Development and validation of the OVATE gene-based functional marker to assist fruit shape selection in tomato

Abstract

Fruit size, shape and colour are important determinants of fruit quality in tomato. Among the different genetic factors, the OVATE gene is a key regulator of fruit elongation in tomato. The loss-of-function recessive ovate allele results from a functional single nucleotide polymorphism (SNP) in the second exon of the gene to produce fruit elongation and variable fruit shapes in different genetic backgrounds. The mutation has also been associated with increased fruit firmness, a desirable trait for processing purpose of tomato. However, the recessive nature of this important mutant allele makes its identification and utilization in breeding programme difficult. Hence, we developed the OVATE gene-based functional marker using the tetra-primer amplification refractory mutation system (T-ARMS) strategy. The developed functional marker was capable of identifying the allelic status at OVATE locus in a co-dominant manner, using routine polymerase chain reaction (PCR) followed by standard agarose gel electrophoresis. Trait–marker association of the developed functional marker was validated in the F₂ segregants bearing elongated and non-elongated fruits. Thus, the functional marker developed and validated in this study will assist the tomato breeders in identification and introgression of the desired allelic version of the OVATE gene in a time-, labour- and cost-effective manner. Moreover, identification of the allelic status at the OVATE locus will help in exploring its interacting partners and modifiers for detailed understanding of the fascinating genetics behind fruit shape variation in tomato.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-021-03029-7.

Introduction

The fruit shape and mature fruit colour of tomato (Solanum lycopersicum L.) are important determinants of fruit quality and consumers’ preference. While wild types bear small and round fruits, domesticated tomatoes exhibit extreme variations for fruit shape and size (Tanksley 2004; Bai and Lindhout 2007; Paran and van der Knaap 2007). Tomato fruit size and shape are predominantly governed by both cell division and cell expansion (Azzi et al. 2015), where the major genetic determinants explaining this variation have been documented (Grandillo et al. 1999; Tanksley 2004; Paran and van der Knaap 2007; Azzi et al. 2015; Adhikari et al. 2020). Inheritance of the ovate/elliptical/pear fruit shape (i.e. having more fruit length than fruit width to give the fruits this characteristic shape) of tomato has been studied since long back (Lindstrom 1927). Using an inter-specific hybridization involving contrasting parents, this trait was documented to be governed by major quantitative trait loci (QTL) on chromosome 2 and minor QTL on chromosome 10 (Ku et al. 1999). Later, the candidate gene OVATE was found to encode a novel class of negative regulator, important for normal plant development (Liu et al. 2002). Through complementation studies, it was documented that OVATE prevents pear-shaped tomato fruit development, whereas the loss-of-function recessive allele ovate, resulting from a non-sense mutation in the second exon of the gene, is responsible for pear-shaped tomato fruits (Liu et al. 2002). Although OVATE was initially proposed as a growth repressor in the neck region of the tomato fruit (that gives the tomato fruits the ‘pear’ shape; Liu et al. 2002), later the mutant ovate allele has been identified in a large number of tomato genotypes bearing rectangular, ellipsoid, obovoid and heart-shaped fruits (Rodríguez et al. 2011), indicating the widespread importance of this gene in governing tomato fruit shape variation. Moreover, the OVATE family proteins (OFPs) regulate diverse plant growth and development events including ovule development, vascular development, fruit ripening and secondary cell wall formation (Wang et al. 2016).

The loss-of-function ovate allele arises from a single nucleotide polymorphism (SNP; G to T transversion in the 2nd exon of the gene) that generates a pre-mature stop codon (TAA in place of GAA; Liu et al. 2002). The recessive nature of this mutant allele makes its identification difficult during breeding programmes, and demands the availability of a functional molecular marker for the same. Depending upon the sequence at or around the SNP, genotyping can be performed through polymerase chain reaction (PCR) followed by restriction enzyme digestion of the amplicon [like, cleaved amplified polymorphic sequence (CAPS) and derived CAPS (dCAPS)]. Utilizing this strategy, a dCAPS marker has been previously reported for the allelic discrimination of the OVATE gene (Rodríguez et al. 2011). But the additional restriction digestion step and a risk of misinterpretation due to possible incomplete restriction digestion in PCR buffers are the major drawbacks for this strategy. On the other hand, the high-throughput genotyping of SNPs utilizes fluorescence-based detection techniques in costly, specialized equipment. Interestingly, utilizing two allele-specific primer pairs in a single PCR for allelic discrimination through amplicon length polymorphism, as done in the tetra-primer amplification refractory mutation system (T-ARMS) strategy (Ye et al. 2001), appears as a cost-, labour- and time-effective viable alternative. Hence, we attempted to develop OVATE gene-based functional marker using this strategy. The marker developed by us could identify the wild-type (functional) and mutant (non-functional) alleles of the OVATE gene present in different tomato genotypes in a co-dominant manner. Furthermore, we utilized this marker in a segregating F₂ population with variable fruit shapes to validate its value in breeding. Our results advocate the utility of this functional marker for selecting desirable fruit shape through marker-assisted breeding (MAB) in tomato, even in the seedling stage.

