Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Contents lists available at ScienceDirect

Insect Biochemistry and Molecular Biology journal homepage: www.elsevier.com/locate/ibmb

Clathrin-dependent endocytosis plays a predominant role in cellular uptake of double-stranded RNA in the red flour beetle Da Xiao a, b, Xiwu Gao b, Jiaping Xu a, c, Xiao Liang a, Qingqing Li a, d, Jianxiu Yao a, Kun Yan Zhu a, * a

Department of Entomology, Kansas State University, Manhattan, KS 66506, USA Department of Entomology, China Agricultural University, Beijing 100094, China School of Life Sciences, Anhui Agricultural University, Hefei, Anhui 230036, China d College of Life Sciences, Yunnan Normal University, Kunming, Yunnan 650092, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 January 2015 Received in revised form 19 March 2015 Accepted 31 March 2015 Available online 9 April 2015

RNA interference (RNAi) is a highly conserved gene regulatory mechanism in eukaryotic organisms; however, an understanding of mechanisms of cellular uptake of double-stranded RNA (dsRNA) in different organisms remains elusive. By using pharmacological inhibitors of different endocytic pathways in conjunction with RNAi of a marker gene (lethal giant larvae, TcLgl) in the red flour beetle (Tribolium castaneum), we demonstrated that two inhibitors (chlorpromazine and bafilomycin-A1) of clathrindependent endocytosis can nearly abolish or significantly diminish RNAi of TcLgl, whereas methyl-bcyclodextrin and cytochalasin-D, known to inhibit other endocytic pathways, showed no effect on RNAi of TcLgl. By using Cy3-labeled TcLgl dsRNA, we observed significantly reduced cellular uptake of TcLgl dsRNA in midgut cells after larvae were injected with each of the two clathrin-dependent endocytosis inhibitors. By using an “RNAi of RNAi” strategy, we further demonstrated that suppression of each transcript of the four key genes encoding clathrin heavy chain (TcChc), clathrin coat assembly protein AP50 (TcAP50), vacuolar (Hþ)-ATPase subunit H (TcVhaSFD) and a ras-related protein (TcRab7) in clathrindependent endocytosis by RNAi can significantly impair RNAi of TcLgl. These results support our conclusion that clathrin-dependent endocytosis is a major mechanism in cellular uptake of dsRNA in T. castaneum. Our study also provides new insights into improving RNAi efficiency by enhancing dsRNA endosomal release. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Clathrin-dependent endocytosis Cellular uptake RNA interference Tribolium castaneum

1. Introduction RNA interference (RNAi) is a highly conserved posttranscriptional gene regulatory mechanism in eukaryotic organisms, including fungi, plants, insects and mammals (Fire et al., 1998;
s, 2010; Zhuang and Hunter, 2011). Mello and Conte, 2004; Belle However, the cellular uptake mechanism of exogenous doublestranded RNA (dsRNA) does not appear to be highly conserved and remains undefined in different organisms. Two pathways for exogenous dsRNA uptake have been identified or implicated, which include passive uptake via a transmembrane channel protein known as systemic RNA interference deficient-1 (SID-1) encoded by sid-1 in the nematode (Caenorhabditis elegans) (Winston et al.,

* Corresponding author. Tel.: þ1 785 532 4721; fax: þ1 785 532 6232. E-mail address: [email protected] (K.Y. Zhu). http://dx.doi.org/10.1016/j.ibmb.2015.03.009 0965-1748/© 2015 Elsevier Ltd. All rights reserved.

2002), the honey bee (Apis mellifera) (Aronstein et al., 2006), and the fish (Siniperca chuatsi) cells (Ren et al., 2011); and endocytosismediated pathway in the fruit fly (Drosophila melanogaster) S2 cells (Ulvila et al., 2006; Saleh et al., 2006), the desert locust (Schistocera gregaria) (Wynant et al., 2014), and the predatory mite (Metaseiulus occidentalis) (Wu and Hoy, 2014). Although a putative transmembrane protein encoded by AmSid1 was found to play an essential role in dsRNA uptake in the honey bee (Aronstein et al., 2006), several studies have shown that SID-1 is unlikely to play a major role in cellular uptake of dsRNA in other insects including the red flour beetle (Tribolium castaneum) (Tomoyasu et al., 2008), the migratory locust (Locusta migratoria) (Luo et al., 2012), and the desert locust (Wynant et al., 2014). Because SID-1-dependent dsRNA uptake does not seem to be common and conserved in insects, many insects must possess additional or different genes with similar functions, or possibly

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

even different mechanisms in cellular uptake of dsRNA (Tomoyasu et al., 2008; Zhang et al., 2010). Ulvila et al. (2006) analyzed 2000 dsRNA fragments from a cDNA library of D. melanogaster S2 cells for their protective effect from lethality induced by RNAi against Ubi-p63E, an essential gene encoding an ubiquitin for cell viability. They identified four genes, one of which encodes clathrin heavy chain (Chc), an important component of clathrin-dependent endocytosis. When Chc transcript in S2 cells is depleted by RNAi, it protects S2 cells from the lethality induced by the Ubi-p63E dsRNA treatment, suggesting that dsRNA molecules are internalized by clathrin-dependent endocytosis. On the other hand, Saleh et al. (2006) screened a dsRNA library of D. melanogaster S2 cells and identified 23 genes likely to be involved in endocytic pathway and required for cellular uptake of dsRNA by S2 cells. Some of these genes have been known to be directly and/or indirectly involved in endocytosis as they encode proteins of the vesicle mediated transport, conserved oligomeric Golgi complex family, cytoskeleton organization and protein transport. Results from previous studies suggest that most insects are likely to rely on different mechanisms rather than C. elegans SID-1dependent dsRNA uptake pathway (Tomoyasu et al., 2008; Zhang et al., 2010), and endocytosis plays an important role in dsRNA uptake in D. melanogaster S2 cells and probably in many insect species such as the desert locust (Wynant et al., 2014). However, the exact mechanism of cellular dsRNA uptake in different insects remains unclear. In this study, we took the advantage of T. castaneum for its robust RNAi to examine: 1) the effect of pharmacological inhibitors of different endocytic pathways on RNAi of a marker gene (TcLgl, lethal giant larvae); 2) the effect of selective inhibitors of clathrin-dependent endocytosis on cellular uptake of Cy3-labeled TcLgl dsRNA in larval midgut; and 3) effect of RNAi targeting each of four key genes in clathrin-dependent endocytosis on RNAi of TcLgl (i.e., a “RNAi of RNAi” strategy). Our studies provided strong evidence that clathrin-dependent endocytosis plays an essential role in cellular uptake of dsRNA in T. castaneum larvae. 2. Materials and methods 2.1. Insect culture The Georgia-1 (GA-1) strain of T. castaneum was reared on whole-wheat flour containing 5% (by weight) of brewers' yeast at 30  C and 65% RH in growth chamber at Kansas State University (Manhattan, KS, USA). 2.2. Subcloning and sequencing of TcChc Total RNA was isolated from the insects by using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and treated with DNase I (Fermentas, Glen Burnie, MD, USA) to remove possible genomic DNA contamination. First-strand cDNA was synthesized from 1.0 mg total RNA by using First Strand cDNA Synthesis Kit (Fermentas). Specific primers designed based on the predicted gene sequences were used to obtain full-length cDNA by reverse transcription PCR (RTPCR) (Table 1). After PCR products were purified and ligated into pGEM-T Easy Vector (Invitrogen), the ligation mixture was used to transform DH5a competent cells. Plasmid DNA was purified and sequenced by the KSU DNA Sequencing and Genotyping Facility (Manhattan, KS, USA). 2.3. Procedures of RNAi dsRNA was synthesized using MEGAscript® RNAi Kit (Ambion, Austin, TX). After each larva was injected in the hemocoel with

