast (TIPs), NOD26-like (NIPs), small basic (SIPs) and unclassified X (XIPs). Martins et al. (2015) found that sweet orange (Citrus aurantium L.) MIP gene family expression under biological stress or abiotic stress could be important genetic resource of resistance and tolerance. In our study, we found the expression of NOD26-like membrane integral protein gene (GenBank Accession AC196472.3) improved significantly with salt treatment continued, although its expression decreased a little after 10 h (Figure 1), which suggested that it was closely related to maize’s physiological responses to salt stress.

Plant receptor-like protein kinase (RLKs) is an important signal for plant signal recognition. The maize receptor-like protein kinases ZmPK1 (Genbank Accession AC225222.3) in our study was the earliest found plant receptor-like protein kinase. So far, although researchers have found and separated a lot of RLKs genes, we still cannot distinguish its function, pattern of signal transmission and ligand recognition. In this study, the expression of this gene increased significantly until 10 h, but then decreased significantly after 10 h. This phenomenon indicates its importance to maize salt tolerance.

In this study, DNA methylation of a salt-tolerant maize inbred line LH196 was analyzed with MSAP method integrated with capillary electrophoreses. DNA methylation variation was detected and some salt tolerance-related genes and their expressions were identified. In future studies, different kinds of maize inbred lines with different tissues will be studied, to deepen the understanding of the molecular mechanism of maize under salt stress.

Acknowledgments

This work was supported by the Key Research and Development Project of Jiangsu Province, China (Modern Agriculture, BE2017365), the State Key Laboratory of Crop Biology (Shandong Agricultural University) Open Fund (No. 2016KF02), and the Practice Innovation Training Program Projects for College Students.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the State Key Laboratory of Crop Biology (Shandong Agricultural University) Open Fund [number 2016KF02]; the Key Research and Development Project of Jiangsu Province, China (Modern Agriculture) [BE2017365].