Molecular approach in elite parents of papaya with a focus on seed physiological quality

Abordagem molecular em genitores elites de mamoeiro com foco na qualidade fisiológica das sementes

Abstract:

Papaya seeds have natural barriers (phenolic compounds, seed structure and genetic factors) that make their germination difficult. And the use of seeds is the main means of propagating the species, whether for commercial cultivation, maintenance of germplasm or selection of genetic materials. Thus, the use of seeds with physiological quality (germination, normal seedlings and germination speed) is fundamental to the production of vigorous plants. However, few studies have been developed with the aim of evaluating and improving seed quality. From this perspective, the study aims to evaluate the quality of seeds in genetic materials widely used to obtain new genotypes, in order to identify superior materials and possible crosses between those that have high physiological seed quality. For this purpose, 44 genotypes were analyzed by molecular characterization, via microsatellite markers (observed and expected heterozygosity, number of alleles, information index, fixation index, inbreeding coefficient, genetic structure and genetic dissimilarity via UPGMA grouping), as well as determining physiological quality of the seed (germination speed index, percentage of germination and abnormal seedlings). Among the genotypes, 10 have germination above 60%, a high rate of germination speed and a low percentage of abnormal seedlings, and can be used in crosses to improve seed quality. The dissimilarity analysis was able to highlight the dissimilarity between the accessions of the Solo and Formosa groups. The use of the dissimilarity coefficient and the number of shared alleles made it possible to identify 18 possible crosses promising to improve the quality of papaya seeds.

Index terms genetic structure; genetic variabilit; germplasm; microsatellites; vigour

Resumo:

Sementes de mamoeiro possuem barreiras naturais (compostos fenólicos, estrutura da semente e fatores genéticos) que dificultam sua germinação. E o uso de sementes é o principal meio de propagação da espécie, seja para o cultivo comercial,manutenção de germoplasma seja para a seleção de materiais genéticos. Assim, o uso de sementes com qualidade fisiológica (germinação, plântulas normais e velocidadede germinação)é fundamental à produção de plantas vigorosas. Contudo, poucos estudos foram desenvolvidos com intuito de avaliar e de melhorar a qualidade das sementes. Nesta perspectiva,o estudo visa a avaliar a qualidade de sementes em materiais genéticos amplamente utilizados na obtenção de novos genótipos, a fim de identificar materiais superiores e possíveis cruzamentos entre aqueles que possuam alta qualidade fisiológica de sementes.Foram analisados 44 genótipos por caracterização molecular, via marcadores microssatélites (heterozigosidade observada e esperada, número de alelos, índice de informação, índice de fixação, coeficiente de endogamia, estrutura genética e dissimilaridadegenética via agrupamento UPGMA), assim como a determinação da qualidade fisiológica da semente (índice de velocidade de germinação, porcentagem de germinação e plântulas anormais).Dentre os genótipos, 10 possuem germinação acima de 60%, alto índice de velocidade de germinação e baixa porcentagem de plântulas anormais, e podem ser utilizados emcruzamentos para a melhoria na qualidade de sementes. A análise de dissimilaridade foi capaz de evidenciar a dissimilaridade entre os acessos dos grupos Solo e Formosa. O uso do coeficiente de dissimilaridade e do número de alelos compartilhados permitiu identificar 18 possíveis cruzamentos promissores à melhoria na qualidade de sementes de mamoeiro.

Termos para indexação estrutura genética; variabilidade genética; germoplasma; microssatélites; vigor

Introduction

Papaya (Carica papaya L.) is considered one of the most cultivated and consumed fruit trees in the subtropical and tropical regions of the globe (CARVALHO et al., 2020) and Brazil is the third largest producer of this crop worldwide, having produced around 43.4 t.ha^-1 in 2020 (FAO, 2022). Its propagation is achieved through seeds, which have high added value and whose price can exceed BRL 6,000.00 per 100 g, as is the case of hybrid Tainung 01. Despite the importance of seeds, as an economic point of view (fruit production and seed trade), environmental (species preservation) and scientific (maintenance of genetic material and obtaining new cultivars) (CARVALHO; NAKAGAWA, 2012) papaya seeds have intrinsic factors its structure and genetic constitution, which make germination and obtaining new cultivars difficult (MELO et al., 2015; CARVALHO et al.,.2020) To overcome the narrow genetics base, papaya breeding programs make use of molecular and classic breeding techniques such as backcrossing, recurrent and mass selection, biparental crossing, segregating populations, and generation advancement, which enable the development of new cultivars with different traits (PEREIRA et al., 2019a; SANTA-CATARINA et al., 2020). The Germplasm Bank of papaya, maintained by the State University of North Fluminense Darcy Ribeiro (UENF) in partnership with Caliman Agrícola S/A, is one of the main papaya germplasm banks in the country, containing over 300 accessions. These are widely used in the production of new hybrids and varieties with desirable traits such as improved fruit quality and disease resistance and/or tolerance (PEREIRA et al., 2019a, PIROVANI et al., 2022; POLTRONIERI et al., 2020).

Obtaining hybrids is a way of increasing the genetic variability of the crop. Therefore, it is essential to determine the genetic variability present in papaya genotypes, allowing the direction of the most appropriate crosses to obtain new genotypes and the characterization of these genotypes represents an important step in this process (NANTAWAN et al., 2019; POLTRONIERI et al., 2020; MORAES et al., 2021; PIROVANI et al., 2022).

Microsatellite markers, due to their multiallelic nature and codominance (TURCHETTO-ZOLET et al., 2017), are ideal markers for the identification and discrimination of genotypes and population genetics studies, being useful tools in various genetic analyzes in plants. In the case of papaya (Carica papaya L.), there are several microsatellite primers designed (SANTOS et al., 2003; EUSTICE et al., 2008) and their uses, such as monitoring the level of homozygosity and transfer of alleles, characterization of lines (RODRIGUES et al., 2023; PIROVANI et al., 2021), development of linkage maps, identification of QTLs (NANTAWAN et al., 2019).

With respect to seed germination in papaya, few advances have been made, and there are, until now, no identified markers that are linked to the expression of characteristics related to the quality of papaya seeds. On the other way, widely used methods to overcome the barriers imposed on seed germination have been adapted and used in the crop.

These techniques are based on the partial or total removal of the sarcotesta and/or seed coat, scarification (physical and chemical) (JESUS et al., 2016; VALE et al., 2020) and the use of plant regulators such as gibberellins and auxins, which stimulate germination (WEBSTER et al., 2016). However, the erratic germination of papaya seeds is often linked to a combination of physical (seed coat, seed size, and weight), chemical (chemical composition of the seed structure) (BUIDES et al., 2017), and physiological (maturity and dormancy) factors intrinsic to the seeds (MELO et al., 2015), besides environmental factors (light, temperature, water, and oxygen) that influence the germination capacity of seeds from maturation to storage. In addition to these factors, the genetic component influences the characteristics of the seed, positively or negatively affecting its germination process (CARVALHO; NAKAGUAWA, 2012).

Genotypes of the Germplasm Bank of UENF/ Caliman S/A, in turn, have been widely evaluated in terms of morpho-agronomic traits aiming at the improvement of characteristics related to the fruit, plant structure, and disease resistance (PIROVANI et al., 2020; MORAES et al., 2021). Despite the importance of seeds for obtaining new plants, there are no studies proposing to examine their physiological quality or the genetic diversity of papaya accessions with the aim of improving the physiological quality of seeds.

Nonetheless, studies involving the evaluation of genetic parameters in papaya seeds, such as the experiment carried out by Santos et al., (2003), revealed high heritability for germination (81.2%), suggesting the possibility of improvements in seed quality, based on genetic analysis. Therefore, research at the molecular level of the accessions of the Germplasm Bank is warranted so that the available genetic resources can be better utilized, taking advantage of this variability for the success of papaya breeding programs.

In this sense, the present study proposes to undertake a molecular characterization of papaya genotypes and investigate the physiological quality of seeds and use this information to define possible hybrid crosses to obtain genotypes with high seed quality.

Materials and Methods

Seed production

Forty-four elite genotypes of papaya were selected from the AGB of UENF/Caliman Agrícola S/A, located between the parallels of 19º06’ and 19º18’ S latitude and meridian of 39º45’ W longitude, at an altitude of 45 m, in Linhares - ES, Brazil. According to the Köppen classification, the climate in the region is the Aw type, with rainy summers and dry winters (ALVARES et al., 2013). Seeds were obtained by self-pollination, when lines; or crossing, when hybrids (Table 1), and belonged to the Formosa, Intermediate, and Solo groups.

