Discussion
Fig wasps are a group of special insects that live in the enclosed syconia of fig trees. In the long evolutionary history, they have maintained a close ecological relationship with the fig fruits. Related to the shelter and relatively safe living environment provided by fig fruits, the evolution of the fig wasps not only shows an adaption to the environment within fig syconia, their adaption to the large-scale geo-climatic environment may also be closely related to the life in the fig syconia. This study focused on TEs in the genomes and tried to explore the evolutionary traces of the fig wasps in response to the changes of the microenvironment of fig syconia and the changes in the geo-climatic macroenvironment.
We first compared the differences in TEs between six pollinator and five non-pollinator species. The results displayed that in non-pollinators, not only the TE types were significantly more than those of pollinators, the total lengths and genome contents of TEs were also significantly higher than those of pollinators. This is inseparable from the relatively complicated lifestyles and living environment of the non-pollinators. Compared with the pollinators, the non-pollinators can usually live on several host fig species, with various parasitic strategies and complex feeding habits, which requires a stronger ability to detect the hosts and foods (Borges 2015). Besides, their longer time living at the external surface of syconia for mating and oviposition is fraught with the risks of predation by ants and diseases infection (Ranganathan, Ghara et al. 2010), and thus also requires a stronger ability to avoid natural enemies and resist diseases. It has been reported that TEs can become active in response to stress (Horvath, Merenciano et al. 2017), and the active state is the prerequisite for TE bursts (Belyayev 2014). In our research results, taking the TEs of the Gypsy family as an example, their expansion is widespread in fig wasp species, so the contents of TEs in the genomes of both pollinators and non-pollinators have increased significantly. However, the patterns of burst of the Gypsy in both groups are quite different, with the degree of expansion higher in non-pollinators, so there is a significant difference in the TE contents in the genomes of both groups. In addition, the more TE contents in non-pollinators may also be because that winged male of non-pollinators can fly out of the fig fruits where they were born in and thus mate outside, which will reduce the probability of inbreeding (Greeff, Jansen van Vuuren et al. 2009, Cook, Reuter et al. 2017) and provide more abundant resources for the increase of TE types; this is also consistent with the results that all non-pollinators had their own unique TE families. On the contrary, in pollinator species, the stricter host specificity, single and sufficient food sources, stable and relatively hidden habitats, and the higher degree of inbreeding may be associated with insufficient motivation for expansion of TEs and thus the lack of TEs in the genomes (Machado, Robbins et al. 2005, Greeff, Jansen van Vuuren et al. 2009, Cruaud, Ronsted et al. 2012). This is not only reflected in the fact that many pollinators did not have species-specific TEs, but also that some TE families were absent in some or all of the pollinator species. For example, hAT-hATm and insect-specific Crypton-I only existed in the genomes of the non-pollinator species.
In order to more clearly reflect the differences in the composition of TEs in evolutionary history, the insertion time distribution of TEs were drawn. Comparing the burst patterns of both groups of fig pollinators and non-pollinators, we found that the burst peaks of non-pollinators were more complex. For example, the burst peaks of the pollinator onF.benjaminapresented a single concentrated “n” shape, while the burst peaks of three non-pollinators on the same host fig species presented a complex and extensive “m” shapes. This implied that both groups of fig wasps experienced different environment changes in the evolutionary history or responded differently to the same environmental changes. It has been reported that the emergence of fig wasps is distinctly seasonal. The pollinators can leave more offsprings in the rainy season, and non-pollinators produce more offsprings in the dry season (Wang, Yang et al. 2005, Dong, Peng et al. 2020). On the contrary, there is a strong competition between non-pollinators and pollinators, and the numbers of both groups show a significant negative correlation. Therefore, non-pollinators have a higher reproduction rate in the hot and rainy season (Weiblen 2002, Zhang and Yang 2009). This suggests that both groups of fig wasps have different response to the same environmental changes (Wang, Compton et al. 2015).
