Regulation of insulin-like peptide expression in adult Blattella germanica females

Insect Biochemistry and Molecular Biology 141 (2022) 103706 Available online 30 December 2021 0965-1748/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Regulation of insulin-like peptide expression in adult Blattella germanica females Claudia V. Domínguez, Viviana Pagone, Jos´ e L. Maestro * Institute of Evolutionary Biology (CSIC.Universitat Pompeu Fabra), Passeig Marítim de la Barceloneta 37-49, 08003, Barcelona, Spain ARTICLE INFO Keywords: Insulin-like peptide Juvenile hormone Insect Blattella germanica Nutritional signalling ABSTRACT The insulin-IGF-signalling (IIS) pathway regulates key processes in metazoans. The pathway is activated through the binding of the ligands, which in insects are usually referred to as insulin-like peptides (ILPs), to a class of receptor tyrosine kinases, the insect insulin receptor. To study the pathway regulation, it is therefore essential to understand how ILPs are produced and released. In this study we analysed the factors that regulate the expression of the seven ILPs (BgILPs) expressed in adult females of the German cockroach, Blattella germanica. The results showed that the starvation-induced expression reduction of brain BgILP3, 5 and 6 and fat body BgILP7 is not due to reduced juvenile hormone (JH) or decreased TOR pathway activity. In addition, depletion of FoxO in starved females did not correct the low levels of these BgILPs, but even reduced further BgILP5 expression, indicating the need to maintain certain basal levels of BgILP5 even during starvation. Furthermore, JH promoted increased BgILP5 and decreased BgILP3 expression in the brain, an effect that required Methoprene-tolerant (Met), the JH receptor, but not Krüppel homolog 1 (Kr-h1), the main JH transducer. On the other hand, JH inhibited the expression of BgILP7 in the fat body, although in this case, the action required both Met and Kr-h1. In addition, JH reduction treatments produced a decrease in the expression of the insulin receptor in the fat body, which suggests an increase in IIS. The results show a peculiar regulation of ILP expression in adult B. germanica females, which is clearly different than that seen in other species. This is understandable given that gene duplications in recent clades have resulted in different sets of ILP genes, involving substantial changes in gene regulatory networks. 1. Introduction For an organism to function correctly, the cells of the various tissues and organs must communicate with each other. In response to a stimulus, the endocrine system enables that information, in the form of a chemical messenger or hormone, to simultaneously reach the entire organism. Hormones can also act in a paracrine or autocrine manner, depending on how locally they act. In the above sense, one of the most important hormonal systems is the insulin-IGF-signalling (IIS) pathway, which is well conserved in metazoans, including insects and vertebrates. In insects, IIS regulates key processes, such as cellular proliferation, growth, longevity, metabolism, reproduction, and caste determination (Claeys et al., 2002; Wu and Brown, 2006). In the German cockroach, Blattella germanica, IIS is involved in regulating growth, reproduction and metabolism (Abrisqueta et al., 2014, 2017; Süren-Castillo et al., 2012, 2014). Thus, in B. germanica, RNAi-triggered depletion of one of the insulin receptors (InR Cluster I according to Smýkal et al., 2020), results in reduced nymphal growth (Abrisqueta et al., 2014). The RNAi of InR and ribosomal S6 protein kinase (S6K) determines an increase in the duration of the last nymphal instar, together with reduced juvenile hormone (JH) and vitellogenin (Vg) production (Abrisqueta et al., 2014, 2017). The transcription factor FoxO is the main mediator in the IIS pathway (Greer and Brunet, 2005). When the pathway is activated, a phosphorylation cascade of different kinases ends with the phosphorylation of FoxO, causing it to be exported from the nucleus and, therefore, inactivated (Greer and Brunet, 2005; Kramer et al., 2003; Puig et al., 2003). When the pathway is inactive, dephosphorylated FoxO moves into the nucleus and activates the transcription of InR, hypertrehalosemic hormone, Brummer lipase, or glycogen phosphorylase, as observed in B. germanica (Abrisqueta et al., 2014; Süren-Castillo et al., 2014), or inhibits Vg transcription in both B. germanica (Süren-Castillo et al., 2012) and the beetle Tribolium castaneum (Sheng et al., 2011). In adult B. germanica females, seven ILPs have been described. Six of * Corresponding author. E-mail address: [email protected] (J.L. Maestro). Contents lists available at ScienceDirect Insect Biochemistry and Molecular Biology journal homepage: www.elsevier.com/locate/ibmb https://doi.org/10.1016/j.ibmb.2021.103706 Received 4 October 2021; Received in revised form 22 December 2021; Accepted 26 December 2021 Insect Biochemistry and Molecular Biology 141 (2022) 103706 2 these (BgILP1, 2, 3, 4, 5, and 6) are mainly expressed in the brain, whereas the fat body expresses BgILP7, and the ovaries also express BgILP2 (Castro-Arnau et al., 2019). Due to the difficulty of identifying orthologies between these genes and those described in other insects (Antonova et al., 2012), the numbering of those found in B. germanica simply corresponds to the order in which they were discovered. Recently, Veenstra (2020) reported the occurrence of a BgILP8, which was found to be abundantly expressed just in transcriptomes of the male reproductive organs and fat body. Although it was extensively tested, we were unable to identify the roles of BgILPs in vitellogenesis or reproduction of adult females (Castro-Arnau et al., 2019). However, we determined that some BgILPs are differentially expressed when starvation and feeding conditions are compared, and a reciprocal compensatory effect in the expression of BgILP3 and BgILP5 has also been reported (Castro-Arnau et al., 2019). Studying the regulation of ILP expression will help us to understand how IIS pathway activity is regulated. In this work we scrutinised the factors that regulate the expression of B. germanica ILPs, in particular during starvation and in response to the action of JH. The results show that various factors regulate BgILP expression, some of which are similar to those described for other species, while others act differently. This points to the difficulty of generalising ILP regulation mechanisms in insects. 