Table 1
Genotypes of C. papaya used for molecular analysis, from the Germplasm Bank of papaya at UENF/Caliman Agrícola S/A.

Genotype	Group	Category		Genotype	Group	Category
Caliman M5	S	L		STZ 03 Pecíolo curto	S	L
Sunrise Solo 783	S	L		STA Helena IIITRA02APLT 08	F	L
Costa Rica	I	L		STA Helena IIITRA14APLT05	F	L
Triwan ET	I	L		Papaya 46 claro	S	L
Grampola	S	L		RN. Selecionado	S	L
Sunrise Solo	S	L		Crioula de Costa Rica	F	L
Caliman AM (amarelinho)	S	L		B5 - Geraldo	S	L
Caliman SG (Super Golden)	S	L		Cimarrom	S	L
Sunrise Solo (SS-72/12)	S	L		THB	S	L
Kapoho polpa amarela	S	L		Aliança	S	L
Baixinho de Sta Amalha	S	L		UCLD08-05II5	F	L
Sunrise Solo TJ	S	L		UCLD08-08II5	F	L
Tailandia	F	L		UCLD08-29II5	F	L
São Mateus	I	L		UCLA08-012	F	L
Kapoho polpa vermelha	S	L		UCLA08-080	F	L
Sunrise solo (Pro Tainung)	S	L		UCLA08-088	F	L
Sekati	F	L		UCLA08-092	F	L
Baixinho Super	I	L		UCLA08-097	F	L
M-4	S	L		UCLA08-101	F	L
JS12 (206)	F	L		Candy	F	H
Califlora 209	F	L		Vitória	F	H
Golden tipo Formosa	S	L		Candy Reciprocal	F	H

• ^*
  F, Formosa; I, Intermediate; S, Solo; L, Lines and H, Hybrids.

After collection, the fruits were remained at rest, at room temperature (25°C), until reaching maturity stage VI. Then, the sarcotesta was removed by hand-rubbing the seeds over a steel wire mesh sieve under running water until its complete disappearance (VALE et al., 2020). Afterwards, the seeds were dried in the shade until reaching a water content of 12%.

Production of Genetic Material

To obtain the young leaves, the seeds were set to germinate in washed sand substrate in a greenhouse. After seedling emergence, ten seedlings of each genotype were transferred to tubes containing Basaplant® substrate.

Sixty days after transplanting, a bulk of leaves was collected to compose a sample of plant material corresponding to each evaluated genotype. The leaf bulk was immediately placed in aluminum envelopes and kept refrigerated at -80 °C until DNA extraction. For genotypes JS12 and Sekati, due to the scarcity of seeds, plant material was obtained from young leaves of plants kept in the field at the AGB of UENF/Caliman Agrícola S/A, which were kept under refrigeration according to the methodology described.

Bulk sampling of leaves was adopted because the evaluated accessions were pure lines and hybrids obtained from these lines, so collecting samples of young leaves from different plants makes the evaluation more comprehensive.

Extraction of genomic DNA

Once refrigerated, the samples were macerated in liquid nitrogen using the CTAB method (DOYLE; DOYLE, 1990), with modifications described next. After maceration, 800 μL of extraction buffer containing 5% CTAB, 5 M NaCl, 0.5 EDTA (pH 8.0), 1.0 M Tris-HCl (pH 8.0), 1% PVP, 0.2% β-mercaptoethanol, and 0.1 mg.mL^-1 proteinase K were added to the tubes. Subsequently, the tubes with the samples and buffer solution were placed in a dry bath at 65 ºC for 40 min, with gentle homogenization performed every 10 min. After 40 min, the samples were cooled (room temperature) and centrifuged for 5 min at 14,000 rpm, 700 μL of the supernatant were transferred to new tubes, 700 μL of chloroform- isoamyl-alcohol (24:1) were added, and gentle homogenization was performed by inversion for 10 min until complete homogenization.

This step was repeated once more and then 550 μL of the supernatant and 550 μL of chloroform-isoamyl alcohol were added. This was followed by another centrifugation (5 min at 14,000 rpm) and transfer of 400 μL of the supernatant to new tubes to which 2/3 of the collected volume of ice-cold isopropanol were added, followed by gentle homogenization by inversion for 10 min and incubation for 30 min in a biofreezer at -80 ºC. After incubation, centrifugation was performed again (14,000 rpm/10 min), generating the pellets (precipitates).

After removing the supernatant, the pellets were subjected to triple washing; 300 μL of 70% ethanol were added, and the material was centrifuged for 10 min at 14,000 rpm (the process was repeated 2x), followed by the addition of 300 μL of 95% ethanol and centrifugation for 5 min at 14,000 rpm.

After the triple wash, the supernatant was removed and the pellets were dried in a dry bath for 30 min at 55 ºC. After drying, the pellets were re-suspended with a solution containing 100 μL of TE and 1 μL of RNAse, incubated in a dry bath for 40 min at 37 ºC, and finally stored at -20 ºC.

Quantification of genomic DNA

The concentrations of genomic DNA of the different papaya genotypes were estimated using agarose gel, by comparing the fluorescence of the samples. This procedure involved 1% agarose gel with 1X TAE buffer (Tris, sodium acetate, EDTA, pH 8.0), using the 100-bp lambda (λ) marker (100 ng.μL^-1) as a molecular weight marker (Invitrogen, USA) and staining by Gel Red™ and Blue Juice solution (1:1). The images were captured by the Mini Bis Pro gel-documenting system (Bio-Imaging Systems). Based on the obtained images, the genomic DNA concentration was estimated in comparison with the 100-bp marker. Finally, DNA samples were diluted to a working concentration of 5 ng.μL^-1 and stored at -20 °C.

Polymerase chain reaction

Fifty-sex pairs of microsatellite primers were selected (Table 2), designed, and described by Eustice et al., (2008). Microsatellite amplification reactions were performed in a final volume of 12 μL, containing 2 μL of DNA (10 ng.μL^-1), 1.2 μL of 10x Taq buffer with (NH₄)SO₄, 1.2 μL of MgCl₂ (25 mM), 1.0 μL of dNTPs (2 mM), 0.2 μL of Taq polymerase (0.5 U/μL), 0.5 μL of each primer, and ultrapure water. The program used for the PCR reaction consisted of a cycle of four minutes at 94°C for the initial denaturation of the DNA, followed by 35 cycles of one minute at 94 °C for denaturation, one minute at Y °C for annealing of the primers, and two minutes at 72 °C for the extension of the primers, plus a later cycle of seven minutes at 72 ºC for the final extension. Amplifications were performed in an Applied Biosystems/Veriti 96 Well Thermal Cycler, where “Y” corresponds to the specific annealing temperature for each primer. PCR reactions were performed for each primer in the population analysis.

Table 2
Diversity parameters for 30 SSR markers in C. papayagenotypes from the AGB of UENF/Caliman Agrícola S/A.

Locus	Na	I	Ho	He	F
CPM1621CC	3	0.74	0.33	0.50	0.35
P6K1117CC	4	1.03	0.16	0.57	0.72
P6K1268CC	3	0.62	0.12	0.39	0.69
P3K6912CC	3	0.60	0.11	0.37	0.69
P3K1850CC	2	0.68	0.14	0.49	0.71
ctg-365A5	2	0.59	0.00	0.40	1.00
P3K149C0	2	0.66	0.00	0.46	1.00
CPM1554C2	3	0.77	0.07	0.50	0.86
P3K1382A5	3	0.91	0.00	0.55	1.00
P6K900CC	4	0.97	0.16	0.53	0.70
ctg-41S5	2	0.69	0.02	0.49	0.95
P3K3511CC	4	0.97	0.14	0.56	0.76
P6K25CC	3	1.01	0.02	0.61	0.96
P3K2152CC	2	0.61	0.00	0.42	1.00
P3K170CC	3	0.79	0.00	0.52	1.00
P3K7483A5	5	1.29	0.16	0.68	0.77
P3K5113C0	5	1.04	0.14	0.56	0.75
P3K418CC	3	0.66	0.00	0.41	1.00
P3K3407CC	3	0.99	0.66	0.59	-0.11
CPM746LCC	4	1.05	0.14	0.59	0.77
P3K4489CC	3	0.74	0.07	0.51	0.87
ctg-203C0	2	0.69	0.07	0.49	0.86
ctg-718CC	4	1.01	0.16	0.54	0.70
P3K4272CC	3	0.84	0.09	0.48	0.81
P3K917C0	3	0.73	0.14	0.44	0.69
P3K86CC	3	0.63	0.12	0.38	0.69
P6K71CC	2	0.59	0.00	0.40	1.00
P6K128CC	3	0.81	0.19	0.53	0.64
P8K187CC	3	0.64	0.23	0.41	0.43
P3K3490CC	2	0.68	0.02	0.49	0.95
Mean	3	0.80	0.12	0.50	0.77

• ^*
  NA: number of alleles; I: information index; He: expected heterozygosity; Ho: observed heterozygosity; and F: fixation index.