Materials and methods

Plant materials and segregating population

The present study used 23 tomato genotypes varying in their fruit shape, size and colour (Supplementary Table 1). Among these genotypes, 17 have red, 1 has pink, 4 have yellow and 1 has red–purple fruit colour. The fruit shape index (FSI) value was calculated from the ratio of average fruit length (polar diameter) to average fruit width (equatorial diameter) of three mature fruits.

A segregating F₂ population was developed from the cross involving two contrasting parents (IIHR 2614 and VRTOLCV-32) out of these 23 genotypes, where the genotype IIHR 2614 was used as the female parent. The resulting F₁ plants were tested for heterozygosity using parental polymorphic molecular marker (data not shown) and their seeds were collected to raise the F₂ population. The F₂ population was transplanted at field in 50 cm × 50 cm plant-to-plant and row-to-row spacing in November, 2020. Standard agronomic practices were adopted for the cultivation. The skewness and kurtosis of frequency distribution of FSI values were determined using web tool (https://www.socscistatistics.com/descriptive/histograms/).

Development of the OVATE gene-based perfect marker

Sequences of the second exon of the OVATE gene (Solyc02g085500 and GenBank: AY140893) were retrieved and compared to develop a marker on the basis of the functional G to T SNP. Sequence alignments were performed using Clustal W (https://www.genome.jp/tools-bin/clustalw) and pictorial representations were prepared using EsPript3 (https://espript.ibcp.fr/ESPript/ESPript). The T-ARMS primers (SlOvate Out F: 5′-GATCGTCGGTTTCTACGTCATCAGATAG-3′, SlOvate Out R: 5′-GATAATGCTTTCCGTTCAACGACAGA-3′, SlOvate InT F: 5′-ATAGTGAAGAAATCTCAGGACCCGTGCT-3′ and SlOvate InG R: 5′-TTCCATCATCGATCTCTTGAAATCCTC-3′) were designed using the Primer1 web-tool (http://primer1.soton.ac.uk/primer1.html).

Genotyping of allelic status of the OVATE gene

The allelic status of the OVATE gene in tomato genotypes and individual plants from the segregating F₂ population was investigated by PCR using the developed primers. The PCR was carried out in a total of 10 μl reaction volume containing 2 μl of genomic DNA isolated by a rapid method (Kumar et al. 2017). Each reaction mix contained 2 μl of 2X Taq premix (BioLit, SRL) and 5 pmol of each primers. The thermal cycling in an automated thermal cycler (Veriti, Applied Biosystems) was programmed at an initial denaturation at 94 °C for 4 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 64 °C for 30 s and extension at 72 °C for 30 s. The programme ended with a final extension at 72 °C for 7 min, followed by a hold at 4 °C for 2 min. The amplicons generated were analysed through 2.5% (w/v) agarose gel electrophoresis in the presence of ethidium bromide and imaged in gel documentation system (Gellite, Cleaver Scientific).

Results

Identification of the allelic status of the OVATE gene in tomato genotypes

The T-ARMS primers were designed to identify the allelic status of the OVATE gene in a co-dominant manner (Fig. 1). The SlOvate Out F and SlOvate Out R primers were designed to amplify a common band of 246 bp, irrespective of the allelic status at the OVATE locus. The SlOvate InT F primer was designed to amplify the non-functional ovate (T) allele-specific 139 bp band, in combination with the SlOvate Out R primer. Similarly, the SlOvate InG R primer was designed to amplify the functional OVATE (G) allele-specific 161 bp band, in combination with the SlOvate Out F primer. The efficacy of these primers was tested in 23 tomato genotypes varying for fruit shape, size and colour (Fig. Sf1). As expected, 20 out of the 23 genotypes contained the functional OVATE allele, as all these genotypes produced the functional OVATE (G) allele-specific 161 bp band along with the 246 bp common band. Only in case of three genotypes (i.e. Superbug SPS, IIHR 2614 and Punjab chhuhara), the ovate mutant allele was identified, as these three genotypes contained the non-functional ovate (T) allele-specific 139 bp band along with the 246 bp common band (Fig. 2A, asterisk marks). This observation is in unison with the fruit shape index (FSI) values recorded in these genotypes (Table 1). The genotypes carrying the ovate mutant allele had elongated fruit shape (Fig. 2B) with FSI values 1.26 (Superbug SPS), 1.74 (IIHR 2614) and 1.63 (Punjab chhuhara). The other 20 genotypes with FSI value ranging from 0.57 (Arka Vikas) to 1.03 (Kashi Aman) contained the functional OVATE allele.

