Identification and characterization of cold-responsive aquaporins from the larvae of a crambid pest Agriphila aeneociliella (Eversmann) (Lepidoptera: Crambidae)

Insects

The larvae of A. aeneociliella were collected from wheat fields of Shandong Province, China (E 120°21, N 36°46), and were reared individually in a 30 mL perforated plastic box in laboratory. The larvae were reared to the third generation in the artificial climate chamber at 23 ± 1 °C, 14 h: 10 h L: D photoperiod, 70% RH (Zhang et al., 2015).

Treatment design and sample preparation

Based on the data from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard NASA’s Terra and Aqua spacecraft, the daytime land surface temperature (LST) in winter in Shandong Province, China ranges from 0 °C to 10 °C, and the nighttime LST from −8 °C to −2 °C (Chen, 2021). Therefore, four temperatures of 4 °C, 0 °C, −4 °C, and −8 °C were chosen as different treatments in the experiment to compare the expression levels of AQPs. Thirty third instar larvae with similar body size were selected and transferred into one petri dishes (diameter = 12 cm) sealed with plastic wrap. Then, the larvae were exposed to one of four temperatures (4 °C, 0 °C, −4 °C, −8 °C) for 3 h, 12 h and 24 h in climate chamber. Each treatment was conducted with three biological replicates. The larvae reared at 23 °C for 3 h, 12 h and 24 h in climate chamber were set as control groups. The treated larvae were immediately collected and immersed in liquid nitrogen and stored at −80 °C until utilized for RNA extraction.

Total RNA extraction and cDNA synthesis

Thirty third instar larvae of A. aeneociliella were grinded into powder in 1.5 mL tubes placed in liquid nitrogen. Total RNA was extracted using TRIzol reagent (TransGen Biotech, Beijing, CHN) following manufacturer’s instructions. The integrity of RNA was verified by 1.5% agarose gel electrophoresis. The quality of total RNA was then detected using ultraviolet spectrophotometry (NanoDrop One; Thermo Fisher, Waltham, MA, USA) (OD260/OD280 = 1.92, OD260/OD230 = 2.0). First strand cDNA was reverse transcribed from RNA using PrimeScript II Rtase with the Oligo (dT) primer and Random 6 primer (PrimeScript^™ II 1^st Strand cDNA Synthesis Kit, Takara Bio, Beijing, China). The cDNA was stored at −20 °C for further experiment.

DNA amplification, cloning and sequence analysis

Specific primers of AQPs sequences were designed by Primer premier 5.0 based on cloned sequence data of A. aeneociliella and all of the primers used in this study were listed in Table S1. The PCR was carried out in Bio Red T100^™ Thermal Cycler by using 2× Accurate Taq Master Mix (Accurate Biology), with the following cycling parameters: initial denaturation for 3 min at 98 °C, followed by 35 cycles of 10 s at 98 °C, 30 s at 45–72 °C, 1 min at 72 °C, and a final extension for 10 min at 72 °C. The PCR products were gel-purified by using Gel Extraction Kit (Beijing CoWin Biotech Co,. Ltd, Beijing, CHN), and inserted into pMD18-T vector (Takara Bio, Beijing, CHN). The recombinant plasmids were transformed into competent E. coli cells (DH5α) for Sanger sequencing by TSINGKE, Qingdao, China.

The DNA sequences were analyzed using the Blastx search program (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and aligned with AQPs from other insect species by using ClustalW. Physicochemical property (pI/MW and Hydrophobicity) of the AQP proteins were predicted using ExPASy tools (http://us.expasy.org/tools/). Transmembrane domains were predicted with TMHMM (https://services.healthtech.dtu.dk/services/TMHMM-2.0/). Functions of the AQP proteins were predicted by InterPro (http://www.ebi.ac.uk/interpro/). Phylogenetic analysis of six A. aeneociliella AQPs and 47 other arthropods AQPs downloaded from GenBank were performed using maximum likelihood method in MEGA 7.

