Introduction
Cadmium (Cd) is a highly toxic, persistent, and non-essential heavy metal that can be released into aquatic environments through various industrial and anthropogenic activities, such as mining and the application of fertilisers (Campoy-diaz et al., Reference Campoy-diaz, Escobar-correas, Canizo, Wuilloud and Vega2020; Shi et al., Reference Shi, Guo, Du, Yang, Liu and Xu2023). Cd can induce oxidative stress, DNA damage, and apoptosis by promoting the generation of reactive oxygen species (Teng et al., Reference Teng, Ren, Chen, Shen, Zhang, Chai and Wang2022). Consequently, even relatively low levels of Cd in aquatic environments can have toxic effects on organisms (Shukla et al., Reference Shukla, Ahmed and Lodhi2023). Organisms in freshwater ecosystems contaminated with Cd are likely to accumulate high levels of Cd, which can negatively impact human health through the food chain (Lin et al., Reference Lin, Yu, Wang and Wu2023).
Although Cd is pervasive in aquatic systems, its concentration in sediment is typically higher than that in water (Dolagaratz Carricavur et al., Reference Dolagaratz Carricavur, Chiodi Boudet, Romero, Polizzi, Marcovecchio and Gerpe2018; Shu et al., Reference Shu, Tang, Cheng, Zhang and Zhang2024). This disparity increases the risk of Cd accumulation in benthic organisms residing near the sediment (Palacio-cortés et al., Reference Palacio-cortés, Horton, Newbold, Spurgeon, Lahive, Pereira, Grassi, Moura, Disner, Cestari, Gweon and Navarro-silva2022; Su et al., Reference Su, Li, Qiu, Zeng, Wang, Jun and Wang2022). Chironomids are common benthic species in aquatic ecosystems, playing a vital role in macroinvertebrate communities. Most of their life cycle occurs in sediment, where they feed on it. Consequently, chironomids exhibit heightened sensitivity to heavy metals in aquatic environments and are therefore often used as model organisms in ecotoxicological studies (Liu et al., Reference Liu, Pei, Shao, Nie, Cao and Yan2025). Research has also shown that some chironomids have a high tolerance to metal pollution, particularly in their late larval stages (Laviad-shitrit et al., Reference Laviad-shitrit, Sharaby, Sela, Thorat, Nath and Halpern2021). Despite extensive studies on the Cd tolerance of chironomids, the mechanisms underlying their adaptation to such conditions remain poorly understood.
α-enolase (Eno1) is one of the three enolase isoenzymes and is expressed in almost all mature tissues. It catalyses the conversion of 2-phosphoglycerate to phosphoenolpyruvate in the glycolysis pathway and facilitates the reverse reaction during gluconeogenesis (Zhu et al., Reference Zhu, Li, Sun, Fan, Hu, Chai, Lin, Wu, Qin, Wang, Hsieh-wilson and Yi2024). In addition to its participation in a multitude of metabolic processes, Eno1 plays a pivotal role in the regulation of diverse stress responses, acting as a stress protein to safeguard the cells from the detrimental effects of adverse environmental conditions, including hypoxia, cold, heat, acidity, salinity, and metal stress (Kim et al., Reference Kim, Lee, Kim, Kim and Moon2015; Santana et al., Reference Santana, Álvarez, Sáenz-martínez, Salinas-parra, Guzmán, Paradela and Mercado2021). Our previous study demonstrated a considerable elevation in Eno1 levels at the protein, mRNA expression, and enzyme activity levels in Propsilocerus akamusi (Diptera: Chironomidae) larvae subjected to Cd2+ exposure. Furthermore, heterologous expression of the Eno1 gene from P. akamusi and Phytolacca americana was found to significantly enhance the Cd2+ tolerance of Escherichia coli (Zheng et al., Reference Zheng, Li and Gao2020; Zhao et al., Reference Zhao, Zhu, Wang, Han, Xu, Feng and Zheng2023). These findings suggest that Eno1 may play a crucial role in the response to Cd2+ stress. However, the potential regulatory mechanisms underlying the response of Eno1 to Cd2+ stress remain poorly understood.
MicroRNA (miRNA), the first endogenous non-coding small RNA to be identified, is approximately 22 nucleotides in length and widely distributed in animals, plants, and microorganisms (Li et al., Reference Li, Xiao, He and Li2019). It regulates gene expression levels after transcription by inhibiting translation or degrading mRNA of the related target genes (Zhang et al., Reference Zhang, Wen, Li, Palli, Li, Wang and Liu2021), thus modulating and facilitating the response of the organism to environmental stress (Li et al., Reference Li, Xiao, He and Li2019). It has been demonstrated that miR-92a plays a pivotal role in Cd2+ resistance in water fleas (Chen et al., Reference Chen, Nichols, Poynton and Sepúlveda2016). MiR-22-3p has been demonstrated to target Eno1 and regulate inflammatory factors, thereby contributing to the development of neuropathic pain (Li et al., Reference Li, Luo, Zhang, Wei, Zhang, Liu and Su2020). Furthermore, the interaction between miR-539-5p and LOC146880 has been associated with a reduction in Eno1 phosphorylation levels (Feng et al., Reference Feng, Feng, Zhu, Li, Zhang, Wang, Gao, Zhou, Yu, Zheng and Qian2020). At present, there is a paucity of research investigating the role of miRNAs in regulating benthic insect chironomids under heavy metal stress. It remains uncertain whether miRNAs can modulate the expression of the Eno1 gene, which plays a role in insect resistance to Cd2+. This will enhance our understanding of the mechanisms underlying Cd2+ resistance in the larvae of P. akamusi.