Electrophoresis

The amplification products were diluted using 6 μL of amplified material for 18 μL of Buffer E from the DNF 900 kit. The samples were distributed in specific plates containing 96 wells for capillary electrophoresis in the Fragment Analyzer (AATI) instrument (Advanced Analytical), in which amplified fragments of 35 to 500 bp were separated with a resolution of approximately 2 bp, allowing a safe detection of differences in base pairs between the analyzed alleles. The DNA Ladder marker with variation from 35 to 400 bp was used during the runs to determine the size of the amplified fragments. The material contained in the plates was transported through the capillaries, where it was subjected to a run of 110 min at a current of 8 Kw. After this step, the images resulting from capillary electrophoresis were analyzed using PROsize 2.0 software.

Cluster analysis

Observations obtained by amplification of the SSR markers were converted into a numerical code for each allele per locus. This numerical matrix was developed by assigning values from 1 to the maximum number of alleles per locus, as described next: for a locus that has three alleles, homozygous forms (A1A1, A2A2, and A3A3) were represented by the numbers 11, 22, and 33; and heterozygotes (A1A2, A1A3, and A2A3) by 12, 13, and 23. From this numerical matrix, three indices were tested: the unweighted index, the weighted index, and the Smouse Peakall index (PEAKALL; SMOUSE, 2012).

Based on the highest cophenetic correlation coefficient, the weighted index was applied and analyses were carried out using GENES software (CRUZ, 2013).

After the distance matrix was obtained, clustering was performed via dendrogram using the UPGMA method (Unweighted Pair-Group Method with Arithmetic Mean), with the aid of Mega software v. 10. The distribution of the genetic variability of the 44 genotypes was estimated using Genalex software v. 6.5 (PEAKALL; SMOUSE 2012), based on the following parameters: number of alleles per polymorphic locus (NA), observed heterozygosity (Ho), expected heterozygosity (He), information index (I), and fixation index (ƒ).

Analysis of the genetic structure of parent lines and hybrids

The method based on Bayesian grouping algorithms was adopted, using Structure software v. 2.3.4 (PRITCHARD et al., 2000). For this purpose, the admixture model and correlated allele frequencies were employed, with a burn-in period of 250,000, followed by an extension (Markov Chain Monte Carlo) of 750,000 repetitions. Twenty simulations were performed with k ranging from 1 to 5.

The Δk statistical test was performed using Structure Harvester software, following the criterion of Evanno et al., (2005). This criterion is based on the mean and standard deviation of the estimated LnP(D) in each of the 20 iterations per k. The Δki values were estimated by the following formula:

1 $$\mathrm{\Delta }\mathrm{k}\mathrm{i}\mathrm{}=\mathrm{}\mathrm{A}\mathrm{B}\mathrm{S}\mathrm{}(\mathrm{k}\mathrm{i}\mathrm{}+\mathrm{}1\mathrm{}–\mathrm{}(2\mathrm{*}\mathrm{k}\mathrm{i})\mathrm{}+\mathrm{}\mathrm{k}\mathrm{i}–1)\mathrm{}/\mathrm{S}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{}\mathrm{d}\mathrm{e}\mathrm{v}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{}\mathrm{K}\mathrm{i}$$

where “i” = number of simulated groups, ranging from 1 to 20; and “ABS” = module.

The value of Δk is estimated for each k, and the one with the highest value is selected.

Seed physiological analyses

Except for JS12, Sekati, and Vitória (due to the absence of seeds), seeds obtained from selected C. papaya elite materials (Table 1) were subjected to physiological analysis.

Accordingly, germination percentage (%G), Germination Speed Index (GSI), and the percentage of abnormal seedlings (%AS) were evaluated in the other accessions (41).

Seed germination

The germination test was performed according to the Rules for Seed Testing (Rules of Seeds Analyses, RAS) (BRASIL, 2009). Thirty days after sowing on paper substrate, germinated seeds and abnormal seedlings were counted and the results were expressed in percentage (%).

Germination Speed Index

This variable was evaluated simultaneously with the germination test, considering the seed that broke the seed coat and produced the radicle. The evaluations were carried out every two days, at the same time, from the day the first seed germinated, and the procedure was repeated until completion at 30 days. Germination Speed Index (GSI) was determined according to the methodology established by Maguire (1962).

Statistical analysis

The experiment was laid out in a completely randomized design with four replications, each of which was composed of 50 seeds and 41 treatments (genotypes). Data were subjected to analysis of variance at a 5% significance level and means were compared using the Scott-Knott test at 5% probability.

Prediction of crosses to improve seed quality

Genotypes with the better seed physiological quality were selected based on %G, %AS, and GSI. As a minimum selection criterion, 60% germination was considered, following the minimum standard established by the MAPA (MAPA, 2019) for the commercialization of papaya seeds. After pre-selection, possible crosses between the selected genotypes were established. The possible crosses were established between genotypes belonging to the Formosa × Formosa, Solo × Solo, and Formosa × Solo groups to obtain fruits of the Formosa, Solo, and Intermediate types, respectively.

The previously established crosses were evaluated for Dissimilarity (%D) and Number of common Alleles (NA), estimated by the UPGMA clustering method. Those with %D above 60% for Formosa × Formosa and Solo × Solo crosses, and above 75% for Formosa × Solo crosses, were admitted as potential crosses for obtaining genotypes with high seed physiological quality.

Results

Molecular analyses Genetic diversity

Of the 58 pairs of primers evaluated, 52% showed polymorphism for the evaluated loci and were used in genetic diversity analysis.

A total of 91 alleles were detected and the number per locus ranged from 2 to 5, with an average of 3 alleles per locus in the evaluated C.papaya accessions. Because they have five alleles, loci P3K7483A5 and P3K5113C0 have a higher information index (I) compared with the other evaluated loci (Table 2).

The information index provides an estimate of the discriminatory power of the locus, taking into account the number of alleles identified and the relative frequencies of these alleles. In this study, I ranged from 0.59 (ctg- 4155 and P6K71CC) to 1.29 (P3K7483A5), averaging 0.80. According to the classification, I values can be grouped into three levels: highly informative (I>0.5), moderately informative (0.25 < I < 0.5), and little informative (I< 0.25) (BOTSTEIN et al., 1980); therefore, those found here can be considered highly informative.

Expected heterozygosity (He) estimates varied between loci, ranging from 0.37 (P3K6912CC) to 0.61 (P6K25CC), with a mean of 0.50 (Table 2). Observed heterozygosity (Ho), on the other hand, ranged from 0.00 (P6K71CC, p3K418CC, P3K170CC, P3K2152CC, P3K1382A5, P3K149C0, and ctg-365A) to 0.66 (P3K3407CC), averaging 0.12. The fixation index (F) or inbreeding coefficient ranged from -0.11 (P3K3407CC) to 1.00 (P6K71CC, P3K418CC, P3K170CC, P3K2152CC, P3K1382A5, P3K149C0, ctg- 365A) with a mean of 0.77.

Genetic variability of C. papaya accessions

Regarding the performance of the genotypes (Table 3), the obtained heterozygosity (H) ranged from 0.00 (Costa Rica, Sekati, STA Helena III TRA 02A PLT 08, and UCLA08-088) to 0.67 (Candy Reciprocal), averaging 0.11.

When we compare the groups, the mean values of H were higher in the Formosa (0.18) compared with the Intermediate (0.06) and Solo (0.05) groups.

The inbreeding coefficient (f), in turn, obtained for the different genotypes (Table 3), ranged from -0.03 (Candy Reciprocal) to 1.00 (Costa Rica, Sekati, STA Helena III TRA 02A PLT 08, and UCLA08-088) and averaged 0.81, corroborating the information obtained for the population. As regards the groups, the mean f was 0.89, 0.91, and 0.70 for the Solo, Intermediate, and Formosa groups, respectively.

Table 3
Diversity parameters in C. papaya genotypes from the AGB of UENF/Caliman Agrícola S/A, based on microsatellite markers.