Molecular modeling

Three-dimensional constructions of AQPs were generated by using SWISS-MODEL (https://swissmodel.expasy.org/) for homology modeling. The crystal structure of AQP1 ortholog (PDB ID 1J4N) and human AQP5 (PDB ID 1LDA) were chosen as the best template because of their high resolution (1.8 Å) and the high level of amino acid sequence identity (32–40%) to the AaAQPs. Briefly, six AaAQP sequences were aligned by using the Phyre2 tool. Pymol was used to visualize the three-dimensional coordinates for the atoms of the model. Pore diameters of homology models were calculated using the HOLE2.0 program of the energy minimized homology model structure using the simple rad van der Waals radius file.

Real-time qPCR analysis

Specific RT-qPCR primers were designed using Primer Premier 5.0 (Table S1). The RT-qPCR analysis was conducted using CFX-96 real-time PCR Detection System (Bio-Rad, CA, USA). First-strand cDNA was used as the templates to examine the relative expression of AQP genes in a 20 μL reaction system containing 3 μL of cDNA, 0.8 μL each of 10 μmol/L forward primer and reverse primer, 10 μL of 2× T5 Fast qPCR Mix (TSINGKE, China), and 5.4 μL of nuclease-free water. Two-step thermocycling conditions were used as follows: 95 °C for 1 min, followed by 40 cycles of 95 °C for 10 s, insert melting curve analysis at 60 °C to 65 °C in increments of 0.55 °C for 15 s. β-actin (accession number: OR526988) was used as reference gene. Three biological replicates were performed, and each biological replicate was conducted with three technical replicates.

Statistical analyses

Relative expression level of AQPs was determined using the 2^−ΔΔCt method (Livak & Schmittgen, 2001), ΔΔCt = ΔCt (cold treatment)—ΔCt (control group). All data obtained from the RT-qPCR were analyzed using analysis of ANOVA and then the results were adjusted using a least-significant different (LSD) multiple comparison test. The statistical analyses were conducted using SPSS 19.0 for windows.

Identification of AQPs from A. aeneociliella

Six AaAQP genes were identified in A. aeneociliella (Table 1). Six AaAQPs shared 57% to 93% identity with amino acid sequences of AQPs from other insect species. AaAQP6 of A. aeneociliella showed the highest similarity with AQP12L2 from the striped stem borer Chilo suppressalis (Walker) (93%). AaAQP3 shares the lowest identity (57%) with the wax moth Galleria mellonella L. The results indicated that AQPs are relatively conserved in Lepidoptera. The encoded protein length of AaAQP genes range from 251 to 292 aa, with their molecular weight (MW) from 26.006 to 32.170 kDa and the isoelectric points (PI) from 5.27 to 8.52 (Table 1).

Sequence analysis of AQPs

As presented in Fig. 1, AaAQPs showed relatively conservative motifs and universal molecular structures in sequence alignment. Each AQP sequence contained six hydrophobic TMHs, which were rich in α-helix, but lack of β-pleated sheet. There were some slightly differences in the family signature motifs between six AQP sequences. AaAQP1, AaAQP2 and AaAQP5 have two conserved NPA (Asn-Pro-Ala) boxes, which are the most common hallmarks among aquaporin family. AaAQP3 and AaAQP4 only have one NPA motif and one NPF motif (Asn-Pro-Phe). While AaAQP6 has the conserved CPY (Cys-Pro-Tyr) and NPV (Asn-Pro-Val) motifs. Residues (Phe, His, Ala, Ser, Asn or Arg) that comprised the ar/R selectivity filter were found in AaAQP sequences except AQP6, the site of Phe⁶³ and Arg²⁰¹ exhibited a highly conservation especially. In the amino acid sequence, all six AQPs sequences had phosphorylation site and N-myristoylation but did not have O-glycosylation. The NISQ and NCSV sites locate at the position 44 of AQP1 and position 165 of AQP3, respectively, which were predicted to be the N-glycosylation.