Here, we hypothesised that Cd stress alters the expression of miR-92a and miR-279d, which participate in the post-transcriptional modulation of PaEno1, and are involved in the molecular response to Cd exposure in chironomid larvae. The differential expression of miRNAs in P. akamusi was screened for using microarray analysis. Subsequently, the roles of miR-92a and miR-279d in the Cd resistance of P. akamusi were investigated, as well as the regulatory relationship between miRNA and PaEno1 and the dynamic expression of four stress-related genes (PaHbV, PaHbVII, alpha-esterase (one type of carboxylesterase (CarEs)), and cholinesterase (COesterase) during exposure to Cd2+. The objective of this study is to elucidate the characteristics and functions of miR-92a and miR-279d in the adaptation of P. akamusi to Cd2⁺ stress. The findings of this research will contribute to a more comprehensive understanding of the mechanisms underlying Cd2⁺ resistance in this species.
Materials and methods
Insect sampling and Cd2+ exposure
P. akamusi larvae samples were obtained from the Qinxian Fish-Bird-Flower market (Xiaodian district, Taiyuan City, China; latitude 42.83; longitude 93.53) in June 2021 and transferred live to the laboratory. The fourth-instar was selected based on the head capsule width and washed thoroughly with dechlorinated tap water. Before exposure in the experiment, the larvae were placed in a beaker containing 500 mL of dechlorinated tap water and 1–2 cm of acid-washed sand. The beaker was maintained at 21 ± 1℃ with a 16:8 h light: dark photoperiod for 48 hours before exposure.
Cadmium chloride (CdCl2, purity 99%) was purchased from Tianjin Reagent Factory (Tianjin, China). The fourth-instar P. akamusi larvae were transferred to a 12-well plate containing 3 mL of sublethal concentrations of CdCl2 (2.4 mmol/L, according to our previous data, Zheng et al., Reference Zheng, Xu, Qin, Wu and Wei2017) for 72 h with the dechlorinated tap water as the control (CK) (three replicates for the Cd and control groups). After the exposure, 10 larvae from each group were collected and transferred to DNase/RNase-Free water for 5 min to remove non-specifically bound Cd2+ and dried with filter paper (Gillis and Wood, Reference Gillis and Wood2008).
Small RNA sequencing
RNA isolation, small RNA sequencing, and data analysis
The RNAiso Plus reagent (Takara Bio Inc, Japan) was used to extract total RNA from the pooled P. akamusi larvae of Cd-exposed and CK treatments. The quality and quantity of total RNA were analysed by NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at A230/A260 (between 1.8 and 2.0) and confirmed by agarose gel electrophoresis. RNA was heated and ligated to a 3′ chimeric oligonucleotide adaptor with T4 RNA ligase, and the products were reverse transcribed using the RT primer. Then a 5′ adaptor was ligated to the small RNAs. Subsequently, adaptor sequence primers were used for polymerase chain reaction (PCR) with 12 cycles. The PCR products were purified and sequenced by Sangon Biotech (Sangon, Shanghai, China) using Solexa technology. Small RNA sequence data are deposited in the Sequence Read Archive under BioProject accession PRJNA1358247.
Initial reads were trimmed and quality-controlled using FastQC, Cutadapt, and Trimmomatic. The clean reads were analysed by BLAST against the Rfam (http://rfam.xfam.org/) database to annotate tRNA, snRNA, and snoRNA. The remaining reads were then aligned to the mature miRNA sequences of all animals in miRBase to detect the miRNA sequences. Reads Per Million (RPM) and edgeR were utilised for analysing the expression of miRNAs in the two samples. In edgeR analysis, genes with a p-value <0.05 and |log2(Fold Change)| >1 were considered significantly expressed in miRNAs.
Verification of differentially expressed miRNAs by poly (A) qRT-PCR and their target prediction on PaEno1
The qRT-PCR method was used to detect the relative expression levels of miR-92a and miR-279d in P. akamusi after 48 hours of Cd2+ exposure (as described in section 2.1). Total RNA isolation and quality control, and reverse transcribed according to the method described in our previously published paper (Zheng et al., Reference Zheng, Gao, Li and Wang2018). miRNA first-strand cDNA synthesis kit (Sangon, Shanghai, China) and PrimeScriptTM RT reagent Kit with gDNA Eraser (Perfect Real-time) (Sangon, Dalian, China) were used to synthesise cDNA according to the manufacturer’s protocol. Then, TB Green™ Premix Ex Taq™ II (Tli RNaseH Plus) (TaKaRa, Dalian, China) was utilised for the qRT-PCR analysis of miR-92a and miR-279d, with the experiments conducted in triplicate (n = 3). U6 was used as the endogenous control. The primer sequences for miR-92a, miR-279d, and U6 are listed in table 1.
Primer sequences used for miR-92a, miR-279d, and metal stress-related gene expression
Table 1 Long description
Two columns list Gene name and Primer Sequences from five prime to three prime. U6 has a forward primer sequence GCTTCGGCTGGACATATACTAAAAT. bmo miR 92a has a forward primer sequence TATATTGCACCAGTCCCGGCC, and tca miR 279d 5p has a forward primer sequence TGGCGATTGTGGGTTTAGTCC. URP has a reverse primer sequence TGGTGTCGTGGAGTCG. PaEno1 lists forward CCAACGCTATTCTCGGTATCTC and reverse AAGTTTCGGCACCCATTTTC. PaHbV lists forward AACTGTGTCAAGAACATTGGCAACG and reverse TAATTGGGCGAGCAACATGAGGAC. PaHbVII lists forward GAGAACAGACCAGCCCTCAAGAATG and reverse AGGAATGCTTCGAGTGCGACTTC. COesterase lists forward CAGCAGGCATTCTCAACTCACCTAG and reverse GCTGGCAGCCCATAGCAACATC. alpha Esterase lists forward GCAACCCAACACCCACCGTATTAG and reverse GGATGGACTTGCTGACTTGCTGTAG. beta Actin lists forward GATGAAGATCCTCACCGAACG and reverse CCTTACGGATATCAACGTCGC.
