AFLP and MSAP Analysis of ‘Lane Late’ Navel Orange and Its Bud Sport Pumpkin-like Navel Orange

Chunwang LAI, Yuling LIN, Xiaojuan ZHOU, Ting PENG

1. National Navel Orange Engineering Research Center, Gannan Normal University, Ganzhou 341000, China; 2. Institute of Horticultural Biotechnology, Fujian Agriculture and Forestry University, Fuzhou 350002, China; 3. Ganzhou Agricultural and Rural Bureau, Ganzhou 341000, China

Abstract [Objectives] The paper was to analyze the differences of genes and genome methylation between ‘lane late’ navel orange and it’s bud sport pumpkin-like navel orange, and to reveal the causes of bud sport. [Methods] Amplification fragment length polymorphism (AFLP) and methylation sensitive amplification polymorphism (MSAP) were used to analyze the differences in genes and genome methylation of ‘lane late’ navel orange and its bud sport pumpkin-like navel orange. [Results] Gene mutation occurred between the ‘lane late’ navel orange and its bud sport pumpkin-like navel orange. A total of 15 differential bands were obtained by AFLP markers, and the percentage of polymorphic bands was 6.000 0%. Meantime, 10 differential bands were obtained by MSAP markers, and the percentage of polymorphic bands was 5.464 5%. The semimethylation rate and permethylation rate of ‘lane late’ navel orange were higher than that of pumpkin-like navel orange, that is, pumpkin-like navel orange had demethylation (demethylation ratio was 1.639 3%). [Conclusions] It is proved at the DNA level that the emergence of bud sport pumpkin-like navel orange is due to gene mutation, and the DNA methylation level of the bud sport material has also changed. This study lays a foundation for further exploring the mechanism of bud sport.

Key words ‘Lane late’ navel orange; Bud sport; AFLP; MSAP; Variance analysis

1 Introduction

Gannan navel orange, as "the first citrus brand in China" and the agricultural characteristic industry in southern Jiangxi Soviet area, has been strongly supported by the national policy. However, there are less varieties of navel orange in southern Jiangxi, and more than 90% of them are medium maturing varieties. Besides, the navel oranges planted at present are mainly fresh varieties, lacking of new varieties suitable for processing and storage. At present, navel orange fruits are mainly spherical or ellipsoidal (fruit shape index ≥1), with smooth peel. It is also a development trend to breed new varieties with different fruit shape and peel from the main cultivars. Pumpkin-like ‘lane late’ navel orange mutant type (MT) was discovered by National Navel Orange Engineering Research Center in an orchard in Anxi Town, Xinfeng County, Jiangxi Province in 2011. In 2009, 2 000 plants of ‘lane late’ navel orange wild type (WT) were planted in this orchard. In 2011, the fruit surface of one navel orange was found to be similar to pumpkin, and there were irregular strip bulges arranged from fruit pedicel to fruit tip. After observation for consecutive 5 years, MT and its progeny maintained this character, indicating that the character of the single plant was stable. Zhouetal.[1]and Youetal.[2]observed and analyzed the phenological period, fruit quality and anatomical structure of peel tissue of WT and MT in the early stage, and found that MT fruits ripened from late January to mid-February of the next year and can keep fresh on trees until late March, significantly later than WT. The solidity-acid ratio is appropriate and the content of vitamin C is high, so MT is an excellent new bud sport material that suitable for further development and promotion. In order to find out whether MT has changed in genetic materials, Zhouetal.[1]conducted ISSR analysis on genomic DNA of WT and MT. As genomic DNA sequences of mutant type from the same variety generally had small differences, fewer polymorphisms were obtained in ISSR analysis, making it difficult to conduct further research on the causes of their differential traits.