Fig. 1.

Designing of the T-ARMS primers for detection of the functional SNP in the OVATE gene. Position of the SNP is shown inside the red ellipse. Position of the out and in primers and the target amplicons generated by their combinations are shown

Fig. 2.

Identification of OVATE allelic status in different tomato genotypes. A Image of 2.5% (w/v) agarose gel showing amplicons for common band (246 bp), OVATE (G) allele-specific band (161 bp) and ovate (T) allele-specific band (139 bp) in the genotypes. Presence of the ovate allele-specific amplicons is indicated by asterisk marks. B Representative fruit shape of wild-type (H-86) and ovate mutants (Superbug SPS, IIHR 2614 and Punjab chhuhara) identified through the developed marker. PC Punjab chhuhara; L = 100 bp DNA ladder (BioLit, SRL)

Table 1.

Haplotype analysis at OVATE locus in parental lines and selected F₂ segregants derived from contrasting parents

Sl. no.	Genotype	Mean fruit length (mm)	Mean fruit width (mm)	Mean FSI	Haplotype at OVATE locus	OVATE functionality as per haplotype
Parental lines
1	H-86	52.6	65.8	0.80	GG	Functional
2	BRDT-1	56.2	73.5	0.77	GG	Functional
3	Superbug SPS	44.2	35.0	1.26	TT	Non-functional
4	Arka Vikas	35.1	61.5	0.57	GG	Functional
5	Arka Alok	61.0	62.4	0.98	GG	Functional
6	CLN-B	35.4	39.7	0.89	GG	Functional
7	CLN-1621-L	41.1	44.5	0.92	GG	Functional
8	Sel 18	26.0	26.0	1.00	GG	Functional
9	Sun Cherry	20.9	21.7	0.96	GG	Functional
10	IIHR 2614	58.3	33.5	1.74	TT	Non-functional
11	IIHR 2612	39.8	47.6	0.83	GG	Functional
12	Kashi Chayan	44.2	57.6	0.77	GG	Functional
13	VRTOLCV-16	46.0	50.5	0.91	GG	Functional
14	VRTOLCV-32	53.4	60.6	0.88	GG	Functional
15	EC520078	13.2	14.1	0.93	GG	Functional
16	H-88-78-1	36.3	47.8	0.76	GG	Functional
17	VRKB 8	39.1	43.6	0.90	GG	Functional
18	VRKB 9	33.0	45.3	0.73	GG	Functional
19	VRKB 12	31.7	37.4	0.85	GG	Functional
20	VRCYT 4	29.1	29.2	0.99	GG	Functional
21	Kashi Aman	55.6	54.2	1.03	GG	Functional
22	Purple Tomato	55.6	59.0	0.94	GG	Functional
23	Punjab chhuhara	76.3	46.8	1.63	TT	Non-functional
Selected F2 segregants with elongated (E) and non-elongated (NE) fruits from the cross IIHR 2614 × VRTOLCV 32
1	E1	65.94	40.93	1.61	TT	Non-functional
2	E2	69.08	53.11	1.30	TT	Non-functional
3	E3	71.41	32.63	2.19	TT	Non-functional
4	E4	68.37	54.23	1.26	TT	Non-functional
5	E10	56.60	29.17	1.94	TT	Non-functional
6	E12	58.33	44.13	1.32	TT	Non-functional
7	E13	57.07	32.63	1.75	TT	Non-functional
8	E14	70.30	38.57	1.82	TT	Non-functional
9	E15	67.80	36.81	1.84	TT	Non-functional
10	E16	65.20	42.90	1.52	TT	Non-functional
11	E18	54.40	35.60	1.53	TT	Non-functional
12	E19	66.40	43.60	1.52	TT	Non-functional
13	E20	55.10	34.33	1.60	TT	Non-functional
14	NE1	44.84	53.27	0.84	GT	Functional/heterozygous
15	NE2	49.60	56.59	0.88	GG	Functional
16	NE3	45.32	53.53	0.85	GG	Functional
17	NE7	45.73	58.30	0.78	GG	Functional
18	NE8	47.00	51.97	0.90	GG	Functional
19	NE9	48.53	65.00	0.75	GG	Functional
20	NE10	41.90	55.03	0.76	GG	Functional
21	NE11	47.43	63.40	0.75	GT	Functional/heterozygous
22	NE12	43.33	47.43	0.91	GT	Functional/heterozygous
23	NE13	47.37	55.00	0.86	GG	Functional
24	NE14	53.17	70.03	0.76	GG	Functional