F, forward primer; R, reverse primer; URP, universal reverse primer for miRNA or U6.
Two online programs, miRanda and RNAhybrid, were used to predict the seed regions of miR-92a and miR-279d, which potentially bind to PaEno1 and the other four metal stress-related genes. Both miRNAs were first predicted by miRanda (http://www.microrna.org/microrna/getDownloads.do) (Enright et al., Reference Enright, John, Gaul, Tuschl, Sander and Marks2003) to target CDS and promoter regions of PaEno1 previously obtained earlier (Zheng et al., Reference Zheng, Gao, Li and Wang2018) and the CDS regions of the other four metal stress-related genes (sequences searched from NCBI). Subsequently, another target detection software, RNAhybrid (https://bibiserv.cebitec.uni-bielefeld.de/rnahybrid/, was also used to predict the binding site of miR-92a and miR-279d on these genes by using the default parameter including the free energy and degree.
Effect of miR-92a and miR-279d on the expression of PaEno1 and four other metal stress-related genes
The miRNA mimics (agomir-92a/agomir-279d), inhibitors (antagomir-92a/antagomir-279d), and agomir/antagomir negative controls (NC) of miR-92a and miR-279d were synthesised in GenePharma (GenePharma, Shanghai, China). To overexpress or inhibit miR-92a and miR-279d in vivo, we administered their respective agomirs (1 µM for agomir-92a and agomir-279d) or antagomirs (1 µM for antagomir-92a, 4 µM for antagomir-279d). Corresponding NCs (agomir-NC and antagomir-NC) were applied at the same concentrations. The fourth-instar P. akamusi larvae were soaked in the five treatments in a 12-well plate for 48 hours, with 10 individuals at each concentration (n = 3). The expressions of miR-92a and miR-279d were monitored through poly (A) real-time quantitative PCR (RT-qPCR) as described above. Subsequently, the expression levels of PaEno1, PaHbV, PaHbVII, alpha-esterase, and COesterase were determined by RT-qPCR analysis to evaluate the regulatory effects of these genes by miR-92a and miR-279d. Primer sequences of PaEno1, PaHbV, PaHbVII, alpha-esterase, and COesterase and the housekeeping gene (β-actin) were designed and synthesised by Sangon Biotech (Sangon, Shanghai, China) (table 1). RT-qPCR was conducted in triplicate (n = 3) using the TB Green™ Premium Ex Taq™ II (Tli RNaseH Plus) (TaKaRa, Dalian, China) on a MyiQ2 real-time PCR system (BIO-RAD, USA), following the manufacturer’s instructions. The relative miRNA and mRNA expression were calculated using the comparative Ct method (2−∆∆Ct) (Schmittgen and Livak, Reference Schmittgen and Livak2008), and the results were expressed as the mean ± SE of three replicates.
Effect of miR-92a and miR-279d on Cd2+ resistance and expression of PaEno1 and four metal stress-related genes under Cd2+ exposure
The larvae were randomly assigned to four groups: agomir-92a, antagomir-92a, agomir-279d, and NC (30 larvae per replicate, three replicates per group). They were treated during the initial 48 hours and subsequently exposed to sublethal concentrations of CdCl2 for 12 days. Larvae survival was monitored daily and assessed according to established criteria after Cd2+ exposure (Zheng et al., Reference Zheng, Gao, Li and Wang2018). The survival percentages (y-axis) were plotted against time (x-axis) using a linear scale. The expression levels of miR-92a, miR-279d, and PaEno1 in P. akamusi larvae under Cd2+ stress were measured as previously described. Additionally, we examined the expression of these metal stress-related genes after larvae were soaked in agomir-92a and agomir-279d for 48 hours and then exposed to Cd2+ for 12 days.
All experiments in this study were conducted with independent biological replicates. For each treatment group and time-point sampling, healthy larvae of similar size and developmental stage were randomly selected and randomly assigned to control and experimental groups to minimise artificial bias and individual differences. Each group contained multiple parallel replicates, and all experimental operations were performed under consistent environmental conditions to ensure the reproducibility and reliability of the data.
Statistical analyses
The expression stability of the two candidate reference genes, U6 and β-actin, was assessed using two established algorithms, geNorm and NormFinder, across all experimental groups and time points, including control, agomir/antagomir treatments, and long-term Cd2⁺ exposure. In the geNorm analysis, genes with an average expression stability value (M) below 1.5 were considered stable. For NormFinder, a lower stability value indicates more consistent gene expression across all experimental conditions. All experiments were performed with three replicates, and the results were expressed as mean ± SE. Statistical analyses were performed using SPSS 25.0 (SPSS, Chicago, IL, USA). Means were evaluated using the Kolmogorov–Smirnov test and Levene’s test to assess normality and variance homogeneity, respectively. One-way Analysis of Variance (ANOVA) and post hoc tests (Tukey’s) were used to analyse significant variations in these parameters conforming to the normal distribution and the variance. When the conditions are not met, the Kruskal–Wallis nonparametric test is carried out to determine whether the means are statistically significant. Statistical significance was indicated by p < 0.05. The significant difference between the survival curves was determined based on the Log-rank (Mantel-Cox) test. Cox proportional hazards regression was performed to calculate hazard ratios (HR) and 95% confidence intervals (CI). Graphics were produced by GraphPad Prism 8.0 (Graphpad Prism, San Diego, CA, USA).