69

100 nl of dsRNA (100 ng/larva) by using Nanoject II injector (Drummond Scientific, Broomall, PA, USA), the larvae were reared under standard conditions. RNAi was performed with three biological replicates (each with at least 40 larvae) for each control and treatment.

2.4. Determination of gene transcript level Developmental stage-dependent expression profiles of TcChc were analyzed in all four life stages at 20 developmental time points, including embryos (1, 2, 3, and 4-day eggs), larvae (1, 5, 10, 15, and 20-day), pupae (1, 2, 3, 4, 5, and 6-day) and adults (1, 5, 10, 15, and 20-day). In each replicate, 60 eggs, 5 larvae, 5 pupae, or 5 adults were pooled as a biological sample for 20 developmental time points. Because our preliminary studies showed a large peak of TcChc transcript during the mid-pupal stage (i.e., 2, 3 and 4-day), we dissected four tissues, including the gut (midgut and hindgut), Malpighian tubule, fat bodies and carcass (the remaining body after the brain, ganglia, gut and fat bodies removed), from 3-day old pupae for profiling tissue-dependent expression of TcChc. In each replicate, Malpighian tubules dissected from 60 pupae were pooled or each of the three remaining tissues dissected from 30 pupae were pooled as a biological sample. Total RNA was isolated from each sample by using TRIzol reagent (Invitrogen) and treated with DNase I (Fermentas). Firststrand cDNA was synthesized from 1.0 mg total RNA by using First Strand cDNA Synthesis Kit (Fermentas). Reverse transcription quantitative PCR (RT-qPCR) was performed by using SYBR Green in iCycler iQTM multi-color real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA) according to Xiao et al. (2014). The specificity of each reaction was evaluated based on the melting temperatures of the PCR products. RT-qPCR analysis of each developmental time point or tissue type was performed with three biological replicates (each with two technical replicates). The gene-specific primers (Table 1) were designed by using the Beacon 7.0 software. The relative expression level of each gene was normalized to ribosomal protein S3 (TcRps3) (Xiao et al., 2014) and calculated by using the 2DDCt method (Livak and Schmittgen, 2001).

2.5. Evaluation of endocytosis inhibitors' effect on RNAi Four endocytosis inhibitors, including chlorpromazine (CPZ), methyl-b-cyclodextrin (MbCD), bafilomycin-A1 (BafA) and cytochalasin-D (CCD), were purchased from SigmaeAldrich (St Louis, MO, USA). CPZ and MbCD were diluted with PBS buffer, whereas BafA and CCD were diluted with 20% dimethylsulfoxide (DMSO). Initial results showed that the injection of 100 nl per larva of CPZ, MbCD, BafA and CCD at 8.0, 13.0, 0.05 and 0.50 mg/ml (i.e., 0.8, 1.3, 0.005 and 0.05 mg/larva), respectively, did not cause larval mortality and marker gene (TcLgl) expression change as evaluated by RT-qPCR (Supplementary Fig. 1). We used these doses and two lower doses (5-fold serial dilutions) of each inhibitor (i.e., 3 doses for each inhibitor) to examine the effect of each inhibitor on RNAi of TcLgl. Specifically, 100 nl of GFP dsRNA (dsGFP, 100 ng) in inhibitor solvent, TcLgl dsRNA (dsTcLgl, 100 ng) in inhibitor solvent, or dsTcLgl (100 ng) containing each of the four inhibitors was injected into the hemocoel of a 16-day-old larva. Each control or treatment for each inhibitor dose and for each time point consisted of three replicates (each with at least 40 larvae). The relative transcript level of TcLgl was determined by RT-qPCR as described above at 24, 48 and 72 h after the injection.

70

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Table 1 Sequences and relevant parameters of the PCR primers used for cDNA subcloning and sequencing, RT-qPCR analysis, and dsRNA synthesis. Primer name

Sequence (50 e30 )

Tm ( C)

Product size (bp)

PCR for cDNA sequencing TcChc-1-F TcChc-1-R TcChc-2-F TcChc-2-R TcChc-3-F TcChc-3-R TcChc-4-F TcChc-4-R TcChc-5-F TcChc-5-R TcChc-6-F TcChc-6-R TcChc-7-F TcChc-7-R

ATCTTGGGTGTCGCAGTTTG GCTGTTACGAAAATTGTCTCGC TACGACGTCATCTACTTAATCACAAAA CGACATTAAGTTCATCTCCAGGA ATCAACCAAATTGTCGATATTTTCA GTGGACGCGACTCTCCAG GTAGACTGCGCCGAAGACAT CTCATACGCACGATCCAAATT GCTCCTGATATCGCAAATATCG ATTATCGTACTCCTCGTACTTATCGT ATTTCGAAGAGCTCATCGGTC GGCGATGCGTCTGAATTC GAGGTCGGTGCAGAATTTG TTTATTAATAGTAGCGTCATTTGAACC

