Fruiting character variability in wild individuals of Malania oleifera, a highly valued endemic species

Weight and dimensions of fruit and stone

The mean weight of a fruit from a particular tree ranged from 21.25 ± 4.26 to 58.26 ± 10.44 g, with the weight of the heaviest mean fruit weight being 2.74 times that of the lightest. Similarly, the mean stone weight ranged from 8.99 ± 2.35 to 20.32 ± 3.14 g, with a 2.26 times difference between the heaviest and lightest stones (Table 2). There were significant differences (p < 0.001) in the mean weights of fruits and stones among individuals, whether in Guannan or Funing (see Supplementary Information Figure S1, Table S1 and Table S4), and there was also a significant difference (p < 0.05) in the mean fruit weight of individuals from Guangnan and Funing, but mean stone weight was not found to be significantly different (p < 0.05) between the two sites (Table 2). The fruit weight, stone weight and outer pericarp thickness showed higher CVs both in Guangnan (15.59%, 11.82% and 13.92%, respectively) and Funing (25.27%, 19.47% and 21.25%, respectively) than the other traits (Table 2). Lv et al.²⁰ reported a mean fruit weight of 21.30 ± 4.25 g and mean stone weight of 9.35 ± 1.68 g from a mixed sample of five M. oleifera individuals from Leye county, Guangxi. Guo et al.¹⁹ collected the fruits from various sites in Guangnan county, Yunnan, and mean fruit weights of individuals from each site ranged from 35.77 ± 7.58 to 47.29 ± 9.55 g, and mean stone weights from 14.03 ± 2.82 to 18.77 ± 3.26 g.

The mean transverse and longitudinal diameters of fruit were from 32.59 ± 3.91 to 48.95 ± 3.04 mm and 31.57 ± 3.83 to 44.99 ± 3.22 mm, respectively. The transverse and longitudinal diameters of the stone ranged from 25.46 ± 3.85 to 35.36 ± 2.66 mm and 22.12 ± 4.00 to 32.91 ± 1.94 mm, respectively. The thickness of the outer pericarp ranged from 2.73 ± 0.58 to 6.79 ± 1.07 mm (Table 2, Fig. 1). The differences in the transverse and longitudinal diameters of both fruits and stones among individuals were significant (p < 0.001) at both Guangnan and Funing (see Supplementary Information Figure S2, Table S2, Table S3, Table S5, Table S6). None of the fruit and stone dimensions were not significantly different (p > 0.05) between Guangnan and Funing, except for the stone transverse diameter (p < 0.05). The CVs in Funing for the traits FTD (9.77%), FLD (9.50%), STD (7.68%), SLD (10.60%), and OPT (21.25%) were higher than those in Guangnan (5.36%, 6.35%, 4.08%, 6.14%, and 13.92%, respectively) (Table 2).

The ranges of fruit transverse and longitudinal diameters have been previously reported to be from 39.31 ± 2.77 to 45.03 ± 3.47 mm and 35.19 ± 2.25 to 40.36 ± 3.49 mm, respectively, and the stone transverse and longitudinal diameters have been reported to range from 28.61 ± 1.55 to 34.16 ± 2.21 mm and 26.12 ± 1.80 to 31.69 ± 2.33 mm, respectively, mixed tree samples^19,20. In these studies, the fruit characters were measured according to mixed sampling from some trees, but the differences among individual trees was under-examined. Characteristics analysis of individual trees is more conducive to the selection of excellent traits for domestication. The assessment of a tree species’ phenotypic variation is a key starting point in any domestication program²⁶. Some reports indicated plant species usually exhibit distinct difference in fruiting characteristics both within and amongst populations^27,28,29. The variations could possibly be related to either genetic or environmental factors²⁶. To a large extent, the flowering and fruiting of trees in their natural habitats are dominated by the environment³⁰. However, the comparison of trait selections in the same environment allowed better evaluation of genotype influence on plant characteristics³¹. Our study suggests that morphological characters were more different among different individual trees than between different geographic localities, even though these had different climates. This means that the different genotypes of the individuals we tested may have largely contributed to the differences in phenotypes between individuals. Consequently, our analysis of morphological characters from individual trees, has allowed us to screen for some excellent individuals with the leading fruiting traits (heavier weight and larger size for fruits and stones). Individuals revealed by this study to have significantly better fruiting traits than the others include GN11 and FN18 (see Supplementary Information Figure S1, Figure S2, Table S1–S6).