Results
P. akamusi small RNA sequencing and differential expression of miR-92a and miR-279d
Small RNA libraries based on the pooled RNA from the control and Cd2+-treated fourth-instar larvae of P. akamusi were sequenced and constructed. We obtained 268.37 million and 583.47 million raw reads in the CK and Cd-treated groups, respectively. After quality control, 147.80 million and 342.24 million clean reads were obtained from the CK and Cd-treated groups, which accounted for 65.13% and 58.66% of the raw sequences, respectively (fig. 1A). The size distribution and abundance of all sequences in the two libraries between 17 and 35 nt was analysed and revealed the peak distribution at 20–21 nt (fig. 1B). A comparison of the sRNA sequence in P. akamusi larvae with the Rfam database revealed that 0.26%, 2.32%, 0.21%, and 16.57% of sRNA in the control group were identified as snRNA, tRNA, snoRNA, and rRNA, respectively. In contrast, the percentages of these four types of sRNA in the Cd2+-treated group were 0.16%, 2.40%, 0.16%, and 23.50% (fig. 1C and 1D). By using the edgeR software with the screening criteria of p < 0.05 and fold change > 2, 125 significantly expressed miRNAs were identified, including 49 up-regulated and 76 down-regulated miRNAs in the Cd2+-treated group (fig. 1E).
Sequence summary of small RNA and differential expression of miR-92a and miR-279d in P. akamusi in the control (CK) and Cd2+-treated groups. (A) Statistics of raw and clean reads number. (B) Size distribution for the unique sequence reads. Read frequency for all types of small RNA reads in the CK (C) and Cd2+-treated groups (D). (E) Numbers of up-regulated (pink dots) and down-regulated (blue dots) miRNAs under Cd2+ stress. (F) Expression levels of miR-92a and miR-279d in the CK and Cd2+-exposed groups using miRNA sequencing and RT-qPCR analysis.
Figure 1 Long description
A grouped bar chart with two categories on the horizontal axis, CK and Cd2 plus treated. The vertical axis label is Total reads. The vertical axis shows tick labels 0, 2 times 10 superscript 7, 4 times 10 superscript 7, 6 times 10 superscript 7 and 8 times 10 superscript 7. A legend lists Raw reads and Clean reads. The CK bars are near 3 times 10 superscript 7 for Raw reads and near 2 times 10 superscript 7 for Clean reads. The Cd2 plus treated bars are near 6 times 10 superscript 7 for Raw reads and near 3.5 times 10 superscript 7 for Clean reads. A bar chart. The horizontal axis label is Length, left parenthesis nt right parenthesis. The horizontal axis tick labels run from 17 to 35 in steps of 1. The vertical axis label is Unique reads. The vertical axis tick labels include 0, 50000, 100000, 150000, 200000 and 250000. A legend lists CK and Cd2 plus treated. Bars for both series are highest around lengths 20 to 22 and then decrease toward length 35. A pie chart labeled CK. A legend lists snRNA, tRNA, snoRNA, rRNA and other. Percentage labels on slices include 0.26 percent, 2.32 percent, 0.21 percent, 16.57 percent and 80.64 percent. A pie chart labeled Cd2 plus treated. A legend lists snRNA, tRNA, snoRNA, rRNA and other. Percentage labels on slices include 0.16 percent, 2.40 percent, 0.16 percent, 23.50 percent and 73.78 percent. A scatter plot. The horizontal axis label is log2, left parenthesis Fold Change right parenthesis, CK vs Cd2 plus treated. The horizontal axis shows tick labels negative 10, negative 5, 0, 5 and 10. The vertical axis label is minus log10, left parenthesis P value right parenthesis. The vertical axis shows tick labels 0, 2, 4, 6, 8 and 10. A legend lists down regulated, no regulated and up regulated. Points form a V shaped distribution with higher vertical values near horizontal values around negative 5 to negative 2 and around positive 5 to positive 10. Two grouped bar charts labeled miRNA 92a and miRNA 279. Each chart has two categories on the horizontal axis, CK and Cd2 plus treated. Each chart has a left vertical axis label, RT qPCR. Each chart also has a right vertical axis label, RNA seq. A legend lists RT qPCR and RNA seq. In the miRNA 92a chart, the RT qPCR bar is higher for Cd2 plus treated than for CK and the RNA seq line decreases from CK to Cd2 plus treated. In the miRNA 279 chart, the RT qPCR bar is higher for Cd2 plus treated than for CK and the RNA seq line decreases from CK to Cd2 plus treated.
To ensure accurate normalisation for qPCR analysis under Cd stress, we evaluated the expression stability of two candidate reference genes, U6 (Supplementary Table S1) and β-actin (Supplementary Table S2), across different treatment groups (CK and AGO) and time points (0–12 days) using geNorm and NormFinder (Supplementary Table S3). The geNorm analysis showed that both genes were stable, with M-values of 0.83 for β-actin and 1.28 for U6, both below the acceptability threshold of 1.5 (Supplementary Table S3). This result was corroborated by NormFinder, which calculated stability values of 0.19 for β-actin and 0.36 for U6 (Supplementary Table S3). Collectively, these results demonstrate that both β-actin and U6 are suitable for use as reference genes in this experimental system.
Of the aforementioned, Cd2+ was observed to significantly elevate and suppress the expression of miR-92a and miR-279d, with a log2 fold change of 1.607 and 3.864, respectively (p < 0.05) (fig. 1F). Similarly, the qRT-PCR results demonstrated a significant increase and decrease in the expression levels of miR-92a and miR-279d in P. akamusi under Cd2+ stress compared to the control group, with expression levels changing by 2.825-fold and 4.225-fold, respectively (p < 0.05, fig. 1F). These results indicated that miR-92a and miR-279d may play crucial roles in the Cd2+-resistance mechanisms of P. akamusi.