2. Material and methods 2.1. Insects Specimens of B. germanica were obtained from a colony reared in the dark on dog food and water at 29 ±1 ◦C and 60–70% relative humidity. For the starvation assays, animals received only water after the imaginal molt. Dissections were carried out in Ringer’s saline on carbon dioxideanesthetized specimens. After dissection, tissues were immediately frozen in liquid nitrogen and stored at −80 ◦C. 2.2. RNA extraction, cDNA synthesis and quantitative real-time PCR analysis Total RNA was extracted using the GenElute™ Mammalian total RNA (Sigma) or HigherPurity™ Tissue Total RNA Purification (canvax) kits. cDNAs were synthetized from total RNA as previously described (Monta˜ n´ es et al., 2021). In the case of fat body and ovary, 1 μ g of total RNA was used, whereas in the case of brain and CC-CA, we lyophilized and concentrated the sample to use the total amount of the RNA. The absence of genomic contamination was confirmed using a control without reverse transcription. Quantitative real-time PCR analyses were carried out as previously described (Ons et al., 2015). Primer sequences to amplify the different BgILPs, FoxO, TOR, Vg, JHAMT, Met, Kr-h1, InR, Actin 5C and eukaryotic initiation factor 4a (EIF4a) have been reported elsewhere (Abrisqueta et al., 2014; Castro-Arnau et al., 2019; Dominguez and Maestro, 2018; Irles and Piulachs, 2014; Lozano and Belles, 2011, 2014; Maestro et al., 2009; Süren-Castillo et al., 2012). In the analysis of fat body expression from the RNAi experiments for TOR, JHAMT and Met, EIF4a was used as the reference gene. In all the other cases, the reference gene was Actin 5C. The total reaction volume was 20 μ L. All reactions were run in duplicate or triplicate. 2.3. RNA interference dsRNAs were synthetized using MEGAscript™ RNAi kit (Invitrogen). To avoid the possible effects of protein depletion during nymphal development we injected 2 μ l of dsRNA at a concentration of 1 μ g/ μ l into the abdomen females on the first day of oothecal transport, using a 5 μ l Hamilton® 75N syringe. Treatment was repeated the treatment on day 7. On the 12, oothecae were manually removed to synchronize the start of the second gonadotrophic cycle for all the experimental animals. The animals were dissected on day 5 of the second gonadotrophic cycle, which is, in all aspects, comparable to the first cycle. The primers used to generate the dsRNAs for FoxO, TOR, JHAMT, Met and Kr-h1 are described elsewhere (Dominguez and Maestro, 2018; Lozano and Belles, 2011, 2014; Maestro et al., 2009; Süren-Castillo et al., 2012). A heterologous 441 bp fragment from the gene sequence of the polyhedrin of Autographa californica nucleopolyhedrovirus was used as negative control. 2.4. Juvenile hormone treatment JH treatment was performed by topical application (Ons et al., 2015). Four days after the adult molt, in the case of the starvation experiment, or after removing the ootheca and triggering the second gonadotrophic cycle, in the case of the JHAMT RNAi experiment, the wings of the animals were cut and 1 μ l of JH III (Sigma) diluted in analytical grade acetone at a concentration of 2 μ g/ μ l or 20 μ g/ μ l was topically applied on the abdominal tergites using a 5 μ l Hamilton® 75N syringe. Controls were equivalently treated with acetone. 2.5. Statistical analysis All data were expressed as mean ±standard error of the mean (S.E. M.). Statistical analyses were performed using IBM SPSS Statistics 24. In the case of experiments which included JH treatments, ANOVA with Tukey’s test was performed. For the remaining experiments, comparison of results from control and treated animals was performed using Student’s t-test. 3. Results 3.1. Regulation of ILP expression in starved females Our first experiment was aimed at checking whether the reduced expression of brain BgILP3, BgILP5 and BgILP6, fat body BgILP7, and the increased expression of ovarian BgILP2 that was observed in starved compared to fed adult B. germanica females (Castro-Arnau et al., 2019), was due to the reduced JH production observed during starvation (Maestro et al., 2009). Thus, we treated 4-day-old starved females with 2 μ g of either JH III or acetone, and quantified BgILP mRNA levels 24 h later. The results showed that BgILP3, 5, 6 and 7 showed reduced mRNA levels in starved animals, although in the case of BgILP6, starvation produced a 48% reduction that didn’t show significant differences compared to the fed animals in the statistical test. In the case of the ovary, starvation induced a 2.3-fold increase in BgILP2 mRNA levels, although again this result didn’t show significant differences compared to the fed animals in the statistical test. In addition, the treatment with JH did not reverse the starvation effect on BgILP expression (Fig. 1), although it did trigger a 9-fold increase in vitellogenin (Vg) expression in the fat body (Fig. S1). One of the potential candidates involved in regulating the expression of ILPs during starvation is the transcription factor FoxO. For this reason, we determined the effect of FoxO depletion on ILP expression in starved females. To avoid possible effects of FoxO depletion on growth and development, we treated B. germanica adult females in the first day of ootheca transport with 2 μ g of dsRNA targeting FoxO (dsFoxO). The treatment was repeated seven days later. We manually removed the ootheca on the twelfth day and kept the animals starved for five days, when we dissected them. The treatment produced a 48%, 78% and 46% reduction of FoxO mRNA levels in brain, fat body and ovary, respectively (Fig. S2). FoxO depletion produced a significant change in brain BgILP5 mRNA. However, this change did not involve an increase, but rather an even greater reduction of BgILP5 mRNA levels (Fig. 2). No significant changes were observed in the levels of brain BgILP3 and BgILP6, fat body BgILP7, and ovarian BgILP2 mRNA (Fig. 2 and Fig. S2). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 3 3.2. Effect of dsFoxO and dsTOR treatments in fed females We then decided to analyse the effect of FoxO silencing on the expression of brain BgILP5 (and the other brain ILPs) in fed females. We followed the same experimental protocol but, in this case, we allowed access to food after ootheca removal. The treatment produced a 49% reduction of brain FoxO mRNA levels (Fig. S3), but no significant expression changes were observed in any of the brain ILPs, although BgILP5 mRNA levels showed a tendency to reduce (51% reduction on average) (Fig. S3). Provided that the target of rapamycin (TOR) pathway is activated by circulating amino acids (Hansen et al., 2004; Kim and Guan, 2019), starvation also produces a reduction in TOR pathway activity. One possibility is, then, that the reduced expression of BgILPs was due to reduced TOR pathway activity. To investigate this possibility, we followed the same experimental protocol as in the previous experiment, this time treating with 2 μ g of dsRNA targeting TOR (dsTOR) and providing food for the second gonadotrophic cycle. Dissections were once again performed five days after oothecal removal, at the peak vitellogenic period for control animals. The treatment resulted in 83%, 92% and 49% reduction of TOR mRNA levels in brain, fat body and ovary, respectively (Fig. S4). The results showed that TOR depletion did not produce the same effect as starvation in terms of ILP expression. In the brain, the treatment elicited an increase in BgILP3 mRNA levels and a tendency towards reduced BgILP5 mRNA levels (Fig. 3). In the fat body, dsTOR treatment produced a 4.3-fold increase in BgILP7 mRNA levels (Fig. 3). Also in the case of the ovary, dsTOR treatment induced a 1.9-fold increase in BgILP2 mRNA levels (Fig. 3). 3.3. Regulation of ILP expression by juvenile hormone One of the effects of dsTOR treatment in B. germanica females is reduced juvenile hormone (JH) synthesis (Maestro et al., 2009). There is, therefore, the possibility that the effects on ILP expression observed after the dsTOR treatment were due to reduced levels of JH. Thus, we analysed the role JH plays in regulating ILP expression. To do this, we depleted the expression of juvenile hormone acid O-methyltransferase (JHAMT) (Fig. S5) by injecting dsRNA targeting JHAMT (dsJHAMT) using a similar protocol as in the previous treatments. The results showed that RNAi-triggered JHAMT depletion (Fig. S5) elicited an 88% decrease in brain BgILP5 mRNA. The treatment also resulted in a tendency to increased BgILP3 mRNA levels, although the differences with the controls were not statistically significant. The expression of the other brain ILPs was practically unaffected after JHAMT depletion (Fig. 4). To ascertain whether the observed effects on ILP expression were really due to the reduced JH levels, we repeated the RNAi treatment, treating the dsJHAMT females with 2 μ g of JH III or acetone 24 h before the dissections. The results showed that JH treatment corrected the effects of Fig. 1. BgILPs expression in fed, starved and starved treated with JH B. germanica females. Starved animals were treated 24 h before the dissections with 2 μ g JH (Starved +JH) or acetone (Starved). Graphs show BgILP3, 5 and 6 mRNA levels in brains, BgILP7 mRNA levels in fat bodies and BgILP2 mRNA levels in ovaries from 5-day-old adult females. Y-axes indicate copies per copy of Actin 5C. The results are expressed as the mean ±S.E. (n =3–4). The different letters (a–b) indicate groups with significant differences according to the ANOVA test (Tukey, p <0.05). Fig. 2. Effect of FoxO RNAi on brain BgILPs expression of starved females. Adult B. germanica females were treated with dsRNA targeting FoxO (dsFoxO) or a heterologous dsRNA (Control) during the oothecal transport period, and dissections were performed on day 5 of the second gonadotrophic cycle (see Material and Methods). The animals were starved from the beginning of the second gonadotrophic cycle. Graphs show BgILP3, 5 and 6 mRNA levels in brains. Yaxes indicate copies per copy of Actin 5C. The results are expressed as the mean ±S.E (n = 4–5). Asterisks represent significant differences between Control and dsFoxO animals (Student’s t-test, *p <0.05). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 4 the JH depletion on brain ILPs, reducing BgILP3 mRNA levels and increasing BgILP5 mRNA levels (Fig. 4). In the ovary, dsJHAMT treatment produced a 3.7-fold increase in the expression of ovarian BgILP2, whereas the JH treatment reverted this increase, and elicited a slight increase of the basal follicle length (Fig. S5). In the fat body, dsJHAMT treatment induced a 27-fold increase in BgILP7 mRNA levels (Fig. 4). In this tissue, the 2 μ g JH treatment did not correct the effect of the dsJHAMT treatment (results not shown). For this reason, we decided to repeat the experiment using a JH dose of 20 μ g; this fully corrected the effect of the dsJHAMT treatment (Fig. 4). 