Genetic mapping and quantitative trait loci analysis for pistillate flowers per node and multi-pistillate flower traits in the F₂ cucumber population

Abstract

This study focused on cucumbers’ multi-pistillate flower (MPF) trait, which is essential for high yields. A genetic linkage map was constructed using a population of 219 F₂ plants to analyze quantitative trait loci (QTL) associated with MPF traits. Crossbreeding of EWSCU-809 (MPF) with EWSCU-989 (single pistillate flower: SPF) generated an F₁ hybrid that self-pollinated to form an F₂ population. Based on 244 single nucleotide polymorphic markers across seven cucumber chromosomes, a linkage map facilitated QTL analysis considering average pistillate flowers (PFs) per node and nodes with MPF traits. The results indicated a 9:6:1 epistatic ratio in the F₂ populations, revealing recessive allele control of the MPF trait in gynoecious plants. Three QTLs (qMP2.1, qMP3, qMP7) on chromosomes 2, 3, and 7 were associated with average PFs per node, explaining 5.6 to 10.3% of phenotypic variation. Four QTLs (qMP2.2, qMP3, qMP4, qMP7) on chromosomes 2, 3, 4, and 7 were linked to the presence of nodes with MPF traits, explaining 5.8 to 10.6% of phenotypic variance. Notably, QTL regions overlapped between the two datasets, suggesting pleiotropic effects, particularly on chromosomes 3 and 7. These reliable QTLs have the potential to improve breeding programs, enhance PF development, and increase cucumber yields.

Introduction

The cucumber (Cucumis sativus L.) belongs to the Cucurbitaceae family and is an economically important vegetable crop worldwide (Simi et al. 2017). While vegetables account for 12% and represent the top three in world crop production by commodity group, in the vegetable group, cucumber has an 8% global share, following tomatoes (16%) and dry onions (9%) (FAO 2021). The cucumber vine displays three main flower types: staminate (male), pistillate (female), and hermaphrodite (bisexual/perfect), and two minor genes with the M gene that regulate unisexual flowers. Numerous studies on cucumber research have identified three genes, F (femaleness), m (andromonoecy), and a (androecy), that control sex determination in cucumbers. These genes are linked to the 1-aminocyclopropane-1-carboxylate synthase (ACS), a crucial regulatory enzyme in the ethylene biosynthetic pathway. Genes involved in ethylene signaling pathways may influence the unique development of sex organs (Li et al. 2009, Matsumura et al. 2014, Mibus and Tatlioglu 2004, Shyam et al. 2020). The most prevalent cucumber plant is the unisexual line, specifically monoecious (MMffAA), which exhibits female (pistillate) and male (staminate) flowers on the same plant. The second type is gynoecious (GY: MMFFA/a), which exclusively produces pistillate flowers; each node may have one or more female flowers. This trait may not be ideal in specific production systems with suboptimal cultural practices or poor production conditions because not all female flowers develop marketable fruits. The third type is predominantly gynoecious (PGY: MMFfA/a), wherein the plant displays pistillate and staminate flowers (Bu et al. 2016, Li et al. 2009, Verma et al. 2018). The other unisexual line is androecious (MMffaa/mmffaa), with only staminate flowers on individual plants. Cucumber plants (bisexual lines) exhibit other sex expression types, including those that produce hermaphrodite (mmFFA/a) flowers containing both staminate and female parts on the same flower. The following common types are andromonoecious (mmffAA), gynomonoecious (gym), and trimonoecious (tr) bearing staminate and hermaphrodite, pistillate and hermaphrodite, and pistillate, staminate, and hermaphrodite flowers on the same plant, respectively (Fig. 1) (Li et al. 2009, Megharaj et al. 2017). Many factors affect sex expression, such as plant growth regulators (Adhikari et al. 2012), light induction based on season and day length, for example, blue light that favors pistillate flower formation (Zhou et al. 2018), and all the factors affecting gene control of F (Mibus and Tatlioglu 2004, Pawełkowicz et al. 2019), m, a, and tr (Kubicki 1969, Li et al. 2009).

Fig. 1.

Sex expression types in cucumber (figure modified from Li et al. 2019).

Cucumbers generally produce SPFs or a single fruit per node. Double and plural pistillate flowering are also called multi-pistillate flowering (MPF). The identification of a single gene initially referred to as mp by Nandgaonkar and Baker (1981), has provided insights into the genetic foundation of MPF per node in specific plants. Subsequent research by Fujieda et al. (1982) suggests a correlation between mp and its designated gene pf, which has three distinct alleles: pf⁺ for a single pistillate, pf^d for a double pistillate, and pf^m for MPF. Remarkably, the incomplete dominance exhibited by these alleles introduces an additional layer of complexity, underscoring the intricate genetic mechanisms governing pistillate flowering variations. This collaborative effort by various research groups provides valuable insights into the genetic regulation of flower development, paving the way for further exploration and comprehension of plant genetics. MPF, which corresponds to the bearing of two or more cucumber flowers/fruits per node, can save some of the work involved in pruning and vine training and provides the advantage of producing a large number of fruits (Fujieda et al. 1982). Yuan et al. (2008) and Dou et al. (2010) reported that the number of fruits per plant is closely related to cucumber yield and has major potential for increasing this. Moreover, the PF ratio influences the fruit per plant or yield.

