Identification of AaAtg8 as a marker of autophagy and a functional autophagy-related protein in Aedes albopictus

Mosquitoes and cells

A. albopictus was provided by Hubei Provincial Center for Disease Control and Prevention (HBCDC). A. albopictus mosquitoes were reared at 28 °C, 80% humidity, and a photoperiod of 16 h light: 8 h dark. Larvae and pupae were grown in cups of distilled water and fed a mixture of finely ground cat food (Nestle, Purina). Research on mosquitoes (A. albopictus) does not require a specific permit in Wuhan, China. All experiments were performed based on standard operating procedures in the College of Life Sciences, Wuhan University, Wuhan, China.

C6/36 cells were provided by the China Center for Type Culture Collection (CCTCC) and were maintained in a minimum essential medium (MEM; HyClone, South Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS; Biological Industries, Kibbutz Beit Haemek, Israel) at 28 °C with 5% CO₂. C6/36 cells were transfected using lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol.

Chemical treatments and starvation treatment

For chemical treatment, C6/36 cells were treated with 3-MA (10 mM; Sigma-Aldrich, St. Louis, MO, USA), Rapa (500 nM; MedChemExpress, Shanghai, China) or CQ (300 µM; Sigma-Aldrich, St. Louis, MO, USA) for 36 h, 6 h and 36 h, respectively. When starvation treatment was performed, the C6/36 cells were maintained in Earle’s Balanced Salt Solution (EBSS; Gibco, Grand Island, NY, USA) for different time period as indicated.

Identification and sequencing of Aaatg8 cDNA

Using primers designed according to some Atg8 homolog genes of insects, fragments of Aaatg8 were got from cDNA of C6/36 cells through PCR. Then, the products of PCR were cloned into pCR-II vectors (TA Cloning® Kit; Invitrogen, Carlsbad, CA, USA) and were sequenced by Sangon Biotech, Shanghai, China. The primers for rapid amplification of cDNA ends (RACE) were designed according to obtained sequences. Then, 5′ and 3′ RACE reactions were operated with SMARTer™ RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA). After purification and sequenced, 5′ and 3′ RACE reaction products were cloned into pCR-II. The open reading frame (ORF) sequence was deduced from the sequencing data. Then, the ORF fragment was amplified and cloned into pCR-II. The sequencing results from positive colonies confirmed the sequences information of previous RACE products. The involved primers are listed in Table 1.

Construction of plasmids

The pIE-GFP-AaAtg8 plasmid was constructed by cloning the coding region of AaAtg8 into the Xho I and Bgl II sites of pIE-GFP vector. The pIE-RFP-GFP-AaAtg8 plasmid was constructed by cloning the coding region of GFP-AaAtg8 into the Kpn I and Sal I sites of pIE-RFP plasmid. The involved primers are listed in Table 1.

Quantitative real-time PCR

Total RNA of C6/36 cells or homogenizing A. albopictus mosquitoes was extracted by using TRIzol regent (Invitrogen, Carlsbad, CA, USA) and used as the template for cDNA synthesis by using M-MLV Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) according to the protocol. The obtained cDNAs were used as templates for quantitative real-time PCR (qPCR) amplification using FastStart universal SYBR master (Roche, Penzberg, Germany) according to the manufacturer’s protocol. Specific gene transcription level was normalized to that of ribosomal gene s7 (Aas7; GenBank: JN132168.1). Quantity values were obtained by means of the 2^−ΔΔCt method (Livak & Schmittgen, 2001). Three biological replicates were performed in each experiment. The involved gene-specific primers are listed in Table 1.

