Pollen tube development in two species of Trithuria (Hydatellaceae) with contrasting breeding systems

Abstract

Trithuria (Hydatellaceae; Nymphaeales) is unique among early-divergent angiosperms in that its species are extremely small and most have exceptionally short, annual life histories. Given the evolution of these extremes of size and development, we sought to understand whether post-pollination processes still varied predictably with breeding system in Trithuria. To address this question, we studied two Western Australian species, Trithuria austinensis (dioecious, obligately outcrossing) and Trithuria submersa (bisexual, highly selfing). To document developmental timing, carpels were hand-pollinated, collected at sequential time points, and examined with light and fluorescence microscopy. In both species, pollen tubes first entered ovules<1 h after pollination, but the pollen tube pathway of outcrossing T. austinensis was almost four times longer and its pollen tube growth rates were up to six times faster (≤2,166 vs. 321 μm/h) than those of T. submersa. T. austinensis also exhibited greater male investment, slower pollen germination, and greater pollen tube attrition. These differences in male gametophyte development are predicted for outcrossers versus selfers in phylogenetically derived angiosperms. These new data for Hydatellaceae reinforce the idea that an acceleration of pollen tube development occurred in the Nymphaeales stem lineage, before the origin of Hydatellaceae. We infer that a recent evolutionary transition to selfing in T. submersa has been accompanied by predictable modifications to reproductive development, which, because of the ancient relationship between Hydatellaceae and all other angiosperms, suggests that traits underlying the lability of flowering plant post-pollination biology were present early in their history.

Ancestral traits and specializations in the flowers of the basal grade of living angiosperms

“…It has been suggested that these structures may represent a primitive, pre floral or "nonfloral" state (Rudall & al., 2009), but parsimony optimization implies instead that they are derived. The female flowers have several long, uniseriate stigmatic hairs on top of the gynoecium, whose large size is disproportionate in view of the minute flowers; pollen tubes grow within the cell walls of these hairs (Prychid & al., 2011;Taylor & Williams, 2012).…”

Divergence in pollen performance between Clarkia sister species with contrasting mating systems supports predictions of sexual selection

“…; Alvarez ). The analogous prediction in plants—that highly outcrossing (i.e., relatively “promiscuous”) taxa will evolve faster‐growing male gametophytes than closely related self‐fertilizing taxa—has been reported in a few taxa (Smith‐Huerta ; Diaz and Macnair ; Kerwin and Smith‐Huerta ; Taylor and Williams ; Hove and Mazer ), but the interpretation of these studies is somewhat problematic because several factors other than mating history may influence pollen performance and must be accounted for when comparing taxa.…”

“…In addition, our estimate of mean PTGR was based on the mean of the distances reached by every pollen tube visible in each style rather than on the distance achieved by the fastest‐growing pollen tube (cf. Taylor and Williams ), on the distance reached by a majority of pollen tubes (cf. Diaz and Macnair ), or on the number of tubes reaching a given distance within a given time period (cf.…”

“…; Shimizu and Tsuchimatsu ; Schneider and Mazer ). Few studies have investigated the effects of mating system on pollen performance (Smith‐Huerta ; Kerwin and Smith‐Huerta ; Taylor and Williams ; Hove and Mazer ), but it has been proposed that intrasexual selection should cause traits that affect the competitive ability of male gametophytes to diverge between outcrossing versus regularly self‐pollinating taxa (Mazer et al. ).…”

“…To date, no studies have compared pollen performance between species with contrasting mating systems in a uniform environment while controlling for variation among populations and for the effects of pollen load, SPR, temperature, and pollen type (self vs. outcross); here, we examined all of these sources of variation. In addition, unlike previous studies (Smith‐Huerta ; Diaz and Macnair ; Kerwin and Smith‐Huerta ; Taylor and Williams ; Hove and Mazer ), we tested for species differences in PTGR using two estimates of this trait. First, we estimated the mean PTGR based on the distance traveled by each pollen tube visible in each style and, second, we estimated PTGR from the length of the longest tube (LLT) in each style.…”