Plant biotechnology is a valuable and practical plant breeding tool. Many techniques have helped breeders identify or confirm genetic control in hybrids, interspecific hybrids, and other trials (Kumchai et al. 2013). Marker-assisted selection (MAS) is an advanced molecular breeding program that supports conventional breeding (Das et al. 2017). Genotyping technology has enabled the cost-effective development of single nucleotide polymorphism (SNP) markers. Moreover, as reported in many crop breeding schemes, the MAS method can be used for various purposes, such as marker-assisted backcrossing (Lee et al. 2022). It can also help identify the Rf/rf locus and normal/sterile cytoplasm in chili breeding programs (Na Jinda et al. 2023). Improving cucumber traits is essential for breeding programs (Yuan et al. 2008). In this regard, the environment has a strong effect on phenotyping; for example, the sex expression of cucumber at high temperatures and long photoperiods results in cucumbers having less PF (Fujieda et al. 1982, Megharaj et al. 2017). Therefore, molecular markers may increase the efficiency and accelerate plant breeding programs. Moreover, due to environmental effects, MAS may help select desirable traits that are not present in the field (Collard et al. 2005). SNPs are the basis for differences between alleles and can be used as simple genetic markers to identify the position of virtually every gene. They have great potential for detecting associations between the allelic forms of genes and phenotypes. Analysis of SNP haplotypes provides an effective way to associate alleles with traits (Rafalski 2002). SNP markers can distinguish between homozygous and heterozygous phenotypes, allowing many samples to be rapidly identified and genotyped (Junta et al. 2020). Therefore, this study aimed to pinpoint SNP markers linked to MPF traits by utilizing the F₂ population resulting from a cross between female MPF and male SPF traits. The primary objective was to identify stable varieties through genotyping and molecular marker techniques in horticultural genetics, providing a guideline for supporting the improvement and development of cucumber varieties alongside conventional methods.

Materials and Methods

Plant materials

The plant materials comprised a wide range of cucumber varieties. Parental lines comprising four MPF and two SPF varieties were obtained from Hortigenetics Research (S.E. Asia) Ltd.

Evaluation and homogeneity testing for parental line selection in cucumber varieties

Two to three young leaves from the upper portion of the cucumber plants were sampled for homogeneity testing. We employed a set of 70 SNP molecular markers derived from the transcriptome resequencing of cucumber varieties from Hortigenetics Research (S.E. Asia) Ltd., utilizing the KASPar technique. This dataset served as the foundation for selecting parent varieties to establish F₁ and F₂ population databases. The F₁ hybrid was the result of crossing an inbred EWSCU-809 (MPF) line as the female parent with EWSCU-989 (SPF) as the male parent. Subsequently, the F₁ generation underwent self-pollination, yielding 219 plants in the F₂ population (Fig. 2), which were sown under greenhouse conditions during the rainy season of 2021, and individual plants were singled out for branch analysis. Data collection for the F₁ and F₂ parental lines involved observing pistillate flower numbers on the main stems from the 5^th to 25^th nodes. Individual plant data included the average number of pistillate flowers per node and the number of nodes with MPF traits. This information facilitated the calculation of heterosis and the identification of dominant and recessive traits.

Fig. 2.

The crossing between parental lines to produce F₁ and F₂ populations for QTL mapping and analysis.

Evaluation and homogeneity testing were conducted to select parental lines of the cucumber varieties. Two to three young leaves from the cucumber plant upper portions were sampled for homogeneity testing. A set of 70 SNP molecular markers derived from the transcriptome resequencing of cucumber varieties from Hortigenetics Research (S.E. Asia) Ltd., was employed using the KASPar technique. This dataset formed the basis for selecting parent varieties to establish the F₁ and F₂ population databases.

The F₁ hybrid was obtained by crossing an inbred line with EWSCU-809 (MPF) as the female parent and EWSCU-989 (SPF) as the male parent. Subsequently, the F₁ generation was self-pollinated, yielding 219 plants in the F₂ population (Fig. 2). These plants were sown under greenhouse conditions during the rainy season of 2021, and individual plants were selected for branch analysis. Data collection for the parental lines F₁ and F₂ involved observing PF numbers on the main stems from the 5^th to 25^th nodes. Individual plant data included the average PF numbers per node and the number of nodes with MPF traits. This information facilitated the calculation of heterosis and the identification of dominant and recessive traits.

DNA extraction and Polymerase Chain Reaction (PCR)

Young true leaves on the main stem of parental lines F₁ and F₂ were collected for DNA extraction using the cetyltrimethylammonium bromide (CTAB) method, and genomic DNA was diluted to a final concentration of 15 ng/μL as described by Nishiguchi et al. (2002).

PCR primers forward 1 (F1), forward 2 (F2), and reverse (R) were used for the PCR. A buffer of KASPar master mix (2X) and primers were added to the assay plate. Next, a DNA 15 ng/μL template was prepared and put in a DNA plate. The assay and DNA plate were centrifuged for 20 min at 3,500 rpm (1178 × g) and placed into a Nexar machine for subsequent dispensing onto array tape^TM (384 holes). The DNA and assay array tape was then sealed with a plastic cover tape.

