The cabbage beetle Colaphellus bowringi Baly is a major pest of crucifers in mountainous areas of China. The field studies did not involve endangered or protected species, and no specific permissions were required for our research activities in these locations.

Cabbage beetles, C. bowringi, were collected from the field in Xiushui County (29°1′N, 114°4′E), Jiangxi Province, China, and maintained in the laboratory since late November 2008 (Xue et al., 2002a; Ma et al., 2011). Experiments were conducted on the adult, female offspring of post-diapause adults that had emerged from the soil in early March 2013. All larvae and adults were kept in incubators (SPX–250IC; Boxun Medical Instruments, Shanghai, China) and fed on fresh radish leaves (Raphanus stativus var. longipinnatus) that were provided daily (Wang et al., 2004; Tan et al., 2016). Diapause-destined (DD) and non-diapause-destined (NDD) beetles were generated by rearing larvae under two different photoperiods; light:dark (L:D) 16:8, and L:D 12:12, at 25°C respectively (Figure 1) (Xue et al., 2002a,b; Wang et al., 2004). Sex was determined during the pupal stage (Wang et al., 2006) and only female adults were used in these experiments. The heads of newly emerged female adults (0 day) that had not yet had the opportunity to feed were collected and placed in 1.5 mL micro centrifuge tubes that were immediately frozen in liquid nitrogen and stored at −80°C. Six replicate samples, comprised of pools of at least fifty DD or NDD heads, were collected. Three of these replicates were used for protein identification and the other three for quantitative Real Time PCR (qRT-PCR). Meanwhile, three replicate groups of beetles were reared continuously at L:D 16:8 and L:D 12:12 to confirm their diapause status. The determination of diapause attributes were made after female adults feed for 7 days at 25°C in transparent plastic containers filled with soil in their respective photoperiod treatments. Diapausing beetles exhibit digging behavior, their ovarian development is arrested, and their abdomens clearly show their hypertrophied fat bodies (Xue et al., 2002b; Wang et al., 2004).

Head samples from cabbage beetles under each photoperiod were ground to a fine powder with a mortar and pestle under liquid nitrogen. Then, Lysis buffer containing phenylmethylsulphonyl fluoride (PMSF) (1 mM final concentration) and ethylenediaminetetraacetic acid (EDTA) (2 mM) was added, and after 5 min, 10 mM dithiothreitol (DTT) was added. The suspension was sonicated at 200 W for 15 min and then centrifuged for 20 min at 4°C, 25,000 × g. The supernatant was mixed well with 5 × volume of chilled acetone and incubated at −20°C overnight. After centrifugation for 20 min at 4°C, 16,000 × g, the supernatant was discarded. The precipitate was washed with chilled acetone three times. The pellet was air-dried and dissolved in Lysis buffer (7M urea, 2 M thiourea, 4% NP-40 Nonidet (NP-40), 20 mM Tris-HCl, pH 8.0–8.5). The suspension was sonicated at 200 W for 15 min and centrifuged for 20 min at 4°C, 25,000 × g. The supernatant was transferred to another tube. Protein disulphide bonds were reduced by treatment with 10 mM DTT at 56°C for 1 h and the cysteine was blocked by using 55 mM iodoacetamide in a dark room for 45 min. Five times the volume of the supernatant in chilled acetone was added and incubated for 2 h at −20°C. After centrifugation for 20 min at 25,000 × g, the supernatant was discarded, and the pellet was air-dried for 5 min. The pellet was dissolved in 200 μL 0.5 M tetraethyl-ammonium bicarbonate and sonicated for 15 min. Samples were centrifuged for 20 min at 25,000 × g and the protein concentration of the supernatant was measured by the Bradford method (Bradford, 1976). Then, a volume containing 100 μg of protein was taken out from each sample solution and treated with Trypsin Gold with the ratio of protein: trypsin = 20:1 to digest the proteins by incubating for 4 h at 37°C. Because the initial 4 h digestion step did not produce complete fragmentation, we additionally exposed the partially digested proteins again to Trypsin Gold with the ratio of protein: trypsin = 20:1, continuing digestion for an additional 8 h at 37°C.

After trypsin digestion, peptides were labeled with iTRAQ reagents. Briefly, desalted samples were reconstituted in 30 μL of dissolution buffer and mixed with two different isobaric tags. Peptides from the heads of DD individuals and heads of NDD individuals were labeled with tags of mass 115 and 116 respectively using the iTRAQ Reagent 8-plex Kit according to the manufacturer's protocol. Labeled samples were pooled together and dried using a speedvac.