Genotype	Group	H	f
Caliman M5	Solo	0.07	0.87
Sunrise Solo 783	Solo	0.07	0.89
Grampola	Solo	0.07	0.87
Sunrise Solo	Solo	0.07	0.86
Caliman AM (amarelinho)	Solo	0.07	0.89
Caliman SG (Super Golden)	Solo	0.07	0.87
Sunrise Solo (SS-72/12)	Solo	0.07	0.87
Kapoho polpa amarela	Solo	0.03	0.94
Baixinho de Sta Amalha (BSA)	Solo	0.03	0.94
Sunrise Solo TJ	Solo	0.03	0.93
Kapoho polpa vermelha	Solo	0.03	0.94
Sunrise solo (Pro Tainung)	Solo	0.04	0.94
M-4	Solo	0.03	0.93
Golden tipo formosa	Solo	0.13	0.76
STZ 03 Pecíolo curto	Solo	0.07	0.86
Papaya 46 claro	Solo	0.07	0.89
RN. Selecionado	Solo	0.03	0.94
B5 - Geraldo	Solo	0.03	0.94
Cimarrom	Solo	0.21	0.70
THB	Solo	0.03	0.94
Aliança	Solo	0.03	0.94
Costa Rica	Intermediate	0.00	1.00
Triwan ET	Intermediate	0.07	0.87
São Mateus	Intermediate	0.03	0.95
Baixinho Super	Intermediate	0.10	0.83
Tailandia	Formosa	0.03	0.95
Sekati	Formosa	0.00	1.00
JS12 (206)	Formosa	0.07	0.85
Califlora 209	Formosa	0.29	0.57
STA Helena III TRA02A PLT 08	Formosa	0.00	1.00
STA Helena III TRA 14A PLT 05	Formosa	0.07	0.89
Crioula de Costa Rica	Formosa	0.30	0.54
UCLD08-05II5	Formosa	0.33	0.44
UCLD08-08II5	Formosa	0.45	0.34
UCLD08-29II5	Formosa	0.00	1.00
UCLA08-012	Formosa	0.03	0.95
UCLA08-080	Formosa	0.04	0.94
UCLA08-088	Formosa	0.00	1.00
UCLA08-092	Formosa	0.07	0.89
UCLA08-097	Formosa	0.10	0.82
UCLA08-101	Formosa	0.03	0.94
Candy	Formosa	0.63	0.03
Vitória	Formosa	0.40	0.26
Candy Reciprocal	Formosa	0.67	-0.03
Mean		0,11	0.81

• ^*
  H, corresponds to heterozygosity and f, to the inbreeding coefficient.

Among the C. papaya lines, only 12.5% showed an H equal to zero and, consequently, an f equal to 1.0. In contrast, 37.5% of the lines had an f between 0.93 and 0.99 and H ranging from 0.03 to 0.04, whereas 50% of the lines showed an H below 0.90 and f ranging from 0.07 to 0.45. Thus, we may conclude that only 12.5% of the lines are in homozygosis, whereas 87.5% of them have an H above and an f below the expected for pure lines, indicating accession segregation.

Genetic structure analysis

Genetic structure analysis was approached by the Bayesian method, using the DK criterion (EVANNO et al., 2005) to make inferences about the genetic structure of the different papaya accessions. The analysis was carried out using Structure software, by selecting an optimal value of K, where K=2 was obtained as the most likely grouping number (Figure 1).

Figure 1
?K peak graph indicating the optimal number of genetic clusters for Bayesian analysis obtained using Structure software v. 2.3.4.

The probability of membership adopted was 70% for each genotype belonging to a certain group. The 44 genotypes of C. papaya, including hybrids and lines, formed two groups (Figure 2).

Figure 2
Bayesian inference clusters of 44 Carica papaya genotypes. Genotypes are represented on the vertical line, and each genetic group is represented by a color. Genotypes belonging to the ¹Formosa, ²Solo, and ³Intermediate groups.

The green group was composed of genotypes belonging to the Formosa group (hybrids and lines) and the accessions Cimarron and Golden type Formosa, of the Solo group, whereas the red group consisted of genotypes of the Solo and Intermediate groups.

The accessions classified on the basis of morphological traits (fruit size and weight) STA Helena III TRA02 PLT08 in Formosa and the accessions Golden type Formosa and Papaya 46 Claro, classified as Solo, showed alleles belonging to both groups, but the set of alleles that define them belongs to the green (STA Helena III TRA02 PLT08 and Papaya 46 Claro) and red (Golden type Formosa) groups, Formosa and Solo, respectively.

The other genotypes, in turn, showed 100% membership to the group.

The cophenetic correlation index was 0.97, indicating a strong correlation between the phenetic and dissimilarity matrices and high representativeness of the dendrogram formed. Meanwhile, diversity analysis, obtained by the average group linkage method (UPGMA), indicated the formation of two groups (Figure 3). We thus have two groups created, where within-group homogeneity and between-group heterogeneity can be observed.

Figure 3
Dendrogram of genetic dissimilarity by the UPGMA clustering method, via SSR markers, based on the analysis of 44 genotypes from the Germplasm Bank of UENF/Caliman Agrícola S/A (cophenetic correlation coefficient = 0.97). ¹Formosa type fruit, ²Solo type fruit, ³Intermediate type fruit.

Cluster I (red) consisted of 18 genotypes, all of which belonged to the Formosa group, whereas Cluster II (blue) contained 26 genotypes: 21 belonging to the Solo group, one to the Formosa group, and four to the Intermediate group. Only genotype STA Helena III TRA02 PLT08, belonging to the Formosa group, was allocated to Cluster II, in which the Solo and Intermediate genotypes are grouped. This behavior may be related to the non-fixation of alleles and the consequent occurrence of “mixing” between accessions, and even if the membership coefficient is considered low (13%), it may have contributed to the allocation of this material in Cluster II. Regarding the Candy hybrid and its reciprocal, the similarity between accessions was 91%, with 55 alleles in common for the 30 evaluated loci. The high similarity observed between Candy and its reciprocal is because they have the same parents.

Seed quality analysis

According to MAPA (2019), germination rates greater than or equal to 60% are considered adequate for papaya seeds. Of the 44 elite genotypes evaluated (Table 4), 25% exhibited germination considered adequate by MAPA for the marketing of seeds.

Table 4
Physiological traits of Carica papaya seeds from the Germplasm Bank of UENF/Caliman Agrícola S/A.

Genótipo	G			AS			GSI
	------- % -------
Sunrise Solo (SS-72/12)	76	b		19	c		5.83	a
Candy	93	a		3	a		4.65	b
Aliança	37	e		15	c		1.89	g
Candy Reciprocal	51	c		3	a		2.72	e
UCLD08-05II5	71	b		11	b		3.17	d
UCLD08-08II5	58	c		11	b		2.27	f
UCLD08-29II5	77	b		16	c		1.90	g
Caliman M5	3	g		2	a		0.17	k
Sunrise Solo 783	0	g		0	a		0.01	k
Costa Rica	4	g		0	a		0.09	k
Triwan ET	43	d		12	b		2.37	f
Grampola	4	g		3	a		0.19	k
Sunrise Solo	5	g		2	a		0.04	k
Caliman AM	20	f		10	b		1.20	i
Caliman SG	7	g		6	b		0.25	k
Kapoho polpa amarela	0	g		2	a		0.04	k
Baixinho de Sta Amalha (BSA)	1	g		2	a		0.07	k
Sunrise Solo TJ	1	g		2	a		0.07	k
Tailandia	78	b		0	a		3.84	c
São Mateus	2	g		0	a		0.07	k
Kapoho polpa vermelha	15	f		9	b		0.80	j
Sunrise solo (Pro Tainung)	2	g		2	a		0.08	k
Baixinho Super	3	g		3	a		0.06	k
M-4	0	g		3	a		0.05	k
Califlora 209	2	g		0	a		0.06	k
Golden tipo formosa	1	g		1	a		0.04	k
STZ 03 pecíolo curto	69	b		7	b		3.09	d
STA Helena III TRA 02A PLT 08	0	g		1	a		0.01	k
STA Helena III TRA 14A PLT 05	0	g		0	a		0.00	k
Papaya 46 claro	21	f		46	f		2.19	f
RN. Selecionado	55	c		5	a		2.49	f
Criola de Costa Rica	0	g		0	a		0.00	k
B5 - Geraldo	0	g		2	a		0.06	k
Cimarrom	2	g		7	b		0.20	k
JS12	0	g		5	b		0.14	k
UCLA08-012	85	a		10	b		4.43	b
UCLA08-013	54	c		10	b		2.19	f
UCLA08-080	37	e		27	d		0.95	j
UCLA08-088	94	a		0	a		3.90	c
UCLA08-092	88	a		2	a		1.72	h
UCLA08-097	87	a		5	b		1.22	i
UCLA08-101	55	c		37	e		1.70	h
THB	85	a		2	a		4.12	c
Vitória	0	g		26	d		1.31	i

• ^*
  G(%), percentage of Germination, AS(%), percentage of Abnormal Seedlings, GSI, Germination Spees Index. Means followed by the same letter belong to the same group according to the Scott-Knott test at 5% probability.