3.4. Mechanism of action of JH in the regulation of ILPs To determine the molecular mechanism through which JH regulates BgILP expression, we analysed the effects of RNAi-triggered depletion of Met (dsMet) and Kr-h1 (dsKr-h1) on BgILP mRNA levels. We used the same dsRNA treatment protocol as in the previous experiments. The dsMet treatment elicited a 50% reduction in brain Met mRNA levels and a 30% reduction in Kr-h1 mRNA levels (Fig. 5). In addition, BgILP5 expression showed an 82% reduction, whereas for BgILP3 there was a tendency to increase (56% increase as average), although differences with respect to the controls were not statistically significant (Fig. 5). In dsKr-h1-treated insects, although the treatment elicited a 52% reduction in Kr-h1 mRNA levels (Fig. 5), we observed no differences in the levels of either BgILP5 or BgILP3 (Fig. 5). In addition, there were no changes in any of the other brain ILP (BgILP1, 2, 4 and 6) mRNA levels after the dsMet or dsKr-h1 treatments (Fig. S6). In terms of the fat body, both dsMet and dsKr-h1 treatments elicited increased BgILP7 expression, similar to that produced by the dsJHAMT treatment (Fig. 5), even though the 33% reduction of Kr-h1 mRNA levels triggered by the dsKr-h1 treatment was not statistically significant. In addition, the dsMet treatment reduced Vg expression, although the dsKrh1 treatment did not (Fig. 5). In the ovaries, dsMet animals did not show a decrease in Met mRNA, although they showed a significant reduction in Kr-h1 mRNA levels (Fig. S6). In addition, dsMet treatment mimicked the increase in BgILP2 mRNA levels triggered by the dsJHAMT treatment but dsKr-h1 did not (Fig. S6). Fig. 3. Effect of TOR RNAi on brain, fat body and ovary BgILPs expression. Adult B. germanica females were treated with dsRNA targeting TOR (dsTOR) or a heterologous dsRNA (Control) during the oothecal transport period, and dissections were performed on day 5 of the second gonadotrophic cycle (see Material and Methods). Graphs show brain BgILP1-6, fat body BgILP7 and ovary BgILP2 mRNA levels. Y-axes indicate copies per copy of Actin 5C in the case of brains and ovaries and copies per copy of EIF4a in the case of fat bodies. The results are expressed as the mean ±S.E (n =7–8 for brains, n =5 for fat bodies and n =3–4 for ovaries). Asterisks represent significant differences between Control and dsTOR animals (Student’s t-test, *p <0.05; **p <0.01; ***p <0.0005). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 5 Fig. 4. Effect of JHAMT RNAi and JH treatment on brain and fat body BgILPs expression. Adult B. germanica females were treated with dsRNA targeting JHAMT (dsJHAMT) or a heterologous dsRNA (Control) during the oothecal transport period, and dissections were performed on day 5 of the second gonadotrophic cycle (see Material and Methods). For A, only Control and dsJHAMT treatment was performed (n =9–10 for brains and n =8–10 for fat bodies). For B and C, animals were treated 24 h before the dissections with JH: 2 μ g in the case of brains and 20 μ g in the case of fat bodies (dsJHAMT +JH), or with acetone (Control and dsJHAMT). (n =5 for brains and n =4 for fat bodies). Graphs show brain BgILP1-6 and fat body BgILP7 mRNA levels. Y-axes indicate copies per copy of Actin 5C in the case of brains and copies per copy of EIF4a in the case of fat bodies. The results are expressed as the mean ±S.E. In A, asterisks represent significant differences between Control and dsJHAMT animals (Student’s t-test, ***p <0.0001). In B and C, the different letters (a–b) indicate groups with significant differences according to the ANOVA test (Tukey, p <0.05). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 6 3.5. Effect of the treatment on the IIS In order to determine the activity of the IIS pathway, we measured InR mRNA levels (InR Cluster I according to Smýkal et al., 2020) in the fat bodies of females from the dsFoxO starved and the dsTOR and dsJHAMT treatments. dsFoxO treatment in starved females induced a 62% reduction of InR mRNA levels, which indicated that the increase in IIS determined by FoxO depletion caused a decrease in InR expression (Fig. 6). Both dsTOR and dsJHAMT treatments also produced a decrease of InR mRNA levels, whereas JH treatment in JHAMT-depleted females tended to restore them. Altogether, these results suggested that the changes in ILP levels resulting from JH reduction induced an increase in fat body IIS, while the application of JH and its concomitant change in ILP levels reduced fat body IIS. 4. Discussion B. germanica ILPs are differentially regulated by nutrition. Starvation Fig. 5. Effect of Met and Kr-h1 RNAi on brain and fat body. Adult B. germanica females were treated with dsRNA targeting Met (dsMet), Kr-h1 (dsKr-h1) or a heterologous dsRNA (Control) during the oothecal transport period, and dissections were performed on day 5 of the second gonadotrophic cycle (see Material and Methods). Graphs show BgILP3, BgILP5, Met and Kr-h1, mRNA levels in brains (n =6) and BgILP7, Vg, Met and Kr-h1 mRNA levels in fat bodies (n =5–6). Y-axes indicate copies per copy of Actin 5C in the case of brains and dsKr-h1 treated fat bodies and copies per copy of EIF4a in the case of dsMet treated fat bodies. The results are expressed as the mean ±S.E. Asterisks represent significant differences between Control and treated animals (Student’s t-test, *p <0.05; **p <0.001; ***p <0.0001). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 7 elicits an 85–90% reduction in BgILP3 and BgILP5 and a ca. 50% reduction in BgILP6 expression in the brain, a more than 95% reduction in fat body BgILP7 expression, and an increase in ovarian BgILP2 mRNA levels (Castro-Arnau et al., 2019). In contrast, no changes are observed in the expression of the other BgILPs. These expression changes resulting from starvation are not due to reduced levels of JH, since a treatment with sufficient JH to increase the expression of Vg in the fat body is unable to restore the mRNA levels of these BgILPs in fed females. One possibility is that, in adult B. germanica females, starvation depletes the IIS pathway by reducing the expression and/or release of some of the BgILPs. In turn, this reduction in IIS would produce FoxO dephosphorylation and activation. FoxO would then inhibit the expression of a number of other BgILPs. If this were the case, starvation plus FoxO-depletion would produce the increased expression of at least some of the BgILPs. The results showed that adding a dsFoxO treatment to the starvation did not change BgILP3 and BgILP6 expression but, unexpectedly, did reduce that of BgILP5. This result suggests that FoxO is necessary for maintaining the low BgILP5 mRNA levels observed during starvation and preventing these levels from dropping further, in order to preserve necessary, although low, expression levels. In the dsFoxO treatment of fed females, BgILP5 was also the gene that presented a greater than 50% reduction, on average, although the difference with the controls was not statistically significant. In adult Drosophila melanogaster females, dilp3 (but not dilp2 or dilp5) mRNA levels were shown to be reduced in FoxO null mutants (Broughton et al., 2008). Also in adult D. melanogaster females, but using a different FoxO mutant, reduced dilp2, dilp3 and dilp5 expression was reported (Slack et al., 2011). In addition, FoxO-specific activation in the adult pericerebral fat body resulted in increased dilp2 expression (Hwangbo et al., 2004). In the case of larvae, however, when FoxO was specifically depleted in brain Insulin Producing Cells (IPCs), dilp5 expression was activated (Okamoto and Nishimura, 2015). In the larvae of the beetle T. castaneum, dsFoxO treatment reduced ilp-2, -3, and -4, although the authors attribute this to the fact that dsFoxO also results in reduced food intake levels (Lin et al., 2018). This cannot be the case in our study since the results were obtained using starved individuals. In the kissing bug, Rhodnius prolixus, dsFoxO reduces insulin growth factor (IGF) adult females expression but, unlike BgILP5, the expression of IGF is greater in unfed than fed females (Leyria et al., 2020, 2021). The differences between the results observed in B. germanica compared to those obtained in other species of insects may be due to differences in the experimental protocol, including the fact of using juvenile vs. adult specimens, but also to differences in the specific regulation of the different genes. The reduced expression of BgILPs observed in starved females is also not due to the presumed reduction in TOR pathway signalling that occurs during starvation (Hansen et al., 2004; Maestro et al., 2009; Oldham et al., 2000), since TOR depletion does not restore BgILP expression levels. In D. melanogaster, several TOR-dependent factors have been reported to send signals from the fat body to the brain IPCs, regulating mainly DILP secretion, but also transcription in some cases (Agrawal et al., 2016; Delanoue et al., 2016; G´ eminard et al., 2009; Ingaramo et al., 2020; Koyama and Mirth, 2016; Sano et al., 2015). In B. germanica, TOR interference results in reduced growth of the developing oocytes (Maestro et al., 2009). In TOR-depleted females, ovarian BgILP2 mRNA levels are similar to those reported during the reproductive cycle in control females with oocytes of the same size (Castro-Arnau et al., 2019). This could correspond to the BgILP2 expression at that stage of maturation, rather than to TOR acting on its expression. JH is the gonadotrophic hormone in cockroaches, as well as many other insects (Belles, 2005), and JH synthesis and JH levels in circulation increase throughout the reproductive cycle of B. germanica females (Cruz et al., 2003; Maestro et al., 1994). To study the effect of JH on the expression of BgILPs, we depleted the expression of JHAMT, a gene coding for a key enzyme in JH biosynthesis (Dominguez and Maestro, 2018). JHAMT depletion reduced JH and Vg synthesis as well as follicle growth to extremely low levels (Dominguez and Maestro, 2018; this work). Moreover, reducing JH levels produced a strong decrease in brain BgILP5 mRNA levels. This reduction tended to be corrected with a treatment with JH 24 h before. In addition, JH depletion produced an increase in brain BgILP3 mRNA levels, which was again corrected with a JH treatment. BgILP3 and BgILP5 show a kind of compensatory regulation, in the sense that the reduced expression of one of them produces an expression increase of the other (Castro-Arnau et al., 2019), possibly as a way to maintain certain levels of activity of such an important pathway. Thus, depleting JH levels could produce, in the first instance, reduced BgILP5 expression or increased BgILP3 expression, and the compensatory regulation would produce the effect on the expression of the other BgILP. In the fat body, the reduction of JH produced a huge increase in BgILP7 expression, the levels of which were reduced after treatment with JH; in the ovary, there was a similar effect on BgILP2 mRNA levels. In D. melanogaster, overexpression of fat body dilp6 results in a reduction in the expression of brain dilp2 and dilp5 (Bai et al., 2012). In B. germancia, it is not the increase in BgILP7 what causes the reduction of BgILP5 expression since BgILP7 RNAi does not modify the levels of BgILP5 mRNA (Castro-Arnau et al., 2019). Thus, the results showed that, in the brain, JH activates the expression of BgILP5 or inhibits the expression of BgILP3, or both, whereas in the fat body, JH inhibits the expression of BgILP7. In terms of the ovary, it is possible that the observed levels, as in the case described above, correspond to the physiological and developmental changes that occur as a consequence of the treatments, rather than to the direct effect of JH on the expression of BgILP2. The increase in the mRNA levels of BgILP3 and BgILP7 and the slight decrease in those of BgILP5 observed in the treatment with dsTOR could be due to the fact that TOR depletion also reduces the levels of JH (Maestro et al., 2009). In terms of BgILP expression, Met depletion produced a phenotype similar to that of JHAMT depletion. As expected, this indicates that the effect of JH on BgILPs operates through the JH receptor. In the case of KrFig. 