61.10 61.14 60.33 60.49 60.77 60.58 60.43 59.97 60.91 57.98 60.73 60.32 59.23 58.38

1089

Reverse transcription quantitative PCR (RT-qPCR) TcLgl(Q)-F GACGGATGGCTTTTGCTA TcLgl(Q)-R CGGCATTCAACTGTCTCT TcChc(Q)-F CGACTTTCCCGTAGCGAT TcChc(Q)-R ATTCGTGCGGAGCTGTTA TcRps3-F CCGTCGTATTCGTGAATTGACTT TcRps3-R TCTAAGAGACTCTGCTTGTGCAATG

61.60 61.20 63.40 63.20 61.10 61.00

141

PCR for dsRNA synthesis dsGFP-F dsGFP-R dsTcLgl(T7)-F dsTcLgl(T7)-R dsTcChc(T7)-F dsTcChc(T7)-R dsTcAP50(T7)-F dsTcAP50(T7)-R dsTcVhaSFD(T7)-F dsTcVhaSFD(T7)-R dsTcRab7(T7)-F dsTcRab7(T7)-R dsTcLgl-50bp-F dsTcLgl-50bp-R dsTcLgl-100bp-F dsTcLgl-100bp-R dsTcLgl-200bp-F dsTcLgl-200bp-R dsTcLgl-396bp-F dsTcLgl-396bp-R dsTcLgl-799bp-F dsTcLgl-799bp-R

60.23 60.40 60.11 60.40 59.96 58.04 60.55 59.88 59.51 60.01 60.44 60.39 61.04 60.68 61.17 59.54 59.60 59.71 59.21 60.10 59.60 60.10

ggatcctaatacgactcactataggGTGACCACCCTGACCTAC ggatcctaatacgactcactataggTTGATGCCGTTCTTCTGC taatacgactcactatagggGACGTTGCAACACGGATTC taatacgactcactatagggTGTCATCGTAAAGCTTGCCA taatacgactcactatagggAAGGGCTTTTCTATTTTCTGGG taatacgactcactatagggGGCTGGGATTTACCTTTTGT taatacgactcactatagggGTACAACCACAAGGGCGAAG taatacgactcactatagggGCCGATTTGATCCCTTGTT taatacgactcactatagggGGCGGACCCTAAAATCAAG taatacgactcactatagggCCCCCATACCGATTCTTTTT taatacgactcactatagggGAAAGTCATCATTTTGGGCG taatacgactcactatagggCAAAAACGCCTGCTCCAC taatacgactcactatagggATGTGCGTCGAAACTGCTG taatacgactcactatagggCCACCCTCTGTGCCTAGAAGT taatacgactcactatagggGGCTATTTTTGAACAACCGGA taatacgactcactatagggTTTGATGGACGTGTTATCGC taatacgactcactatagggAGGCCGTGTGGTCACTCTAT taatacgactcactatagggTAAAGCTTGCCAAATCCAAAA taatacgactcactatagggTCTATGTGACGACAACTCCCTC taatacgactcactatagggGCCTCCAACTGTTGATTTGAA taatacgactcactatagggAGGCCGTGTGGTCACTCTAT taatacgactcactatagggGGCCAATACGACAAGAGCAT

2.6. Examination of cellular uptake of Cy3-labeled dsTcLgl in larval midgut Labeling of a 385-bp dsRNA of TcLgl with Cy3 was performed using Silencer siRNA Labeling kit (Ambion). After Cy3 dye was diluted to have the same absorbance as the Cy3-labeled dsTcLgl solution, it was used for florescence background control. Each of 20-day-old larvae was first injected in the hemocoel with 100 nl of an inhibitor solvent in a background control and in a positive control, and 100 nl each of the two inhibitors (CPZ, 0.8 mg/larva; BafA, 0.005 mg/larva) in two inhibitor treatments. At 6 h after the first injection, each larva was re-injected with 100 nl of Cy3 dye (i.e., solvent þ Cy3) in the background control or 100 nl of Cy3labeled dsTcLgl (100 ng/larva) in both the positive control (i.e., solvent þ Cy3-dsTcLgl) and the two inhibitor treatments (i.e., CPZ þ Cy3-dsTcLgl, BafA þ Cy3-dsTcLgl). The injected larvae were reared under the standard conditions without flour supplied to minimize the presence of gut contents. At 24 h after the second injection, midguts were dissected in sterile 1X PBS and washed twice with 1X PBS. Three biological replicates were carried out with eight midguts in each replicate. After the midguts

966 966 828 966 650 761

158 130

305 388 306 402 391 442 50 100 200 396 799

were fixed in 4% paraformaldehyde for 1 h at room temperature, they were washed three times with 1X PBST (PBSþ0.1% Tween) for 5 min each. The middle parts of the midguts were mounted onto a glass slide containing SlowFade Antifade reagent with DAPI (Invitrogen). Images were captured using a 550 nm (Cy3) and 360 nm (DAPI) laser for excitation and 570 nm (Cy3) and 450e460 nm (DAPI) for emission filter setting under a confocal microscope. Scanned images were processed using LSM (Carl Zeiss AIM; version 4.2). 2.7. RNAi of RNAi procedures All dsRNA samples were synthesized by using MEGAscript® RNAi Kit (Ambion). Each of 16-day-old larvae was injected in the hemocoel with 100 nl of dsGFP (100 ng/larva) in a negative control and in a positive control, and 100 nl of each of four dsRNA (dsTcChc, dsTcAP50, dsTcVhaSFD and dsTcRab7; 100 ng/larva) in each of the four treatments. After 48 h, each larva was re-injected with 100 nl of dsGFP in the negative control, but 100 nl of dsTcLgl in the positive control and all the four treatments. After 48 h, four larvae were collected from each replicate and pooled as a sample for

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

determining relative TcLgl transcript level by RT-qPCR. Each control and treatment was performed with five replicates (each with two technical replicates). 2.8. Statistical analysis The percent data of the relative gene expression were transformed using arcsine square root transformation, and then the transformed data were subjected to ANOVA followed by Tukey's honestly significant difference (HSD) multiple comparisons to separate the means among the treatments using ProStat software (Poly Software International, Pearl River, NY).