Among the accessions with %G greater than 60% are UCLA08-088 (94%), UCLA08- 092 (90%), UCLA08-097 (83%), UCLA08012 (85%), UCLD08-29II5 (76%), UCLD08-05II5 (71%), and Candy Reciprocal (93%), belonging to the Formosa group; and Sunrise Solo SS72/12 (76%), THB (85%), Tailandia (78%), and STZ03 Pecíolo Curto (68%), of the Solo group. None of the accessions belonging to the Intermediate group had a germination percentage above 60%, and Triwan ET (%G = 43%) was the accession with the highest germination rate in the group.

The different papaya accessions had an average %AS of 6.86%. On average, the Formosa, Solo, and Intermediate groups showed %AS values of 7.0, 7.0, and 2%, respectively.

Among the 41 accessions of C. papaya, 73.2% had an %AS below 10%, and of these, only 23.3% exhibited a %G above 60%, meaning they are not good-quality seeds.

The average GSI of the evaluated accessions was 1.43. Between the groups, the Formosa, Solo, and Intermediate types showed average GSI of 2.0, 1.0, and 1.0, respectively.

Germination speed index reflects the average speed of seed germination, with higher values meaning a faster germination process.

Therefore, among the 41 genotypes evaluated, 11 (UCLA08-088, UCLA08-092, UCLA08- 097, UCLA08-012, UCLD08-05II5, UCLD08- 29II5, Candy Reciprocal, Sunrise Solo SS72/12, THB, Tailandia, and STZ03 Pecíolo Curto) have important traits for obtaining seeds with high physiological quality. As such, these accessions can be used in the papaya breeding program aiming at improved seed quality, since the above-mentioned genotypes—which belong to the different groups (Solo, Intermediate, and Formosa)—have an average germination rate of 82%, which is above the standard set at 60% by MAPA (2019) for sale.

Prediction of crosses aiming at improved seed quality

By jointly analyzing the results obtained in the different analyses carried out (dissimilarity, population structure, and seed physiological quality), it is possible to establish the possible crosses between the most divergent genotypes and those that possess the best seed traits (Table 5).

Table 5
Possible crosses between Carica papaya genotypes from the AGB of UENF/Caliman Agrícola S/A aiming at improved seed quality.

Cross
Formosa × Formosa			%D	CA
UCLA08-088	×	UCLA08-092	39.28	38
UCLA08-088	×	UCLA08-012	24.49	45
UCLA08-088	×	UCLA08-097	29.08	43
UCLA08-088	×	UCLD08-05II5	44.39	33
UCLA08-088	×	Candy Reciprocal	58.16	24
UCLA08-088	×	Tailandia	57.65	27
UCLA08-088	×	UCLD08-29II5	34.69	42
UCLA08-092	×	UCLA08-012	50.51	30
UCLA08-092	×	UCLA08-097	26.53	45
UCLA08-092	×	UCLD08-05II5	55.57	27
UCLA08-092	×	Candy Reciprocal	59.18	24
UCLA08-092	×	Tailandia	57.65	26
UCLA08-092	×	UCLD08-29II5	35.20	40
UCLA08-012	×	UCLA08-097	45.41	33
UCLA08-012	×	UCLD08-05II5	42.86	34
UCLA08-012	×	Candy Reciprocal	57.65	25
UCLA08-012	×	Tailandia	59.69	27
UCLA08-012	×	UCLD08-29II5	41.84	37
UCLA08-097	×	UCLD08-05II5	45.40	32
UCLA08-097	×	Candy Reciprocal	54.08	27
UCLA08-097	×	Tailandia	62.75	24
UCLA08-097	×	UCLD08-29II5	36.22	40
Candy Reciprocal	×	UCLD08-29II5	66.33	20
Candy Reciprocal	×	Tailandia	67.35	22
Tailandia	×	UCLD08-29II5	44.89	34
Solo × Solo
Sunrise Solo	×	THB	64.28	23
Sunrise Solo	×	STZ 03 Pecíolo Curto	12.24	54
THB	×	STZ 03 Pecíolo Curto	67.35	21
Solo × Formosa
Sunrise Solo	×	UCLA08-088	82.65	10
Sunrise Solo	×	UCLA08-092	74.49	14
Sunrise Solo	×	UCLA08-012	78.57	14
Sunrise Solo	×	UCLA08-097	80.1	11
Sunrise Solo	×	UCLD08-05II5	55.61	27
Sunrise Solo	×	Candy Reciprocal	57.14	27
Sunrise Solo	×	Tailandia	56.63	27
Sunrise Solo	×	UCLD08-29II5	79.09	13
THB	×	UCLA08-088	24.49	45
THB	×	UCLA08-092	50.51	30
THB	×	UCLA08-012	79.59	14
THB	×	UCLA08-097	45.41	33
THB	×	UCLD08-05II5	68.88	20
THB	×	Candy Reciprocal	67.35	21
THB	×	Tailandia	76.02	16
THB	×	UCLD08-29II5	70.41	20
STZ 03 Pecíolo Curto	×	UCLA08-088	85.71	9
STZ 03 Pecíolo Curto	×	UCLA08-092	81.12	11
STZ 03 Pecíolo Curto	×	UCLA08-012	81.63	13
STZ 03 Pecíolo Curto	×	UCLA08-097	81.12	11
STZ 03 Pecíolo Curto	×	UCLD08-05II5	60.71	25
STZ 03 Pecíolo Curto	×	Candy Reciprocal	60.2	26
STZ 03 Pecíolo Curto	×	Tailandia	61.73	24
STZ 03 Pecíolo Curto	×	UCLD08-29II5	81.12	12
Total				91

• ^*
  %D, corresponds to the percentage of dissimilarity; CA, corresponds to the number of alleles in common between the accessions.

In view of this, to obtain genotypes that produce fruits of the standard Formosa type (Formosa × Formosa), we suggest crossing between lines Tailandia × UCLA08-097 or Candy Reciprocal and UCLD08-29II5 × Candy Reciprocal. For standard Solo fruits, we recommended crossing THB × Sunrise Solo or STZ 03 Pecíolo Curto. Finally, for standard Intermediate fruits (Formosa × Solo), we recommend crosses between accessions THB × UCLA08-012, Tailandia, or UCLD08-29II5 and crosses between the Sunrise Solo or STZ 03 Pecíolo Curto lines with the following lines: UCLA08-088, UCLA08-092, UCLA08-012, UCLA08-097, and UCLD08-29II5.

Discussion

Genetic diversity

Greater genetic variability is expected in germplasm banks, given the existence of accessions with different traits, which can be related to the fruit, the plant, disease resistance, among others. Therefore, greater molecular variation is expected within AGBs compared with isolated accessions. However, the variability found in the Germplasm Bank of UENF/Caliman was lower than the average of 4.53 and 4.0 alleles per locus found in AGBs by Matos et al., (2013) and Oliveira et al., (2010), respectively, with the use of microsatellite markers. This low number of alleles observed in the present study is the result of restricted genetic variability and the genetic structure of the evaluated population. In fact, the Germplasm Bank of UENF/Caliman is maintained through controlled pollination to obtain hybrids (controlled crosses) and lines (controlled self-pollination), which tends to maintain or reduce the number of alleles per locus, resulting in the low variability observed. Furthermore, new accessions are commonly obtained in the Germplasm Bank of UENF/Caliman by backcrossing and by crossing related individuals. Thus, the method adopted for maintaining and obtaining new accessions causes the genetic variability estimated for the Germplasm Bank of UENF/Caliman to be lower than that of AGBs that contain accessions from different countries and unrelated individuals, as is the case of the AGBs evaluated by Matos et al., (2013) and Oliveira et al., (2010).

The 30 microsatellite markers used (Table 1) showed an Information index (I), an indicator of marker quality in genetic studies (segregation, population and paternity control) upside 0.5. According to the classification established by Botstein et al., (1980), 100% of these markers are considered highly informative and, therefore, have great utility in molecular characterization studies. The obtained results were superior to those observed by Shivkumar et al., (2014), who used 20 SSR markers to estimate genetic diversity in seven accessions of Indian papaya (C.papaya) and obtained I values that ranged from 0.215 to 0.370, with a mean of 0.240, considered uninformative.

