Screening of Diploid Wild Potatoes Pollen Traits under Heat Stress

1. Introduction

As F1-hybrid potato breeding continues to progress, the assessment of pollen viability becomes an essential trait in germplasm screening [1]. In regions with tropical climates characterized by high temperatures, the identification of genotypes with robust blooming and appropriate pollen viability under supraoptimal temperatures can be advantageous for the cultivation of True Potato Seeds (TPS) propagated cultivars [2]. In the context of global climate change, exploring the phenotypic plasticity of crops in response to rising temperatures is crucial for anticipating potential heat-related effects[3]. Considering future climate change scenarios projecting a 1 to 3°C increase in global average temperatures during the twenty-first century [4,5], there is a potential for substantial reductions in agricultural production and impacts on natural populations of crop wild relatives in their natural distribution areas, including potatoes[6,7].

Globally, potato (Solanum tuberosum L., Solanaceae) ranks as the third most important crop, trailing only rice and wheat[8,9]. Despite the genetic diversity of the potato crop is relatively limited[10], potatoes and their wild relatives encompass a phylogenetic branch comprising 107 wild species, four domesticated species, and their natural and artificial hybrids[11]. Natural populations of potato wild species have evolved in extreme environmental conditions such as in arid regions, high temperatures, intense sunlight, water scarcity, and high altitudes, given their extensive geographical distribution from the southwestern United States to Mexico, Peru, and the coastal regions of Chile, including Brazil[12,13]. The genetic variability is crucial for maintaining and improving the genetic foundation of agricultural crops[14], and can be found in the crop wild relatives, which are considered natural sources of genes with agronomic, industrial, and nutritional significance[15,16,17].

The domesticated and wild potatoes are members of Solanum sect. Petota, which can reproduce sexually, involving botanical seeds, also referred to as true-potato-seeds (TPS), and vegetatively, via stolons and tubers[18,19]. Particularly, the ability to generate diploid inbred lines through TPS[9,20,21] stands in contrast to the prevailing practice of predominantly tetraploid commercial potato crops (2n=4x=48), which are propagated through vegetative manner by tubers[20]. Diploid potato cultivars propagated through TPS help prevent the transmission of viruses to the next generation and facilitates the storage and conservation of propagative material[21,22]. Even more, along with the potential utility of an F1-hybrid population for genetic mapping studies[23,24,25]. This contrast has piqued significant interest among breeders, farmers and for the seed industry towards the development of F1-hybrid potato cultivars[22,26,27]. This heightened interest is further fueled by the accessibility of modern plant breeding techniques and genetic tools[1,28,29].

The effect of heat stress can lead to the unreduced (2n) gametes formation in many crop species such as tomato (Solanum lycopersicon L.)[30] and wheat (Triticum turgidum L.)[31], causes pollen unviability[32], and can cause disruptions in the stability of the natural populations[33].

By investigating the impact of temperature on pollen viability and 2n pollen production in diploid potato wild relatives, our study aims to contribute to the identification of genotypes that can thrive in high-temperature environments. This research is crucial for the development of potato cultivars that can withstand the challenges posed by global climate change.

2. Materials and Methods

2.1. Plant Material

This study used nine wild potato accessions from the species S. chacoense (BRA 00167447-2, BRA 00167017-3, BRA 00167023-1, BRA 00167028-0), S. commersonii (BRA00167007-4, BRA00167420-9, BRA00183760-8), and S. malmeanum (BRA 00183755-8), as well as a control accession from of the commercial cultivar of S. tuberosum named BRSIPR-BEL (BRA 00167251-8) from the Embrapa Potato Gene Bank (Table 1).

Tubers of uniform size were planted in a 5L plastic container with TurfaFértil® organo-mineral substrate and supplemental fertilizer according to the crop instructions. The plants were transported to the growth chambers after 15 days of growth and exposed to two temperature gradients: control temperature (Figure 1A), with thermal amplitude ranging from 14 to 27°C; and supraoptimal temperature (Figure 1B), with amplitude ranging from 24 to 34°C (Figure 1B). The photoperiod was 12 hours (7:00 to 19:00) with a light intensity of about 400 mol m^-2 s^-1. The plants stayed under these two distinct treatments until harvest.

A double factorial experimental scheme 9 × 2 (genotype vs. temperature), was set up in randomized blocks with two replications each consisting of one plant.

2.2. Pollen Viability Assessment

Five flowers in the anthesis stage (when the floral bud opens) from each accession were collected on the same day for anther extraction and pollen processing[34]. All the accessions from the two treatments opened the flowers in the same week, and pollen was collected seven days after the anthesis. Pollen from each plant was collected and preserved in 1.5 mL tubes in a freezer (-4^oC) until analysis. To determine pollen viability, the percentage of staining pollen (PSP) was accessed in 400 grain samples dyed with red aceto-carmine glycerol[35]. The examination of samples, as described in the technique provided by the protocol of CIP (International Potato Center)[36] and demonstrated in Figure 1C, was conducted using an optical microscope with a 200x magnification.