Seed oil content and fatty acid composition

The ANOVA revealed significant differences (p < 0.001) in seed oil content among individuals at both in Guangnan and Funing (see Supplementary Information Table S7, Table S8). The percentage of seed oil renged from 48.31 ± 0.47% to 67.93 ± 1.07% (Table 2). The CVs were 5.92% and 4.57% in Guangnan and Fujing, respectively. It is worth noting that the seed oil content in individuals was significantly different (p < 0.01) between Guangnan (58.15 ± 3.44%) and Funing (64.16 ± 2.93%). Previous studies have suggested that seed oil content in M. oleifera was about 60.25% ± 2.5%¹⁶, 64.5%¹⁵, 51.3% and 52.7%²¹, and 51.7% and 56.8%²². However, these previous studies used a mixed seed sample from trees and did not investigate the differences in seed oil concentration among individual trees or between sites with divergent climates.

Eleven compounds were identified from the analysis of the fatty acids in the seed oil. Together, these eleven compounds contributed between 91.39 to 96.34% of the total lipids. The most abundant fatty acid was nervonic acid (38.93–47.24%), and the other major fatty acids were octadecenoic acid (26.79–32.08%), docosenoic acid (10.94–17.24%), dodecenoic acid (1.98–3.11%), tetracosanoic acid (1.39–2.61%), octadecadienoic acid (0.66–1.79%), docosanoic acid (0.90–1.37%), hexadecanoic acid (0.50–0.77%), octadecanoic acid (0.31–0.58%), arachidic acid (0.22–0.32%), and hexadecenoic acid (0.06–0.15%) (Table 3). Trees in Funing had significantly higher nervonic acid content in their seed oils than did those at Guangnan (p < 0.01), but the docosenoic acid content in the seed oil was significantly higher in trees at Guangnan (p < 0.01) than those at Funing, and the concentration of octadecenoic acid in the seed oil did not differ significantly between the two sites (p > 0.05) (Table 3). That these 11 fatty acids are the most prevalent in M. oleifera seed is largely consistent with the findings of other studies. Tang et al.¹⁶ identified ten of eleven fatty acids above in seed oil, but did not find hexadecenoic acid. The largest fraction was represented by nervonic acid (55.70%), followed by octadecenoic acid (23.81%), docosenoic acid (13.13%), tetracosanoic acid (2.65%) and dodecenoic acid (1.28%). Peng et al.³² identified the same ten fatty acids as Tang et al.¹⁶, with the major components being nervonic acid (33.17 ± 2.75%), octadecenoic acid (27.82 ± 2.04%), docosenoic acid (12.90 ± 0.96%), arachidic acid (3.91 ± 1.28%), dodecenoic acid (3.04 ± 0.70%), and tetracosanoic acid (2.78 ± 0.37%). Cao et al.²³ reported ten fatty acids in M. oleifera seed oil, but in this case the analysis identified linolenic acid (0.55%), but neither octadecanoic acid nor arachidic acid, and the fatty acid at the highest concentration was octadecenoic acid (37.88%) rather than nervonic acid (32.15%). However, across all of these studied, the three fatty acids present in the largest amounts were nervonic acid, octadecenoic acid and docosenoic acid, which together made up more than 73% (to as much as 92%) of the total lipids. Ou¹⁵ reported that the nervonic acid content alone could reach up to 67% of the total lipids.

Of all the fatty acids, nervonic acid had the lowest CVs (3.95% and 3.09% at Guangnan and Funing , respectively), followed by octadecenoic acid (4.41% and 4.37%), and docosenoic acid (7.57% and 8.10%), docosanoic acid (6.50% and 10.48%), arachidic acid (6.90% and 11.54%), hexadecanoic acid (9.09% and 9.68%), dodecenoic acid (9.20% and 11.20%), tetracosanoic acid (16.24% and 13.15%), octadecanoic acid (14.63% and 15.56%), hexadecenoic acid (18.18% and 15.38%), and octadecadienoic acid (15.04% and 19.47%) (Table 3). Peng et al.³² reported the following CVs from fatty acids in M. oleifera seed oil: octadecenoic acid (7.32%), docosenoic acid (7.44%), nervonic acid (8.30%), docosanoic acid (12.07%), tetracosanoic acid (13.14%), hexadecanoic acid (20.66%), dodecenoic acid (22.95%), arachidic acid (32.68%), octadecadienoic acid (43.34%), and octadecanoic acid (82.79%). Taken together with our results, this suggests that the content of some fatty acids, including nervonic acid, octadecenoic acid, docosenoic acid, is relatively stable in M. oleifera seed oil.