The observed variation in Ho and He values results from the amplitude of the number of alleles per locus and the distribution of allele frequencies. The Ho value of 0.00 in loci P6K71CC, p3K418CC, P3K170CC, P3K2152CC. P3K1382A5, P3K149C0, and ctg- 365A can be explained by the low number of alleles in these loci (2 to 3 alleles), as well as their homozygous status (Oliveira et al., 2010). Therefore, when comparing the Ho and He estimates, with the exception of locus P3K3407CC, a deficit of heterozygotes is observed, since expected heterozygosity is higher than that observed in the evaluated C.papaya genotypes. This behavior (i.e. homozygosity > heterozygosity) is expected, given that the AGB is maintained under field conditions with self-pollination in hermaphrodite plants for the conservation and maintenance of pure lines. Oliveira et al., (2010), Matos et al., (2013), and Pirovani et al., (2020) obtained similar results in studies evaluating papaya accessions from Germplasm Banks of papaya.

Genetic variability

For F, values close to zero indicate random crosses, while negative values indicate excessive heterozygosity and that inbreeding for that locus is null in the population, as can be seen in locus P3K3407CC, where Ho was higher than He, resulting in a low F. Therefore, for this locus, inbreeding is not occurring. In contrast, high positive values indicate high inbreeding (PEAKALL; SMOUSE, 2012). The high F (1.0) found in loci P6K71CC, P3K418CC, P3K170CC, P3K2152CC, P3K1382A5, P3K149C0, and ctg-365A reflects a decrease in the frequency of heterozygotes and, consequently, an increase in the frequency of homozygotes and allele fixation. For pure lines, as is the case of the genetic material under study, with the exception of hybrids Vitória, Candy, and Candy Reciprocal, F=1 is expected; i.e. the individuals are homozygous and have their allele fixed. Therefore, there is no segregation of these lines. Regarding the performance of hybrids Vitória, Candy, and its reciprocal, their heterozygosity is expected to be superior to that shown by the other genotypes of C.papaya, which are pure lines. The average H of hybrids and isolated reciprocals was 0.57, indicating the contribution of both parents.

For the maintenance of the AGB, lines are obtained from the self-pollination of accessions, which is achieved by protecting the hermaphrodite flowers with paper bags to avoid free pollination. In the case of hybrids, parents are crossed by removing the closed hermaphrodite flower of the male plant and subsequently inserting it in the female flower of the mother plant; then, the pollinated flowers are protected with paper bags.

In this way, free pollination does not occur and the accessions keep their alleles fixed.

However, due to the occurrence of diseases, bird attacks, and adverse weather, the fruit obtained from controlled pollination may be lost, and in these situations, fruits obtained from free pollination are collected to avoid losing the accession. Although papaya is an autogamous plant with cleistogamy (i.e. it was already self-pollinated by the time the hermaphrodite flower completely open), even if at low frequency, cross-pollination can occur in free-pollinated individuals, which culminates in the segregation observed in accessions of the Germplasm Bank of UENF/Caliman S/A.

Segregation of accessions in a Germplasm Bank must be taken into account during the process of obtaining hybrids and commercial lines, since allele fixation in parental lines, as well as in commercial lines, is essential to ensure the uniformity of lines and hybrids developed. The use of seeds obtained from materials whose alleles are not fixed leads to the occurrence of segregation, resulting in fruits and plants outside the desired pattern, thereby affecting the quality of the orchard and the uniformity of fruits. This lack of uniformity leads to financial damages as well as loss of genetic material or important traits. It is worth mentioning that the papaya germplasm bank of UENF/Caliman is one of the main germplasm banks in Brazil, which highlights the importance of fixing alleles in the accessions kept by the germplasm bank so that there is no loss of traits such as pulp color, firmness, flavor, plant height, and disease resistance.

This observed segregation of genotypes can, nonetheless, be used to improve the quality of a registered cultivar or help in the development of new lines with different traits, thus increasing the diversity of materials that can be used in breeding programs of the species. Pirovani et al., (2022) reported that there is a possibility of purifying the parents JS12, Sunrise Solo (SS-72/12), and Sekati and that segregation in the parents can be exploited in the papaya breeding program to obtain hybrids such as UC10 and Candy, with different fruit weights and sizes, to better meet the market demands (PEREIRA et al., 2019b, c).

Genetic structure analysis

Due to the narrow genetic base of papaya, new cultivars are commonly obtained from segregating populations of this crop.

These populations have sufficient genetic variability to identify promising genotypes that carry important characteristics such as disease resistance, quality, and fruit production (PIROVANI et al., 2022), which can thus provide significant genetic resources, increasing variability and contributing to the development of new papaya cultivars.

Among its accessions, the Germplasm Bank of UENF/Caliman Agrícola S/A has genotypes resulting from segregation and pollination between related and unrelated individuals, and the high membership coefficient revealed by genetic structure analysis informs that the genetic structure of the accessions belonging to the germplasm bank is well defined.

However, some accessions have “mixtures” because their alleles are not completely fixed, as is the case with genotypes STA Helena III TRA02 PLT08, Papaya 46 Claro, and Golden tipo Formosa. For the other genotypes, the observed absence or low sharing of alleles is due to controlled pollination for the maintenance of pure lines and hybrids in the Germplasm Bank of UENF/Caliman Agrícola S/A. These self-pollinations prevent the exchange of alleles between genotypes; therefore, the groups formed from the evaluated accessions have a well-defined genetic constitution.

In the above-described scenario, Bayesian analysis allowed us to identify individuals who share the same genomic regions analyzed.

The formation of groups associated with the high membership of the genotypes to their respective groups is sufficient to guide the selection of contrasting parents for crosses between the C. papaya genotypes.

Diversity analysis as obtained by the average group linkage method (UPGMA)

Cluster analysis based on molecular traits tends to separate the two main heterotic groups: Solo and Formosa. Therefore, it can be stated that the Formosa and Solo heterotic groups have divergent genetic traits and the existing variability between groups was sufficient for them not to fit into the same group. Pirovani et al., (2020) observed that papaya genotypes tend to separate according to their heterotic group (Solo and Formosa), and differ basically by the size and weight of their fruit. The fruit of the Solo heterotic group has an average weight of 0.5 kg and is smaller than Formosa group, which is relatively larger, with an average weight above 1.0 kg. Santa-Catarina et al., (2020) undertook a phenotypic characterization of papaya genotypes and reported that fruit size was the trait that most contributed to the formation of groups. Thus, the characterization of diversity in papaya genotypes via molecular markers, as well as morphological analysis, is able to separate individuals from the Solo and Formosa groups.

Molecular characterization contributed to the detection of existing genetic variability and allocation of genotypes into heterotic groups, allowing greater genetic gains to be achieved through selection, as it indicates crosses between divergent parents. In this way, it facilitates the process of parental selection.

An example of this is the commercial hybrids Candy and Vitória. Therefore, dissimilarity analysis via SSR markers facilitates the process of selecting divergent genotypes between groups as well of the most dissimilar genotypes within the heterotic group. In the future, this technique may be used to indicate crosses between Formosa × Formosa and Solo × Solo lines to obtain standard Formosa and Solo hybrids, respectively, and Formosa × Solo crosses to obtain Intermediate fruits (PEREIRA et al., 2019a).

Seed quality and prediction of hybrid crosses to improve seed quality

In terms of germination, the 41 accessions of C. papaya vary widely (0 to 94%) and the process occurs slowly and unevenly. Factors such as time of pollination, storage time, water content, dormancy, as well as genetic factors, are linked to the low germination percentage of papaya. In this respect, the behavior shown by the genotypes highlights the importance of assessing seed quality during the papaya breeding process.

Seeds from Formosa type fruits tend to show a higher GSI than those of the Solo and Intermediate groups. This behavior is related to morphological traits such as seed size and weight (BUIDES et al., 2017). Formosa fruits have larger and heavier seeds when compared with the Solo and Intermediate groups, and the combination of these factors gives these seeds greater energy reserves.

However, the effect of seed size on GSI occurs with greater intensity during the initial growth of the seedling and tends to decrease throughout its development (CARVALHO; NAKAGAWA, 2012).

Seeds with high physiological quality have the ability to germinate, emerge, and produce vigorous and healthy plants, and the lower the %AS, the greater the ability of a seed to form normal seedlings, whereas higher GSIs mean faster germination.