Identification of miR-92a and miR-279d targeting and their regulation on the expression of PaEno1 and metal stress-related genes in P. akamusi
Bioinformatic analysis predicted putative complementary binding sites for miR-92a in the CDS region of PaEno1 (1556–1592 bp) with a free energy of −11.4 kcal/mol (fig. 2A). Similarly, putative binding sites of miR-279d were predicted in the PaEno1 CDS region (981–1022 bp) with a free energy of −23.5 kcal/mol (fig. 2B). Following a 48-hour incubation period with agomir 92a, the expression level of miR-92a in the larvae exhibited a 246.4-fold increase, while the relative expression of PaEno1 in P. akamusi larvae demonstrated an 80% reduction (fig. 2C and 2E). In contrast, the relative expression of miR-92a in the larvae was found to decrease by 97.8%, while the expression level of PaEno1 increased by 10.5-fold following the administration of a 48-hour miR-92a antagomir treatment (fig. 2C and 2E). Similarly, miR-279d agomir successfully induced the expression of tca-miR-279d-5p (fig. 2D). However, no inhibitory effect was observed even in larvae treated with the miR-279d antagomir up to 4 μM (fig. 2D). It is noteworthy that the expression of PaEno1 was inhibited and induced, respectively, following treatment with agomir-279d and antagomir-279d for a soaking of 48 hours (fig. 2F).
MiR-92a and miR-279d targeting and their regulation on the expression of PaEno1. MiR-92a (A) and miR-279d (B) target sites were predicted in the coding sequence (CDS) of PaEno1 of P. akamusi. Expression levels of miR-92a (C) and miR-279d (D) in P. akamusi after administration of agomir/antagomir by soaking. (E) Expression levels of PaEno1 after miR-92a agomir and antagomir treatment and (F) Expression levels of PaEno1 after miR-279d agomir and antagomir treatment. Data are presented as mean ± SD. *p < 0.05, **p < 0.01.
Figure 2 Long description
Predicted pairing diagram between PaEno1 and bmo-miR-92a. Text states, mfe: minus 21.3 kcal per mol and Position: 365. Two aligned letter strings are labeled PaEno1 and bmo-miR-92a, with 3 prime and 5 prime markers at the right ends. Predicted pairing diagram between PaEno1 and tca-miR-279d-5p. Text states, mfe: minus 23.5 kcal per mol and Position: 981. Two aligned letter strings are labeled PaEno1 and tca-miR-279d-5p, with 3 prime and 5 prime markers at the right ends. Bar charts titled, miR-92a, with two conditions per chart and significance brackets. Chart 1 y-axis label is, Relative expression of miR-92a, with labeled ticks 0, 100, 200, 300, 400. The x-axis categories are, NC agomir and miR-92a agomir. The miR-92a agomir bar is taller than the NC agomir bar and the comparison is marked with three asterisks. Chart 2 y-axis label is, Relative expression of PaEno1, with labeled ticks 0.00, 0.01, 0.02, 0.03, 0.04, 0.05. The x-axis categories are, NC antagomir and miR-92a antagomir. The miR-92a antagomir bar is taller than the NC antagomir bar and the comparison is marked with two asterisks. Units beyond the phrase, Relative expression, are not shown. Bar charts titled, miR-279d, with two conditions per chart and significance brackets. Chart 1 y-axis label is, Relative expression of miR-279d, with labeled ticks 0, 500, 1000, 1500. The x-axis categories are, NC agomir and miR-279d agomir. The miR-279d agomir bar is taller than the NC agomir bar and the comparison is marked with two asterisks. Chart 2 y-axis label is, Relative expression of PaEno1, with labeled ticks 0, 5, 10, 15, 20, 25. The x-axis categories are, NC antagomir and miR-279d antagomir. The miR-279d antagomir bar is taller than the NC antagomir bar and the comparison is marked with one asterisk. Units beyond the phrase, Relative expression, are not shown. Grouped bar chart with y-axis label, Relative expression of PaEno1 and labeled ticks 0, 4, 8, 12. The x-axis categories are, agomir and antagomir. The legend lists two series, NC and bmo-miR-92a. In the agomir group, the bmo-miR-92a bar is lower than the NC bar, with a bracket labeled three asterisks. In the antagomir group, the bmo-miR-92a bar is higher than the NC bar, with a bracket labeled three asterisks. Units beyond the phrase, Relative expression, are not shown. Grouped bar chart with y-axis label, Relative expression of PaEno1 and labeled ticks 0.0, 0.5, 1.0, 1.5, 2.0. The x-axis categories are, agomir and antagomir. The legend lists two series, NC and tca-miR-279d-5p. In the agomir group, the tca-miR-279d-5p bar is lower than the NC bar, with a bracket labeled one asterisk. In the antagomir group, the tca-miR-279d-5p bar is higher than the NC bar, with a bracket labeled one asterisk. Units beyond the phrase, Relative expression, are not shown.
Furthermore, no additional complementary binding sites for miR-92a were identified in the remaining four metal stress-related genes. The expression of PaHbV and PaHbVII was down-regulated in the miR-92a agomir group, whereas the expressions of PaHbV, alpha-esterase, and COesterase were up-regulated in the miR-92a antagomir group (Supplementary Fig. S1A–1D). For miR-279d, two distinct binding sites for the PaHbV gene were identified, located at 365–398 bp and 300–327 bp (Supplementary Fig. S2A and S2B). The expression patterns of the PaHbV, PaHbVII, and alpha-esterase genes were similar to that of miR-279d following treatment with agomir-279d and antagomir-279d for 48 hours (Supplementary Fig. S2C–S2E), while the relative expression of COesterase significantly increased in the miR-279d agomir group (p < 0.05, respectively) (Supplementary Fig. S2F).