For SCX chromatography we used a Shimadzu LC-20AB high performance liquid chromatography (HPLC) system where peptides from the digestion step were reconstituted with 4 mL of buffer A (25 mM NaH₂PO₄ in 25% acetonitrile (ACN), pH 2.7) and loaded onto a 4.6 × 250 mm Ultremex SCX column containing 5-μm particles (Phenomenex). Peptides were eluted at a flow rate of 1 mL/min with a gradient of buffer A for 10 min, 5–60% buffer B (25 mM NaH₂PO₄, 1 M KCl in 25% ACN, pH 2.7) for 27 min, and then 60–100% buffer B for 1 min. The system was then maintained in 100% buffer B for 1 min before equilibrating with buffer A for 10 min prior to the next injection. Elution was monitored by measuring absorbance at 214 nm, and fractions were collected every 1 min. Eluted peptides were pooled into 20 distinct fractions, desalted with a Strata X C18 column (Phenomenex), and vacuum-dried.

Each fraction was resuspended in buffer A (5% ACN, 0.1% formic acid) and centrifuged at 20,000 x g for 10 min. In each fraction, the final concentration of peptides was ~0.5 μg/μL. Ten μL of supernatant was loaded on an Shimadzu LC-20AD nanoHPLC by an autosampler onto a 2 cm C18 trap column (inner diameter 200 μm) and the peptides were eluted onto a custom 10 cm analytical resolving C18 column (inner diameter 75 μm) made in-house. The samples were loaded at 8 μL/min for 4 min, then the 35 min gradient was run at 300 nL/min starting from 2 to 35% buffer B (95% ACN, 0.1% FA), followed by 5 min linear gradient to 60%, then, followed by 2 min linear gradient to 80%, and maintenance at 80% B for 4 min, and finally returning to 5% B in 1 min.

Data acquisition was performed with a Triple TOF 5600 System (AB SCIEX, Concord, ON) fitted with a Nanospray III source (AB SCIEX, Concord, ON) and a pulled quartz tip as the emitter (New Objectives, Woburn, MA). Data were acquired using an ion spray voltage of 2.5 kV, curtain gas of 30 PSI, nebulizer gas of 15 PSI, and an interface heater temperature of 150°C. The MS was operated with a RP of greater than or equal to 30,000 FWHM for TOF MS scans. For IDA, survey scans were acquired in 250 ms and as many as 30 product ion scans were collected if exceeding a threshold of 120 counts per second (counts/s) and with a 2+ to 5+ charge-state. Total cycle time was fixed to 3.3 s. The Q2 transmission window was 100 Da for 100%. Four time bins were summed for each scan at a pulse frequency value of 11 kHz by monitoring the 40 GHz multichannel TDC detector with four-anode channel detections. A sweeping collision energy setting of 35 ± 5 eV adjusted rolling collision energy was applied to all precursor ions for collision-induced dissociation. Dynamic exclusion was set for 1/2 of the peak width (15 s), and then the precursor was refreshed off the exclusion list.

Mascot software (version 2.3.02, Matrix Science, London, UK) was used to identify and quantify proteins. Only unique peptides were selected for quantification. After proteins had been quantified, the de novo transcriptome database of C. bowringi deposited into the NCBI Sequence Read Archive (Accession: SRP026471) (Tan et al., 2015), NCBI Tribolium castaneum database (54364 seqs, download from http://www.ncbi.nlm.nih.gov/protein?term=txid7070%5BOrganism%5D&cmd=search), NCBInr, SwissProt, and UniProt, were all searched for matches. The MGF (Mascot generic format) search file consisted of Spectra from 10 fractions. We parameterized searches to include single missed cleavages with trypsin; carbamidomethylation at Cys and oxidation at Met selected as fixed modifications and variable modifications, respectively. The tolerances for peptides and MS/MS were set at 10 ppm and 0.05 Da, respectively. Filters were set as follows: significance threshold P < 0.05 (with 95% confidence) and ion score or expected cut-off less than 0.05 (with 95% confidence) for protein identification. The setting “median” was chosen for protein ratio type (http://www.matrixscience.com/help/quant_config_help.html); minimum precursor charge was set to 2+, and the minimum peptides value was set to 2 with only unique peptides used to quantify proteins. Median intensities were used for normalization and outliers were removed automatically. The peptide threshold was set as above for identity. We only used ratios with p < 0.05, and only fold changes of ≥1.2 were considered as significant.