However, the isolated evaluation of %AS and GSI cannot provide information about seed vigor, warranting a combined analysis of these variables for its assessment. In this respect, the high GSI (5.84) exhibited by genotype Sunrise Solo SS 72/12 reflects its high percentage of abnormal seedlings (19%) and high germination percentage (76%), since GSI is determined from the production of the radicle, whereas %AS is determined by the formation of abnormal seedlings at the end of 30 days. In this context, genotypes such as UCLA08-088, which has a high germination percentage (94%), coupled with low formation of abnormal seedlings (1%) and adequate GSI (3.9), produce high-vigor seeds, which ultimately facilitate the establishment of the crop. Vigorous seeds tend to form seedlings that are stronger and more resistant to stress, thus being less susceptible to bad weather during the establishment of the crop in the field. Genotypes such as accession UCLA08-097 (83% G, 7% AS, and 1.19 GSI), in turn, which have a low GSI, make plants more susceptible to competition with weeds, as they require more time to emerge and establish themselves, thus hindering the establishment of the crop and formation of the plant stand in the field. In this way, based on the pre-selection of genotypes with high seed quality (UCLA08- 088, Candy Reciprocal, THB, UCLA08-012, UCLA08-097, UCLA08-092, Sunrise Solo SS 72/12, UCLD08-29II5, Tailandia, UCLD08- 05II5, STZ 03 Pecíolo Curto), it was possible to define three crosses to obtain standard Formosa type fruits (Formosa × Formosa); two for standard Solo type fruits (Solo × Solo); and 13 crosses to obtain Intermediate type fruits (Solo × Formosa), totaling 18 potential crosses to obtain new genotypes aiming at improved seed quality.

Of the 41 accessions evaluated, 11 showed good seed quality. It is worth mentioning that identifying genotypes that comprise high germination percentage + adequate GSI + low percentage of abnormal seedlings increases the probability of obtaining new accessions with high seed quality. It also makes it possible to improve genetic materials with high fruit quality, for instance, but whose seeds’ quality is a limiting factor. Therefore, knowledge about the genetic structure and physiological quality of seeds belonging to the AGB of UENF/Caliman Agrícola S/A is fundamental for the selection of individuals as well as the prospecting of future crosses and production of new varieties that meet the standard established by MAPA for seed production and marketing.

Conclusion

The use of microsatellite markers revealed the existing genetic diversity between the papaya accessions and the excess of heterozygotes present in the lines of the AGB of UENF/Caliman Agrícola S/A, warranting the purification of these lines.

The UPGMA clustering method and genetic structure analysis via microsatellite markers were sensitive enough to discriminate two heterotic groups, identifying the individuals that most diverge and thus generating genetic information that can be used as a basis for the papaya breeding program.

Based on divergence and genetic structure analyses and the evaluation of seed physiological traits, 18 potential crosses can be recommended to improve seed quality.

Acknowledgments

This study was financially supported by the Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES), Research Support Foundation of the State of Rio de Janeiro (FAPERJ), and Caliman Agrícola S/A.

• This study was financially supported by the Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES), Research Support Foundation of the State of Rio de Janeiro (FAPERJ), and Caliman Agrícola S/A.