Afterward, the array tape was placed in a Thermal Cycler (Soellex) to increase the DNA and centrifuged for 10 min at 5,000 rpm (2404 × g) (total 2 hours). The results were analyzed using Nexar-Araya system Intellics^® software v 1.0.0.0. All processes used ultra-high-throughput SNP technology (Douglas Scientific).

Primer screening for polymorphic markers and identification and substitution mapping of quantitative trait loci (QTL) for MPFs

A total of 1,914 SNP markers were designed to cover all seven cucumber chromosomes and screened to identify polymorphism in female, male, and F₁ hybrids. SNP genotyping was performed using the KASPar^TM technology by KBioscience^®. In the two fluorescence resonance energy transfer (FRET) cassettes, a primer was conjugated to a fluorescent dye (VIC, FAM, or heterozygous) (Raitio et al. 2012), and the DNA was amplified using specific primers. When the FRET cassette primer hybridized to DNA, the quencher was separated (Saeko et al. 2020).

The average number of PFs per node and the number of nodes with MPF traits of individual plants from 219 F₂ plant populations were used for QTL mapping, resembling linkage analysis with the JoinMap4.0 program and MapQTL 6 software. Out of the 1,914 SNPs screened, 244 were found to be polymorphic. Marker data were calculated using the logarithm of odds (LOD) with a multiple QTL model.

Results

Evaluation and homogeneity testing for parental line selection of cucumber varieties

EWSCU-809 and EWSCU-989 were chosen as the female and male lines, respectively, demonstrating a GY-type with MPF and PGY with SPF. Furthermore, they exhibited impressive homogeneity, with 97.14% in the female line and 95.24% in the male line (Table 1, Fig. 3).

Table 1.Variety evaluation and homogeneity testing of six cucumber varieties

Variety	Phenotype description of flowering	Sex expression	Homogeneity (percentage)	Selecting evaluattion
EWSCU-809	MPF	Gynoecious	97.14a	Female line
EWSCU-173	MPF	Predominantly Gynoecious	95.95ab	Discard
EWSCU-735	MPF	Predominantly Gynoecious	94.76bc	Discard
EWSCU-536	MPF	Gynoecious	93.57c	Discard
EWSCU-989	SPF	Predominantly Gynoecious	95.24abc	Male line
EWSCU-076	SPF	Predominantly Gynoecious	90.24d	Discard
F-test			0.00
CV (%)			1.19

Means followed by the same letters indicate no differences at P ≤ 0.05 by the least significant.

Multi-pistillate flowers (MPF).

Single-pistillate flowers (SPF).

Fig. 3.

Single-pistillate flowers shown by the red circle (A), multi-pistillate flowers shown by the dark blue circle (B).

MPF traits in parental lines, F₁, and F₂ populations

P1 exhibited the highest number of female flowers per node (3.76) and the highest number of nodes with MPF traits (19.50), results markedly different from those of P2 and F₁. Heterosis yielded negative results in both sets of data. Within the F₂ population, there was a segregation in the flowering phenotype, distinguishing between MPF and SPF traits and variations in the number of female flowers per node and the number of nodes with MPF traits (Table 2, Fig. 4).

Table 2.Mean of number of pistillate flowers (PF) per node, number of nodes with multi-pistillate flowers (MPF) and the mid-parent heterosis

Accession/Heterosis	Phenotype of flowering	Generation	Average of PF per node	Num. of nodes with MPF
EWSCU-809 (P1)	Female line (MPF)	F6 (inbred line)	3.76 ± 0.54a*	19.50 ± 1.18a
EWSCU-989 (P2)	Male line (SPF)	F8 (inbred line)	0.46 ± 0.17c	0.00 ± 0.00c
EWSCU-991	F1 hybrid	F1	1.05 ± 0.07b	1.30 ± 1.16b
EWSCU-992	Segregating population	F2	Segregation	Segregation
Mean	—	—	1.76	6.77
F-test	—	—	0.00	0.00
CV (%)	—	—	18.63	16.44
Heterosis	—	—	–50.23%	–90.08%

*Means followed by the same letters indicate no differences at P ≤ 0.05 by the least significant.

Fig. 4.

The frequency distribution histograms of four populations (10 plants of the female and male lines, F₁ hybrid, and 219 F₂ populations). The average number of pistillate flowers per node (A) and the number of nodes with multi-pistillate flowers (B).

The distribution and chi-square (χ²) estimates of F₂ populations comprising 219 plants were analyzed for the average number of MPFs per node and the presence of MPF traits in nodes. The populations were categorized based on the ratio of non to partial to strong MPFs, and the observed ratio was 9:6:1 in both the two-parameter studies (Table 3).

