The coexistence of hermaphroditic and dioecious plants is associated with polyploidy and gender dimorphism in Dasiphora fruticosa

Abstract

Dasiphora fruticosa comprises male, female and hermaphrodite plants, which are distributed sympatrically in some populations on the Qinghai-Tibet Plateau. To explore what governs the coexistence of these three sexual phenotypes, we investigated the DNA contents, pollen and ovule production, pollen deposition, and performed hand-pollination in both hermaphroditic and dioecious individuals of D. fruticosa. Flow cytometry confirmed that the DNA content of males and females were almost twice as much as that of the hermaphrodites. Male and female flowers produced more pollen grains and ovules than hermaphroditic flowers. Hand-pollinated treatments showed that unisexual flowers were sterile in one sexual function and bisexual flowers were fertile for both functions, but no sterile seeds were produced between unisexual and bisexual flowers. Our findings imply that polyploidy is related to gender dimorphism, and both are likely to play a strong role in the coexistence of two cryptic biological species of D. fruticosa (low ploidy hermaphroditic species and high ploidy dioecious species) in the Qinghai-Tibet Plateau.

Keywords: Dasiphora fruticosa; Dioecy; Gender polymorphism; Polyploidy; Reproductive isolation.

Fig. 1

Three different sexual flowers (A: male flower, B: female flower, C: hermaphrodite flower), inflorescence part of plants (D: males, E: females, F: hermaphrodites), and habitat (G) of the Dasiphora fruticosa population in the Qinghai-Tibet Plateau, China.

Fig. 2

The DNA contents of combined and separate sexes of Dasiphora fruticosa. The average DNA contents of males (1.35 ± 0.026, N = 5) and females (1.50 ± 0.045, N = 22) was roughly twice that of the hermaphrodites (0.74 ± 0.019, N = 19) (χ² = 34.142, df = 2, P < 0.001). Different letters beside the bars indicate significant differences at P = 0.05 level. Error bars indicate SE.

Fig. 3

The pollen number (A) and ovule number (B) per flower of combined and separate sexes of Dasiphora fruticosa. There were more pollen grains in males (5.35 ± 0.39 × 10⁵, N = 32) than in hermaphrodites (3.85 ± 0.36 × 10⁵, N = 36) (F = 219.82, df = 1, P < 0.001). Female flowers (97.58 ± 4.79, N = 40) also produced substantially more ovules than hermaphrodites (78.5 ± 3.43, N = 36) (F = 120.80, df = 1, P < 0.001). Different letters beside the bars indicate significant differences at P = 0.05 level. Error bars indicate SE.

Fig. 4

(A) Percentages of pollen removal during the total male phase in the hermaphrodite (96.63 ± 0.36%, N = 99) and male (96.91 ± 0.25%, N = 100) flowers; there is no significant difference between female flowers and bisexual flowers (F = 0.277, df = 1, P = 0.599). (B) Mean number of pollen grains deposited on the stigmas in natural hermaphrodite flowers (64.78 ± 3.78, N = 100) and female flowers (9.78 ± 1.05, N = 101); the difference between the two sexual morphs was significant (F = 200.187, df = 1, P < 0.001). Different letters beside the bars indicate significant differences at P = 0.05 level. Error bars indicate SE.

Fig. 5

The mean seed numbers of control and hand-pollination treatments of hermaphrodites and females. (A) Hermaphrodite plants: Control (HC), Hand self-pollination (HSP), Hand cross-pollination (HCP), Hand cross-pollination with pollen grains from males (HCPM); (B) Female plants: Control (FC), Hand-pollination with pollen grains from hermaphrodites (FPH), Supplemental hand-pollination with pollen grains from males (FM). Different letters beside the bars indicate significant differences at P = 0.05 level. Error bars indicate SE.