Introduction

It is well known that the impact of pollen, also referred to as the xenia effect1, varies in its influence on fruit size across different plant species. Numerous studies have demonstrated that pollen from different sources significantly affects the physical characteristics of fruit2,3,4. Efficient pollen can often enhance fruit set rate and quality while shortening the fruit development period5,6. For instance, cross-pollination affects the nut diameter and fruit shape index of Macadamia7, increases fruit size, enhances blueberry fruit set rate, and shortens the development period6. Furthermore, the need for assessing viability of pollen used is important in determining the strategies to be used in hybridization. Pollen storage capacity is related to its own genetic characteristics and external factors, and different pollen exhibit different longevity with various storage conditions, as well as different viability after certain periods of storage8,9.Temperature is one of the key factors affecting pollen viability. Therefore, exploring the storage capacity of pollen and selecting suitable hybrid parents are critical for enhancing hybridization potential.

Combining ability of varieties is one of the main methods used to identify superior parental combinations. It is an important indicator for evaluating the utility value of parents and forming the basis for developing highly advantageous hybrid offspring10. Pongpitak et al. evaluated the combining ability of six cassava varieties and determined that the general combining ability (GCA) effects predominantly influenced fresh shoot yield, harvest index, and starch content11. Similarly, Damião et al. used combining ability as a strategy of upland cotton selection for high fiber quality and identified the optimal hybrid varieties through specific combining ability12. Therefore, research on the heritability and genetic components of comprehensive breeding traits can not only improve the predictability of breeding work and reduce its randomness but also provide a scientific basis for selecting appropriate breeding strategies13.

  • In China, the genus Gleditsia comprises six distinct species along with two varieties, among which G. sinensisis the most well-known one. It is widely used in afforestation, landscaping, and traditional medicine, and has high economic value14. The fruit of G. sinensisis not only an important ingredient in traditional chinese medicine but also plays a significant role in detergents, cosmetics, and food additives15. G. sinensisis a dioecious plant with male and bisexual flowers on separate plants, predominantly outcrosses and hermaphrodite flowers are functionally female flowers16. In the practice of cultivating G. sinensis, inadequate consideration has been given to the configuration of male and female varieties, and pollination often relies on natural processes with unclear pollen sources. This has led to inconsistent quality among Gleditsia varieties and low yields.

  • Despite these challenges, efficient pollination strategies, including intraspecific and interspecific hybridization, could potentially improve fruit size, set rate, and overall yield. Therefore, in this study, we compared the pollen viability of different Gleditsia varieties, observed the fruit set rate and fruit size of various Gleditsia hybrid combinations, and analyzed genetic parameters such as combining ability and heritability of key breeding traits. This study aimed to explore the hybridization potential of G. sinensis with other Gleditsia species and elucidate the genetic influence of parental lines on G. sinensis fruit traits. Information from this study can provide important insights for optimizing pollination strategies in Gleditsia and offer a scientific basis for broader hybridization practices within the Gleditsia genus.

Materials and methods

Overview of the experimental site and materials

The hybridization experiments were conducted at the Maocang Gleditsia Base in Maocang Town, Guizhou Province (26°32’19.554” N, 106°2’14.455” E) at an altitude of 1398.3 m. The region has a northern subtropical highland monsoon humid climate, with an annual average temperature of 11.6–15.4 °C and annual precipitation of 938.4–1340.6 mm. The maternal plants were selected from well-growing female G. sinensis trees within the base. These trees were of similar age and size (approximately 12 years old), with an average trunk diameter of 15.4–16.1 cm and heights ranging from 6 to 8 m. All maternal plants were cultivated under uniform conditions, including regular irrigation, fertilization, and pest control, to ensure optimal growth and consistent health. Paternal pollen was collected from high-quality specimens of G. sinensis, Gleditsia japonica var. delavayi (G. delavayi), G. japonica, and G. fera located in Huaxi District and Wudang District of Guiyang City, and Anlong County of Southwest Guizhou Autonomous Prefecture. Among them, G. delavayi is a variant of G. japonica, with inflorescences that are axillary or terminal and covered with short soft hairs. The pods are irregularly twisted or curved, forming a sickle shape, with a smooth surface and no hairs. The female inflorescences are 7–9 cm in length. The number of flower bases in the male inflorescences is significantly higher than that in G. japonica. The male inflorescence structure for each variety used in the experiment is shown in Supplementary Fig. S1, and the detailed information about the paternal sources can be found in Supplementary Table 1.