71

6 h with Cy3-dsTcLgl, significant fluorescence signals (red) were detected in midgut cells at 24 h as compared with the controls, in which larvae were injected with the inhibitor solvent followed by equivalent amount of Cy3 dye alone (background control) (Fig. 2). These results indicated that Cy3-dsTcLgl, but not Cy3 dye alone, was effectively up-taken by the midgut cells. However, when larvae were first injected with CPZ (0.8 mg/larva) or BafA (0.005 mg/larva) and 6 h later with Cy3-dsTcLgl, the signals were significantly reduced in the midgut dissected from the larvae at 24 h. These results clearly indicated that either CPZ or BafA can effectively diminish cellular uptake of dsTcLgl in midgut cells, which were in agreement with our results of inhibitory effects of CPZ and BafA on RNAi of TcLgl (Fig. 1).

3. Results 3.1. Effect of different endocytosis inhibitors on RNAi

3.3. Effect of silencing genes in clathrin-dependent endocyctosis on RNAi of marker gene

We used TcLgl as a marker gene and four relatively selective pharmacological endocytosis inhibitors (CPZ, MbCD, BafA and CCD) to identify specific endocytic pathways involved in cellular uptake of TcLgl dsRNA (dsTcLgl) in T. castaneum larvae. The functions of Lgl are known to be remarkably conserved in maintenance of cell polarity and regulation of cell proliferation in various organisms. We chose TcLgl as a marker gene in this study due to its broad expression in different developmental stages and tissue types, and high amenability to RNAi in T. castaneum (Xiao et al., 2014). Our initial studies indicated that all five different lengths (between 50 and 799 bp) of dsTcLgl can effectively suppress the expression of the target gene at 48 and 96 h (Supplementary Fig. 2). Therefore, we used a 385-bp dsTcLgl in our subsequent studies. When 100 ng of TcLgl dsRNA (dsTcLgl, 385 bp) in 100 nl of the inhibitor solvent was injected to each of 16-day-old larvae, the transcript level of TcLgl was suppressed by 91.9, 94.0 and 89.7% at 24, 48 and 72 h (Fig. 1), respectively, as compared with those of the control larvae injected with the same amount of the green fluorescent protein gene dsRNA (dsGFP, 305 bp) in the same inhibitor solvent. These results indicated an effective RNAi-induced suppression of TcLgl expression at all the three time points. However, when each larva was injected with a mixture of 0.8 mg CPZ and 100 ng dsTcLgl in a total volume of 100 nl, the transcript level of TcLgl was suppressed only by 9.9, 33.0 and 35.4% at 24, 48 and 72 h (Fig. 1), respectively, as compared with those of control larvae injected with dsTcLgl in inhibitor solvent. Similarly, when each larva was injected with a mixture of 0.005 mg BafA and 100 ng dsTcLgl in a total volume of 100 nl, the TcLgl transcript was suppressed only by 22.6, 26.2 and 48.2% at 24, 48 and 72 h (Fig. 1), respectively, as compared with those of the control larvae injected with dsTcLgl in inhibitor solvent. Because all the four inhibitors did not show any significant effect on TcLgl expression (Supplementary Fig. 1), our results indicated that both CPZ (0.8 mg/larva) and BafA (0.005 mg/larva) were potent inhibitors to effectively block RNAi of TcLgl. In contrast, neither MbCD nor CCD can effectively block RNAi of TcLgl because the transcript levels of TcLgl in the larvae injected with the mixture of 100 ng dsTcLgl and MbCD or CCD at all three doses were not significantly different from those of the control larvae injected with dsTcLgl in inhibitor solvent (Fig. 1).

To provide further evidences of clathrin-dependent endocytosis as a mechanism of cellular uptake of dsRNA in T. castaneum larvae, we used an “RNAi of RNAi” strategy to examine if suppression of the transcript level of each of four selected key genes in clathrindependent endocytosis (Table 2) by RNAi can diminish RNAi of TcLgl. TcChc encodes clathrin heavy chain as major component of clathrin-coat vesicle. Other key genes were TcAP50 (GenBank accession No. KJ476827) encoding medium chain (m2) of adaptor protein 2 (AP2) also known as clathrin coat assembly protein, TcVhaSFD (GenBank accession No. KJ476830) encoding the subunit H of vacuolar (Hþ)-ATPase responsible for acidification of intracellular compartments, and TcRab7 (GenBank accession No. KJ476829) encoding a ras-related protein (small GTPase) responsible for regulating the transport of uncoated vesicle from early to late endosomes. TcChc is a relatively large gene with 26,253 bp and putatively encodes a protein of 1684 amino acid residues (Fig. 3A; GenBank accession No. KJ476828). The deduced Chc protein sequences among different insect species are highly conserved (Fig. 3B). TcChc was expressed during all the developmental stages with peak expressions in the 2- to 4-day pupae and 5-day adults (Fig. 3C). TcChc was also highly expressed in all the four tissues examined, including the gut (midgut and hindgut), Malpighian tubules, fat bodies and carcass (Fig. 3D). TcChc was highly amenable to RNAi. Injections of dsTcChc at 50, 100 and 200 ng/ larva significantly suppressed the TcChc transcript up to 96 h in late larvae (Fig. 4). When 16-day-old larvae were injected with dsGFP (100 ng/ larva) and 48 h later with dsTcLgl (100 ng/larva), the TcLgl transcript level was reduced by 69.6% as compared with that of the control larvae injected with dsGFP and 48 h later again with dsGFP (100 ng/ larva each time) at 48 h (Fig. 5). However, when 16-day-old larvae were injected with each of the four dsRNAs (100 ng/larva) targeting TcChc, TcAP50, TcVhaSFD and TcRab7 and 48 h later with dsTcLgl (100 ng/larva), the TcLgl transcript levels were 33.1, 52.1, 39.4, and 25.7%, respectively, higher than that of the corresponding control larvae injected with dsGFP (100 ng/larva) and 48 h later with dsTcLgl (100 ng/larva). Thus, the pre-injections of all the four dsRNAs targeting clathrin-dependent endocytosis significantly diminished RNAi of TcLgl in the larvae (Fig. 5).