6. Effect of FoxO, TOR and JHAMT RNAi on fat body InR expression. Adult B. germanica females were treated with dsRNA targeting FoxO (dsFoxO), TOR (dsTOR), JHAMT (dsJHAMT) or a heterologous dsRNA (Control) during the oothecal transport period, and dissections were performed on day 5 of the second gonadotrophic cycle (see Material and Methods). In the case of the dsFoxO experiment, the animals were starved from the beginning of the second gonadotrophic cycle. In the case of the dsJHAMT experiment, animals were treated 24 h before the dissections with 20 μ g JH (dsJHAMT +JH), or with acetone (Control and dsJHAMT). Y-axes indicate copies per copy of Actin 5C in the case of the dsFoxO experiment and copies per copy of EIF4a in the case of the dsTOR and dsJHAMT experiments. The results are expressed as the mean ±S.E (n =4–6). Asterisk represents significant differences between Control and treated animals (Student’s t-test, *p <0.05) and the different letters (a–c) indicate groups with significant differences according to the ANOVA test (Tukey, p <0.05). C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 8 h1, the depletion elicited no effect on the expression of brain BgILPs, although, as observed in the treatments with dsJHAMT and dsMet, dsKrh1 treatment increased fat body BgILP7 mRNA levels. This indicates that Kr-h1 is involved in regulating the expression of fat body BgILP7, but not of that of brain BgILP3 and BgILP5. On the other hand, although the reduction of Kr-h1 mRNA levels in the fat body is not statistically significant (although sufficient to produce an effect on BgILP7 expression), we did not observe an effect on Vg expression, which could suggest that in B. germanica, JH action on Vg expression requires Met but not Kr-h1. In T. castaneum adult females, dsRNA-targeting JHAMT reduces brain and fat body levels of ilp2 and ilp3 mRNAs, whereas JH treatment tends to restore normal levels (Sheng et al., 2011). Also in T. castaneum, ilp2 expression in starved adult males is reduced in Met RNAi beetles (Xu et al., 2013). Similarly, in the locust Schistocerca gregaria, dsMet reduces fat body insulin-related peptide (SgIRP) expression (Gijbels et al., 2019). In D. melanogaster, a hypomorphic mutant of Kr-h1 reduces brain dilp2 and dilp5 expression (Kang et al., 2017), and in the mosquito Aedes aegypti, JH treatment increases ilp 2, 6 and 7 and represses ilp 1, 3, 4, 5 and 8 expression, whereas dsMet and dsKr-h1 treatment elicited the opposite effect (Ling and Raikhel, 2021). In addition, these authors demonstrated that Kr-h1 binds to the promoter of all the ilps, regardless of their activatory or inhibitory effect, while Met only binds to the promoter of ilp 6 (Ling and Raikhel, 2021). To ascertain IIS activity, we measured fat body InR mRNA levels in different treatments. In D. melanogaster, reduction of IIS produces the activation of FoxO, which results in an activation of InR expression as a feedback mechanism (Puig et al., 2003; Puig and Tjian, 2005). Also in B. germanica, we already demonstrated that starvation increases fat body InR mRNA levels whereas dsFoxO treatment abolishes this increase (Abrisqueta et al., 2014), which enables the use of InR mRNA levels as a marker of IIS activity also in this species. The present results again showed that FoxO depletion and the consequent IIS increase reduced fat body InR mRNA levels. In addition, treatments that produced a reduction of JH (dsTOR and dsJHAMT) showed reduced fat body InR mRNA levels whereas JH treatment increased them, which indicated IIS activation and inhibition, respectively. These results suggested that ILPs levels induced by those treatments will be responsible of the changes in IIS, at least in fat bodies. As we discussed above, RNAi for JHAMT in T. castaneum reduces ilp expression and it is ilp reduction that triggers the decrease in Vg transcription (Sheng et al., 2011). This does not appear to be exactly the case in B. germanica where, although InR activity is necessary for full activation of Vg expression (Abrisqueta et al., 2014), JH reduction increases fat body IIS (this work), while Vg expression is low (Dominguez and Maestro, 2018). We then conclude that in B. germanica, both JH and IIS are necessary for activating Vg expression. In summary, the reduction of BgILP3, 5 and 6 expression in the brain and BgILP7 in the fat body observed in starvation is not due to low levels of JH or TOR pathway activity, or to the action of FoxO, but to unknown regulatory processes whose activity outweighs the effect of the factors indicated above. BgILP3 and BgILP5 are the brain ILPs that are subject to more dynamic regulatory activity. In addition, they show more reduced expression in starvation and are those for which compensatory regulation has been demonstrated (Castro-Arnau et al., 2019). On the other hand, it is remarkable that the action of JH on the expression of BgILP7 in the fat body requires the activity of Kr-h1, while this transcription factor is not necessary in the case of brain BgILPs. As for BgILP2 expression regulation in the ovary, the observed results seem more related to the effect of the treatment on the maturation of the ovaries than to the regulation of BgILP2 specifically. The results presented here point to a number of signalling pathways, in particular the FoxO and JH pathways, that contribute to regulating the expression of B. germanica ILPs. As discussed above, the results obtained in other species show similar regulation in some cases, although in others the regulation is very different. When analysing the sequences of the different insect