Experiment method

Hybrid bagging

On clear or cloudy days, we visited the base to select suitable inflorescences from all directions of the maternal trees. We removed the male flowers (including opened male flowers and male flower buds) and any opened female flowers from the inflorescences. Flowers that were well-developed but with petals not yet fully open (i.e., not yet pollinated) were chosen as the pollination targets. We then used 500-mesh pollen isolation bags to cover these flowers, preventing contamination from external pollen(Fig. 1).

Fig. 1
figure 1

Diagram of hybrid bagging of female parent trees using pollen isolation nets before pollination.

Full size image

Pollen collection and preservation

On clear or cloudy days, inflorescences of each paternal Gleditsia variety in the large bud stage were collected and placed into sulfuric acid paper bags with labels. These bags were immediately placed into a cooler with ice packs and transported back to the laboratory. Using fine-tipped tweezers, the anthers were extracted onto sulfuric acid paper. The anthers were dried in a desiccator for 12 h and then placed into small drying vials containing color-changing silica gel. After determining pollen viability, the vials were stored in dark conditions at 4℃, −20℃, and − 80℃ for long-term preservation.

Cross-pollination

When the stigmas of the bagged female flowers were receptive, pollen of each hybrid male parent was dusted onto them on a sunny and windless day. Before pollination, the viability of the paternal pollen is measured. After completing the manual pollination, the pollen isolation bags are immediately sealed. The pollination date and number of pollinated flowers are recorded, and the flowers are tagged accordingly. The pollination details for each hybrid combination are provided in Supplementary Table 2.

Determination of pollen viability

The TTC(2%), MTT(1%) staining and pollen germination in vitro were used to test the pollen viability. Add 1 mL of dyeing solution and culture solution to a 1.5 mL centrifuge tube respectively, and pollen was collected using a fine brush, transferred to centrifuge tubes for processing, and stored under controlled conditions for subsequent use. Next, we transferred a drop of the mixed solution onto a microscope slide and used an Olympus DP74 microscope (10x magnification) to observe the pollen staining and pollen tube growth. Pollen grains with pollen tube elongated longer than the diameter of the pollen grain were scored as successful germination. Each method was repeated three times, three visual fields were randomly selected from each experimental treatment, and no less than 80 pollen grains were counted in each visual field.

Pollen viability = (number of viable pollen/the total number of observed pollen grains) × 100%.

Pod phenotype and fruit setting rate

After pollination, pod phenotypic changes and fruit setting were regularly observed (Fig. 2). A tape measure was used to measure the pod length (accurate to 0.1 cm), and a vernier caliper was used to measure the pod width and thickness (accurate to 0.01 mm). The number of fruits set was also recorded.

Fruit setting rate =(number of pods/number of pollinations)× 100%.

Fig. 2
figure 2

Pods and seeds after hybrid pollination.

Full size image

Data processing and analysis

This study used R software to analyze hybrid fruit traits. One-way analysis of variance (ANOVA) was used to evaluate the differences in combining ability and pollen viability between the hybrid parents. Prior to conducting the analysis of variance (ANOVA), we performed normality and homogeneity of variance tests on the data. The results indicated that the data met the assumptions required for ANOVA. Detailed results of these tests have been included in Supplementary materials. Mixed linear model analysis was performed using the ‘’lme4’’ software package, and breeding values were estimated using the best linear unbiased prediction (BLUP) method17.

The combining ability was calculated using the following equation, according to Shuchun Li and Griffing18,19:

The general combining ability (GCA): \(\:\text{G}\text{C}\text{A}\text{i}=\text{X}\text{i}-\text{X}.\)

The specific combining ability (SCA): \(\:\text{S}\text{C}\text{A}\text{i}\text{j}=\text{X}\text{i}\text{j}-\text{X}.-\text{G}\text{C}\text{A}\text{i}-\text{G}\text{C}\text{A}\text{j}\)

In the formula, GCAi is the estimated GCA value of parent i; Xi is the average value of hybrid fruits of parent i, and X. is the overall average value of hybrid fruits. SCAij is the estimated SCA value of parent i and j combination, Xij is the average fruit value of parent i and j hybrid combination, GCAi is the GCA value of parent i, and GCAj is the GCA value of parent j.

An ANOVA (analysis of variance) was estimated using following mixed linear model:

$$\:\text{Y}\text{i}\text{j}={\upmu\:}+\text{F}\text{i}+\text{M}\text{j}+\text{F}\text{M}\text{i}\text{j}+\text{e}\text{i}\text{j}$$

In the formula, Yij is the hybrid pod between the i-th male parent and the j-th female parent, µ is the mean of all observed values, Fi is the male parent effect, Mj is the female parent effect, FMij is the interaction effect between the female parent and the male parent, eij is the random error.

