Ancient pollination and multi-modal partner-encounter signals
Cycads have maintained obligate insect pollination since before the rise of flowering plants. While cycads were historically thought to be exclusively wind- pollinated, brood-site pollination mutualists have now been found in most cycad species, mostly involving Coleoptera (beetles and weevils), but sometimes Thysanoptera (thrips) or Lepidoptera (moths) (Tang, 2004; Terry et al ., 2012; Salzmanet al ., 2020). These brood-site pollinators feed, breed and develop on plant reproductive tissue - almost exclusively on pollen cones although one obligate ovule parasite may play a small role in pollination of at least one species (Donaldson, 1997). Pollinators must be induced to leave the pollen cone and visit an ovulate cone of a nearby conspecific to complete pollination. In both thrips and weevils, this occurs through a ‘push-pull’ pollination process whereby pollinators are attracted to lower quantities of cone VOCs and repelled by high VOC quantities (Terry et al ., 2007; Salzman et al ., 2020; Salzman et al ., 2021; Fig. 1a). Pollen and ovulate cones undergo a daily process of increased respiration and thermogenesis that is followed closely by an increase in cone VOCs (Terry et al ., 2016; reviewed in Salzman et al ., 2020). Ovulate cones mimic pollen cone scent and emission patterns but have much lower emission rates, making them attractive at peak VOC emission times when pollen cones become repellent (Fig. 1a). This has been hypothesized to be a form of “pollination by mistake” whereby pollinators are tricked into visiting the ovulate cone where they would not otherwise aggregate in large numbers (Tang 2004; Salzman et al., 2020). This push-pull pollination appears to be ancestral in the lineage (Salzmanet al., 2020) and fossil evidence dates Coleopteran-cycad brood site pollination to at least 185 Mya (Cai et al ., 2018), placing this pollination mechanism before the rise of flowering plants and as the oldest pollination mechanism yet documented (Salzman et al ., 2020). The likely antiquity of this pollination strategy makes cycad pollination a rich case-study on the mechanisms and maintenance of plant-insect mutualisms.
Cycads also provide an excellent system to investigate mechanisms of partner-encounter signaling with implications for understanding early insect pollination prior to the overt visual signaling of flowers. Thermogenesis produces many potential cues (temperature, CO2, humidity) that insects are innately capable of perceiving and likely preceded the evolution of visual or chemical signals. Indeed, cycad cone humidity has been shown to affect pollinator behavior (Terry et al ., 2014; Salzman et al ., 2023) and has recently been suggested to perform a signal-like function in pollination in general (Dahake et al ., 2022; Salzman et al ., 2023). As a byproduct of respiration, humidity cannot be under selective pressure, but morphological traits can and cycad cone morphology that amplifies or disseminates a humidity gradient matches pollinator humidity preferences in at least the two species that have been tested (Terry et al ., 2014; Salzmanet al ., 2023). It is becoming increasingly clear that studies of plant signaling should be extended beyond visual and chemical signals to include humidity (Arx et al., 2012; Terry et al., 2014; Dakahe et al., 2022; Salzman et al., 2023), temperature (Seymour & Matthews 2006; Terry et al., 2014) and CO2 (Goyret et al., 2008). Given the evolutionary history, pollination mutualisms, and variety of plant signals, research on cycad pollination mutualisms has the potential to provide new insight into the evolution of insect pollination through the lens of partner-encounter and multi-modal signaling.