ILPs, it has not been possible to identify orthologies between peptides, even when they belong to closely related groups (Antonova et al., 2012; Veenstra, 2020). This suggests that the gene duplications that gave rise to the different ILPs occurred relatively late on in evolution and, in most cases, independently in different lineages. This may have determined the specific regulation of ILPs, even in closely related insect groups, making it difficult to find general trends. Acknowledgments This work was supported by Agencia Estatal de Investigaci´ on (grant number PID2019-104483GB-I00/AEI/10.13039/501100011033), Secretaria d’Universitats i Recerca, Catalan Goverment (grant number 2017 SGR 1030) and Agencia Estatal Consejo Superior de Investigaciones Científicas (grant number 2019AEP029). Thanks to Xavier Bell´ es for critical reading of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.ibmb.2021.103706. References Abrisqueta, M., Süren-Castillo, S., Maestro, J.L., 2017. S6 protein kinase activates Juvenile Hormone and vitellogenin production in the cockroach Blattella germanica. Physiol. Entomol. 42, 10–16. https://doi.org/10.1111/phen.12156. Abrisqueta, M., Süren-Castillo, S., Maestro, J.L., 2014. Insulin receptor-mediated nutritional signalling regulates juvenile hormone biosynthesis and vitellogenin production in the German cockroach. Insect Biochem. Mol. Biol. 49, 14–23. https:// doi.org/10.1016/j.ibmb.2014.03.005. Agrawal, N., Delanoue, R., Mauri, A., Basco, D., Pasco, M., Thorens, B., L´ eopold, P., 2016. The Drosophila TNF eiger is an adipokine that acts on insulin-producing cells to mediate nutrient response. Cell Metabol. 23, 675–684. https://doi.org/10.1016/J. CMET.2016.03.003. Antonova, Y., Arik, A.J., Moore, W., Riehle, M.A., Brown, M.R., 2012. Insulin-like peptides: structure, signaling, and function. In: Gilbert, L.I. (Ed.), Insect Endocrinology. Elsevier, pp. 63–92. https://doi.org/10.1016/B978-0-12-3847492.10002-0. Bai, H., Kang, P., Tatar, M., 2012. Drosophila insulin-like peptide-6 (dilp6) expression from fat body extends lifespan and represses secretion of Drosophila insulin-like peptide-2 from the brain. Aging Cell 11, 978–985. https://doi.org/10.1111/ acel.12000. Belles, X., 2005. Vitellogenesis directed by juvenile hormone. In: Reproductive Biology of Invertebrates, XII, pp. 157–197. Part B, Progress in Vitellogenesis. Broughton, S., Alic, N., Slack, C., Bass, T., Ikeya, T., Vinti, G., Tommasi, A.M., Driege, Y., Hafen, E., Partridge, L., 2008. Reduction of DILP2 in Drosophila triages a metabolic phenotype from lifespan revealing redundancy and compensation among DILPs. PLoS One 3, e3721. https://doi.org/10.1371/journal.pone.0003721. Castro-Arnau, J., Marín, A., Castells, M., Ferrer, I., Maestro, J.L., 2019. The expression of cockroach insulin-like peptides is differentially regulated by physiological conditions and affected by compensatory regulation. J. Insect Physiol. 114, 57–67. https://doi. org/10.1016/j.jinsphys.2019.02.010. Claeys, I., Simonet, G., Poels, J., Van Loy, T., Vercammen, L., De Loof, A., Vanden Broeck, J., 2002. Insulin-related peptides and their conserved signal transduction pathway. Peptides 23, 807–816. https://doi.org/10.1016/S0196-9781(01)00666-0. Cruz, J., Martín, D., Pascual, N., Maestro, J.L., Piulachs, M.D., Bell´ es, X., 2003. Quantity does matter. Juvenile hormone and the onset of vitellogenesis in the German cockroach. Insect Biochem. Mol. Biol. 33 https://doi.org/10.1016/j. ibmb.2003.06.004. Delanoue, R., Meschi, E., Agrawal, N., Mauri, A., Tsatskis, Y., McNeill, H., L´ eopold, P., 2016. Drosophila insulin release is triggered by adipose Stunted ligand to brain Methuselah receptor. Science 353, 1553–1556. https://doi.org/10.1126/science. aaf8430. Dominguez, C.V., Maestro, J.L., 2018. Expression of juvenile hormone acid O -methyltransferase and juvenile hormone synthesis in Blattella germanica. Insect Sci. 25, 787–796. https://doi.org/10.1111/1744-7917.12467. G´ eminard, C., Rulifson, E.J., L´ eopold, P., 2009. Remote control of insulin secretion by fat cells in Drosophila. Cell Metabol. 10, 199–207. https://doi.org/10.1016/j. cmet.2009.08.002. Gijbels, M., Lenaerts, C., Vanden Broeck, J., Marchal, E., 2019. Juvenile Hormone receptor Met is essential for ovarian maturation in the Desert Locust, Schistocerca gregaria. Sci. Rep. 9, 10797. https://doi.org/10.1038/s41598-019-47253-x. Greer, E.L., Brunet, A., 2005. FOXO transcription factors at the interface between longevity and tumor suppression. Oncogene. https://doi.org/10.1038/sj. onc.1209086. Hansen, I.A., Attardo, G.M., Park, J.H., Peng, Q., Raikhel, A.S., 2004. Target of rapamycin-mediated amino acid signaling in mosquito anautogeny. Proc. Natl. Acad. Sci. U.S.A. 101, 10626–10631. https://doi.org/10.1073/pnas.0403460101. C.V. Domínguez et al. Insect Biochemistry and Molecular Biology 141 (2022) 103706 9 Hwangbo, D.S., Gershman, B., Gersham, B., Tu, M.-P., Palmer, M., Tatar, M., 2004. Drosophila dFOXO controls lifespan and regulates insulin signalling in brain and fat body. Nature 429, 562–566. https://doi.org/10.1038/nature02549. Ingaramo, M.C., S´ anchez, J.A., Perrimon, N., Dekanty, A., 2020. Fat body p53 regulates systemic insulin signaling and autophagy under nutrient stress via Drosophila Upd2 repression. Cell Rep. 33, 108321 https://doi.org/10.1016/J.CELREP.2020.108321. Irles, P., Piulachs, M.