Table 3.The segregation and chi-square (χ²) analysis of the F₂ populations (219 plants) were carried out considering the average number of multi-pistillate flowers (MPF) per node and the number of nodes with MPF traits, categorized into non MPF, partial MPF, and strong MPF

Trait	Accession	Observed				Expected			χ2 (9:6:1)	P-value
		No. of non MPF (plant)	No. partially MPF (plant)	No. of strong MPF (plant)		No. of non MPF (plant)	No. partially MPF (plant)	No. of strong MPF (plant)
The average no. of MPF /node	EWS-809 (P1)	0	0	10		0	0	10
	EWS-989 (P2)	10	0	0		10	0	0
	EWS-991 (F1)	10	0	0		10	0	0
	EWS-992 (F2)	139	67	13		123.1875	82.125	13.6875	4.850	0.0885
The no. of nodes with MPF	EWS-809 (P1)	0	0	10		0	0	10
	EWS-989 (P2)	10	0	0		10	0	0
	EWS-991 (F1)	10	0	0		10	0	0
	EWS-992 (F2)	113	85	21		123.1875	82.125	13.6875	4.850	0.0885

df = 2.0, α = 0.05, χ² = 5.99.

Markers screening

In total, 1,914 markers were screened for female and male lines and F₁ hybrids. These were separated into polymorphic (Fig. 5A) and monomorphic (Fig. 5B) variants. Distributed across seven chromosomes, 244 polymorphic SNP markers (Fig. 6) were selected for subsequent QTL analysis.

Fig. 5.

Polymorphic (A), and monomorphic (B) marker.

Fig. 6.

A physical map of 244 polymorphic markers on 7 chromosomes.

The QTL of the average number of MPFs per node and the number of nodes with MPF traits were analyzed using LOD scores. The results of the average number of MPFs per node revealed three notable QTLs on chromosomes 2, 3, and 7, which are qMP2.1, qMP3, and qMP7, with LOD scores of 4.23, 7.53, and 4.53, PVE values of 5.6, 10.3, and 5.9%, and DOM values of –0.068, –0.083, and –0.004, respectively. Based on the number of nodes with MPF traits, four important QTLs were detected on chromosomes 2, 3, 4, and 7, which are qMP2.2, qMP3, qMP4, and qMP7, with LOD scores of 7.58, 4.99, 4.29, and 6.9, PVE values of 10.6, 6.8, 5.8, and 9.6%, and DOM values of –0.101, –0.108, –0.526, and 0.118, respectively (Table 4, Fig. 7).

Table 4.The statistically significant quantitative trait locus (QTL) scores were determined based on the average number of pistillate flowers (PF) per node and the presence average number of multi-pistillate flower traits identified in the F₂ population

Traita	QTL*b	Chrc	Flanking markersd	Start-end position (cM*)	Position (bp*)	LOD peak	Additive effecte	DOMf	PVE (%)g
APF	qMP2.1	2	CUCSNP-0009679–CUCSNP-0009699	87.525–92.598	19,036,261	4.23	0.132226	−0.06823	5.6
	qMP3	3	CUCSNP-0009679–CUCSNP-0009699	46.878–51.269	12,522,694	7.53	0.183583	−0.08286	10.3
	qMP7	7	CUGSNP-7-60–CUCSNP-0032394	49.914–52.922	13,452,982	4.53	0.152063	−0.00418	5.9
NNMPF	qMP2.2	2	CUGSNP-2-96–CUGSNP-2-95	83.258–86.717	8,238,869	7.58	2.05504	−0.10091	10.6
	qMP3	3	CUCSNP-0011238–CUCSNP-0011410	46.878–51.269	12,522,694	4.99	1.79866	−0.10814	6.8
	qMP4	4	CUCSNP-0017116–CUCSNP-0017117	6.892–14.892	1,814,861	4.29	1.59174	−0.52615	5.8
	qMP7	7	CUGSNP-7-60–CUCSNP-0032394	49.914–52.922	13,452,982	6.9	1.95556	0.117892	9.6

* cM: Centi Morgan, QTL: quantitative trait locus, bp: base pair.

^a APF: Average number of pistillate flower/node, NNMPF: Number of nodes with MPF.

^b QTL: Designation of the QTL is in accordance with the rules recommended by McCouch (2008).

^c Chr: Chromosome.

^d LOD peak: Logarithm of the odds ratio (LOD) score of ≥2.5 was set as threshold for this data.

^e Additive effect: The value indicates the decrement in the trait value.

^f Dom: Dominance effect.

^g PVE (%): Total phenotypic variance (PVE%) in percentage explained by the QTL.

Fig. 7.

A physical mapping of 7 QTL positions, using the average number of pistillate flowers per node and the number of nodes with multi-pistillate flowers from 219 F₂ population plants, detected on chromosome 2 (A), chromosome 3 (B), chromosome 7 (C), and chromosome 4 (D).