The lipid biosynthesis is influenced by environmental factors such as temperature and rainfall^33,34. Many studies indicated that temperature was negatively correlated with seed oil concentration^35,36,37. Qiao et al.³⁸ reported that seed oil content of Acer truncatum showed significant negative correlations with both annual average temperature and annual rainfall. Nevertheless, some studies showed that seed oil content increased with the optimum temperature, above which the oil concentration declined^39,40,41. We found that seed oil content was significantly higher in the samples from Funing than those from Guangnan (Table 2). This showed higher temperature could improve seed oil content of M. oleifera. Composition of fatty acid varied with temperature and rainfall. Temperature plays a leading role in modifying oilseed rape quantity and the balance between saturated, mono-unsaturated and polyunsaturated fatty acids⁴². The saturated fatty acids such as octadecanoic acid, arachidic acid, docosanoic acid, and tetracosanoic acid, were higher in warm climates, while mono-unsaturated fatty acids did not demonstrate a uniform response to temperature in Camelina sativa, as demonstrated by increasing octadecenoic acid and docosenoic acid with temperature, and decreasing hexadecenoic acid and dodecenoic acid with temperature³³. Octadecenoic acid was negatively correlated, while hexadecanoic acid and Octadecadienoic acid were positively correlated with the mean daily temperature at the time of intense growth and ripening of the olive fruits³⁴. Other reports showed that nervonic acid content was not influenced by temperature in Acer truncatum and Camelina sativa^33,38,43. However, precipitation exhibited a significant positive correlation with nervonic acid in Acer truncatum³⁸. It was also found that weather conditions deviating from the long-term average, both in terms of temperature and precipitation, did not affect the quantitative profile of fatty acids⁴⁴. Our studies showed nervonic acid content was significantly higher in M. oleifera in Funing, while the other main fatty acids (content > 1%) was significantly lower except for octadecenoic acid and tetracosanoic acid (Table 3). The results suggest that both seed oil production and quality may be improved by planting M. oleifera in warmer areas.

Correlations between fruit traits

There were significant positive correlations between every two traits (fruit weight, stone weight, fruit transverse diameter, fruit longitudinal diameter, stone transverse diameter, and stone longitudinal diameter) both at Guangnan or Funing (Fig. 2). The thickness of the outer pericarp was positively and significantly correlated with both the weight and dimensions of fruits at both of the sites, however, the thickness of the outer pericarp was only positively and significantly correlated with the weight and transverse diameter of the stone at Guangnan, having no correlation with stone characters at Funing (Fig. 2). Lv et al.²⁰ reported that the proportion of the outer pericarp weight increased with increased fruit weight. This conclusion corresponds with our findings that the thickness of the outer pericarp correlates more with the weight and dimensions of the fruit than with those of the stone. Seed oil content did not correlate with the weights or dimensions of either the fruit or the stone. This demonstrates that the morphological characters of the fruit and stone do not have a great effect on the seed oil content. There was usually an extremely significant correlation between the phenotypic fruiting traits in drupe plants, such as fruit weight and dimension, stone weight and dimension, and pulp thickness^26,45. Therefore, when selecting plants with elite traits, we should mainly focus on some representative indicators. For the selection of high-quality individuals of M. oleifera, large fruit, large stone and high oil content are indicators of elite traits.

The amounts of nervonic acid and tetracosanoic acid were significantly and negatively correlated with the major fatty acids (octadecenoic acid, dodecenoic acid and docosenoic acid), but there was a significant and positive correlation between the content of nervonic acid and tetracosanoic acid in M. oleifera seed oil both at Guangnan and Funing (Fig. 3). Docosenoic acid had a significant and positive correlation with octadecenoic acid and dodecenoic acid at both sites. The correlation between octadecenoic acid and dodecenoic acid was not significant at Guangnan, but it was both significant and positive at Funing (Fig. 3). It was worth noting that the correlation between nervonic acid and the other major fatty acids was very similar in the samples from both Guangnan and Funing. The biosynthetic pathway of nervonic acid includes de novo fatty acid synthesis in plant plastids and fatty acid elongation starting from C18:1, and nervonic acid as a long-chain fatty acids is synthesized in the form of acyl-CoAs by the fatty acid elongation enzyme complex, C20:1 and C22:1 are the precursors in the synthesis of nervonic acid⁴⁶. This may be the reason why nervonic acid is highly and negatively correlated with these major fatty acids.

With rigorous investigation and research, some endemic and endangered plant species are subsequently discovered and re-understood, some of which possess great utilization value in medicine, food, and as ornamental, etc^47,48. Due to the vulnerability of endemic species in environmental adaptation, it is not easy to propagate and cultivate these species⁴⁹, restricting the development and utilization of these species. Usually, some wild plant resources are collected in large quantities for direct use or economic purposes, which could accelerate the endangerment and extinction of these species^50,51. By selecting good genotype suitable for domestication and using appropriate cultivation techniques, there are good prospects in using these plants and serving the local economy, and preserving wild resources.

Source: Ecology - nature.com