-D., 2014. Unlike in Drosophila meroistic ovaries, hippo represses notch in Blattella germanica panoistic ovaries, triggering the mitosis-endocycle switch in the follicular cells. PLoS One 9, e113850. https://doi.org/10.1371/journal. pone.0113850. Kang, P., Chang, K., Liu, Y., Bouska, M., Birnbaum, A., Karashchuk, G., Thakore, R., Zheng, W., Post, S., Brent, C.S., Li, S., Tatar, M., Bai, H., 2017. Drosophila Kruppel homolog 1 represses lipolysis through interaction with dFOXO. Sci. Rep. 71 (7), 1–15. https://doi.org/10.1038/s41598-017-16638-1, 2017. Kim, J., Guan, K.L., 2019. mTOR as a central hub of nutrient signalling and cell growth. Nat. Cell Biol. https://doi.org/10.1038/s41556-018-0205-1. Koyama, T., Mirth, C.K., 2016. Growth-blocking peptides as nutrition-sensitive signals for insulin secretion and body size regulation. PLoS Biol. 14, e1002392 https://doi. org/10.1371/journal.pbio.1002392. Kramer, J.M., Davidge, J.T., Lockyer, J.M., Staveley, B.E., 2003. Expression of Drosophila FOXO Regulates Growth and Can Phenocopy Starvation. Leyria, J., Orchard, I., Lange, A.B., 2021. The involvement of insulin/ToR signaling pathway in reproductive performance of Rhodnius prolixus. Insect Biochem. Mol. Biol. 130, 103526. https://doi.org/10.1016/j.ibmb.2021.103526. Leyria, J., Orchard, I., Lange, A.B., 2020. Transcriptomic analysis of regulatory pathways involved in female reproductive physiology of Rhodnius prolixus under different nutritional states. Sci. Rep. 10 https://doi.org/10.1038/s41598-020-67932-4. Lin, X., Yu, N., Smagghe, G., 2018. FoxO mediates the timing of pupation through regulating ecdysteroid biosynthesis in the red flour beetle. Tribolium Castaneum. Gen. Comp. Endocrinol. 258, 149–156. https://doi.org/10.1016/J. YGCEN.2017.05.012. Ling, L., Raikhel, A.S., 2021. Cross-talk of insulin-like peptides, juvenile hormone, and 20-hydroxyecdysone in regulation of metabolism in the mosquito Aedes aegypti. Proc. Natl. Acad. Sci. Unit. States Am. 118 https://doi.org/10.1073/ PNAS.2023470118. Lozano, J., Belles, X., 2014. Role of methoprene-tolerant (Met) in adult morphogenesis and in adult ecdysis of Blattella germanica. PLoS One 9, e103614. https://doi.org/ 10.1371/journal.pone.0103614. Lozano, J., Belles, X., 2011. Conserved repressive function of Krüppel homolog 1 on insect metamorphosis in hemimetabolous and holometabolous species. Sci. Rep. 1, 163. https://doi.org/10.1038/srep00163. Maestro, J.L., Cobo, J., Bell´ es, X., 2009. Target of rapamycin (TOR) mediates the transduction of nutritional signals into juvenile hormone production. J. Biol. Chem. 284, 5506–5513. Maestro, J.L., Dan´ es, M.D., Piulachs, M.D., Cassier, P., Bell´ es, X., 1994. Juvenile hormone inhibition in corpora allata from ovariectomized Blattella germanica. Physiol. Entomol. 19, 342–348. https://doi.org/10.1111/j.1365-3032.1994.tb01061.x. Monta˜ n´ es, J.C., Rojano, C., Ylla, G., Piulachs, M.D., Maestro, J.L., 2021. siRNA enrichment in Argonaute 2-depleted Blattella germanica. Biochim. Biophys. Acta - Gene Regul. Mech. 1864, 194704 https://doi.org/10.1016/J. BBAGRM.2021.194704. Okamoto, N., Nishimura, T., 2015. Signaling from glia and cholinergic neurons controls nutrient-dependent production of an insulin-like peptide for Drosophila body growth. Dev. Cell 35, 295–310. https://doi.org/10.1016/j.devcel.2015.10.003. Oldham, S., Montagne, J., Radimerski, T., Thomas, G., Hafen, E., 2000. Genetic and biochemical characterization of dTOR, the Drosophila homolog of the target of rapamycin. Genes Dev. 14, 2689–2694. https://doi.org/10.1101/gad.845700. Ons, S., Bell´ es, X., Maestro, J.L., 2015. Orcokinins contribute to the regulation of vitellogenin transcription in the cockroach Blattella germanica. J. Insect Physiol. 82, 129–133. https://doi.org/10.1016/j.jinsphys.2015.10.002. Puig, O., Marr, M.T., Ruhf, M.L., Tjian, R., 2003. Control of cell number by Drosophila FOXO: downstream and feedback regulation of the insulin receptor pathway. Genes Dev. 17, 2006–2020. https://doi.org/10.1101/gad.1098703. Puig, O., Tjian, R., 2005. Transcriptional Feedback Control of Insulin Receptor by dFOXO/FOXO1. https://doi.org/10.1101/gad.1340505. Sano, H., Nakamura, A., Texada, M.J., Truman, J.W., Ishimoto, H., Kamikouchi, A., Nibu, Y., Kume, K., Ida, T., Kojima, M., 2015. The nutrient-responsive hormone CCHamide-2 controls growth by regulating insulin-like peptides in the brain of Drosophila melanogaster. PLoS Genet. 11, e1005209 https://doi.org/10.1371/ journal.pgen.1005209. Sheng, Z., Xu, J., Bai, H., Zhu, F., Palli, S.R., 2011. Juvenile hormone regulates vitellogenin gene expression through insulin-like peptide signaling pathway in the red flour beetle, Tribolium castaneum. J. Biol. Chem. 286, 41924–41936. https:// doi.org/10.1074/jbc.M111.269845. Slack, C., Giannakou, M.E., Foley, A., Goss, M., Partridge, L., 2011. dFOXO-independent effects of reduced insulin-like signaling in Drosophila. Aging Cell 10, 735–748. https://doi.org/10.1111/j.1474-9726.2011.00707.x. Smýkal, V., Pivarˇ ci, M., Provazník, J., Bazalov´ a, O., Jedliˇ cka, P., Lukˇ san, O., Hor´ ak, A., Vanˇ eˇ ckov´ a, H., Beneˇ s, V., Fiala, I., Hanus, R., Doleˇ zel, D., 2020. Complex evolution of insect insulin receptors and homologous decoy receptors, and functional significance of their multiplicity. Mol. Biol. Evol. 37, 1775–1789. https://doi.org/ 10.1093/MOLBEV/MSAA048. Süren-Castillo, S., Abrisqueta, M., Maestro, J.L., 2014. FoxO is required for the activation of hypertrehalosemic hormone expression in cockroaches. Biochim. Biophys. Acta Gen. Subj. 1840, 86–94. Süren-Castillo, S., Abrisqueta, M., Maestro, J.L., 2012. FoxO inhibits juvenile hormone biosynthesis and vitellogenin production in the German cockroach. Insect Biochem. Mol. Biol. 42, 491–498. Veenstra, J.A., 2020. Arthropod IGF, relaxin and gonadulin, putative orthologs of Drosophila insulin-like peptides 6, 7 and 8, likely originated from an ancient gene triplication. PeerJ 2020, e9534. https://doi.org/10.7717/peerj.9534. Wu, Q., Brown, M.R., 2006. Signaling and function of insulin-like peptides in insects. Annu. Rev. Entomol. 51, 1–24. https://doi.org/10.1146/annurev. ento.51.110104.151011. Xu, J., Sheng, Z., Palli, S.R., 2013. Juvenile hormone and insulin regulate trehalose homeostasis in the red flour beetle, Tribolium castaneum. PLoS Genet. 9, e1003535 https://doi.org/10.1371/journal.pgen.1003535. C.V. Domínguez et al.