The phylogenetic tree was conducted to compare six A. aeneociliella AQPs and 46 arthropod AQPs using maximum likelihood method. As shown in Fig. 2, six AaAQPs of A. aeneociliella were distinct classified into five major subfamilies: DRIP, PRIP, RPIP, Eglp and AQP12L, but no BIB-like sequence was identified. The five subfamilies were further classified into four clades (A–D). The A-D clades represented the classical aquaporins, the aquaglyceroporins (Glp), AQP12L and BIBs, respectively. Based on the phylogenetic analysis, AQP1 and AQP2 were water-specific classical aquaporins, AQP3, AQP4 and AQP5 were Glps. The BIBs group was not identified in A. aeneociliella AQPs, which mainly responsible for cation transport (Fig. 2).

Three-dimensional structure of AaAQPs

A three-dimensional (3D) construction of AQP1 was generated with SWISS-MODEL and Pymol for homology modeling by using the crystallographically resolved AQP1 ortholog (PDB ID 1J4N) as a template. The structure of AQP1 shared the highest homology with that of AaAQP1 (40% identity). Extracellular view of the structure showing that, AQPs usually function as a tetramer unit with each monomer proving an independent water channel (Fig. 3A). Combined with hydrophobicity analysis, the transmembrane region of each channel was hydrophobic, and both sides out of the membrane were hydrophilic (Fig. S1). Water molecules were transported through the hydrophobic membrane channels, establishing a coordinated water transport pathway (Fig. 3B). The 3D models of AaAQP1 monomer comprise six transmembrane (TMD1-6) and two half-transmembrane helices, forming a typical hourglass-shaped structure of AQPs, and each of the half-transmembrane helices contains a conserved NPA motif that predicted to site one side of the constriction pore (Fig. 3C).

Imperceptible changes happened in four ar/R residues of AaAQPs sequences had critical effects on constriction structure and transport characteristics. Representative sequence (AQP1, AQP3 and AQP6) from each of the A-C groups was selected separately to construct a 3D homology model to observe the spatial distribution of ar/R region (Figs. 3D–3F). Clearly, water-specific classical aquaporins (D) equipped with a tighter pore which contraction region radius close to 1.5 Å (Fig. 4), allowing water molecules (diameter = 2.8 Å) to pass through. In more relaxed structure of Glps, the ar/R shrinkage diameter was enlarged to a radius of 1.9 Å through certain amino acid replacement of His¹⁸⁶ and Ser¹⁹⁵, which made it possible to infiltrate glycerol, urea and other molecules.

In order to investigate the relationship between constriction structure and permeability properties of ar/R under freezing stress, an overview of the ar/R residues’ composition, corresponding permeability and predicted functions that have been confirmed in previous studies were presented in Table 2. Classical aquaporin subfamily that only allows water pass through exhibited a highly conservative features in constitution, which is responsible for insect cold tolerance by maintaining water homeostasis. However, a diversified residue variation took place in Glps. As the diversity of permeability, Glps allowed cryoprotectant such as glycerol and trehalose to enter the cells, improving the tolerance of insects to low temperature. AQP12L exhibited conservative features in constitution, and is also involved in osmoregulatory processes.

Expression profiles of AaAQPs exposed to low temperatures

All six AaAQP genes showed different expression profiles when the larvae were exposed to 4 °C, 0 °C, −4 °C, −8 °C during the same cooling duration. The expression levels of AQP1 had a significant up-regulation at −8 °C for the time points of 12 h (P = 0.029) and 24 h (P = 0.016). Similar expression trends were also found in AQP6 expression, which significantly increased under 0 °C, −4 °C and −8 °C for 24 h (0 °C: F = 22.170, P = 0.02; −4 °C: F = 42.350, P < 0.001; −8 °C: F = 91.591, P < 0.001), and no significant differences were found in response to 3 h cooling time under all temperatures (F = 1.333, P = 0.412) (Fig. 5). The transcription levels of AaAQP1, AaAQP3, and AaAQP6 increased significantly with the extension of treatment time below 0 °C. However, the expressions of AQP4 (F = 8.420, P = 0.018) and AQP5 (F = 31.523, P = 0.001) decreased significantly when the cooling time extended to 24 h under −8 °C (Fig. 5).