References

• ALVARES, C.A.; STAPE, L.J.; SENTELHAS, P.C.; GONÇALVES, J.L.M.; SPAROVEK, G. Koppen’s climate classification map for Brazil. Meteorologische Zeitschritt, Stuttgart, v.22, p.711-28, 2013. https://doi.org/10.1127/0941-2948/2013/0507
  » https://doi.org/10.1127/0941-2948/2013/0507
• BOTSTEIN, D.; WHITE, R.L.; SKOLNICK, M.; DAVIS, R.W. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. American Journal of Human Genetics, Cambridge, v.32, p.314–31, 1980.
• BRASIL, Ministério da Agricultura, Pecuária e Abastecimento - Regras para Análise de Sementes MAPA/ACS, 399p., 2009.
• BUIDES, J.A.F.; ÁLVAREZ, A.G.; LA FÉ, P.L.C.; SÁNCHEZ, G.G. Aspectos anatómicos y viabilidad de semillas de papaya (Carica papaya L.) variedad “Maradol Roja” sometidas a almacenamento prolongado. Agrotecnia de Cuba, Havana, v.41, p.41–51, 2017.
• CARVALHO, N. M.; NAKAGAWA, J. Sementes: ciência, tecnologia e produção. 5.ed. Jaboticabal: Funep, 2012. 590 p.
• CARVALHO, E.M.L.; REIS, R.C.; BORGES, V.P.; LEDO, C.A.S.; ARAÚJO, E.S.; DANTAS, J.L.L. Physicochemical and sensory properties of papaya fruits of elite lines and hybrids. Semina Ciências Agrárias, Londrina, v.41, p.121-30, 2020. https://doi.org/10.5433/1679-0359.2020v41n1p121
  » https://doi.org/10.5433/1679-0359.2020v41n1p121
• CRUZ, C.D. Acta Scientiarum GENES - a software package for analysis in experimental statistics and quantitative genetics. Acta Scientiarum - Agronomy, Maringá, v.35, p.271-6, 2013. https://doi.org/10.4025/actasciagron.v35i3.21251
  » https://doi.org/10.4025/actasciagron.v35i3.21251
• DOYLE, J.J.; DOYLE, J.L. A rapid total DNA preparation procedure for fresh plant tissue. Focus, Washington, v.12, p.13–5, 1990
• EUSTICE, M.; YU, Q.; LAI, C.W. Development and application of microsatellite markers for genomic analysis of papaya. Tree Genetics and Genomes, Heidelberg, v.4, p.333-41, 2008. https://doi.org/10.1007/s11295-007-0112-2
  » https://doi.org/10.1007/s11295-007-0112-2
• EVANNO, G.; REGNAUT, S.; GOUDET, J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology, Oxford, v.14, p.2611-20, 2005. https://doi.org/10.1111/j.1365-294X.2005.02553.x
  » https://doi.org/10.1111/j.1365-294X.2005.02553.x
• FAOSTAT - Food and Agriculture Organization. Corporate statistical database 2022. Disponível em: https://www.fao.org/faostat/en/#data/QCL Acesso em: 10 jan 2023.
  » https://www.fao.org/faostat/en/#data/QCL
• JESUS, V.A.M.; ARAÚJO, E.F.; SANTOS, F.L.; DIAS, L.A.S.; SILVA, R.F. Sodium hypochlorite for removal of the sarcotesta from newly extracted and stored papaya seeds. Journal of Seed Science, Londrina, v.38, p.358-64, 2016. https://doi.org/http://dx.doi.org/10.1590/2317-1545v38n4159502
  » https://doi.org/http://dx.doi.org/10.1590/2317-1545v38n4159502
• MAGUIRE, J.D. Speed of germination: aid in selection and evaluation for seedling emergence and vigor. Crop Science, Madison, v.2, p.176-7, 1932.
• MAPA – Ministério da Agricultura, Pecuária e Abastecimento. Normas para a produção e a comercialização de sementes e mudas de espécies olerícolas, condimentares, medicinais e aromáticas Instrução normativa nº 42, de 17 de setembro de 2019. Brasília, 2019. Disponível em: https://sidago.agrodefesa.go.gov.br/site/adicionaisproprios/protocolo/arquivos/1088820.pdf Acesso em: 10 jan 2023.
  » https://sidago.agrodefesa.go.gov.br/site/adicionaisproprios/protocolo/arquivos/1088820.pdf
• MATOS, E.L.S.; OLIVEIRA, E.J.; JESUS, O.N.; DANTAS, J.L.L. Microsatellite markers of genetic diversity and population structure of Carica papaya Annals of Applied Biology, Warwick, v.163, p.298–310, 2013. https://doi.org/10.1111/aab.12053
  » https://doi.org/10.1111/aab.12053
• MELO, A.P.C.; SELEGUINI, A.; PEREIRA, J.; AURÉLIO NETO, R.; WISINTEINER, C.; NEVES, R.G.; CAMILO, Y.M. V. Maturação de frutos e tratamento pré-germinativo na produção de mudas de mamão. Revista de Ciências Agrarias, Recife, v.38, p.330-7, 2015. https://doi.org/10.19084/rca.16937
  » https://doi.org/10.19084/rca.16937
• MORAES, R.; VIVAS, M.; CRUZ, D.P.; SANTA-CATARINA, R.; ALMEIDA, R.F.; SOUZA, Y.P.; DAHER, R.F.; VIANA, A.P.; PEREIRA, M.G. Selection for papaya resistance to multiple diseases in a base population of recurrent selection. Euphytica, Dordrecht, v.217, p.288, 2021. https://doi.org/10.1007/s10681-021-02919-4
  » https://doi.org/10.1007/s10681-021-02919-4
• NANTAWAN, U.; KANCHANA-UDOMKAN, C.; BAR, I.; FORD, R. Linkage mapping and quantitative trait loci analysis of sweetness and other fruit quality traits in papaya. BMC Plant Biology, London, v.19, p.449, 2019. https://doi.org/10.1186/s12870-019-2043-0
  » https://doi.org/10.1186/s12870-019-2043-0
• PEAKALL, R.; SMOUSE, P.E. GenAlEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research--an update. Bioinformatics, Brsibane, v.28, p.2537-9, 2012. https://doi.org/10.1093/bioinformatics/bts460
  » https://doi.org/10.1093/bioinformatics/bts460
• PEREIRA, M.G.; POLTRONIERI, T.P.D.S.; PEREIRA, T.N.S.; RAMOS, H.C.C.; SANTA-CATARINA, R.; VETORRAZZI, J.C.F.; ARÊDES, F.A.; BOECHAT, M.S.B.; VENÂNCIO, T.M.; AZEVEDO, A.O.N.; CORTES, D.F.M.; MOREIRA, N.F.; BOHRY, D.; SILVEIRA, S.F.; VIVAS, M.; FORREGUETTI, G.A. Twenty-two-year papaya breeding program: From breeding strategy establishment to cultivar. Functional Plant Breeding Journal, Campos dos Goytacazes, v.1, p.9–27, 2019a. http:// dx.doi.org/10.35418/2526-4117/v1n2a2
  » http:// dx.doi.org/10.35418/2526-4117/v1n2a2
• PEREIRA, M.G.; LUZ, L.N.; SANTA-CATARINA, R.; RAMOS, H.C.C.; PEREIRA, T.N.S.; BARROS, G.B.; FERREGUETTI, A.G.; CORTES, D.F.M.; VETTORAZZI, J.C.F.; AZECEDO, A.O.N.; SILVEIRA, S.F.; OLIVEIRA, J.G.; VIANA, A.P. ‘UC14’: a new papaya cultivar with intermediate fruit size. Crop Breeding and Applied Biotechnology, Londrina, v.19, p.226-9, 2019b. https://doi.org/10.1590/1984-70332019v19n2c31
  » https://doi.org/10.1590/1984-70332019v19n2c31
• PEREIRA, M.G.; LUZ, L.N.; SANTA-CATARINA, R.; RAMOS, H.C.C.; PEREIRA, T.N.S.; BARROS, G.B.; FERREGUETTI, G.A.; VIVAS, M.; CORTES, D.F.M.; VETTORAZZI, J.C.F.; AZECEDO, A.O.N.; SILVEIRA, S.F.; OLIVEIRA, J.G.; VIANA, A.P.UC10: a new early Formosa papaya cultivar. Crop Breeding and Applied Biotechnology, Londrina, v.19, p.131-4, 2019c. https://doi.org/10.1590/1984-70332019v19n1c18
  » https://doi.org/10.1590/1984-70332019v19n1c18
• PIROVANI, A.A.V.; RAMOS, H.C.C.; DUARTE, R.P.; MIRANDA, D.P.; BOHRY, D.; SANTANA, J.G.S.; SANTOS, P.H.A.D.; PEREIRA, M.G. Molecular characterization of elite lines of papaya (Carica papaya L.) via SSR markers. Functional Plant Breeding Journal, Campos dos Goytacazes, v.3, p.49–58, 2020. http://dx.doi.org/10.35418/2526-4117/v3n1a5
  » http://dx.doi.org/10.35418/2526-4117/v3n1a5
• PIROVANI, A.A.V.; RAMOS, H.C.C.; SANTA-CATARINA, R.; MIRANDA, D.P.; RODRIGUES, A.S.; VETTORAZZI, J.C.F.; SILVA JUNIOR, I.R.; PEREIRA, M.G.; VIANA, A.P. Is there a possibility to improve a developed hybrid? A current demand on papaya (Carica papaya L.). Euphytica, Dordrecht, v.218, p.64, 2022. https://doi.org/10.1007/s10681-022-03014-y
  » https://doi.org/10.1007/s10681-022-03014-y
• POLTRONIERI, T.P.S.; SILVEIRA, S.F.da; VIVAS, M.; SANTA-CATARINA, R.; AZEVEDO, A.O.N.; dos SANTOS, P.H.D.; VETORRAZI, J.C.V., RAMOS, H.C.C.; PEREIRA, M.G. Topcross hybrids in papaya: Genes derived from backcrossing provide resistance to multiple diseases. Crop Protection, Amsterdam, v.37, n.105240, 2020. https://doi.org/10.1016/j.cropro.2020.105240
  » https://doi.org/10.1016/j.cropro.2020.105240
• PRITCHARD, J.K.; STEPHENS, M.; DONNELLY, P. Inference of population structure using multilocus genotype data. Genetic Society of America, Rockwille, v.155, p.945–959, 2000.
• RODRIGUES, A.S.; RAMOS, H.C.C.; MIRANDA, D.P.; PIROVANI, A.A.V.; SANTA-CATARINA, R.; DUATE, R.P.; BRAGA C.S.; BOHRY, D.; PEREIRA, M.G. Intra-genotypic variability in elite parent lines of papaya. Celullar and Molecular Biology, Paris, v.1, p.95, 2023. https://doi.org/10.1590/0001-3765202320220468
  » https://doi.org/10.1590/0001-3765202320220468
• SANTA-CATARINA, R.; VETTORAZI, J.C.F.; CORTES, D.F.M.; SANTANA, J.G.S.; POLTRONIERI, T.P.S.; MIRANDA, D.P.; PIROVANI, A.A.V.; MOREIRA, N.F.; VIVAS, M.; PEREIRA, M.G. Phenotypic characterization of recurrent selection S1 Papaya (Carica papaya L.) families by multivariate approach. Euphytica, Dordrecht, v.1, p.117, 2020. https://doi.org/10.1007/s10681-020-02655-1
  » https://doi.org/10.1007/s10681-020-02655-1
• SANTOS, S.C.; RUGGIERO, C.; SILVA, C.L.S.P.; LEMOS, E.G.M. A microsatellite library for Carica papaya L. cv.Sunrise solo. Revista Brasileira de Fruticultura, Jaboticabal, v.25, n.2, p.263-7, 2003. https://doi.org/10.1590/S0100-29452003000200020
  » https://doi.org/10.1590/S0100-29452003000200020
• SHIVKUMAR, P.; SHAMPRASAD, P.; ROHINI, B.; PETER, A. Molecular diversity analysis and sex determination in papaya (Carica Papaya L.) using molecular markers. Supplement on Genetics and Plant Breeding, v.9, p.1815-20, 2014.
• TURCHETTO-ZOLET, A.C.; TURCHETTO, C.; ZANELLA, C.M.; PASSIA, G. Marcadores na era genômica: metodologias e aplicações. Ribeirão Preto: Sociedade Brasileira de Genética, 2017. 180 p.
• VALE, L.; SÉRGIO, R.; MARTINS, P.H.M.; FÉLIX, M.J.D.; WINDER, A.R.S.; MARQUES, M.L.S.; ASSIS, E. Métodos de remoção da sarcotesta para superação de dormência em sementes de mamão. Brazilian Journal of Development, Curitiba, v.6, p.41161-74, 2020. https://doi.org/10.34117/bjdv6n6-598
  » https://doi.org/10.34117/bjdv6n6-598
• WEBSTER, R.E., WATERWORTH, W.M., STUPPY, W., WEST, C.E., ENNOS, R.,BRAY, C.M., PRITCHARD, H.W. Biomechanical, biochemical, and morphological mechanisms of heat shock-mediated germination in Carica papaya seed. Journal of Experimental Botany, v. 67, p. 6373–6384, 2016.

Edited by

• Willian Krause

Data availability

Data citations

FAOSTAT - Food and Agriculture Organization. Corporate statistical database 2022. Disponível em: https://www.fao.org/faostat/en/#data/QCL Acesso em: 10 jan 2023.

MAPA – Ministério da Agricultura, Pecuária e Abastecimento. Normas para a produção e a comercialização de sementes e mudas de espécies olerícolas, condimentares, medicinais e aromáticas Instrução normativa nº 42, de 17 de setembro de 2019. Brasília, 2019. Disponível em: https://sidago.agrodefesa.go.gov.br/site/adicionaisproprios/protocolo/arquivos/1088820.pdf Acesso em: 10 jan 2023.

Publication Dates

• Publication in this collection
  04 Nov 2024
• Date of issue
  2024

History

• Published
  14 Oct 2024
• Received
  11 July 2023
• Accepted
  07 Aug 2024