Western blot analysis

Prepared cell lysates were boiled for 5 min in SDS loading buffer. Proteins were separated by SDS-PAGE and transferred to Immobilon-P membrane (Merck Millipore, Darmstadt, Germany). Membranes were blocked in Tris-buffered saline (TBS) with 5% nonfat dry milk (Sangon Biotech, Shanghai, China) and 0.1% Tween 20 (Sangon Biotech, Shanghai, China) for 1 h and then incubated with mouse anti-actin antibody (1:10,000; Proteintech, Rosemont, IL, USA), rabbit anti-AaAtg8 antibody which was antibody against GABARAP (1:1,000; MBL, Tokyo, Japan) or mouse anti-GFP antibody (1:3,000; Santa Cruz Biotechnology, Dallas, TX, USA) for 1 h. After washed three times with TBST, the membrane was incubated by an anti-rabbit or anti-mouse antiserum conjugated to horseradish peroxidase (anti-rabbit 1:5,000 and anti-mouse 1:5,000; Thermo Fisher Scientific, Waltham, MA, USA) for 1 h. Then the membrane was washed three times with TBST. After incubating the membrane with HRP substrate (Merck Millipore, Darmstadt, Germany), the bands were visualized by LAS 4000 (Fujifilm, Tokyo, Japan). The images’ brightness and contrast were adjusted using ImageJ 1.51k (National Institutes of Health, Bethesda, MD, USA), and the backgrounds were subtracted (Schindelin et al., 2012). To measure the integrated densities, the intensities of regions of interest (ROIs) were selected for all bands and normalized to the corresponding Actin.

Fluorescence microscopy and software-aided counting

In MDC (Sigma-Aldrich, St. Louis, MO, USA) staining assay, C6/36 cells were incubated with 50 µM MDC at 28 °C for 10 min. After incubation, cells were washed three times with 0.1 M PBS (pH 7.4) and fixed with 4% paraformaldehyde in 0.1 M PBS (pH 7.4) for 10 min at room temperature. Then, cells were washed three times with PBS. Cell images were taken with an automated microscope (FV1000, Olympus, Tokyo, Japan).

In fluorescent AaAtg8 assay, after GFP-AaAtg8 or RFP-GFP-AaAtg8 was overexpressed in C6/36 cells for the time period as indicated, cells were fixed with 4% paraformaldehyde in 0.1 M PBS (pH 7.4) for 15 min at room temperature. Then, cells were washed three times with PBS and cell nuclei were counterstained with Hoechst 33342 (Sigma-Aldrich, St. Louis, MO, USA). Cell images were taken with a scanning confocal microscope (SP8, Leica, Wetzlar, Germany).

We used ImageJ 1.51k as a tool for fluorescent spots counting in cells (Schindelin et al., 2012; Guillery & August, 2002). Fluorescent spots were counted in ROIs of each C6/36 cell.

Data analysis

Multiple-sequence alignment was performed with DNAMAN Version7 (Lynnon Corporation, San Ramon, CA, USA). Phylogenetic analysis was performed through MEGA 6 (Tamura et al., 2013). Secondary structures were predicted with ESPript3.0 (Robert & Gouet, 2014). The pictures were edited with Adobe Photoshop CC 2017 (Adobe, San Jose, California, USA). GraphPad Prism (GraphPad Software, Inc., San Diego, CA, USA) was used for plotting graphs. All data was presented as Mean ± SEM of triplicate experiments. Group comparisons were performed by t test or one-way ANOVA with the Duncan test with SPSS 19 system software (SPSS Inc. Chicago, IL, USA). A value of P less than 0.05 was considered to be significant differences.

Sequence analysis of AaAtg8

Primers were designed according to Aedes aegypti Atg8/GABARAP gene (Aeatg8/AeGABARAP, NCBI Reference Sequence: XP_001652571.1) and cDNA of C6/36 cells as templates, partial sequence of Aaatg8 was amplified by PCR. Based on that partial sequence information, a complete ORF fragment containing of 357 bp as well as a 5′ UTR of 151 bp and a 3′ UTR of 747 bp was obtained by 5′ and 3′ RACE reactions. The obtained ORF sequence encoded a protein containing 118 amino acids (Fig. 1A).