Discussion

The female line, characterized by GY, is commonly employed as the female parent to create hybrid lines and for cucumber seed production. This choice was made to enhance fruit production per plant, benefiting farmers through increased yields and higher incomes. It is advantageous for cucumber plants to have MPFs per node and numerous nodes with abundant PFs (More 2001). This study focused on two parental lines exhibiting distinct phenotypes. The first was the MPF-type EWSCU-809 (female line: inbred line), characterized by 3.76 PFs per node. In contrast, the SPF-type EWSCU-989 (male line: inbred line) features 0.46 PFs per node. EWSCU-809 had 19.50 nodes with multiple PFs, whereas EWSCU-989 had one PF per node. Following crossbreeding, EWSCU (809 × 989: F₁; EWSCU 991) manifested a heterozygous phenotype. The presence of one SPF (one flower per node) indicated that its genotype closely resembled that of a homozygous dominant male parental line. MPFs in the female line signify a homozygous recessive trait (Miko 2008a). The F₂ population (EWSCU 992) exhibited pronounced segregation of SPFs and MPFs. Identifying a single gene initially referred to as mp by Nandgaonkar and Baker (1981), has provided insights into the genetic foundation of MPFs per node in specific plants.

The segregation pattern in the F₂ generation adhered to Mendel’s law of segregation, with selfing of the F₁ generation resulting in a segregation ratio of 1:2:1. There were two alleles in a gene pair, one of which was completely dominant, resulting in a 3:1 ratio. Mendel’s law explained the relationship between these phenotypes based on observations of both the F₁ and F₂ generations. This study revealed a 9:6:1 segregation ratio for both characteristics in the phenotype of the F₂ generation, indicating the influence of epistatic interactions, which implies complete dominance of both gene pairs. However, when one gene was dominant, the other was suppressed. The epistatic ratio type comprises duplicate genes with cumulative effects or additive gene interactions (Huang et al. 2021, Miko 2008a). Hence, MPF traits in gynoecious cucumber lines are governed by a recessive allele, and genetic analysis suggests the involvement of one (mp) or two major genes and several modifying factors that influence this characteristic. In contrast, the dominant allele controls the SPF (Nandgaonkar and Baker 1981). The genotype determines the phenotype, with the environment contributing as another influencing factor (Miko 2008b). These results indicate that the F₂ population serves as a fundamental and advanced model for studying Mendelian genetics, making it well-suited for genetic investigations in both conventional and molecular breeding. It illustrates key genetic concepts, such as dominance, epistasis, linkage, and statistical predictions (Miko 2008c).

Cucumber horticultural characteristics are important traits controlled by QTL. With the exponential increase in cucumber QTL mapping studies, the naming of quantitative traits and the corresponding genes and QTL responsible for key phenotypic traits (Sheng et al. 2019, Wang et al. 2020). Numerous traits associated with flowering time (FT) and sex expression in cucumber plants are correlated, suggesting potential regulation by common ethylene-related pathways, as exemplified by the identification of QTL clusters on chromosomes 1, 3, 5, 6, and 7 in cucumber (Wang et al. 2020). Here, examination of flowering types, male flowers per node (MPF), and nodes with MPFs in GY cucumber plants revealed consistent chromosome locations on 3 and 7, denoted as qMP3 and qMP7, respectively, implying potential pleiotropic effects. Additionally, other cucumber traits exhibited similar pleiotropic effects, as evidenced by Wang et al. (2021), who identified a pleiotropic B locus on Chr4 controlling fruit ground color at maturity and fruit spine color in Sikkim cucumbers. Furthermore, studies by Somta et al. (2020) on black gram (Vigna mungo (L.) Hepper) confirmed the pleiotropic effect of the mog gene on linkage group 6. The study was observed across various traits, including seed length, seed width, leaf length, leaf width, leaf area, pod width, straight pod length, and stem plant height represented by qSdl6.1 and qSd100wt6.1, qSdwa6.1, qLfl6.1, qLfw6.1, qLfa6.1, qPdw6.1, qPdl6.1, and qPlh6.1, respectively. Notably, QTL mapping of cucumber horticultural traits validated 36 QTLs associated with FT, fruit length (FL), fruit diameter (FD), fruit shape (LD), fruit number (FN), and powdery mildew resistance. Among these, five moderate or major effect QTLs for FL, FD, LD, and FN within the inversion likely indicate the pleiotropic effects of andromonoecy (m) or the cn locus (Pan et al. 2020). The findings presented in Table 4 indicate that LOD scores >3 signify the presence of a molecular marker at a position correlated with the QTL. The QTL of MPFs of chromosomes 3 exhibited a phenotypic variance of 10.3% (Table 4), indicating a marked genetic segregation in the F₂ generation, particularly in the homozygous recessive genotype. Han et al. (2010) demonstrated that the five QTL of MPFs in cucumber progeny exhibited substantial phenotypic variance, surpassing 10% across six QTL cultivated during both the autumn and spring seasons. Klosinska et al. (2006) reported that a pair of PFs was recessive and remained unexpressed in the F₁ hybrid generation. This study observed the absence of a major QTL, possibly because this trait is influenced by environmental factors such as day length, air temperature, and photoperiod. For example, elevated air temperatures may decrease the presence of PFs, whereas lower air temperatures may increase their occurrence (Hikosaka and Sugiyama 2004). Han et al. (2010) highlighted that QTL and LOD scores in cucumber recombinant inbred lines of the F₉ generation varied between seasons, specifically between spring and autumn, and across different years. Sheng et al. (2019) indicated that QTL FT6.3 contributed to photoperiod-sensitive flowering time during domestication, whereas QTL FT1.1 played an important role in regulating flowering time in cultivated cucumbers. Sex expression in Cucurbitaceae is influenced by several factors (Li et al. 2019). Ethylene is a crucial element that induces PF in cucumber (Manzano et al. 2013). Consistent with prior research, Li et al. (2019) emphasized that ethylene, which functions as a plant hormone, plays a pivotal role in initiating PF formation and determining the PF ratio (Megharaj et al. 2017, Pan et al. 2018). Moreover, Dou et al. (2015) demonstrated that the ratio of pistillate to staminate flowers was season-dependent. Therefore, strategically selecting markers could contribute to developing cucumber varieties with stable yields and consistent sex expression. Considering these findings, the identified QTL could serve as valuable tools for MAS in breeding programs focused on MPF traits. Furthermore, QTL qMP3 was identified at the same genomic position on chromosome 3 and QTL qMP7 at the corresponding position on chromosome 7. This observation applies to data patterns involving the average number of PF per node and the incidence of MPF traits in the F₂ population. Genetic and environmental factors influence sex expression in cucurbits. Therefore, the breeding program needs to consider and control flower sex for a successful program. The MPF per node is a recessive trait less pronounced in F₂ populations. To further advance the study of MPF for cucumber improvement, an increased population of plant samples for thorough examination and additional screening of primers for polymorphic markers and identification is necessary for result validation. Moreover, investigating QTL related to PF on the cucumber chromosome is a focal point for studying cucumber breeding programs. Identifying high yields per plant through this observation could benefit plant breeders aiming to create new varieties by modifying conventional methods, such as backcross breeding and hybrid cross. Incorporating molecular techniques would expedite gene transfer to target plants, streamlining the overall process.