3.2. Effect of endocytosis inhibitors on dsRNA uptake in larval midgut cells

4. Discussion

We examined whether the two clathrin-dependent endocytosis inhibitors (CPZ and BafA) can block cellular uptake of dsTcLgl in larval midgut by using Cy3 dye-labeled dsTcLgl. When 20-day-old larvae were first injected with the inhibitor solvent alone and after

Since the detailed mechanism of RNAi was first reported in C. elegans about 17 years ago, RNAi has rapidly emerged as a powerful tool to study gene function, regulation and interaction at the cell and whole-organism levels in various organisms (Fire et al., 1998). The current rapid paces of identification of target genes and

72

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Fig. 1. Determination of endocytic pathway in dsRNA cellular uptake using selective inhibitors. Four selective endocytosis inhibitors, including chlorpromazine (CPZ), methyl-bcyclodextrin (MbCD), bafilomycin A1 (BafA) and cytochalasin-D (CCD), at three different doses (mg/larva) were used to examine their effect on RNAi of the marker gene TcLgl at 24, 48 and 72 h in 16-day-old larvae. Results are the mean and standard errors of three biological replicates; each with two technical replicates in RT-qPCR analysis. Different letters on the bars indicate statistically significant difference based on ANOVA followed by Tukey's HSD multiple comparison test (P < 0.05).

development of new dsRNA delivery systems will also soon lead to the applications of RNAi-based technologies for human disease therapeutics (Jain et al., 2014) and pest management (Zhu, 2013; Kim et al., 2015). Currently, however, our knowledge on cellular uptake of dsRNA in different organisms is very limited. In insects, it has been speculated that significant difference in RNAi efficiency among different taxonomic groups could be at least in part due to possible difference in cellular uptake of dsRNA (Huvenne and Smaggle, 2010). Because dsRNA must be first internalized in insect cells in order to trigger RNAi process regardless of dsRNA delivery methods, better understanding of the mechanism of cellular uptake of dsRNA in insects will help researchers develop effective RNAi-based technologies for managing various insect pests in different taxonomic groups. By using pharmacological inhibitors of different endocytic pathways in conjunction with RNAi of a marker gene (TcLgl), we demonstrated that RNAi of TcLgl can be dramatically diminished by both CPZ and BafA (Fig. 1), two selective inhibitors of clathrindependent endocytosis (Ivanov, 2008; Xu et al., 2003; Gagliardi et al., 1999). CPZ is a cationic amphiphilic drug that inhibits the

formation of clathrin-coated pits by reversible translocation of clathrin and its adaptor proteins from the plasma membrane to intracellular vesicles (Wang et al., 1993; Dudleenamjil et al., 2010), whereas BafA inhibits vacuolar H (þ)-ATPase responsible for acidifying a variety of intracellular compartments to varying degrees in eukaryotic cells (Dettmer et al., 2006). In contrast, both MbCD and CCD did not show any significant effect on RNAi of TcLgl. Because MbCD is known to inhibit caveolaedependent endocytosis whereas CCD is an inhibitor of phagocytosis in different organisms, our results suggest that caveolae-dependent endocytosis and phagocytosis don't seem to play any important role in cellular uptake of dsTcLgl in T. castaneum larvae. However, the lack of the effect on RNAi of TcLgl by MbCD and CCD could also be due to their lack of inhibition to caveolae-dependent endocytosis and phagocytosis in T. castaneum. Therefore, further studies would be necessary to demonstrate that MbCD and CCD under the given conditions can in fact inhibit caveolae-mediated endocytosis and phagocytosis, respectively. Nevertheless, because both the clathrin-dependent endocytosis inhibitors CPZ and BafA can significantly block the RNAi for TcLgl, our results suggest that

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

73

Fig. 2. Evaluation of cellular uptake of Cy3-labeled dsTcLgl in larval midgut. Two selective clathrin-dependent endocytosis inhibitors including chlorpromazine (CPZ) and bafilomycin-A1 (BafA) were used to examine their effect on cellular uptake of Cy3-labeled dsTcLgl. Solvent þ Cy3 represented a negative (background) control with injections of inhibitor solvent and then Cy3 dye 6 h later, whereas solvent þ Cy3-dsTcLgl represented a positive control with injections of inhibitor solvent and then Cy3-dsTcLgl (100 ng/larva) 6 h later. The two treatments were the pre-injections of CPZ (0.8 mg/larva) and BafA (0.005 mg/larva) followed by Cy3-dsTcLgl (100 ng/larva) 6 h later. The arrowhead in the positive control indicates the fluorescence signal of Cy3-dsTcLgl in midgut cells as detected at 24 h. Nuclei DAPI staining was used to visualize the midgut cells.

clathrin-dependent endocytosis plays a key role in cellular uptake of dsTcLgl. Indeed, the role of endocytic pathway in cellular uptake of dsRNA from the medium has been studied in D. melanogaster S2 cells (Saleh et al., 2006). BafA can effectively inhibit RNAi of luciferase reporter gene, whereas neither MbCD nor CCD can affect the RNAi. Apparently, caveolae-mediated endocytosis or phagocytosis does not seem to play any important role in dsRNA uptake from the medium by S2 cells. Thus, results from our in vivo studies are consistent with those of the in vitro studies with D. melanogaster S2 cells. Although the scavenger receptors have been known to play an important role in phagocytosis of bacterial pathogens (ErturkHasdemir and Silverman, 2005; Kocks et al., 2005) and dsRNA uptake in ticks (Aung et al., 2011), our results don't seem to support the involvements of phagocytosis in cellular uptake of dsRNA in T. castaneum larvae. By using Cy3-labeled TcLgl dsRNA (Cy3-dsTcLgl) in conjunction with each of the two selective inhibitors (CPZ and BafA) of clathrindependent endocytosis in the midgut of T. castaneum larvae, we

further demonstrated that the midgut cells can effectively take up Cy3-dsTcLgl (Fig. 2), which is consistent with a recent report that the midgut of Diabrotica virgifera virgifera can effectively uptake a Cy3-dsRNA (Bolognesi et al., 2012). However, such a cellular uptake can be effectively blocked by either CPZ or BafA. This result is in agreement with our first evidence that the injection of either CPZ or BafA can significantly diminish the dsTcLgl-induced repression of TcLgl transcript in T. castaneum larvae. Because our pharmacological studies using selective endocytosis inhibitors strongly suggested that clathrin-dependent endocytosis played a major role in cellular uptake of dsRNA, we took a different approach to examine if suppression of each transcript of the key genes in clathrin-dependent endocytosis by RNAi can diminish RNAi of TcLgl (i.e., an “RNAi of RNAi” strategy). Saleh et al. and Miller reported 23 and 28 candidate genes potentially involved in systemic RNAi in D. melanogaster S2 cells and T. castaneum, respectively (Saleh et al., 2006; Miller, 2009). We chose four genes known to play essential roles in clathrin-dependent endocytosis in this study.