Multiple-sequence alignment revealed that AaAtg8 had high sequence identities with other Atg8s. The protein sequence of AaAtg8 shared 96%, 93%, 93%, 92%, 90%, 90%, 57%, 29% and 28% sequence identities with Atg8 from A. aegypti (AeGABARAP/AeAtg8, NCBI Reference Sequence: XP_001652571.1), D. melanogaster (DmAtg8a, NCBI Reference Sequence: NP_727447.1), Bombyx mori (BmAtg8, NCBI Reference Sequence: NP_001040244.1), Galleria mellonella (GmAtg8, GenBank: AFP66874.1), Homo sapiens (HsGABARAP, NCBI Reference Sequence: NP_009209.1), Rattus norvegicus (RtGABARAP, NCBI Reference Sequence: NP_742033.1), Saccharomyces cerevisiae (ScAtg8, NCBI Reference Sequence: NP_009475), Homo sapiens (HsLC3B, NCBI Reference Sequence: NP_073729.1) and Rattus norvegicus (RtLC3B, NCBI Reference Sequence: NP_074058.2), respectively. Then, the secondary structure was predicted with ESPript3.0 based on HsGABARAP (PDB: 1KOT) (Gouet, Robert & Courcelle, 2003). The predicted secondary structure of AaAtg8 consisted of a series of alpha helices and beta sheets that are highly conserved in other Atg8 proteins (Fig. 1B).

Phylogenetic analysis of AaAtg8

Phylogenetic tree of AaAtg8 and nine selected Atg8 from other species was constructed on the basis of distances of amino acid sequences. The obtained dendrogram placed AaAtg8 in the insect group but separated from mammalian LC3B proteins (Fig. 2). The phylogenetic analysis also illustrated that AaAtg8 and AeGABARAP/AeAtg8 formed a single clade inside the insect group, indicating that AaAtg8 was closer to GABARAP than to LC3B.

AaAtg8 expression profile in developmental and adult stages of Aedes albopictus

In order to research the expression profile of AaAtg8, the transcription levels of Aaatg8 in different developmental stages of A. albopictus were analysed by qPCR. A. albopictus ribosomal gene s7 was used as a control gene. There was a ubiquitous expression of AaAtg8 in all the developmental stages. The transcription level of AaAtg8 increased progressively from 2nd instar larvae to pupae, which implied that AaAtg8 might play a role in development (Fig. 3A). Also, transcription level of AaAtg8 in female adults was about 3 times as that in male adults (Fig. 3B).

Responses of A. albopictus C6/36 cells to chemicals commonly used in autophagy study

Since there is no research on autophagy process in A. albopictus C6/36 cells published so far, it is important to establish the basic method to study autophagy in C6/36 cells. We first studied how to induce or inhibit autophagy in C6/36 cells. For that purpose, C6/36 cells were treated with 3-MA (an inhibitor that effectively blocks early stage of autophagy), Rapa (a kind of widely used autophagy inducer) or CQ (an acidification inhibitor blocking late stage of autophagy) and the effect of each treatment was evaluated by fluorescence microscopy after monodansylcadaverine (MDC) staining. Mock and DMSO treated C6/36 cells were used as controls. MDC is an acidotropic dye routinely used as an indicator in autophagy assays due to its capacity to stain acidic autolysosomes. The data showed that a few MDC fluorescent puncta were observed in mock treated C6/36 cells, indicating a basal level of autophagy in C6/36 cells. DMSO treatment did not change the number of fluorescent puncta. Compared to that in DMSO treated C6/36 cells, the number of fluorescent puncta decreased in 3-MA treated C6/36 cells, whereas increased in Rapa treated C6/36 cells. The number of fluorescent puncta in CQ treated C6/36 cells was higher than that in the DMSO or Rapa treated C6/36 cells (Figs. 4A and 4B). The above data suggested that Rapa could be used to induce autophagy, while 3-MA and CQ were effective agents to inhibit autophagy at early and late stages of autophagy in C6/36 cells, respectively (see ‘Discussion’). Moreover, to determine whether these chemicals affect the C6/36 cells viability, we performed the MTT assay. The result revealed that there were not significant effects on the viability of C6/36 cells treated with these chemicals (P > 0.05) (Fig. 4C).