The genetic components were determined following the methodology proposed by Kearsey and Pooni, and the heritability was calculated using the following equation20:

narrow sense heritability: \(\:{\text{h}}_{2}=\frac{{\text{V}}_{\text{f}}+{\text{V}}_{\text{m}}}{{\text{V}}_{\text{f}}+{\text{V}}_{\text{m}}+{\text{V}}_{\text{f}\text{m}}+{\text{V}}_{\text{e}}}\)

broad sense heritability: \(\:{\text{H}}_{2}=\frac{{\text{V}}_{\text{f}}+{\text{V}}_{\text{m}}+{\text{V}}_{\text{f}\text{m}}}{{\text{V}}_{\text{f}}+{\text{V}}_{\text{m}}+{\text{V}}_{\text{f}\text{m}}+{\text{V}}_{\text{e}}}\)

In the formula, Vf, Vm, Vfm, and Ve are the variance components of the male parent, the female parent, the interaction between the male parent and the female parent, and the random error, respectively.

Results

Viability determination and storage of pollen

From Fig. 3 it can be seen that except for TTC staining, the other two methods can detect the viability of pollen. Among them, the viable pollen grains are stained purple by MTT or pollen tubes are elongated, and the TTC method failed to stain pollen grains or the staining was not obvious. With extended storage time, the pollen viability decreased significantly at all storage temperatures (Fig. 4). After 60 days of short-term storage, the differences in pollen viability among G. sinensis-W, G. delavayi and G. japonica under different temperatures were minimal. In contrast, the pollen viability of G. sinensis-T and G. sinensis-Q was significantly higher at −80℃ compared to 4℃, while the pollen viability of G. fera was lowest at −80℃.

Fig. 3
figure 3

Measurement results of different pollen viability detection methods, including (a) using TTC staining method, (b) using MTT staining method, and (c) using in vitro germination method.

Full size image
Fig. 4
figure 4

Effect of different storage temperatures on the vitality of male parent pollen.

Full size image

Changes in fruit setting rate

The dynamic changes in fruit set rate after pollination for each hybrid combination are shown in Fig. 5a. Most of the flower and fruit drop occurred within the first 35 days after pollination, with a significant decline in fruit set rate, up to 76%, occurring within the first 15 days. After 35 days, the fruit set rate stabilized. Through variance and correlation analyses (Table 1; Fig. 5b), it was found that significant differences in fruit set rates among different maternal plants, while the influence of the paternal parent on fruit set rate was less significant compared to the maternal parent. Additionally, no clear linear relationship was observed between pollen viability and fruit set rate, indicating that when paternal pollen viability is relatively high, the maternal parent was the primary factor affecting the fruit set rate.

Fig. 5
figure 5

(a) The dynamic changes in fruit set rate, (b) The impact of pollen viability on the final fruit set rate.

Full size image
Table 1 Variance analysis and correlation analysis of the fruit set rate in G. sinensis.
Full size table

Combining ability of pod phenotypic traits

It can be seen from Table 2 that the GCA of different parents has obvious differences in the length, width and thickness of the pod. The female parent GCA has a significant impact on the length, width and thickness of the pod, while the male parent GCA has a significant impact on the thickness of the pod, and has no significant impact on the length and width. There are great differences in SCA effect values of the same trait in different hybrid combinations or different traits in the same hybrid combination (Table 3). G. sinensis−4 X G. sinensis-T has the largest SCA effect value on pod length and pod width, while G. sinensis−1 X G. sinensis-T has the smallest. For pod thickness, G. sinensis−3 X G. sinensis-W has the largest SCA effect value, while G. sinensis−5 X G. fera has the smallest SCA effect value.

Table 2 Significance analysis of combining ability of pod phenotypic traits.
Full size table
Table 3 Analysis of combining ability effects on pod phenotypic traits.
Full size table

Genetic variation and heritability of pod phenotypic traits

Through the analysis of genetic variation in pod phenotypic traits (Table 4), it was found that for pod length and width, the variance component of the maternal parent was the largest, while the paternal variance component was zero, and the interaction effect between parents was minimal. Both traits exhibited high heritability, with broad-sense heritability reaching 94.0% and 83.4%, and narrow-sense heritability at 90.8% and 63.4%, respectively. In contrast, for pod thickness, the variance components of both the female and male parents and their interaction effects were relatively small, with broad-sense and narrow-sense heritability being relatively low, at 49.5% and 38.2%, respectively. This indicates that the maternal genetic contribution is the most important factor, with pod width and length being more stably inherited, while pod thickness is more susceptible to other influencing factors.