74

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Table 2 Four key genes identified in T. castaneum and their roles in clathrin-dependent endocytosis. Genes identified in T. castaneum

Most similar genes found in other insect species by BLASTP search

Roles in clathrin-dependent endocytosis (Doherty and McMahon, 2009)

Species

Accession number

Amino acid identity (%)

TcChc

Bombyx mori Drosophila melanogaster Acyrthosiphon pisum Apis mellifera

NP_001136443.1 NP_477042.1 XP_001945333.2 XP_623111.3

87.9 87.1 83.5 88.2

Clathrin heavy chain assisting the formation of clathrin-coated pits on the inner surface of the cell plasma membrane.

TcAP50

Aedes aegypti Acyrthosiphon pisum Nasonia vitripennis Pediculus humanus corporis

XP_001663168.1 XP_001947094.1 XP_001606373.1 XP_002427003.1

96.6 91.4 94.6 95.9

Medium chain (m2) also known as AP50 of adaptor protein 2 (AP2) that links the embedded cargo with clathrin layer.

TcVhaSFD

Aedes aegypti Acyrthosiphon pisum Bombyx mori Apis florae

XP_001652018.1 XP_001949116.2 NP_001040488.1 XP_003692426.1

80.8/73.6a 73.0/67.1 74.3/68.5 75.3/75.2

Subunit H of vacuolar (Hþ)-ATPase responsible for acidification essential for ligand-receptor dissociation and receptor recycling.

TcRab7

Aedes aegypti Bombyx mori Acyrthosiphon pisum Nasonia vitripennis

XP_001663804.1 NP_001040368.1 NP_001155484.1 XP_001607917.1

81.6 82.1 80.8 87.2

A small GTPase that plays a crucial role in the regulation of transport cargo from early to late endosomes.

a

Alternative splicing variants: TcVhaSFD1/TcVhaSFD2.

When we injected 16-day-old T. castaneum larvae with the dsRNA of TcChc, TcAP50, TcVhaSFD or TcRab7, and 48 h later with the dsRNA of TcLgl, we found significantly increased TcLgl transcript levels as compared with the larvae injected with GFP dsRNA and then TcLgl dsRNA (Fig. 5). These results indicated that suppressing the transcript levels of the clathrin-dependent endocytosis related genes by RNAi can significantly diminish RNAi of TcLgl. These results also support our conclusion that clathrin-dependent endocytosis plays an important role in

cellular uptake of TcLgl dsRNA. Our results are also supported by the report that scavenger receptor-mediated endocytosis contributes to environmental RNAi in the desert locust (Wynant et al., 2014). In T. castaneum, three homologs of C. elegans sid-1 have been identified to share high sequence identities with another C. elegans gene (tag-130), but they do not seem to play any important role in dsRNA uptake because silencing of these genes doesn't affect RNAi of other genes (Tomoyasu et al., 2008). Nevertheless, we were able

Fig. 3. Molecular analyses of the clathrin heavy chain gene (TcChc) in T. castaneum. (A) Schematic diagram showing the organization of TcChc gene (26,253 bp). (B) Rooted phylogenetic tree of deduced Chc amino acid sequences from Metaseiulus occidentalis and 19 insect species constructed by the neighbor-jointing method. Following sequences were used in the analysis: TcChc (KJ476828, T. castaneum in this study); AaChc (XP_001656876.1, Aedes aegypti); AeChc (EGI60613.1, Acromyrmex echinatior); AfChc (XP_003690808.1, Apis florae); AgChc (XP_311856.3, Anopheles gambiae); AmChc (XP_623111.3, Apis mellifera); ApChc (XP_001945333.2, Acyrthosiphon pisum); BiChc (XP_003486702.1, Bombus impatiens); BmChc (NP_001136443.1, Bombyx mori); BtChc (XP_003402443.1, Bombus terrestris); CfChc (EFN61414.1, Camponotus floridanus); CqChc (XP_001864930.1, Culex quinquefasciatus); DmChc (NP_477042.1, Drosophila melanogaster); DpChc (EHJ79063.1, Danaus plexippus); HsChc (EFN80585.1, Harpegnathos saltator); MoChc (XP_003740941.1, Metaseiulus occidentalis); MrChc (XP_003706348.1, Megachile rotundata); NvChc (XP_001607995.1, Nasonia vitripennis); PhChc (XP_002424037.1, Pediculus humanus corporis); and SiChc (EFZ18330.1, Solenopsis invicta). (C) Developmental stage-dependent expression pattern of TcChc determined by RT-qPCR. E1, E2, E3, and E4 represent 1, 2, 3 and 4-day eggs; L1, L5, L10, L15, and L20 represent 1, 5, 10, 15 and 20-day larvae; P1, P2, P3, P4, P5, and P6 represent 1, 2, 3, 4, 5 and 6-day pupae; and A1, A5, A10, A15, and A20 represent 1, 5, 10, 15 and 20day adults, respectively. (D) Tissue-dependent expression pattern of TcChc determined by RT-qPCR. Each of the four tissue types, the gut (midgut and hindgut), Malpighian tubule, fat body and carcass (the remaining body after the brain, ganglia, gut and fat bodies removed), were dissected from late pupae.