qRT-PCR analysis was used to quantify transcript abundance for target proteins. Total RNA was extracted using TRIzol (TaKaRa Bio., Dalian, China) following the manufacturer's protocol. Concentration and purity of the total RNA isolated from different samples were determined using a NanoDrop 2000 (Thermo Scientific, Wilmington, DE, USA). One microgram of total RNA was used to synthesize first-strand cDNA using the PrimeScriptRT reagent kit with gDNA Eraser (Perfect Real Time) (TaKaRa Bio, Dalian, China), according to the manufacturer's protocol. Synthesized cDNA was stored at −20°C. Quantitative PCR reactions were performed using a MyIQ2 Two-color Real-time PCR Detection System (Bio-Rad, USA) and SYBR Premix Ex Taq II (TaKaRa, Dalian, China). Ribosomal protein L19 (RPL19) was amplified for internal standardization based on our previous evaluation of reference genes in C. bowringi (Tan et al., 2015). Primers for gene amplification used in qRT-PCR are shown in Table S1. cDNA samples were derived from each of the three independent biological replicates for each experimental condition, and each biological replicate was analyzed with three technical replicates. Relative qRT-PCR expression data were calculated using the 2^−ΔΔCT method (Schmittgen and Livak, 2008). Eight proteins, namely FABP (Gerstner et al., 2008), alcohol dehydrogenase (Fan et al., 2013), U6 snRNA-associated Sm-like protein LSm3, protein phosphatase 1 (Gonsalvez et al., 2008), catalytic subunit (Bao and Xu, 2011), cytoglobin-1-like (Hoogewijs et al., 2007), tricarboxylate transport protein (Dwivedi et al., 2014), dolichyl-diphosphooligosaccharide (Bao et al., 2010), and protein disulfide isomerase (Filonova et al., 2013), were selected for verifying the protein expression patterns.

The gene function databases complied for the Gene Ontology (GO), Cluster of Orthologous Groups of proteins (COG), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were used to infer putative functions of identified proteins. Detailed information can be found on http://www.geneontology.org. COG is the database for orthologous protein classification. We compared identified proteins with the COG protein database to predict the function of proteins and conduct statistical analyses. For KEGG pathways (Kanehisa et al., 2004), we deduced the main biochemical metabolism and signal transduction pathways of identified proteins by pathway analysis. Hypergeometric tests were applied to assess the significance of over-representation for predefined protein sets (p ≤ 0.05).

A 321-bp fragment of C. bowringi FABP was amplified by PCR using the corresponding primers (Table S1) and purified with phenol-chloroform. One microgram of PCR product was used as the template for double-stranded RNA (dsRNA) synthesis using the T7 transcription kit (Fermentas, Lithuania) according to the manufacturer's protocol. dsRNA was precipitated in ethanol overnight, resuspended in RNase-free water, and quantitated at 260 nm using a NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific Inc.) before microinjection. The quality and integrity of dsRNA were determined by agarose gel electrophoresis. Microinjection was performed using a Nanoliter 2010 microinjection system (Word Precision Instruments INC, Sarasota, FL 34240-9258 USA). The dsRNA of green fluorescent protein (dsGFP) was used as the control. Two microgram of either FABP or GFP dsRNA in 200 nL was microinjected into the ventral abdomen of each newly emerged female adult before they had fed (day 0). Total RNA from the fat body or head was extracted after 2 and 4 days post-treatment to determine whether RNAi treatments affected FABP or heat shock protein transcript abundance. After 7 days post-treatment, the survival rates and diapause incidence were determined for beetles in this study.

The fat body was dissected from individual beetles in a Petri dish on wax containing phosphate buffer solution (PBS) (Zhou et al., 2010) and photographed with a digital camera (Nikon D5100, Nikon Imaging (China) Sales, China) mounted on a stereo-microscope (SMZ-t4, Chong Qing Optec Instrument, China).

The number of lipid droplets in fat body cells were visualized after Oil Red O staining that can be used to quantify fat concentration by staining intracytoplasmic lipids (Ramirez-Zacarias et al., 1992; Koopman et al., 2001; Singh et al., 2009). Fat bodies were fixed in 4% paraformaldehyde solution for over 24 h at 4°C and the fixed tissues were then cut into 8–10 μm thick sections at −20°C (Tissue-Tek; Sakura Finetechnical, Tokyo, Japan) and mounted on slides. The slides were air dried for 10 min at room temperature, fixed in ice cold 4% paraformaldehyde for 15 min, and then washed 3 times with PBS. Slides were incubated in pre-warmed Oil Red O solution for 10–15 min and again washed 3 times with PBS. Slides were then stained in Oil Red O solution for 1–2 h in a 37°C oven, differentiated in 75% alcohol solution for 2 s, and washed with distilled water for 1 min. Slides were then counterstained for 1–2 min in Harris hematoxylin, rinsed in tap water and differentiated in 1% hydrochloric acid alcohol solution for a few seconds. After mounting in glycerogelatin, sections were observed under a phase contrast microscope (Nikon eclipse CI). Lipid droplets in adipocytes stained red whereas cell nuclei stained blue. Lipid droplets were quantified by measuring the intensity of the red signal using ImageJ software (Schneider et al., 2012).