The lowest and highest mean FSI values recorded in tomato genotypes carrying functional OVATE alleles are underlined. The lines carrying functional OVATE allele in homozygous and heterozygous conditions are mentioned

Fruit shape variation in F₂ population segregating for the OVATE alleles

Variation in fruit shape was studied in a segregating F₂ population derived from contrasting parents IIHR 2614 (FSI = 1.74, containing the mutant ovate allele) and VRTOLCV-32 (FSI = 0.88, containing the functional OVATE allele). A great extent of fruit shape variation was observed in this segregating population (Fig. 3). The population containing 200 plants had a mean fruit length and mean fruit width of 50.73 mm and 47.07 mm, respectively. The population mean value for the FSI was 1.11, indicating the impact of the mutant ovate allele in this population. The fruit length ranged from 35.47 to 76.17 mm, whereas the fruit width ranged from 29.17 to 70.07 mm. Significant variation for the FSI value was recorded, which ranged from 0.65 to 2.19. The frequency of FSI classes, as revealed by the histogram, followed a positively skewed (skewness = 1.20) leptokurtic (kurtosis = 2.12) distribution (Fig. Sf2), where 101 plants had FSI values up to 1.07 and only 11 plants had FSI values more than 1.55. Fruit shape variation data of individual plants is summarized in Supplementary Table 2.

Fig. 3.

Variation of fruit shape in the F₂ plants derived from the IIHR 2614 (ovate) × VRTOLCV-32 (OVATE) cross-combination

Validation of the OVATE gene-based functional marker in F₂ segregants

The developed OVATE gene-based functional marker was utilized to find out the allelic status of this gene in the IIHR 2614 × VRTOLCV-32-derived F₂ segregants bearing elongated (E) or non-elongated (NE) fruits. The E-type plants had fruits with FSI values ranging from 1.26 to 2.19; on the other hand, the NE-type plants had fruits with FSI values ranging from 0.75 to 0.91 (Table 1, Fig. Sf3). As expected, all the E-type plants (13) were found to be homozygous for the non-functional ovate allele (Fig. 4A) like the female parent IIHR 2614. Among the tested NE-type plants, 8 out of the 11 plants were found to be homozygous for the functional OVATE allele (Fig. 4B) like the male parent VRTOLCV-32. The remaining three (NE#1, 11 and 12) plants were observed to be heterozygous for this gene, which is quite natural considering the recessive nature of the ovate mutation.

Fig. 4.

Validation of the functional marker in F₂ segregants derived from the IIHR 2614 (ovate) × VRTOLCV-32 (OVATE) cross-combination. A Identification of the ovate (T) mutant alleles in 13 segregants (E1–E20) bearing elongated fruit shape (FSI values ranging from 1.26 to 2.19). B Identification of the OVATE (G) wild-type allele in 11 segregants (NE1–NE14) bearing non-elongated fruit shape (FSI values ranging from 0.75 to 0.91). NE1, NE11 and NE12 are heterozygous at the OVATE locus. C Fruit shape of the representative segregants from the elongated (E12, E16 and E19) and non-elongated (NE10, NE13 and NE14) groups. L = 100 bp DNA ladder (BioLit, SRL)

For comparison, the OVATE haplotypes, as revealed by the marker in parental and segregating genotypes, were summarized and analysed with the corresponding FSI values (Table 1). The non-functional ovate haplotype TT yielded elongated fruits with higher FSI values, ranging from 1.26 to 2.19 including the parental genotypes and selected segregants. On the other hand, the functional OVATE haplotype (GG or GT) resulted in non-elongated fruits with lower FSI values, ranging from 0.57 to 1.03.