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Fig. 4. Time (48 and 96 h) and TcChc dsRNA dose (50, 100 and 200 ng/larva)-dependent suppression of TcChc in late larvae. In all the RT-qPCR analyses, ribosomal protein S3 gene (TcRps3) was used as an internal reference. Different letters on the bars of the histograms in figure panels c and d indicate statistically significant difference based on ANOVA followed by Tukey's HSD multiple comparison test (P < 0.05).

to almost completely abolish RNAi targeting TcLgl at 24 and 48 h when the larvae were injected with a mixture of CPZ (0.8 mg/larva) and dsTcLgl (100ng/larva) (Fig. 1). Taken together, these results clearly indicate that clathrin-dependent endocytosis rather than

75

SID-1 plays a predominant role in cellular uptake of dsRNA in T. castaneum larvae. Indeed, clathrin-dependent endocytosis has been known as a major pathway for the internalization of lipids and receptorbound macromolecules into eukaryotic cells and is critical for many biological processes including nutrient uptake and cell signaling (Conner and Schmid, 2003). For cellular uptake of dsRNA, we adapt the general clathrin-dependent endocytic pathway (Doherty and McMahon, 2009) as shown in Fig. 6. The process includes: 1) dsRNA molecules first bind to unknown receptor proteins on the plasma membrane of the cell; 2) the dsRNA and receptor protein complexes interact with adaptor protein complex 2 (AP-2) and clathrin to form a clathrin-coated pit on inner surface of the plasma membrane; 3) the pit then buds into the cell to form a clathrin-coated vesicle; 4) the vesicle is uncoated and fuses with early endosome; and 5) the early endosome matures into a late endosome before fusing with lysosome. In this pathway, dsRNA molecules must escape form endosomes into the cytosol, where they trigger the RNAi process through the RNAi core machinery (Fig. 6). Thus, our finding of clathrin-dependent endocytosis as a major mechanism of cellular uptake of dsRNA has also provided insights into the development of novel strategies to improve RNAi efficiency by promoting endosomal release of dsRNA, which could include the use of fusogenic lipids, polymers with high buffering capacity, and membraneinteracting peptides (Dominska and Dykxhoorn, 2010; Liang and Lam, 2012).

Fig. 5. Role of clathrin-dependent endocytosis genes in RNAi. Effect of four key genes (TcChc, TcAP50, TcVhaSFD and TcRab7) in clathrin-dependent endocytic pathway on RNAi of marker gene (TcLgl) was evaluated in 16-day-old larvae. Results are the mean and standard errors of five replicates (each with two technical replicates in RT-qPCR analysis). Different letters above the standard error bars indicate significant differences based on ANOVA followed by Tukey's HSD multiple comparison test (P < 0.05).

76

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77

Fig. 6. Clathrin-dependent endocytosis in cellular uptake of dsRNA. Clathrin, a large three-legged cytosolic protein consisting of three heavy (Chc) and three light chains, assists in the formation of clathrin-coated pits on the inner surface of the cell plasma membrane. The pit then buds into the cell to form a clathrin-coated vesicle. The formation of clathrincoated vesicle also requires adaptor protein complex 2 (AP-2) which is a heterotetramer composed of two large adaptins (alpha-type subunit and beta-type subunit), a medium adaptin (mu-type subunit known as AP50) and a small adaptin (sigma-type subunit). After extracellular dsRNA molecules have been internalized in the cell, lysosomes and endosomes are acidified by the proton pumps, vacuolar H(þ)-ATPase, which is a heteromultimeric enzyme composed of a peripheral catalytic V1 complex attached to an integral membrane V0 proton pore complex, where VhaSFD is a regulatory subunit of vacuolar H(þ)-ATPases. Rab7 is a ras-related protein (a small GTPase) and plays a key role in regulating endo-lysomal trafficking and early-to-late endosomal maturation. The dsRNA molecules must escape form endosomes into the cytosol, where they trigger the RNAi process through the RNAi core machinery.

Acknowledgments We thank Yoonseong Park (KSU Department of Entomology) and Joel Sanneman (KSU CVM Confocal Microscopy Core Facility) for providing confocal microscopy facilities and technical assistance. The Confocal Microscopy and Microfluorometry Core are supported by KSU-CVM. This research was supported by the Kansas Agricultural Experiment Station and the U.S. Department of Agriculture (USDA/NIFA 2014-67013-21714) to KYZ, and the China Scholarship Council to DX. This manuscript is contribution No. 15202-J from the Kansas Agricultural Experiment Station. The T. castaneum voucher specimens (voucher No. 159) are located in the Kansas State University Museum of Entomlogical and Prairie Arthropod Research, Manhattan, Kansas, USA. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ibmb.2015.03.009. References Aronstein, K., Pankiw, T., Saldivar, E., 2006. SID-1 is implicated in systemic gene silencing in the honey bee. J. Apicult. Res. 45, 20e24.

Aung, K.M., Boldbaatar, D., Umemiya-Shirafuji, R., Liao, M., Xuenan, X., Suzuki, H., Galay, R.L., Tanaka, T., Fujisaki, K., 2011. Scavenger receptor mediates systemic RNA interference in ticks. PLoS One 6, e28407.
s, X., 2010. Beyond Drosophila: RNAi in vitro and functional genomics in insects. Belle Annu. Rev. Entomol. 55, 111e128. Bolognesi, R., Ramaseshadri, P., Anderson, J., Bachman, P., Clinton, W., Flannagan, R., Ilagan, O., Lawrence, C., Levine, S., Moar, W., Mueller, G., Tan, J., Uffman, J., Wiggins, E., Heck, G., Segers, G., 2012. Characterizing the mechanism of action of double-stranded RNA activity against western corn rootworm (Diabrotica virgifera virgifera LeConte). PLoS One 7, e47534. Conner, S.D., Schmid, S.L., 2003. Regulated portals of entry into the cell. Nature 422, 37e44. Dettmer, J., Hong-Hermesdorf, A., Stierhof, Y.D., Schumacher, K., 2006. Vacuolar HþATPase activity is required for endocytic and secretory trafficking in Arabidopsis. Plant Cell 18, 715e730. Doherty, G.J., McMahon, H.T., 2009. Mechanisms of endocytosis. Annu. Rev. Biochem. 78, 857e902. Dominska, M., Dykxhoorn, D.M., 2010. Breaking down the barriers: siRNA delivery and endosome escape. J. Cell Sci. 123, 1183e1189. Dudleenamjil, E., Lin, C.Y., Dredge, D., Murray, B.K., Robison, R.A., Johnson, F.B., 2010. Bovine parvovirus uses clathrin-mediated endocytosis for cell entry. J. Gen. Virol. 91, 3032e3041. Erturk-Hasdemir, D., Silverman, N., 2005. Eater: a big bite into phagocytosis. Cell 123, 190e192. Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello, C.C., 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabtidis elegans. Nature 39, 806e811. Gagliardi, S., Rees, M., Farina, C., 1999. Chemistry and structure activity relationships of bafilomycin A1, a potent and selective inhibitor of the vacuolar Hþ-ATPase. Curr. Med. Chem. 6, 1197e1212. Huvenne, H., Smaggle, G., 2010. Mechanisms of dsRNA uptake in insects and potential of RNAi for pest control: a review. J. Insect Physiol. 56, 227e235.