Our results indicate that among the five metal-stress-related genes, miR-92a has a potential targeting and regulatory effect on PaEno1 in vivo, while no targeting or regulatory effects are observed for the other four genes. Furthermore, miR-279d may be involved in the regulation of PaEno1. Although miR-279d harbours two specific target sites in the PaHbV gene, antagomir-279d failed to suppress the expression of endogenous miR-279d. Intriguingly, both agomir-279d and antagomir-279d treatments triggered an inverse expression pattern of PaEno1 in larvae.
The role of miR-92a and miR-279d in Cd2+ resistance and their regulatory effect on PaEno1 and four other metal stress-related genes under Cd2+ exposure
For miR-92a, survival curves under Cd2⁺ exposure were plotted using the Kaplan–Meier method and compared by the Log-rank test (χ2 = 1.524, df = 2, p = 0.4668, fig. 3A). Cox proportional hazards regression (Supplementary Table S4) revealed no significant differences in mortality risk relative to the control group for either agomir-92a (HR = 0.810, 95% CI: 0.399–1.642, p = 0.558) or antagomir-92a (HR = 0.869, 95% CI: 0.434–1.740, p = 0.691). Similarly, for miR-279, Log-rank analysis showed no significant divergence among survival curves (χ2 = 2.362, df = 2, p = 0.307, fig. 3B). Cox regression (Supplementary Table S4) indicated that neither agomir-279 (HR = 0.715, 95% CI: 0.266–1.912, p = 0.506) nor antagomir-279 (HR = 0.609, 95% CI: 0.217–1.712, p = 0.347) significantly altered larval survival compared with the CK group. Overall, these results demonstrated that neither overexpression nor inhibition of miR-92a or miR-279 significantly affected the survival of Propsilocerus akamusi larvae under Cd stress.
Functional study of miR-92a and miR-279d on PaEno1 expression in the P. akamusi larvae exposed to Cd2+ stress. The survival rate of agomir-92a, antagomir-92a (A), and agomir-279d, antagomir-279d (B) soaked P. akamusi larvae after a sublethal dose of Cd2+ exposure (n = 30 larvae, three replicates for each group). Relative expression of miR-92a and miR-279d in the P. akamusi larvae exposed to Cd2+ after 48 hours of miR-92a agomir (C), miR-92a antagomir (E), and miR-279d agomir (G). Relative expression of PaEno1 in the P. akamusi larvae exposed to Cd2+ after 48 hours of miR-92a agomir (D), miR-92a antagomir (F) and miR-279d agomir (H) (n = 3). An asterisk represents significant differences between comparisons. Ns, p > 0.05, *p < 0.05; **p < 0.01; ***p < 0.001.
Figure 3 Long description
A line graph with the vertical axis label, Survival percent, ranging from 70 to 100 and the horizontal axis label, Time left parenthesis Days right parenthesis, ranging from 0 to 12. A text label reads, P equals 0.4668. A legend lists, miR 279d agomir, miR 279d antagomir and CK. The plotted lines step down over time. The miR 279d agomir line is at 100 at day 0, about 95 at day 1, about 90 at day 2, about 85 at day 8, about 80 at day 10 and about 75 at day 12. The miR 279d antagomir line is at 100 at day 0, about 95 at day 1, about 90 at day 2, about 90 at day 8, about 85 at day 10 and about 85 at day 12. The CK line is at 100 at day 0, about 95 at day 1, about 90 at day 2, about 85 at day 3, about 85 at day 8, about 80 at day 10 and about 80 at day 12. The image B showing a line graph with the vertical axis label, Survival percent, ranging from 70 to 100 and the horizontal axis label, Time left parenthesis Days right parenthesis, ranging from 0 to 12. A text label reads, P equals 0.9619. A legend lists, miR 92a agomir, miR 92a antagomir and CK. The plotted lines step down over time. The miR 92a agomir line is at 100 at day 0, about 90 at day 1, about 85 at day 2, about 80 at day 6, about 80 at day 10 and about 75 at day 12. The miR 92a antagomir line is at 100 at day 0, about 90 at day 1, about 85 at day 2, about 80 at day 6, about 75 at day 8, about 75 at day 10 and about 75 at day 12. The CK line is at 100 at day 0, about 90 at day 1, about 85 at day 2, about 80 at day 3, about 75 at day 6, about 75 at day 10 and about 70 at day 12. The image C showing a grouped bar chart with the vertical axis label, Relative expression of miR 92a, ranging from 0 to 40 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 3, 6, 9, 12. A legend lists, agomir NC and agomir 92a. Bars at day 0 are about 10 for agomir NC and about 30 for agomir 92a. Bars at day 3 are about 10 for agomir NC and about 25 for agomir 92a. Bars at day 6 are about 10 for agomir NC and about 15 for agomir 92a. Bars at day 9 are about 10 for agomir NC and about 20 for agomir 92a. Bars at day 12 are about 10 for agomir NC and about 15 for agomir 92a. The image D showing a grouped bar chart with the vertical axis label, Relative expression of PaEno1, ranging from 0 to 1.5 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 3, 6, 9, 12. A legend lists, agomir NC and agomir 92a. Bars at day 0 are about 1.0 for agomir NC and about 0.2 for agomir 92a. Bars at day 3 are about 1.0 for agomir NC and about 0.5 for agomir 92a. Bars at