Amplification fragment length polymorphism (AFLP) is an efficient molecular labeling technique, which is featured by small amount of template DNA, high stability, large number of amplified bands, good polymorphism and strong identification ability, which is suitable for identification of varieties with close relationship. During RAPD, SSR and AFLP identification of genomes from 30 kinds of ‘Misket’ apple, it was found that AFLP markers were more suitable for analyzing genetic relationships among apple genotypes[3]. Roncalloetal.[4]analyzed the genetic diversity of 168TriticumturgidumL. var.durumfrom different origins via SNP and AFLP, and found that only two subgroups were detected by SNP (KASP) analysis of the whole series, while six subspecies were detected in the structure analysis of 119 germplasm subgroups using AFLP markers. Chandraetal.[5]found that AFLP markers could effectively perform cluster analysis on 36Cenchrusspecies with apomixis and sexual reproduction. The application of AFLP markers can provide guarantee for the fine classification of fruit tree varieties, and can also be used as the basis to further study the occurrence and development mechanism of fruit tree traits, which greatly shorten the breeding period of bud sport, and create conditions for breeding fruit tree varieties with more abundant traits. DNA methylation is one of the main modification methods of eukaryotic genomic DNA. Cytosine methylation has epigenetic and mutational effects, which can be inherited or reversed[6]. Currently, methylation sensitive amplification polymorphism (MSAP) is a commonly used method to detect DNA methylation. In 1999, Xiongetal.[7]proved for the first time that MSAP was a reliable method to detect DNA methylation in plants. MSAP analysis is an improvement of AFLP technology, which is relatively convenient and can detect a large number of methylation loci in DNA samples. Lawetal.[8]studied the methylation patterns of animals and plants, and found that DNA methylation played an important role in maintaining the stability of organism genome structure and regulating gene expression. The amount and degree of DNA methylation varies from organism to organism and is highly susceptible to external biotic and abiotic factors. Gongetal.[9]believed that epigenetic regulation mediated by DNA methylation enhanced the environmental adaptability of organisms. Sahuetal.[10]also believed that plants could resist and adapt to adversity through changes in DNA methylation. The processes of fruit trees, such as bud sport[11], dormancy[12], fruit peel coloring[13], fruit maturity[14]and polyploidization[15], are closely related to DNA methylation. Therefore, the study of DNA methylation of bud sport materials is of great significance to elucidate the genesis and development mechanism of bud sport.

In this study, the AFLP technology with higher resolution and MSAP analysis were used to analyze the genomic DNA of MT and WT, in order to further explore the cause and development mechanism of MT traits.

2 Materials and methods

2.1 MaterialsA total of 6 test samples, 3 MT and 3 WT, were all collected from orchards in Xinfeng County, Jiangxi Province where MT bud sport was found (Table 1).

Table 1 Test materials

2.2 Methods

2.2.1Extraction and purification of genomic DNA. The extraction and purification of genomic DNA referred to the method of Su[16], with some modifications. Genomic DNA was extracted from leaves by CTAB-Triton method, and the quality of genomic DNA was detected by 0.8% agarose gel electrophoresis combined with ultra low volume spectrometer (Thermo Scientific, USA). Part of genomic DNA was diluted to 100 ng/μL, and stored in a refrigerator at 4 ℃. The remaining DNA was stored at -20 ℃.

2.2.2AFLP and MSAP molecular marker technology. The adaptors and primers used in AFLP referred to that proposed by Han[17]. All adaptors and primers of MSAP referred to that proposed by Meng[18]. Both adaptors and primers were synthesized by BGI (Table 2). Both AFLP and MSAP molecular marker technologies included double digestion, ligation, pre-amplification and selective amplification[19-20]. After the selective amplification reaction, 5 μL of dilution buffer was added in the system, and 20 μL of mineral oil was added to prevent evaporation. The diluted AFLP and MSAP amplified products were loaded into automatic nucleic acid protein analyzer (Qsep 100TM, Houze Bio-tech). High resolution cartridge(S1) was used for electrophoresis analysis, and Alignment Marker 20 bp & 1 000 bp was selected as the alignment marks. Clearly distinguishable bands in the electropherogram were marked as "1" and vacancies as "0".