One-way analysis of variance (ANOVA) was used to evaluate FABP and heat shock protein transcript abundance. Differences between means were examined using Tukey's Least Significant Difference (LSD) test. Differences in transcript abundance and lipid droplet content were evaluated using the independent-samples t-test. P < 0.05 were considered statistically significant.

Under diapause-inducing conditions (L:D 16:8 and 25°C), 100% of DD females entered diapause, these diapausing beetles dug into the soil, their ovarian development was arrested, and their abdomens were full of hypertrophied fat bodies. At the same time, 8.9% of NDD females entered diapause under conditions not expected to induce diapause (L:D 12:12 and 25°C). Three replicate iTRAQ-LC-MS/MS assays identified a total of 3,175 unique proteins (confidence > 95%) from 62,030 spectra, of which 9,512 peptides were unique (Table 1). The full list of these proteins is available in Table S2. Comparison with the GO database indicated that 23, 16, and 14 protein groups could be assigned to the Biological process (BP), Cellular component (CC) and Molecular function (MF) categories, respectively (Figure S1). Among these categories, cellular process, cell part, and protein binding were the most abundant sub-categories in BP, CC, and MF respectively.

We identified 297 differentially expressed proteins (9.4% of total proteins ≥1.2-fold change, P < 0.05) (Table S3), 141 of which were up-regulated and 156 were down-regulated in the heads of DD females.

To verify these apparent differences in protein expression, we used qRT-PCR to analyze transcript abundance of 8 selected proteins in the heads of another cohort of cabbage beetles.

For proteins that were in greater abundance in diapause-destined females, FABP, protein phosphatase 1, catalytic subunit, tricarboxylate transport protein, and protein disulfide isomerase, all seven showed higher transcript abundance. For the three proteins we selected that were down-regulated in diapause-destined females, alcohol dehydrogenase and dolichyl-diphosphooligosaccharide, both had lower transcript abundance as expected, but transcript abundance for U6 snRNA-associated Sm-like protein LSm3 was not detectably higher in females not destined for diapause, although there was a trend toward higher transcript abundance in females not destined for diapause that is consistent with our proteomic data (Figure 2).

Of the differentially expressed proteins we identified, 138 could be classified in the COG database. Among these, 82 up-regulated proteins could be assigned to 19 COG categories and 56 down-regulated proteins to 17 COG categories (Figure 3). Many abundant up-regulated proteins in the heads of DD females were involved in energy production and conversion, carbohydrate metabolism and transport, and lipid metabolism (Table 2, Table S3). This, in turn, suggests that many energetic metabolic pathways are vigorously triggered even early during diapause preparation.

Because lipid metabolism pathways are particularly important during the diapause preparation phase (Sim and Denlinger, 2009; Hahn and Denlinger, 2011), we matched differentially expressed protein sequences to the reference canonical lipid metabolism pathways in the KEGG database to identify potential diapause-associated proteins involved in lipid metabolism. A total of 9 proteins could be assigned to 6 lipid metabolic pathways, of which 7 proteins were up-regulated in the heads of DD females (Table 3). Interestingly, we found that FABP was assigned to both the peroxisome proliferator-activated receptor (PPAR) signaling pathway, which can regulate lipid metabolism, and the fat digestion and absorption pathway. In addition, when mapping to vertebrate proteins, the beetle FABP was assigned to both FABP1 (liver) and FABP7 (brain) in the PPAR signaling pathway (Schoonjans et al., 1996; Gilde et al., 2003). This suggests that lipid metabolism pathways are upregulated, and that FABP may regulate lipid metabolism, during the onset of diapause preparation.

