In the present study, twenty-three NRs were identified in P. pseudoannulata and grouped into six subfamilies based on their phylogenetic analysis with no members in NR0 subfamily. It was the first complete identification and characterization of NR family in spiders. The number of NRs in P. pseudoannulata was close to that in insects with 19–22 members (Adams et al. 2000; Bertrand et al. 2004; Bonneton et al. 2008; Cheng et al. 2008; Christiaens et al. 2010; Cruz et al. 2009; King-Jones and Thummel 2005; Tan and Palli 2008; Velarde et al. 2006; Xu et al. 2017). Interestingly, each NR had single copy in all investigated insects, except for duplicates of tailless (Tll) in A. gambiae (Bertrand et al. 2004) (Table S4), whereas in P. pseudoannulata, six NRs had two duplicates each, including E78, HR3, USP, HR51, SVP, and HR4. NR duplication was generally occurred in arachnids (Table S4). For example, T. urticae had 8 hormone receptor-like in 96 (HR96s), 2 USP/RXRs, and 2 hormone receptor-like in 38 (HR38s) (Grbic et al. 2011). Both A. americanum (Guo et al. 1998)d citri (Li et al. 2017) had 2 USP/RXRs. Therefore, it raised concern about the functional differentiation between NR duplicates in arachnids, such as USP/RXRs. A. americanum was the first arachnid with focus on functional investigations of two USP/RXRs (Guo et al. 1998). Although the transcriptional level of EcR and two USP/RXRs were correlated with ecdysteroid titer in the developmental stages, the electrophoretic gel mobility shift assay showed that EcR/USP/RXR-1 but not EcR/USP/RXR-2 exhibited broad DNA binding specificity (Palmer et al. 2002). In P. citri, EcR had the similar temporal expression patterns to USP/RXR-2 but different from USP/RXR-1, and differential expression genes in deutonymphs indicated that EcR/USP/RXR-2 and USP/RXR-1 might regulate different physiological processes to control the mite moulting (Li et al. 2022). An EcR and an USP/RXR were cloned from L. australasiae, and the ligand-binding assay showed that EcR had high binding ability to PA while the USP/RXR did not enhance the binding ability (Nakagawa et al. 2007). USP/RXR is required for EcR binding to ecdysteroids (Thomas et al. 1993; Yao et al. 1993), so whether another USP/RXR in L. australasiae playing this role remains to be further explored. As we found here in the present study that two P. pseudoannulata USPs had differential functions in the development and reproduction, the duplication of NRs gave a more comprehensive network for the gene expression regulation in arachnids.
We thoroughly quantified the spatiotemporal expression of EcR and two USPs in P. pseudoannulata. EcR and two USPs were in accordance with the expression pattern of Halloween genes with high expression in the end of the 2nd instar spiderlings followed by a rapid dropdown once moulting was completed (Yang et al. 2021), which indicated they might involve in the moulting process. EcR and two USPs exhibited different expression patterns in eleven developmental stages, with EcR and USP-2 being significantly expressed in spiderlings while USP-1 being significantly expressed in adult females. Besides, EcR and two USPs were highly expressed in six tissues from adults, except for USP-2 that was mostly undetectable in fat body, ovary, and testis. These results strongly indicated the functional differentiation of two USPs in P. pseudoannulata. Therefore, RNAi against two USPs showed that USP-1 knockdown remarkably delayed spiderlings’ moulting and increased the number of invalid eggsacs in females, while the development and reproduction of P. pseudoannulata were not affected by dsUSP-2 treatment, except for the reduced egg numbers. In addition, the phenotypes in spiderlings and females treated with dsCYP307A1 were similar to the dsEcR and dsUSP-1 treatments. Meanwhile, EcR and USP-1 responded to the changes of ecdysteroid with the upregulation by PA application and downregulation by CYP307A1 silencing. Therefore, we suggested that EcR/USP-1 mediated ecdysteroid signaling to regulate the development and reproduction in P. pseudoannulata.
An unusual case that the female ate her invalid eggsac occurred in dsEcR, dsUSP-1, and dsCYP307A1 treatments, in which EcR/USP-1-mediaed ecdysteroid signaling was suppressed, to stop the useless maternal care. The wolf spiders, including P. pseudoannulata, express parental care to their offsprings by carrying their eggsacs and juvenile spiderlings (Ruhland et al. 2016b). Spider mothers termly detected their juveniles’ statuses and opened the eggsac at a certain time to release the juveniles, the exact timing of eggsac opening partially depends on stimuli from juvenile movements in eggsac (Ruhland et al. 2019; Viera et al. 2007). In this study, the ecdysteroid signaling-suppressed P. pseudoannulata mothers might not receive the signals from the eggsac full of nonviable eggs. Carrying an invalid eggsac did not give any benefit for population growth of P. pseudoannulata. Alternately, they selected to eat the invalid eggsacs to terminate the useless reproduction cycle and provide chances for the generating of a new eggsac in advance. This selection of eating invalid eggsac shortened the useless reproduction cycle, which partially compensated for the reproduction loss due to the suppression of EcR/USP-1-mediaed ecdysteroid signaling from the point of view of time. Eating invalid eggsac also partially recovered the energy costs because maternal care was costly for females, even to invalid eggsacs (Ruhland et al. 2016b). In fields, some pesticides may disrupt the EcR/USP-mediaed ecdysteroid signaling and lead to the generating of the nonviable eggs, such as ecdysteroid analogues (Borchert et al. 2005; Zhang et al. 2021).
In summary, the suppression of ecdysteroid signaling severely constrained the population growth of P. pseudoannulata by generating the invalid eggsacs. The wolf spider forwardly selected to eat the nonviable eggs as a compensatory strategy. The spider adopted this compensatory strategy to reduce costs in population growth and consumption from carrying invalid eggsacs.