Table 2 Adaptors and primer sequence of AFLP and MSAP

2.3 Data analysis of AFLP and MSAPIn AFLP markers,EcoRI was combined withMseI, so each sample corresponding to each primer had only one lane, and each pair of primers had two lanes. In MSAP markers,EcoRI was paired withHpaII andMspI, respectively, so each sample corresponding to each primer had 2 lanes, and each pair of primers had WT-H, WT-M, MT-H and MT-M4 lanes. According to the presence or absence of lanes, there were four types of bands, which can be subdivided into 15 subtypes: A1, A2, A3, A4, B1, B2, B3, B4, C1, C2, C3, C4, D1, D2 and D3 (Table 3).

DNA methylation rate was calculated following the method proposed by Du[2]. Permethylation rate(%)=number of permethylation bands/total number of amplified loci×100%. Hemimethylation rate(%)=number of hemimethylation bands/total number of amplified loci×100%. Total methylation rate=permethylation rate+semimethylation rate, that is, polymorphic amplification of methylation sensitivity.

Table 3 Classification of cytosine methyl variation pattern at CCGG loci

3 Results and analysis

3.1 Extraction results of genomic DNAThe results of ultramicro nucleic acid protein analyzer showed that the concentration of total DNA among samples had little difference, ranging from 489.7 to 586.5 ng/μL. TheOD260/280andOD260/230met the test requirements.

3.2 Analysis of AFLP resultsA total of 120 pairs of primer combination were screened from AFLP test, and 9 pairs of primer combination with clear bands and good polymorphism were obtained, including E01-MC02, E01-MC03, E01-MC10, E01-CMC02, EO2-MC03, E02-MC10, E03-MC01, E06-MC01 and E06-MC03. WT and MT were analyzed by AFLP with 9 pairs of selected primers (Fig.1). Each pair of primers obtained 24 (MT E01-MC03) to 32 (WT E01-MC10) clear bands, with an average of 27.7 bands. The size of bands was 43-870 bp. A total of 499 bands were obtained (250 for WT and 249 for MT). The polymorphism of each pair of primer combinations was statistically counted (Table 4), and a total of 15 polymorphic bands were obtained, with the percentage of polymorphic bands of 6.00%. The bands obtained from MT was 8 less than that obtained from WT (E01-MC02 341 bp, E01-MC03 444 bp, 699 bp, E01-MC10 288 bp, E01-CMC02 382 bp, E02-MC10 309 bp, E03-MC01 304 bp and E06-MC03 693 bp), and 7 bands were emerged (E01-MC02 150 bp, EO2-MC03 183 bp, 480 bp, E03-MC01 181 bp, 247 bp, E06-MC01 154 bp and E06-MC03 110 bp). Each pair of primers had 1-3 polymorphic bands, among which E01-MC10, E01-CMC02, E02-MC10 and E06-MC01 had one polymorphic band; E01-MC02, E01-MC03, EO2-MC03 and E06-MC03 had two polymorphic bands; only E03-MC01 had three polymorphic bands. Most of the differential bands were 100-500 bp, accounting for 86.67% of the total polymorphic bands.

Table 4 Differential bands of materials detected by AFLP

Fig.1 Amplification map of AFLP markers

3.3 Analysis of MSAP resultsIn MSAP test, 81 pairs of primer combinations were screened, and 9 pairs of primer combinations with clear bands and good polymorphism were obtained, including E s1 H-M s1, E s2 H-M s2, E s3 H-M s3, E s4 H-M s4, E s5 H-M s5, E s6 H-M s6, E s7 H-M s8, E s8 H-M s8 and E s9 H-M s9. WT and MT were analyzed by MSAP with 9 pairs of selected primers (Fig.2), and each pair of primers obtained 10 (E s1 H-M s1 H, E s3 H-M s3 H) to 33 (E s5 H-M s5 M) clear bands, with an average of 15.78 H bands and 19.50 M bands. The size of bands was 43-900 bp. A total of 183 amplification loci were obtained (182 for WT and 178 for MT), and 635 clear bands were obtained (143 for WT H, 176 for WT M, 141 for MT H and 175 for MT M). Ten polymorphic bands were obtained from 6 pairs of primers; there were bands in WT H and WT M lanes at 499 bp of E s1 H-M s1, while no bands were found in MT M lane; at 466 bp of E s2 H-M s2, there was no band in WT H lane, but there were bands in WT M lane, while there were bands in both MT H and MT M lanes; at 230 bp of E s4 H-M s4, there were bands in WT H lane, while there was no band in MT H lane; at