Beyond automated enrichment of functional groups and our interrogation of lipid metabolism, we also inspected the lists of differentially expressed proteins to detect additional patterns. One apparent difference is that non-diapause destined individuals had greater abundance of cuticular proteins expressed, perhaps indicating a trajectory of faster post-emergence growth than in the diapause-destined individuals. Inspection of proteins significantly more abundant in the heads of diapause-destined females revealed several previously associated with increased stress hardiness including a putative small heat shock protein, a putative peptidoglycan recognition protein that could be involved in increased pathogen resistance, as well as glutathione S-transferase and thioredoxin, two proteins that have been implicated in ameliorating both oxidative stress and xenobiotic stress. Similar processes of increased hardiness to pathogens, oxidative stress and heat shock protein expression have commonly been found in other “omics” studies of insect diapause (Ragland et al., 2010; Poelchau et al., 2013; King and MacRae, 2015; Poupardin et al., 2015; Qi et al., 2015; Meyers et al., 2016 as some examples). Two proteins that stand out as being associated with development of synaptic activity in nervous systems, bruchpilot and syntaxin (Broadie et al., 1995; Wagh et al., 2006; Short, 2013), were also more abundant in diapause-destined females and have previously been associated with diapause in other insects and C. elegans worms (Ailion and Thomas, 2003; Ragland et al., 2011). Last, a protein that stood out as being more abundant in diapause-destined females is a putative farnesoic acid o-methyltransferase, the enzyme that catalyzes the last step in juvenile hormone synthesis (Shinoda and Itoyama, 2003; Noriega, 2014).

Because FABP peptides were clearly more abundant in DD females than that in NDD females, we examined spatial patterns of FABP transcript abundance in the head, fat body (the adipose tissue and liver analog of insects), and ovaries of DD females to investigate the function of FABP in diapause preparation. The results showed that FABP transcripts were more highly expressed in the fat body and head than in the ovary (Figure 4A). We then examined the temporal pattern of FABP transcript abundance in the head and fat body during the diapause preparation phase. FABP transcript abundance in both the head and the fat body increased with feeding, peaked 2 days after eclosion, and then declined on day 4 after eclosion (Figures 4B,C). These results suggested that FABP may play an important role during diapause preparation, specifically considering that FABP transcript abundance increases at the same time in the diapause preparatory period that we have previously shown fat is rapidly accumulated (Tan et al., 2016, Figure 1). FABP transcript abundance also declined coincident with the time in the diapause preparatory period that we have previously shown fat storage plateaus just prior to the time in which beetles have stopped feeding and dig themselves into the soil to enter diapause (Tan et al., 2016, Figure 1).

To investigate the potential function of FABP in diapause preparation, we silenced it by injecting FABP dsRNA (dsFABP) into newly emerged DD females that had not yet had the opportunity to feed (0 day) and measured FABP transcript abundance in the head and fat body at 2 and 4 days after dsFABP injection. qRT-PCR showed that FABP transcript abundance in the dsFABP treatment group was significantly lower in both the head and fat body than that in a control check (CK) or dsGFP group (Figure 5A). In addition, the survival rates of the different treatment groups at 7 days post-dsRNA injection were all above 95%. The diapause incidence of the CK, dsGFP, and dsFABP groups were 91.15, 90.75, and 89.83%, respectively. Therefore, knockdown of FABP did not affect either survival or diapause incidence of DD C. bowringi.

A previous study by our group showed that DD beetles accumulate greater fat than NDD beetles during the diapause preparation phase (Tan et al., 2016). In this study, photomicrographs revealed hypertrophy and clumping of the fat body in the CK and dsGFP groups at 4 days after dsRNA injection. Hypertrophy of fat body was, however, inhibited in the dsFABP group and the fat body became detached (Figure 5B). Suppressing FABP transcript abundance during diapause preparation substantially decreased lipid accumulation. We further investigated the effect of FABP suppression on lipid droplet accumulation with Oil Red O staining. Red-stained lipid droplets were significantly lower in the dsFABP group than that in the CK and dsGFP groups (Figures 5B,C), suggesting that accumulation of lipid droplets was reduced by FABP transcript suppression.

Because most insects do not feed during the long diapause period, increased lipid stores are often associated with hardiness against metabolic stress (Arrese and Soulages, 2010; Hahn and Denlinger, 2011). In addition to metabolic stress, diapausing insects also face thermal stress, desiccation, and other environmental challenges (MacRae, 2010; Storey and Storey, 2012; Colinet et al., 2015). Heat shock proteins are known regulators of environmental stress and are often associated with increased stress hardiness in diapausing insects (Rinehart et al., 2007; Gkouvitsas et al., 2008, 2009). Here, we found that knocking down FABP not only affected fat accumulation, but also Hsps. Transcript abundance of Hsp21 and Hsp23 in the fat body decreased at 4 days after FABP dsRNA injection. Hsp70 transcript abundance trended toward decreasing but the effect was not statistically significant (Figure 5D). These results indicate that FABP suppression affects both lipid accumulation and transcript abundance of two heat shock proteins during diapause preparation.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.