AaAtg8 was involved in autophagy and a useful marker for autophagy assay in C6/36 cells

As described before, during the process of autophagy, Atg8-I is transformed to Atg8-II that specifically associates with premature as well as completed autophagosomes. So, Atg8-II levels correlate with the number of autophagosomes and Atg8-II is widely used as a marker to monitor autophagosomes and autophagic activity (Mizushima, Yoshimori & Levine, 2010). In order to study whether AaAtg8 could indicate autophagic level in C6/36 cells similar to other research systems, the levels of AaAtg8-I and AaAtg8-II in 3-MA, Rapa or CQ treated C6/36 cells were assessed by immunoblotting using an antibody against AaAtg8. Mock treated and DMSO treated cells were used as controls. The data showed that in mock treated C6/36 cells, high levels of AaAtg8-I and low levels of AaAtg8-II were detected, suggesting that there was a basal level of autophagy in C6/36 cells. The pattern remained unchanged in DMSO treated C6/36 cells. Compared to that in DMSO treated C6/36 cells, the levels of AaAtg8-II were decreased in 3-MA treated C6/36 cells whereas increased in Rapa treated C6/36 cells, indicating that 3-MA inhibited the early stage of autophagy and Rapa induced autophagy in C6/36 cells (see ‘Discussion’). In CQ treated C6/36 cells, levels of AaAtg8-I and AaAtg8-II were up-regulated. In C6/36 cells treated by both Rapa plus CQ, the levels of AaAtg8-II were even higher than those in C6/36 cells treated by Rapa or CQ alone (Fig. 5). These results further confirmed that Rapa could successfully induce autophagy and CQ could inhibit the late stage of autophagy in C6/36 cells (see ‘Discussion’). Taken together the above results and previous data, these findings indicated that the turnover of AaAtg8 was a useful marker of autophagy in C6/36 cells and could be used to monitor the levels of autophagy in A. albopictus.

Moreover, as shown in previous MDC staining assay, the number of fluorescent puncta was decreased in 3-MA treated C6/36 cells. Here, in immunoblotting analysis, the levels of AaAtg8-II were decreased in 3-MA treated C6/36 cells. Similarly, in Rapa or CQ treated C6/36 cells, the number of fluorescent puncta in MDC staining assay and the levels of AaAtg8-II in immunoblotting analysis were increased. These results proved that AaAtg8 was a functional Atg protein involving in autophagy in A. albopictus C6/36 cells.

Fluorescent AaAtg8 could be used to monitor autophagy in C6/36 cells

When being observed under fluorescence microscopy, fluorescent Atg8 can be visualized as a diffused cytoplasmic pool when cells are under normal conditions or as punctate structures when cells undergo autophagy process. Thus, fluorescent Atg8 provides a convenient way to monitor autophagy and it has been successfully applied to study autophagy in many model organisms including Drosophila, nematodes, plants, zebrafish and mouse (Melendez et al., 2003; Mizushima et al., 2004; Rusten et al., 2004; Scott, Schuldiner & Neufeld, 2004; Yoshimoto et al., 2004; He et al., 2009). In order to determine whether this method could be used in C6/36 cells, GFP-AaAtg8 overexpressing C6/36 cells were treated with 3-MA, Rapa or CQ, and levels of GFP-AaAtg8 were monitored via immunoblotting and the number of GFP-AaAtg8 dots was observed by fluorescence microscopy. Mock or DMSO treated C6/36 cells were used as controls. According to the data of immunoblotting, levels of GFP-AaAtg8-I and GFP-AaAtg8-II showed a pattern similar to that of endogenous AaAtg8-I and AaAtg8-II under each treatment (Fig. 6A). This result suggested that, similar to endogenous AaAtg8, GFP-AaAtg8 was also a useful marker of autophagy in C6/36 cells. When analyzed by fluorescence microscopy, a few fluorescent puncta were observed in mock treated C6/36 cells, indicating a basal level of autophagy in C6/36 cells. DMSO treatment did not change the number of fluorescent puncta. Compared to that in DMSO treated C6/36 cells, the number of fluorescent puncta was decreased in 3-MA treated C6/36 cells whereas increased in Rapa treated C6/36 cells. In CQ treated C6/36 cells, the number of fluorescent puncta was higher than that in DMSO or Rapa treated C6/36 cells. Moreover, the number of fluorescent puncta in C6/36 cells increased when cells were treated with Rapa plus CQ compared to the treatments with Rapa or CQ alone (Figs. 6B and 6C). The above data showed that the number of GFP-AaAtg8 puncta was positively correlated with the abundance of intrinsic AaAtg8-II and GFP-AaAtg8-II detected by immunoblotting analysis. Thus, similar to endogenous AaAtg8, GFP-AaAtg8 could be used to detect autophagy in C6/36 and monitoring GFP-AaAtg8 via immunoblotting and fluorescence microscopy were applicable methods to evaluate the levels of autophagy in C6/36 cells.