The formula for calculating viability (percent) is according to:

$$Viability \left(\%\right)=\frac{Number of stained pollen grains}{Total number of pollen grains}\times 100$$

Based on the viability range data obtained, four distinct categories were defined to classify pollen viability[36]. The percentage of pollen viability was considered sterile (0%), low (> 0% and ≤ 50%), moderate (< 50% and ≤ 80%), or high (< 80% and ≤ 100%).

2.3. Pollen Size

To distinguish 2n pollen, pollen grains were stained with a droplet of 2% (w/v) aceto-carmine solution on a microscope slide. Enlarged pollen grains were classified as 2n pollen if their diameter exceeded 1.2 times the average diameter of the pollen grains in the control group[36]. For analysis of pollen size, pollen from each accession subjected to both temperature treatments were collected, following the protocol described by CIP[36].

2.4. Statistical Analysis

The standard error and significant differences (p≤0.05) for temperature, genotype, and genotype interaction were determined using analysis of variance (ANOVA), for pollen viability and pollen size. Following that, the Tukey mean comparison test was used, and the analyses were carried out using the function fat2.dbc of the ‘ExpDes.pt’ package[37] in the RStudio environment[38].

2.4.1. Pollen Viability-Based Heat Susceptibility Index (HSIpv)

The pollen viability-based heat susceptibility index (HSIpv) was calculated in Microsoft Excel 2023 using percent pollen viability values of heat stress and normal conditions following the formula below[39,40]:

$$HSIpv=1-\left(\frac{Xstress}{Xnormal}\right)/1-\left(\frac{\stackrel{-}{\mathrm{x}}\mathrm{stress}}{\stackrel{-}{\mathrm{x}}\mathrm{normal}}\right)$$

Building upon, HSIpv genotypes were grouped into three classes: tolerant (HSIpv <0.5), moderately tolerant (HSIpv 0.5-0.99), and susceptible (HSIpv >1.0).

2.4.2. Genetic Parameters

The basic genetic parameters – PCV%, GCV%, h² (broad sense), genetic gain (GG) and (genetic advance) GA were calculated. The phenotypic coefficient of variance (PCV) and genotypic coefficient of variance (GCV) were calculated using the formulas: $PCV \left(\%\right)=\frac{\surd Vp}{mean}\times 100$ and $GCV \left(\%\right)=\frac{\surd Vg}{mean}\times 100$, respectively.

Where PCV is phenotypic coefficient of variance, VP is phenotypic variance, GCV is genotypic coefficient of variance, and Vg is genotypic variance. To categorize the GCV and PCV values, the classification proposed by Burton[41] was employed. GCV and PCV values falling within the range of 0 to ≤10% were classified as low, values ranging from 10 to ≤20% were categorized as moderate, and values exceeding 20% were considered high.

2.4.3. Heritability

Estimated as the ratio of total genotypic variance to the phenotypic variance[42]:

$$H^2=\frac{Vg}{Vp}\times 100$$

where 𝐻2 = % broad sense heritability. The heritability percentage was categorized as low (0–30%), moderate (30–60%), and high ≥60% as given[43]:

$$GG\%=\frac{GA}{\sum }\times 100$$

The genetic gain was calculated by the ratio of genetic advance ($GA=(k\times \sqrt{V_p}\times H^2$), k represents the constant value 2.06 at 5% selection intensity and overall mean (∑) of the population used.

3.1. Analysis of Variance

The results of the ANOVA indicated that both main effects, block, genotype and temperature, and the G×E interaction (Table 2). A non-significative reduction in pollen viability was observed with increasing temperature in genotypes BRA00167007-4, BRA00167420-9, BRA 00167447-2, BRA 00167017-3, BRA 00167028-0, BRA 00183755-8 and BRA00183760-8. Conversely, BRA 00167023-1 failed to produce pollen, and BRA 00167251-8 did not bloom under elevated temperatures; hence, these two genotypes were not included in this analysis.

3.2. Pollen Viability-Based Heat Susceptibility Index (HSIpv)

In the Table 3 are presented the genetic parameters estimated for pollen viability. GCV (15.76%) was lower than PCV (17.84%). Following Johnston et al.[44] genetic advance (% average) is categorized as low (<10%), moderate (10 – 20%) and high (>20%). In our study the genetic advance for pollen viability was high (23.30%) (Table 3).