Discussion

Domesticated tomatoes show significant variation for fruit shape and size due to the purposeful artificial selections during domestication. The desirable fruit shape of tomato depends on the purpose; for example, flat, round juicy fruits are preferred for culinary purpose, whereas elongated fruits with thick pericarp are good for processing purpose (van der Knaap et al. 2014; Sun et al. 2017). The genetic factors determining tomato fruit growth and shape have been well documented (Rodríguez et al. 2011; Azzi et al. 2015). The LC (Muños et al. 2011) and FAS (Cong et al. 2008) loci regulate the locule number in tomato, which is pleiotropic to fruit shape and size. SUN encodes a positive regulator to cause elongated fruit type through gain-of-function mutation (Xiao et al. 2008), but OVATE encodes a negative regulator of fruit elongation, and the loss-of-function recessive mutant allele (ovate) leads to elongated fruit development (Liu et al. 2002). Further analyses have revealed the role of additional loci, like fs8.1 and sov (suppressor of ovate), in regulating fruit elongation in tomato (Ku et al. 2000; Rodríguez et al. 2013; Sun et al. 2015; Wu et al. 2018). Interestingly, the ovate mutation has been recently documented to dramatically increase the fruit firmness (Liu et al. 2017), a character highly desirable for storage and long distance transport of processing quality tomato. Naturally, the ovate mutant allele is quite important not only for fruit shape, but also for fruit processing quality improvement in tomato. Furthermore, the effect of the ovate mutation varies according to the genetic background, indicating its epistatic interaction with other gene(s) (Gonzalo and van der Knaap 2008). Supporting the involvement of epistatic interaction, we observed positively skewed leptokurtic frequency distribution for FSI values in segregating F₂ population derived from parents differing in allelic status of the OVATE gene (Fig. Sf2). Naturally, molecular identification of the loss-of-function ovate allele is also important to decipher the complex genetic regulation behind fruit shape variation in tomato.

The T-ARMS strategy (Ye et al. 2001) is a low-cost robust method for reliable identification of SNPs. In tomato, T-ARMS primers have been successfully applied for identification of Tm2 disease resistance allele (Arens et al. 2010). The strategy should have tremendous application in fruit quality improvement in tomato, as information about numerous SNPs associated with fruit carotenoid and flavonoid pigments’ variation is available (reviewed in Chattopadhyay et al. 2021a). Very recently T-ARMS-based functional markers for the identification of the high pigment-2^{dark green} (hp-2^dg) mutant allele capable of significantly increasing fruit pigmentation in tomato (Mukherjee et al. 2021) and for identification of null allele of a Pseudomonas-responsive receptor like protein gene of tomato (Chattopadhyay et al. 2021b) have been developed.

The OVATE gene-based functional marker developed in this study could discriminate the wild-type and mutant alleles present in different tomato genotypes (Fig. 2). The F₂ population segregating for the OVATE alleles was observed to have a high level of variation for fruit shape (Fig. 3), where the FSI values ranged from 0.65 to 2.19 (Supplementary Table 2). The yet unknown allelic status of the parental lines (i.e. IIHR 2614 and VRTOLCV-32) at the SUN, LC, FAS, fs8.1 and sov loci and possible interactions thereof might explain this fruit shape variation. Nevertheless, the lines with elongated fruits (having FSI value ranging from 1.26 to 2.19) were found to be homozygous for the recessive ovate allele (Fig. 4A, Table 1), justifying the utility of the developed marker. In a similar way, lines with non-elongated fruits (having FSI values ranging from 0.75 to 0.91) had the functional OVATE allele in homozygous or heterozygous condition (Fig. 4B, Table 1), which is in unison with the recessive nature of this mutation. The overall haplotype–FSI association is in complete agreement with the negative regulation imparted by OVATE on tomato fruit length (Table 1), where the non-functional TT haplotype always resulted in elongated fruit type with higher FSI values. Hence, this result clearly demonstrates the utility of the developed OVATE gene-based co-dominant functional marker in genotyping for early assessment of tomato fruit shape and selection of genotypes with firm and elongated fruits, even at seedling stage.

Conclusion

Gene-based functional markers are the most valuable tools in molecular breeding programmes. The OVATE gene-based functional marker developed and reported in this study will not only assist the breeders in identification and introgression of this important allele for fruit shape and quality improvement in tomato, but also will be helpful to decipher the complex genetics behind fruit shape variation in tomato.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

TC conceived the idea. DM, AM and TC designed the primers and performed the laboratory experiments. DM, AM, SA, and TC carried out plant hybridization, population development, and field and fruit shape data recording works. DM and AM contributed equally. All the authors took part in the preparation and correction of the manuscript. All the authors read the final manuscript and approved it.

Declarations

Disclosure of potential conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Research involving human participants and/or animals

No human participants and/or animals were involved in this study.

Informed consent

Not applicable as no human participants and/or animals were involved in this study.

Accession numbers

Tomato OVATE gene (Solyc02g085500 and GenBank: AY140893).