Since GFP protein is not stable in low pH condition, GFP fluorescence will be quenched after autophagosomes fuse with lysosomes. So, Atg8 is usually fused with GFP plus RFP (a fluorescent protein which is stable in autolysosomes) to monitor autophagy by fluorescence of Atg8. RFP-GFP-Atg8 is observed as yellow signals generated by merged RFP and GFP when associated with autophagosomes. However, the fluorescence of RFP-GFP-Atg8 will be observed as red signals generated by RFP when associated with acidic autolysosomes (Kimura, Noda & Yoshimori, 2007). Thus, if the maturation of autophagosomes into autolysosomes is blocked, yellow puncta will be accumulated without a concomitant increase in red puncta. If autophagosomes can fuse with lysosomes to form autolysosomes, the signal of yellow puncta will decrease, while the signal of red puncta will increase. Thus, monitoring the number of yellow and red puncta provides a convenient and widely used approach to distinguish autophagosomes and autolysosomes in autophagic flux. In order to determine whether this method could be used in C6/36 cells, RFP-GFP-AaAtg8 overexpressing C6/36 cells were treated with Rapa or CQ and RFP-GFP-AaAtg8 was detected by a confocal fluorescence microscopy. DMSO treated C6/36 cells were used as a control. The data indicate that the red and yellow puncta were increased in the Rapa or CQ treated C6/36 cells compared to that in DMSO treated C6/36 cells. However, the number of yellow puncta was lower than that of red puncta in Rapa treated C6/36 cells but higher in CQ treated C6/36 cells. These results suggested that RFP-GFP-AaAtg8 could be used to measure complete autophagy induced by Rapa treatment and incomplete autophagy induced by CQ treatment (Figs. 7A and 7B). All the above data indicated that RFP-GFP-AaAtg8 could be used to monitor flux of autophagy in C6/36 cells.

Starvation could induce autophagy in C6/36 cells

In addition to the Rapa treatment, starvation is also frequently used to induce autophagy (Mizushima, Yoshimori & Levine, 2010). Thus, it would be of interest to know whether starvation could induce autophagy in C6/36 cells. Our result revealed that starvation could induce autophagy in C6/36 cells because starvation increased the levels of AaAtg8-II at 0.5 h after EBSS treatment (Fig. 8A). Interestingly, the levels of AaAtg8-II decreased from 1 h to 6 h post starvation treatment. Moreover, C6/36 cells were stained by MDC after EBSS treatment. It was obviously observed that the number of fluorescent puncta was increased at 0.5 h and 1 h after EBSS treatment but decreased from 2 h to 6 h post EBSS treatment (Fig. 8B). These results proved that starvation could induce autophagy in C6/36 cells, but C6/36 cells responded to starvation very rapidly.