Pollen viability of the potato wild genotypes from the two experimental conditions are presented in Table 4. As the G×E interaction was non-significative, the Tukey test was only applied for each treatment. At the CT, genotypes BRA 00183755-8, BRA 00167017-3, BRA 00167028-0, BRA 00167447-2, and BRA00167007-4 exhibited the highest pollen viability values. Conversely, genotypes BRA00167420-9 and BRA00183760-8 displayed the lowest pollen viability values. At the HS conditions, the maximum pollen viability values were observed in genotypes BRA 00167017-3, BRA 00183755-8, and BRA 00167028-0. However, the genotypes BRA 00167447-2, BRA00183760-8, BRA00167007-4, and BRA00167420-9 demonstrated lower pollen viability at the HS conditions.

The heat susceptibility (HSIpv) is a useful criterion to select heat-tolerant genotypes[40]. The genotypes with HSIpv < 0.5 were considered as heat tolerant, HSIpv 0.5 – 0.99 were considered as moderately tolerant, and genotypes with HSIpv > 1.0 were considered as susceptible[40]. The genotypes categorized as tolerant were BRA00183760-8 (0.06), BRA 00167017-3 (0.36), BRA 00183755-8 (0.66) and BRA 00167028-0 (1.01). While BRA 00167447-2 (1.57), BRA00167420-9 (1.59), and BRA00167007-4 (1.89) were moderately tolerant (Table 4).

3.3. Heritability for Pollen Viability

The estimated heritability for pollen viability varied from 58.82% to 90.79%. The lowest h² was in BRA00167007-4 (58.82%) and the highest value was observed in BRA 00167023-1 (90.79%), followed by BRA 00183755-8 (91.32%), BRA 00167028-0 (88.56%), BRA 00167017-3 (88.35%), BRA00167420-9 (87.72%), BRA00183760-8 (83.82%) and BRA 00167447-2 (83.16%). All genotypes were classified as high heritability. Heritability is classified as low (<30%), medium (30 – 60%), and high (>60%)[42,44]. High heritability indicates that the selection for the trait is effective and is less influenced by environmental effects[42,44]. Broad-sense heritability becomes particularly intriguing when it applies to situations where the full genetic variability can be harnessed in clones or single-cross hybrids[45].

Non-significant variation in pollen viability was observed as the temperature increased, indicating that the genotypes maintained their performance under both treatments. Heat stress has been demonstrated to adversely affect or even inhibit pollen viability in a diverse range of agricultural crops, including corn (Zea mays L.)[46], wheat (Triticum aestivum L.)[47], rice (Oryza sativa L.)[48], tomato (Solanum lycopersicum L.)[49], and wild potatoes (Solanum sect. Petota)[50]. Similar to our findings, under elevated temperature stress conditions, Bamberg and collaborators[50] observed that the percentage of pollen viability above 45% was seen in genotypes of different wild potatoes, including: S. stoloniferum Schltdl. (cited as S. fendleri A. Gray, PI 275156 with 50%, PI 497998 with 56%, PI 498004 with 63%), S. stoloniferum Schltdl. (cited as S. polytrichon Rydb. (PI 255547 with 49%), S. demissum Lindl. (PI 160208 with 70% and PI 498232 with 50%) and S. jamesii Torr. (PI 458425 and PI 195190, both with 45%), S. chacoense (PI 320293 with 49%), S. commersonii (PI 243503 with 53%), S. infundibuliforme Phil. (PI 498351 with 68%), and S. boliviense Dunal in DC. (cited as S. megistacrolobum Bitter (PI 473133 with 47%). Except for the genotypes that did not produce pollen and did not bloom under heat stress, S. chacoense (BRA 00167023-1) and S. tuberosum (BRA 00167251-8), respectively, the genotypes of the species S. chacoense (BRA 00167017-3, BRA 00167028-0, BRA 00167447-2), S. commersonii (BRA00167007-4, BRA00183760-8, BRA00167420-9), and S. malmeanum (BRA 00183755-8) exhibited similar pollen viability results to the study by Bamberg and collaborators[50], with values higher than 54.5%.

3.4. Pollen Size

Pollen size was compared using a non-parametric Wilcoxon-Mann-Whitney statistical test. The average of pollen size was 20 μm in all wild potato genotypes. Pollen size in BRA00167007-4 was 20 μm at the CT and 22 μm at the HS. BRA00167420-9, BRA 00167447-2, BRA 00167017-3, BRA 00167028-0, BRA 00183755-8, BRA00183760-8 exhibited pollen with 20 μm in size from both treatments. BRA 00167023-1 produces pollen with 20 μm and BRA 00167251-8 with 24 μm in size at the CT, also these two genotypes did not produce pollen and did not bloom at the HS treatment, respectively (Figure 3).

The increase in temperature did not result in the production of non-reduced (2n) pollen grains in any of the genotypes evaluated in this study. To consider the production of 2n pollen grains, there needs to be 1.2 times increase compared to the normal[36].