Table 4 Estimation of genetic parameters for pod phenotypic traits.
Full size table

Breeding values of pod phenotypic traits

Based on the breeding value estimates in Table 5, the female parent significantly contributes to the improvement of pod length and width, particularly the G. sinensis−4, which shows significantly higher pod length and width attributes compared to the overall average effect. For pod thickness, the female parent G. sinensis−5 has the highest breeding value. The breeding values of the male parents, from highest to lowest, are G. delavayi, G. fera, G. sinensis-Q, G. sinensis-W, G. japonica, and G. sinensis-T. Among the breeding values of the various hybrid combinations, the intra-specific hybrid combination of G. sinensis−4 X G. sinensis-T exhibits higher pod length and width attributes than other combinations. In interspecific hybrids, combinations with G. fera as the male parent show positive effects on pod length, width, and thickness.

Table 5 Estimated breeding values of pod phenotypic traits.
Full size table

Discussion

As an important carrier of plant germplasm resources, pollen is one of the important forms of plant germplasm resource preservation. Understanding the pollen viability of existing germplasm is crucial for selecting hybrid donors21. For long-term preservation, it is crucial to select an appropriate storage method to maintain the pollen viability of Gleditsia. It is reported that low temperature can slow down pollen metabolism, reduce water evaporation rate and oxidative damage, and extend the storage life of pollen22,23,24. In our study, with the extension of storage time, the viability of all pollen decreased significantly. Under 60 days of short-term storage, most pollen tends to maintain higher viability at lower temperatures. Notably, G. ferapollen showed high viability at −20 °C, but its activity at −80 °C was lower than at 4 °C. This result may be related to the biological characteristics and adaptability of this specific pollen species. After a storage period of up to one year, pollen viability was highest at −80°C, while it was extremely low at 4 °C (Supplementary Fig. S2). Similar results have also been reported in pollen preservation studies of pecans25, apples26, peonies27and other plants. This may be because higher temperatures accelerate pollen metabolism and moisture loss, reducing physiological activity, while extremely low temperatures significantly slow down metabolism and oxidation reactions, allowing pollen to maintain high viability28. Therefore, different pollen varieties may have varying sensitivity to temperature, leading to differences in optimal storage conditions.

The fruit setting rate of a plant is one of the important indicators to measure its yield. A high fruit setting rate means that the plant can make full use of resources and increase yield and quality29. Previous studies have found that the selection of paternal pollen significantly affects fruit set rates, as observed in wolfberry and oranges30. However, in the hybridization experiments of G. sinensis, significant differences in fruit set rates were observed among different maternal plants. The pollen used in the experiment exhibited high viability, and no clear linear relationship was observed between pollen viability and the final fruit set rate, indicating that the influence of the paternal variety on fruit set rate is relatively smaller compared to the maternal parent. This may be because the paternal plant primarily transfers genetic information through pollen without directly participating in the nutrient supply and later development of the pods, thus having a small impact on the fruit set rate. The genetic differences among paternal individuals are substantial, influencing their affinity in close or distant hybridizations, where combinations with lower affinity often result in lower fruit set rates31. In this experiment, when G. fera is hybridized with G. sinensis, it exhibits a high fruit set rate and positive effects on pod phenotypic traits. This indicates that G. fera has strong affinity with G. sinensis and can serve as an excellent parent for interspecific hybridization. Therefore, in addition to close hybridization within species, G. sinensis can also hybridize distantly between different Gleditsia species. In this experiment, the maternal plant, as the primary source of nutrients, is the key factor influencing the fruit set rate of G. sinensis. Within the first 15 days after pollination, each hybrid combination can still maintain a high fruit setting rate. However, during the subsequent 20 days, there was a significant physiological fruit drop, leading to a decreased fruit set rate. This may be due to insufficient nutrients from the maternal or environmental stressors such as high temperature and high humidity, which hinder the continued development of the pod. Therefore, taking timely measures to support flower and fruit retention on the maternal plants within the first 15 days after pollination is crucial for improving the fruit set rate of G. sinensis.