Fig.2 Amplification map of MSAP markers

268 bp of E s4 H-M s4, there was no band in WT H and WT M lanes, while there were bands in MT M lane; at 224 bp of E s7/H-M s8, there were bands in WT M, while there was no bands in MT lanes; for E s8/H-M s8, there were bands in WT H lane at 66 bp, in WT M lane at 107 bp, and in WT H and WT M lanes at 243 bp, but there was no band in MT lanes; at 323 bp of E s9 H-M s9, there were bands in WT M lane, and in MT H and MT M lanes; at 828 bp, there were bands in WT H and WT M lanes, while there were bands only in MT M lane. In conclusion, WT had 4 more bands than MT in H and M lanes; a new band appeared in H and M lanes, respectively (Table 5). The percentage of polymorphic bands was 5.464 5% (10/183).

3.3.1Analysis of methylation levels of genomic DNA. A total of 182 and 178 amplification loci were obtained from WT and MT, respectively. There were 44 methylation bands obtained from WT (5 hemimethylation bands, 39 permethylation bands), and the total methylation rate was 24.175 8% (hemimethylation rate 2.747 3%, permethylation rate 21.428 6%). The unmethylation loci and unmethylation rate were 135 and 74.175 8%, respectively. There were 41 methylation bands obtained from MT (4 hemimethylation bands, 37 permethylation bands), and the total methylation rate was 23.033 7% (hemimethylation rate 2.247 2%, permethylation rate 20.786 5%). The unmethylation loci and unmethylation rate were 135 and 75.842 7%, respectively (Table 6). The permethylation rates of WT and MT were significantly higher than their semimethylation rates, that is, the main methylation mode was double intra-strand methylation. The frequency of methylation was 2.497 2%-21.107 5% (2.497 2% was the average semimethylation rate, 21.107 5% was the average permethylation rate). The semimethylation rate and permethylation rate of WT were both greater than those of MT, that is, WT had significant changes in methylation when MT bud sport was formed.

3.3.2Analysis of DNA methylation pattern. According to the classification method of DNA methylation pattern proposed by Duetal.[21], only 10 of the 15 DNA methylation pattern subtypes appeared, namely, A1, A2, A3, A4, B1, B4, C2, C3, C4 and D3. Most loci of MT maintained their original methylation types, accounting for 94.535 5% of the total number of loci, including 135 loci for A2 (73.770 5%), 3 loci for B1 (1.639 3%) and 35 loci for C3 (19.125 7%). Only a few loci changed, accounting for 5.464 5% of the total number of loci, among which there was 1 locus in A1, A3, A4 and D3, respectively, accounting for 0.546 4%, and 2 loci in B4, C2 and C4, accounting for 1.092 9%. The DNA methylation patterns were summarized and classified into four types: CG hypermethylation, CHG hypermethylation, CG demethylation and CHG demethylation. The hypermethylation and demethylation of WT at CCGG sites changed to varying degrees when MT bud sport was formed. There were 7 hypermethylation bands (4 CG hypermethylation bands and 3 CHG hypermethylation bands), and the hypermethylation ratio was 3.825 1% (CG hypermethylation ratio was 2.185 8% and CHG hypermethylation ratio was 1.639 3%). There were 3 demethylation bands (2 CG demethylation bands and 1 CHG demethylation band), and the demethylation ratio was 1.639 3% (CG demethylation bands were 1.092 9% and CHG demethylation bands were 0.546 4%). Hypermethylation and demethylation occurred simultaneously during bud sport formation, and the proportions of CG hypermethylation and CHG hypermethylation were significantly higher than those of CG demethylation and CHG demethylation.