day 6 are about 1.0 for agomir NC and about 0.8 for agomir 92a. Bars at day 9 are about 1.0 for agomir NC and about 1.0 for agomir 92a. Bars at day 12 are about 1.0 for agomir NC and about 0.5 for agomir 92a. The image E showing a grouped bar chart with the vertical axis label, Relative expression of miR 92a, ranging from 0 to 2.0 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 6, 8, 10, 12. A legend lists, antagomir NC and antagomir 92a. Bars at day 0 are about 1.2 for antagomir NC and about 0.2 for antagomir 92a. Bars at day 6 are about 1.2 for antagomir NC and about 0.2 for antagomir 92a. Bars at day 8 are about 1.2 for antagomir NC and about 0.2 for antagomir 92a. Bars at day 10 are about 1.2 for antagomir NC and about 0.2 for antagomir 92a. Bars at day 12 are about 1.2 for antagomir NC and about 0.2 for antagomir 92a. The image F showing a grouped bar chart with the vertical axis label, Relative expression of PaEno1, ranging from 0 to 1.5 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 6, 8, 10, 12. A legend lists, antagomir NC and antagomir 92a. Bars at day 0 are about 0.1 for antagomir NC and about 1.1 for antagomir 92a. Bars at day 6 are about 0.1 for antagomir NC and about 0.1 for antagomir 92a. Bars at day 8 are about 0.1 for antagomir NC and about 0.1 for antagomir 92a. Bars at day 10 are about 0.1 for antagomir NC and about 0.1 for antagomir 92a. Bars at day 12 are about 0.1 for antagomir NC and about 0.1 for antagomir 92a. The image G showing a grouped bar chart with the vertical axis label, Relative expression of miR 279d, ranging from 0 to 2.0 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 3, 6, 9, 12. A legend lists, agomir NC and agomir 279d. Bars at day 0 are about 1.0 for agomir NC and about 1.8 for agomir 279d. Bars at day 3 are about 1.0 for agomir NC and about 1.7 for agomir 279d. Bars at day 6 are about 1.0 for agomir NC and about 2.0 for agomir 279d. Bars at day 9 are about 1.0 for agomir NC and about 1.6 for agomir 279d. Bars at day 12 are about 1.0 for agomir NC and about 1.5 for agomir 279d. The image H showing a grouped bar chart with the vertical axis label, Relative expression of PaEno1, ranging from 0 to 2.0 and the horizontal axis label, Time left parenthesis Days right parenthesis, with tick labels 0, 3, 6, 9, 12. A legend lists, agomir NC and agomir 279d. Bars at day 0 are about 1.0 for agomir NC and about 0.8 for agomir 279d. Bars at day 3 are about 1.0 for agomir NC and about 0.2 for agomir 279d. Bars at day 6 are about 1.0 for agomir NC and about 0.3 for agomir 279d. Bars at day 9 are about 1.0 for agomir NC and about 1.2 for agomir 279d. Bars at day 12 are about 1.0 for agomir NC and about 1.0 for agomir 279d.
We observed the overexpression and inhibition of miR-92 following 48 hours of treatment with agomir-92a and antagomir-92a during the entire Cd2+ exposure period (fig. 3C and 3E). This indicates that agomir-92a and antagomir-92a consistently influenced the expression of miR-92a throughout the Cd2+ exposure. However, the inhibited expression of PaEno1 in the agomir-92a group was only detected on day 0 (before Cd2+ exposure), as well as on days 6 and 12 (fig. 3D). Furthermore, when the expression of miR-92a was inhibited in larvae treated with antagomir-92a throughout the Cd2+ exposure (fig. 3E), PaEno1 expression was significantly up-regulated before Cd2+ exposure (day 0) and subsequently downregulated on day 8 and 10 after exposure (fig. 3F). We also observed the overexpression of miR-279d after 48 hours of treatment with agomir-279d throughout the Cd2+ exposure period (fig. 3G) and a gradual decline in the expression levels of PaEno1 between the 3rd and 6th day (fig. 3H) then increased but not significant levels on days 9 and 12 (fig. 3G). Collectively, these findings suggest that the regulatory interaction between miR-92a/miR-279d and PaEno1 in response to Cd2+ exposure is not persistent. This implies that Cd2+ may temporally interfere with the regulatory effect of miR-92a and miR-279d on PaEno1.
We also examined the response of four other metal stress-related genes after the larvae were soaked in agomir-92a for 48 hours and then exposed to Cd2+ for 12 days (fig. 4A–4D). It is important to note that although miR-92a does not have target sites on these four genes and does not regulate their expression, we found that the expression of PaHbV, PaHbVII, and alpha-esterase was also inhibited on days 6 and 12 (fig. 4A–4C). However, the sole induced gene expression of COesterase was observed on day 12 of Cd2+ exposure (fig. 4D). It appears that COesterase may play an important role in the Cd2+ resistance of P. akamusi larvae when the expression of the other four genes (PaEno1, PaHbV, PaHbVII, and alpha-esterase) was inhibited.
Relative expression of four metal stress-related genes in the P. akamusi larvae exposed to Cd2+ stress after 48 hours of miR-92a agomir (A–D) treatment (n = 3). (A) PaHbV, (B) PaHbVII, (C) alpha-esterase, and (D) COesterase. Data are presented as mean ± SD. *p < 0.05, **p < 0.01.