In summary, the investigation of the cucumber parental lines EWSCU-809 and EWSCU-989 revealed a notable level of homogeneity but displayed negative heterosis in the F₁ generation. The F₂ generation exhibited marked segregation in MPF traits and the number of nodes with MPFs. Identification of four QTL on chromosomes 2, 3, 4, and 7 in cucumber revealed that only qMP2.1, qMP3, and qMP7 are associated with MPF traits. The number of nodes with MPF traits were detected in four QTL: qMP2.2, qMP3, qMP4, and qMP7. Notably, qMP2.2 on chromosome 2 had the most notable impact, with an LOD score of 7.58, closely followed by qMP3, on chromosome 3, with an LOD score of 7.53. This study highlights the effectiveness of QTL analysis regarding the number of female flowers per node, and the nodes with female flowers indicated that nearly all QTL were associated with the same SNP molecular markers or had closely adjacent positions. The only exception was qMP4 on chromosome 4, which exhibited a specific association with the trait related to the data on the number of nodes with more than one female flower. These QTL show promise for their potential use in supporting conventional breeding efforts to enhance efficiency and expedite selection and improvement processes, ultimately leading to higher cucumber yields and increased farmer’s income.

Author Contribution Statement

Conceptualization, N.A., W.S., N.O., S.C. and J.K.; methodology, N.A., N.O., W.S., S.C. and J.K.; software, N.A.; validation, J.K., W.B. and N.A.; formal analysis, N.A.; N.O. and W.S.; investigation, J.K. and N.A.; resources, N.A., W.S., N.O. and S.C.; data curation, N.A., W.S., N.O. and S.C.; writing original draft preparation, N.A., W.S. and N.O.; writing review and editing, J.K. and W.B.; visualization, J.K., N.O. and N.A.; supervision, J.K., W.B., N.O. and N.A.; project administration, N.O., S.C. and N.A.; funding acquisition, N.O., W.S. and S.C. All authors have read and agreed to the published version of the manuscript.

 Acknowledgments

The faculty of Agriculture at Chiang Mai University partially supported this research work, and the major was supported by Hortigenetics Research (S.E. Asia) Limited, Thailand. We thank them for their kind support.