Combining ability is an important indicator to measure the quality of parents and directly affects the performance of hybrid offspring32,33. General combining ability (GCA) is mainly affected by the cumulative genetic effect of the parents and can be inherited stably, while specific combining ability (SCA) is the relative performance of a cross that is associated with non-additive gene action, predominantly contributed by dominance, epistasis, or genotype-environment interaction effects34,35. In this experiment, the GCA of the maternal showed significant differences in pod length, width, and thickness, while the GCA of the paternal showed significant differences primarily in pod thickness. This indicates that pod length and width are mainly influenced by the additive effects of the maternal, while the paternal has some potential for improving pod thickness, especially as the GCA of G. delavayi and G. feraexhibited high positive effects on pod thickness. SCA provides an assessment of the non-additive effects of parental combinations ability, aiding breeders in selecting and utilizing the best parent combinations to enhance target traits and breeding efficiency36. In this experiment, the non-additive effects of the parents significantly influenced pod length, width, and thickness. However, the combinations with high SCA values did not always involve parents with high GCA, such as the combination of G. sinensis-2 and G. sinensis-T, which had low GCA for pod length and width but high SCA. This suggests that the GCA values of parents are not sufficient to predict the SCA values of hybrid combinations, which is consistent with previous research results37,38. Therefore, in the hybridization practice of G. sinensis, priority should be given to selecting maternal with high GCA to achieve greater advantages in pod length, width, and thickness. Additionally, paternal should be selected based on the specific combining ability of the combination, especially those that show significant positive effects on specific traits such as pod thickness, like G. delavayi and G. fera.

Heritability, which reflects the proportion of phenotypic variance attributable to genetic variance, is categorized into broad-sense heritability and narrow-sense heritability. Broad-sense heritability refers to the ratio of genotypic variance to phenotypic variance, while narrow-sense heritability refers to the ratio of additive genetic variance to phenotypic variance39,40. In our experiment, the factors affecting pod length and width were predominantly contributed by the maternal parent, with both broad-sense and narrow-sense heritability for these traits being high. This indicates that additive effects are greater than non-additive effects, and these traits are mainly controlled by genetics and have little environmental influence. Conversely, the broad-sense and narrow-sense heritability for pod thickness were relatively low and showed a larger difference, indicating that thickness is more influenced by non-additive genetic effects and environmental factors. Additionally, phenotypic correlation analysis revealed a high correlation between pod length and width (Supplementary Table 3). This indicating that pod length and width may be controlled by a common set of polygenes or closely linked gene clusters that exhibit positive coordinated effects in phenotypic expression. In contrast, the genes controlling thickness may be independent of those controlling length and width, or thickness may respond independently to specific environmental conditions. Therefore, in the selection of parental combinations, it is crucial to choose parents with strong heritability for the primary breeding traits and consider the correlations between traits. Otherwise, even if the best strains are used as parents, they may not produce the ideal traits41,42. The level of breeding value provides us with the possibility of selecting individuals with the best potential for genetic contribution, and more accurately predicts and selects excellent genetic combinations43,44. In this experiment, the estimated breeding values also indicated that pod phenotypic traits were mainly influenced by the maternal parent, while the paternal parent had a certain impact on pod thickness. This may be due to the influence of the paternal on seed growth, which leads to differences in the thickness of the pods. The significant differences in seed size observed in subsequent observations support this hypothesis, but further research is needed to understand the specific reasons behind this phenomenon. Moreover, in interspecific hybridization, the combination of G. fera and G. sinensis exhibited positive effects on length, width, and thickness traits, aligning with the combining ability results for pod phenotypic traits. Therefore, G. fera, as an interspecific hybrid parent of G. sinensis, can not only improve the phenotypic traits of the pods but also provide the possibility of greater genetic diversity and potential.

Conclusions

By analyzing the hybridization experiments of Gleditsia, the results revealed that the dry storage conditions at −80℃ are suitable for most Gleditsia pollen, while the G. fera pollen is more suitable to be stored at −20℃. G. sinensis exhibits significant hybrid potential in interspecific hybridization, particularly when hybridized with G. fera, it not only exhibited a high fruit set rate but also showed positive effects on pod phenotypes, making it a superior interspecific hybrid parent. Additionally, the length and width of the hybrid pods are mainly influenced by the female parent and have high heritability and correlation. The heritability of pod thickness is relatively low and significantly affected by non-additive genetic effects and environmental factors. Therefore, in G. sinensis hybridization practices, selecting superior female parents can yield greater advantages in pod length, width, and thickness. The choice of male parents should consider the combining ability, and interspecific hybridization with other species’ pollen can enhance pod phenotypic traits while providing greater genetic diversity and potential. However, this study utilized only a limited number of Gleditsia species for hybridization experiments, and the genetic analysis primarily focused on the phenotypic traits of pod size, width, and thickness. Future studies should integrate molecular marker techniques to conduct a more in-depth analysis of the genetic background of hybrid offspring and explore the hybridization potential among a broader range of species. This would provide valuable insights for further research on intra- and interspecific hybridization in Gleditsia.