D. Xiao et al. / Insect Biochemistry and Molecular Biology 60 (2015) 68e77 Ivanov, A.I., 2008. Pharmacological inhibition of endocytic pathways: is it specific enough to be useful? Methods Mol. Biol. 440, 15e33. Jain, C.K., Gupta, A., Dogra, N., Kumar, V.S., Wadhwa, G., Sharma, S.K., 2014. MicroRNA therapeutics: the emerging anticancer strategies. Recent Pat. Anticancer Drug Discov. 9, 286e296. Kim, Y.H., Soumaila Issa, M., Cooper, A.M.W., Zhu, K.Y., 2015. RNA interference: applications and advances in insect toxicology and insect pest management. Pestic. Biochem. Physiol. http://dx.doi.org/10.1016/j.pestbp.2015.01.002. Kocks, C., Cho, J.H., Nehme, N., Ulvila, J., Pearson, A.M., Meister, M., Strom, C., Conto, S.L., Hetru, C., Stuart, L.M., 2005. Eater, a transmembrane protein mediating phagocytosis of bacterial pathogens in Drosophila. Cell 123, 335e346. Liang, W., Lam, J.K.W., 2012. Endosomal escape pathways for non-viral nucleic acid delivery systems. In: Ceresa, B. (Ed.), Molecular Regulation of Endocytosis. InTech, Rijeka, Croatia, pp. 421e467. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2-DDCt method. Methods 25, 402e408. Luo, Y., Wang, X., Yu, D., Kang, L., 2012. The SID-1 double-stranded RNA transporter is not required for systemic RNAi in the migratory locust. RNA Biol. 9, 663e671. Mello, C.C., Conte, D., 2004. Revealing the world of RNA interference. Nature 431, 338e342. Miller, S.C., 2009. RNA Interference in the Red Flour Beetle Tribolium castaneum. Ph.D. Dissertation. Kansas State University, Manhattan, KS. Ren, R., Xu, X., Lin, T., Weng, S., Liang, H., Huang, M., Dong, C., Luo, Y., He, J., 2011. Cloning, characterization, and biological function analysis of the SidT2 gene from Siniperca chuatsi. Dev. Comp. Immunol. 35, 6692e6701. Saleh, M.C., van Rij, R.P., Hekele, A., Gillis, A., Foley, E., O'Farrell, P.H., Andino, R., 2006. The endocytic pathway mediates cell entry of dsRNA to induce RNAi silencing. Nat. Cell. Biol. 8, 793e802. Tomoyasu, Y., Miller, S.C., Tomita, S., Schoppmeier, M., Grossmann, D., Bucher, G., 2008. Exploring systemic RNA interference in insects: a genome-wide survey for RNAi genes in Tribolium. Genome Biol. 9, R10.

77

€met, M., Ulvila, J., Parikka, M., Kleino, A., Sormunen, R., Ezekowitz, R.A., Kocks, C., Ra 2006. Double-stranded RNA is internalized by scavenger receptor-mediated endocytosis in Drosophila S2 cells. J. Biol. Chem. 281, 14370e14375. Wang, L.H., Rothberg, K.G., Anderson, R.G., 1993. Mis-assembly of clathrin lattices on endosomes reveals a regulatory switch for coated pit formation. J. Cell Biol. 123, 1107e1117. Winston, W.M., Molodowitch, C., Hunter, C.P., 2002. Systemic RNAi in C. elegans requires the putative transmembrane protein SID-1. Science 295, 2456e2459. Wu, K., Hoy, M.A., 2014. Clathrin heavy chain is important for viability, oviposition, embryogenesis and, possibly, systemic RNAi response in the predatory mite Metaseiulus occidentalis. PLoS One 9, e110874. Wynant, N., Santos, D., Van Wielendaele, P., Broeck, J.V., 2014. Scavenger receptormediated endocytosis facilitates RNA interference in the desert locust, Schistocerca gregaria. Insect Mol. Biol. 23, 320e329. Xiao, D., Liang, X., Gao, X., Yao, J., Zhu, K.Y., 2014. The lethal giant larvae gene in Tribolium castaneum: molecular properties and roles in larval and pupal development as revealed by RNA interference. Int. J. Mol. Sci. 15, 6880e6896. Xu, J., Feng, H.T., Wang, C., Yip, K.H., Pavlos, N., Papadimitriou, J.M., Wood, D., Zheng, M.H., 2003. Effects of Bafilomycin A1: an inhibitor of vacuolar H (þ)-ATPases on endocytosis and apoptosis in RAW cells and RAW cell-derived osteoclasts. J. Cell. Biochem. 88, 1256e1264. Zhang, X., Zhang, J., Zhu, K.Y., 2010. Advances and prospects of RNAi technologies in insect pest management. In: Liu, T.X., Kang, L. (Eds.), Recent Advances in Entomological Research: from Molecular Biology to Pest Management. Higher Education Press, Beijing, pp. 202e208. Zhu, K.Y., 2013. RNA interference: a powerful tool in entomological research and a novel approach for insect pest management. Insect Sci. 20, 1e3. Zhuang, J.J., Hunter, C.P., 2011. Tissue-specificity of Caenorhabditis elegans enhanced RNAi mutants. Genetics 188, 235e237.