Literature Cited

•  Adhikari,  S.,  T.K.  Bandyopadhyay and  P.  Ghosh (2012) Hormonal control of sex expression of cucumber (Cucumis sativus L.) with the identification of sex linked molecular marker. Nucleus 55: 115–122.
•  Bu,  F.,  H.  Chen,  Q.  Shi,  Q.  Zhou,  D.  Gao,  Z.  Zhang and  S.  Huang (2016) A major quantitative trait locus conferring subgynoecy in cucumber. Theor Appl Genet 129: 97–104.
•  Collard,  B.C.Y.,  M.Z.Z.  Jahufer,  J.B.  Brouwer and  E.C.K.  Pang (2005) An introduction to markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop improvement: The basic concepts. Euphytica 142: 169–196.
•  Das,  G.,  J.K.  Patra and  K.H.  Baek (2017) Insight into MAS: A molecular tool for development of stress resistant and quality of rice through gene stacking. Front Plant Sci 8: 985.
•  Dou,  X.,  X.  Zou,  X.  Li,  D.  Shen and  H.  Wang (2010) Genetic analysis on node rates of multi-pistillate flowers in cucumber. Acta Hortic 871: 111–118.
•  Dou,  X.,  S.  Di,  Z.  Xiaohui,  C.  Jiaqi,  L.  Sigeng,  S.  Jiangping,  W.  Haiping,  Z.  Wei and  L.  Xixiang (2015) Diversity of sex type and seasonal sexual plasticity in a cucumber germplasm collection. Hortic Plant J 1: 61–69.
• FAO (2021) World food and agriculture statistical year book. Rome.
•  Fujieda,  K.,  Y.  Fujita,  Y.  Gunji and  K.  Takahashi (1982) The inheritance of plural-pistillate flowering in cucumber. J Japan Soc Hort Sci 51: 172–176.
•  Han,  M.,  X.G.  Fang,  S.Z.  Ping,  W.  Jian,  F.Z.  Yuan and  Z.Z.  Xian (2010) Mapping QTLs for multiple pistillate flowers in cucumber. Acta Hortic Sin 37: 1449–1455.
•  Hikosaka,  S. and  N.  Sugiyama (2004) Characteristics of flower and fruit development of multi-pistillate type cucumbers. J Hortic Sci Biotechnol 79: 219–222.
•  Huang,  B.,  W.  Wu and  Z.  Hong (2021) Genetic interactions of awnness genes in barley. Genes 12: 606.
•  Junta,  K.,  O.  Saritnum,  R.  Juwattanasamran,  P.  Puddhanon,  D.  Struss,  J.  Puttarach and  S.  Songchanthuek (2020) Identification of single nucleotide polymorphism markers associated with northern corn leaf blight resistance in sweet corn. CMU J Nat Sci 19: 155–162.
•  Klosinska,  U.,  E.U.  Kozik and  T.C.  Wehner (2006) Inheritance of a new trait twin fused fruit in cucumber. HortScience 41: 313–314.
•  Kubicki,  B. (1969) Investigation of sex determination in cucumber (Cucumis sativus L.). VII Trimonoecism. Genet Pol 10: 123–143.
•  Kumchai,  J.,  Y.C.  Wei,  C.Y.  Lee,  F.C.  Chen and  S.W.  Chin (2013) Production of interspecific hybrids between commercial cultivar of eggplant (Solanum melongena L.) and its wild relative S. torvum. Genet Mol Res 12: 755–764.
•  Lee,  E.S.,  H.B.  Yang,  J.  Kim,  H.E.  Lee,  Y.R.  Lee and  D.S.  Kim (2022) Development of SNP marker sets for marker-assisted background selection in cultivated cucumber varieties. Agronomy 12: 487.
•  Li,  D.,  Y.  Sheng,  H.  Niu and  Z.  Li (2019) Gene interactions regulating sex determination in cucurbits. Front Plant Sci 10: 231.
•  Li,  Z.,  S.  Huang,  S.  Liu,  J.  Pan,  Z.  Zhang,  Q.  Tao,  Q.  Shi,  Z.  Jia,  W.  Zhang,  H.  Chen et al. (2009) Molecular isolation of the M gene suggests that a conserved-residue conversion induces the formation of bisexual flowers in cucumber plants. Genetics 182: 1381–1385.
•  Manzano,  S.,  C.  Martínez,  Z.  Megías,  D.  Garrido and  M.  Jamilena (2013) Involvement of ethylene biosynthesis and signalling in the transition from male to female flowering in the monoecious Cucurbita pepo. J Plant Growth Regul 32: 789–798.
•  Matsumura,  H.,  N.  Miyagi,  N.  Taniai,  M.  Fukushima,  K.  Tarora,  A.  Shudo and  N.  Urasaki (2014) Mapping of the gynoecy in bitter gourd (Momordica charantia) using RAD-Seq analysis. PLoS One 9: e87138.
•  McCouch,  S.R. (2008) Gene nomenclature system for rice. Rice (NY) 1: 72–84.
•  Megharaj,  K.C.,  P.S.  Ajjappalavara,  Revanappa,  D.C.  Manjunathagowda and  J.C.  Bommesh (2017) Sex manipulation in cucurbitaceous vegetables. Int J Curr Microbiol App Sci 6: 1839–1851.
•  Mibus,  H. and  T.  Tatlioglu (2004) Molecular characterization and isolation of the F/f gene for femaleness in cucumber (Cucumis Sativus L.). Theor Appl Genet 109: 1669–1676.
•  Miko,  I. (2008a) Epistasis: Gene interaction and phenotype effects. Nature Education 1: 197.