Table 5 Differential bands of materials detected by MSAP

Table 6 DNA methylation level of materials tested

4 Discussion

4.1 AFLP and MSAP molecular markers can quickly identify bud sport materialsBud sport is a kind of somatic mutation, although the molecular breeding and other new breeding methods emerge in endlessly, bud sport plays an extremely important role in plant variety improvement because of simple and practical characteristics, and good bud sport varieties can be directly bred with clone strain and rapidly put into production, overcoming the disadvantages of long juvenile period of perennials[22]. Bud sport tends to occur when the natural environment is not suitable for plant growth (stress)[23]. The planting area of navel orange is huge in China, and the navel orange planting area is more than 126 700 hm2only in southern Jiangxi. There are a lot of bud sport materials in the field, which creates conditions for improving the variety structure of navel orange. The generation of bud sport is often accompanied by changes in genetic material and phenotype[24-25], and leaves are the most direct basis for finding bud sport materials in the field. For example, the leaves of Gannan navel orange, which is the Newhall bud sport material, have prominent veins and pale leaf color[26], but environmental changes easily cause changes in plant leaf phenotypes. In order to shorten the breeding time, molecular markers can be used to identify the phenotypic variation materials obtained from initial screening in the field to determine whether the genetic material has changed, thus greatly shortening the breeding period[22].

4.2 WT and MT are significantly different at the genetic levelAmong the 9 pairs of primers screened in AFLP test, there were 1-3 differential bands in each pair of combinations, and a total of 15 primers were obtained, with the percentage of polymorphic bands of 6.000 0%. In MSAP test, 5 pairs of primer combinations had differential bands, and a total of 10 differential bands were obtained, with the percentage of polymorphic bands of 5.464 5%. The results of AFLP and MSAP showed that MT was significantly different from WT in genomic DNA, suggesting that the difference in MT traits was caused by changes in its genetic materials. Hongetal.[27]analyzed 24 navel orange varieties by MSAP, and the results showed that the average methylation polymorphism ratio of these navel orange varieties was 6.7%, that is, methylation occurred frequently in navel oranges, which was consistent with the conclusion of this study. In terms of percentage of polymorphic bands, the percentage of polymorphic bands of AFLP was slightly higher than that of MSAP, but there was no significant difference between the two techniques, that is, the two molecular markers were suitable for the identification of bud sport materials with close genetic relationship.

4.3 Changes in DNA methylation may be responsible for the strip bulges of MTDNA methylation plays an important role in plant growth and development. In order to maintain normal growth and development, higher plants have different amounts and degrees of DNA methylation[28]. MSAP is a commonly used technique for detecting plant methylation, and the methylation rate detected by this technique is mostly 4.7%-60%[29]. The average methylation rate in this study was 23.604 7%, which was within this range. The main forms of DNA methylation at CCGG sites are semimethylation (single extra-strand methylation) and permethylation (double intra-strand methylation), and the latter is reported to be greater than the former[21,27]. The average semimethylation rate and average permethylation rate in this study were 2.497 3% (WT 2.747 3%, MT 2.247 2%) and 21.107 6% (21.428 6% and 20.786 5%), respectively. The permethylation rate was significantly higher than the semimethylation rate, that is, permethylation was the main methylation form of WT and MT. The semimethylation rate and permethylation rate of MT were lower than those of WT, that is, the DNA methylation of the bud sport material MT was changed compared with its parent WT. Only 10 of the 15 methylation pattern types of DNA methylation appeared in the experiment, and most of the loci kept the original methylation type, accounting for 94.535 5% of the total number of loci, which narrowed the scope for later analysis of the generation of mutations from DNA methylation levels.