The variations on pollen viability among the tested genotypes allows the selection of desirable genotypes for future potato crop improvement. This variation indicated that there is a way to identify promising genotypes based on the pollen viability trait. However, the PCV and GCV for all genotypes in both locations were moderate. A medium level of the coefficient of variation implies an equal influence of additive and nonadditive gene action[51]. In this present investigation, the PCV was higher than the GCV, but the differences between the PCV and GCV ranges were low, indicating the minimal impact of the environment on the expression of the traits, which is a sign of the heritable nature of the traits[51]. Several studies have also observed a higher PCV than GCV[52,53]. Hence, there is significant potential for precise trait selection based on the phenotypic expression[54,55,56,57].

Heritability is a powerful tool used to estimate the degree of variation within a population[58]. In this current investigation, a high heritability was observed for pollen viability. Previous research has consistently shown that high heritability implies minimal environmental influence on the genotype[59,60,61]. Therefore, minimized environmental effect on the studied trait allowed the high accuracy of 94% in the selection of genotypes under the HS conditions[40,62,63,64]. This information from the current study related to heritability is helpful for selecting the best trait for the improvement of crops[40,62,63,64]. The current study also revealed a high level of heritability for the pollen viability trait in potato wild genotypes. Nevertheless, placing exclusive reliance on heritability-based trait selection may not consistently lead to success, since broad-sense heritability accounts for the total genetic variance, which includes additive, dominant, and epistatic variances[65]. Therefore, estimation of the heritability of a group of genotypes coupled with high genetic advance is more reliable and efficient for the selection of desirable traits for a group of the population[66]. High heritability coupled with high genetic gain were found for the pollen viability in our study, which gave information to select superior genotypes[67,68]. Pollen viability is predominantly governed by additive gene action and can be improved by simple selection, hence, would be favorable for a potato breeding program[69].

Pollen viability studies in wild potato species grown under heat stress conditions are scarce in the literature. The identification of heat-tolerant wild potato genotypes regarding pollen viability favors the direction of future studies that relate stress, as plants respond to heat stress triggering a cascade of physiological, biochemical, and molecular processes and adapt by activating several stress-responsive genes[70,71]. Conducting a screening for the ability to produce viable pollen under heat stress conditions, as performed in this study, is crucial for identifying heat tolerance. It paves the way for further molecular characterization of genes involved in this process and facilitates the selection of promising genotypes for future stages of a potato genetic improvement program.

The abortion of pollen development observed in BRA 00167023-1 at the HS condition can be attributed to several factors, including the degradation of membrane integrity, the accumulation of reactive oxygen species (ROS), alterations in carbohydrate metabolism[72], disruptions in protein and lipid metabolism[73], and shifts in phospholipid profiles[74], which led to degeneration and abnormalities in tapetum cells[75].

Some causes for the non-development of the flowers in BRA 00167251-8 under heat stress conditions can be hypothesized. The transition to flowering and its timing are regulated by endogenous and environmental cues such as light and temperature[76]. Signaling pathways, in response to various endogenous cues such as hormones, the circadian clock, carbohydrate source-to-sink ratios, as well as autonomous and environmental stimuli like photoperiod and temperature, converge toward a limited number of floral integrator genes. Notably, these genes include SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1), FLOWERING LOCUS T (FT), and AGAMOUS-LIKE 24 (AGL24)[77]. These genes play a pivotal role in activating meristem identity genes, namely LEAFY (LFY), APETALA1 (AP1), SEPALLATA3 (SEP3), and FRUITFULL (FUL), which instigate the irreversible transition from a vegetative to a floral meristem[78] (Lohani et al. 2020).

To accomplish this goal, a detailed understanding of events underlying response to supra-optimal temperature during sexual reproduction and identification of reproductive traits related to thermotolerance will provide better tools for breeders to develop heat stress-resilient crops with enhanced crop productivity[79,80,81,82,83,84,85,86,87].

The identification of heat-tolerant wild potato genotypes that demonstrated enhance pollen viability is a significant milestone in the advancement of pre-breeding efforts aimed at improving the potato crop. Heat stress induces a cascade of physiological, biochemical, and molecular processes in plants, prompting them to activating numerous stress-responsive genes as a mean of adaptation[1,70]. The capacity to produce viable pollen under elevated temperatures, as evidenced in this study, plays a pivotal role in the identification of heat tolerance, and facilitates the subsequent molecular characterization of genes involved in this intricate process.

3. Conclusions

Pollen viability in wild potato germplasm is genetically variable whether grown at optimum temperature (control) and high temperature (stress).

The S. malmeanum genotype (BRA 00183755-8) and S. chacoense genotypes (BRA 00167017-3, BRA 00167028-0 and BRA00183760-8) were classified as tolerant for pollen viability under heat stress, signifying their potential as valuable candidates for future improvement studies.

Elevated temperatures do not lead to 2n pollen formation in the potato wild genotypes evaluated.