•  Miko,  I. (2008b) Genetic dominance: Genotype-phenotype relationships. Nature Education 1: 140.
•  Miko,  I. (2008c) Gergor mendel and the principles of inheritance. Nature Education 1: 134.
•  More,  T.A. (2001) Development and exploitation of tropical gynoecious lines in F₁ hybrid of cucumber. Acta Hortic 588: 261–267.
•  Na Jinda,  A.,  M.  Nikornpun,  N.  Jeeatid,  S.  Thumdee,  K.  Thippachote,  T.  Pusadee and  J.  Kumchai (2023) Marker-assisted selection of male-sterile and maintainer line in chili improvement by backcross breeding. Horticulturae 9: 357.
•  Nandgaonkar,  A.K. and  L.R.  Baker (1981) Inheritance of multi-pistillate flowering habit in gynoecious pickling cucumber. J Am Soc Hortic Sci 106: 755–757.
•  Nishiguchi,  M.K.,  P.  Doukakis,  M.  Egan,  D.  Kizirian,  A.  Phillips,  L.  Prendini,  H.C.  Rosenbaum,  E.  Torres,  Y.  Wyner,  R.  DeSalle et al. (2002) DNA isolation procedures. In: Wheeler, W., G. Girbet and R. DeSalle (eds.) Techniques in molecular systematics and evolution. Springer, New York, pp. 249–287.
•  Pan,  J.,  G.  Wang,  H.  Wen,  H.  Du,  H.  Lian,  H.  He,  J.  Pan and  R.  Cai (2018) Differential gene expression caused by the F and M loci provides insight into ethylene-mediated female flower differentiation in cucumber. Front Plant Sci 9: 1091.
•  Pan,  Y.,  C.  Wen,  Y.  Han,  Y.  Wang,  Y.  Li,  S.  Li,  X.  Cheng and  Y.  Weng (2020) QTL for horticulturally important traits associated with pleiotropic andromonoecy and carpel number loci, and a paracentric inversion in cucumber. Theor Appl Genet 133: 2271–2290.
•  Pawełkowicz,  M.E.,  A.  Skarzyńska,  W.  Pląder and  Z.  Przybecki (2019) Genetic and molecular bases of cucumber (Cucumis sativus L.) sex determination. Mol Breed 39: 50.
•  Rafalski,  A. (2002) Application of single nucleotide polymorphisms in crop genetics. Curr Opin Plant Biol 5: 94–100.
•  Raitio,  M.,  A.  Perala,  R.R.  Harinen and  J.  Lampinen (2012) Validation of KASParTM dual FRET based SNP genotyping assay. In: Thermoscientific.com. Thermo fisher sciencetific, Finland, pp. 1–4.
•  Saeko,  A.,  S.K.  Lithanatudom,  D.  Struss and  S.  Pongjaroenkit (2020) QTL mapping specific to Thrips palmi resistance in Capsicum annuum. Asia Pac J Sci Technol 25: APST-25-03-09.
•  Sheng,  Y.,  Y.  Pan,  Y.  Li,  L.  Yang and  Y.  Weng (2019) Quantitative trait loci for fruit size and flowering time-related traits under domestication and diversifying selection in cucumber (Cucumis sativus). Plant Breed 139: 176–191.
•  Shyam,  S.D.,  T.K.  Behera,  R.  Bhatia and  A.D.  Munshi (2020) Accelerated breeding in cucumber using genomic approaches. In: Gosal, S.S. and S.H. Wani (eds.) Accelerated plant breeding, Volume 2: Vegetable crops. Springer Nature, Switzerland, pp. 271–299.
•  Simi,  F.,  N.A.  Ivy,  H.B.  Saif,  S.  Akter and  M.F.A.  Anik (2017) Heterosis in cucumber (Cucumis sativus L.). Bangladesh J Agric Res 42: 731–747.
•  Somta,  P.,  J.  Chen,  T.  Yimram,  C.  Yundaeng,  X.  Yuan,  N.  Tomooka and  X.  Chen (2020) QTL mapping for agronomic and adaptive traits confirmed pleiotropic effect of mog gene in black gram (Vigna mungo (L.) Hepper). Front Genet 11: 635.
•  Verma,  N.,  R.  Kumar and  J.  Kaur (2018) Maintenance of gynoecious lines of cucumber through modification of sex expression using gibberellic acid, silver nitrate and silver thiosulfate in cucumber (Cucumis sativus L.). Int J Curr Microbiol Appl Sci 7: 320–328.
•  Wang,  Y.,  K.  Bo,  X.  Gu,  J.  Pan,  Y.  Li,  J.  Chen,  C.  Wen,  Z.  Ren,  H.  Ren,  X.  Chen et al. (2020) Molecularly tagged genes and quantitative trait loci in cucumber with recommendations for QTL nomenclature. Hortic Res 7: 3.
•  Wang,  Y.,  B.  Jiang,  R.  Dymerski,  X.  Xu and  Y.  Weng (2021) Quantitative trait loci for horticulturally important traits defning the Sikkim cucumber, Cucumis sativus var. sikkimensis. Theor Appl Genet 134: 229–247.
•  Yuan,  X.J.,  J.S.  Pan,  R.  Cai,  Y.  Guan,  L.Z.  Liu,  W.W.  Zhang,  Z.  Li,  H.L.  He,  C.  Zhang,  L.T.  Si et al. (2008) Genetic mapping and QTL analysis of fruit and flower related traits in cucumber (Cucumis sativus L.) using recombinant inbred lines. Euphytica 164: 473–491.
•  Zhou,  Y.,  G.J.  Ahammed,  Q.  Wang,  C.  Wu,  C.  Wan and  Y.  Yang (2018) Transcriptomic insights into the blue light-induced female floral sex expression in cucumber (Cucumis sativus L.). Sci Rep 8: 14261.