Our study detected that almost all the fig wasps had significant TE burst events in the time range of 32-34 Mya. This period coincided with the special Eocene-Oligocene (E/O) transisiton period of ~33.7 Mya’s, which was considered to have been an important stage of the global climate transition in geological history. Due to the action of the Oi-1glaciation, the global temperature dropped significantly. With the Antarctic ice sheet growing rapidly, the climate became colder and drier, causing different degrees of extinction of organisms and the decline of broad leaf forests (Zachos, Pagani et al. 2001, DeConto and Pollard 2003). The major burst peaks in non-pollinators and the minor burst peaks in pollinators in this period indicated that both groups were all sensitive to the climate during this period, but the non-pollinators retained more TEs than the pollinators. These ice sheets persisted until the late Oligocene (26 to 27 Mya), when a warming current reduced the extent of Antarctic ice. This warm climate lasted until the mid-Miocene (~15 Mya), and reached its peak in the late mid-Miocene climate optimum (17 to 15 Mya). Thereafter, the East Antarctic Ice Sheet(EAIS)began to expand ~ 14 million years ago, and the global climate underwent a major reorganization (Groeneveld, Henderiks et al. 2017). The major burst peaks of pollinators were located in the time range of 13-28 Mya, which coincided with the above-mentioned high temperature phase, and this implied that pollinators might be more sensitive to high temperature climate changes. The early Pliocene was marked by a subtle warming trend until the onset of Northern Hemisohere Glaciation (~3.2 Mya), the gradual formation of Northern Hemisphere Ice Sheet, and present of bipolar state of the global ice sheet (Tiedemann, Sarnthein et al. 1994, Zachos, Pagani et al. 2001). During this period, the obvious recent TE bursts in non-pollinators (0-2 Mya) further demonstrated the sensitivity of this group to low-temperature climates. To figure out whether the temporal overlap between the TEs bursts of the fig wasps and the geological climate changes is widespread, we also investigated the TE burst events of five species that are not related to fig trees but are widespread and live in a relatively open environment, and found that the major burst peaks of these species were distributed in 0-3 Mya. These comparable results indicate that the fig wasps that are confined to the fig trees in tropical and subtropical regions have retained more traces of TE bursts related to geo-climate changes than the free-living species that are more widespread.
In order to further clarify the relationship between the TE bursts of fig wasps and the geo-climate changes, for all the 11 species of fig wasps, we analyzed the function of genes near TEs that were inserted in the four periods representing the growth or decay of the major continental ice sheet of the Cenozoic. The results showed that the functions of these genes were mainly involved in the pathways such as rhythm, signal transduction, and neural activity. Among them, the Circadian entrainment pathway (map 04713) was enriched in many species in all periods. Especially at 31-35 Mya, when the common burst peaks occurred, this pathway was significantly enriched in up to nine fig wasps. Circadian entrainment is the process to activate the internal biological clock of the organism by recurring exogenous signals, so that the organism’s endocrine and behavioral rhythms will be synchronized to environmental cues (Johnson, Elliott et al. 2003). There are interesting and open questions whether the periodic climate changes will have a direct impact on the fig wasps, or they will exert selective pressure on the fig wasps through their influence on the fig trees. Generally, climate changes have great impacts on plants because they will change the flowering cycle (Ehrlen and Valdes 2020). The year-round phenological variations of fig trees living in the tropics are generally limited, with fig production featuring intra-individual flowering synchrony and inter-individual asynchrony throughout the year (Bronstein, Gouyon et al. 1990). However, for the higher latitude regions and the edges of the species’ distribution range, the synchronization of flowering synchrony, and the productions of wasps and seeds production display seasonal fluctuations, which are accompanied by a reduction in reproductive success (Chen, Zhang et al. 2018). Therefore, considering that both parts of the symbiosis system of the fig and fig wasps have different responses to the geographic climate differences, it is not difficult to speculate that, in the evolutionary history of the symbiosis system, the changes in the flowering periods induced by the rapid paleoclimate changes would exert great evolutionary pressure on the fig wasps, because they need to keep pace with the flowering cycle of the fig trees. In this adaptive process, the adaptation of the Circadian entrainment genes may play a pivotal role. In other words, periodic climatic changes can have a huge impact on plants, which in turn can affect the evolution of symbiotic organisms closely related to plants.
Finally, by analyzing the correlation between the distributions of TEs and CRMs in the genomes of the fig wasps, we speculated a potential molecular mechanism for the adaptive evolution of these fig wasps through the action of TEs. There are three main ways for TEs to regulate host genes. One is that newly inserted or deleted TEs may change the regulation or coding environment at a specific location, thereby affecting gene expression. For example, the adaptive “industrial melanism” phenotype ofBiston betularia has been evolved through the intronic insertion of a DNA transposon, increasing transcript abundance of the cortex gene during early wing development (van’t Hof, Campagne et al. 2016). Second, TE insertions might also lead to modifications at the epigenetic level, which can affect the expression of neighbouring genes. For example, inArabidopsis , gene expressions are negatively correlated with the density of methylated TEs (Hollister and Gaut 2009). Third and the last, TEs could be evolved into regulatory sequences which might affect the downstream gene expression (Oliver and Greene 2011). In the view of the positive role and universality of TEs as regulatory sequences, we analyzed the relationship between TEs and CRMs based on homology prediction results, and we found that many TEs in the genomes may have evolved into functional CRM sequences. The genes near those CRM sequences that overlapped with TEs were mainly the genes enriched in pathways related to environmental information processing such as Circadian entrainment. This suggested that in these fig wasp species, some TEs might have evolved into CRMs to affect nearby gene expression to achieve gene environmental adaptability.