Figure 4 Long description
A grouped bar chart with the vertical axis label, Relative expression of PaHbV, with tick labels 0, 0.5, 1.0, 1.5. The horizontal axis label is, Time (Days), with tick labels 0, 6, 12. A legend lists agomir-NC and agomir-92a. At each time point 0, 6 and 12, the agomir-NC bar is near 1.0 and the agomir-92a bar is lower, near 0.2 to 0.5. Brackets with asterisks appear above each time point comparison: three asterisks at 0, three asterisks at 6 and two asterisks at 12. The image B showing a grouped bar chart with the vertical axis label, Relative expression of PaHbVII, with tick labels 0, 0.5, 1.0, 1.5. The horizontal axis label is, Time (Days), with tick labels 0, 6, 12. A legend lists agomir-NC and agomir-92a. At time 0 and 6, the agomir-NC bars are near 1.0 and the agomir-92a bars are lower, near 0.6. At time 12, the agomir-NC bar is near 1.0 and the agomir-92a bar is lower, near 0.2. Brackets with one asterisk appear above the comparisons at 0 and 6. The image C showing a grouped bar chart with the vertical axis label, Relative expression of alpha-esterase, with tick labels 0, 0.5, 1.0, 1.5, 2.0, 2.5. The horizontal axis label is, Time (Days), with tick labels 0, 6, 12. At time 0, the agomir-92a bar is higher than the agomir-NC bar, with the agomir-92a bar near 1.8 and the agomir-NC bar near 1.0. At time 6 and 12, the agomir-NC bars are near 1.0 and the agomir-92a bars are lower, near 0.2 to 0.3. Brackets with three asterisks appear above the comparisons at 6 and 12. The image D showing a grouped bar chart with the vertical axis label, Relative expression of COesterase, with tick labels 0, 1, 2, 3. The horizontal axis label is, Time (Days), with tick labels 0, 6, 12. A legend lists agomir-NC and agomir-92a. At time 0 and 6, both bars are near 1.0. At time 12, the agomir-92a bar is higher, near 2.3 and the agomir-NC bar is near 1.0. A bracket with three asterisks appears above the comparison at 12. Across A to D, each chart uses the same time points 0, 6 and 12 and compares agomir-NC with agomir-92a for the labeled target.
After treating the larvae with agomir-279d followed by 12 days of Cd2⁺ exposure, we observed that a 48-hour miR-279d agomir treatment prior to Cd2⁺ exposure up-regulated the expression of PaHbV, PaHbVII, alpha-esterase, and COesterase, as shown on day 0 (fig. 5A–5D). From day 3 to day 6, the expression levels of PaHbVII and alpha-esterase gradually decreased (fig. 5B–5C). In contrast, the expression of PaHbV, PaHbVII, and alpha-esterase increased by day 12 (fig. 5A–5C). Meanwhile, COesterase expression was up-regulated on day 9 but downregulated on day 12 (fig. 5D). Overall, agomir-279d treatment prior to Cd2⁺ exposure induced dynamic changes in the expression of PaHbV, PaHbVII, alpha-esterase, and COesterase over the experimental period.
Relative expression of four metal stress-related genes (A: PaHbV, B: PaHbVII, C: alpha-esterase, D: COesterase) in P. akamusi larvae after exposure to Cd2⁺ stress and 48-hour treatment with miR-279d agomir (n = 3). Data are presented as mean ± SD. *p < 0.05, **p < 0.01.
Figure 5 Long description
Energy flow within a greenhouse structure is illustrated, depicting six labeled processes. At the top left, the sun emits short wavelength radiation, shown by a solid arrow entering the greenhouse. Inside, plants are depicted with arrows indicating long wavelength radiation, represented by dashed lines, moving upwards. Convection is shown with curved arrows circulating within the greenhouse. Soil evaporation is indicated by wavy arrows rising from the soil bed. Heat loss by ventilation is depicted with a curved arrow exiting the greenhouse. Conduction is represented by double-headed arrows along the soil bed. A cloud is shown outside the greenhouse, with a dashed arrow indicating long wavelength radiation escaping towards it. The key on the right explains the symbols: convection, short wavelength, long wavelength, soil evaporation, heat loss by ventilation and conduction.
Discussion
In this study, we hypothesised that miRNAs inhibit the transcription of PaEno1, potentially serving as a key mechanism for enhancing resistance to Cd2+ stress in this species. The detection of 49 up-regulated and 76 down-regulated miRNAs in the Cd-treated groups indicates their sensitivity to Cd2+ and suggests their potential as biomarkers for environmental monitoring, influencing the response of P. akamusi to Cd2+ stress. Through small RNA sequencing, differential expression analysis using edgeR software, and qRT-PCR analysis, we identified miR-92a and miR-279d as potential miRNAs that target and regulate the expression of PaEno1 in P. akamusi.
MiR-92a is a member of the highly conserved miRNA-17-92 gene cluster. It has been documented to exert pivotal functions in multiple stress tolerance and insecticide resistance processes: it modulates cytochrome P450 to participate in DDT detoxification in Drosophila melanogaster (Seong et al., Reference Seong, Coates and Pittendrigh2019), mediates deltamethrin resistance by negatively regulating CpCPR4 expression in female mosquitoes (Ma et al., Reference Ma, Li, Hu, Zhou, Sun, Ma, Zhu and Shen2017), and contributes to Cd tolerance in Daphnia pulex (Chen et al., Reference Chen, Nichols, Poynton and Sepúlveda2016). Environmental stress can also regulate the expression of target genes by modulating miR-279, which further improves organismal adaptability to adverse environments. For instance, miR-279 mediates the deltamethrin resistance of Culex pipiens pallens by targeting cytochrome P325BB1 (Li et al., Reference Li, Hu, Zhang, Yin, Wang, Zhou, Sun, Ma, Shen and Zhu2021). It also suppresses cell growth and proliferation via the GTPase and keratin signaling pathways, consequently enhancing the Cd tolerance of Daphnia pulex (Chen et al., Reference Chen, Nichols, Poynton and Sepúlveda2016).
Our results confirmed the induction of miR-92a and inhibition of miR-279d under Cd2+ stress in P. akamusi larvae and also their function to target and regulate the expression of PaEno1. Although in vivo antagomir treatment failed to significantly suppress miR-279d expression, an opposite expression pattern of PaEno1 was observed in larvae co-treated with agomir-279d and antagomir-279d. This discrepancy may be explained by multiple underlying mechanisms. First, antagomirs can sequester endogenous miRNAs without triggering their degradation (Esau, Reference Esau2008). Second, miRNAs mediate cell- or organ-specific regulation of mRNA levels, which frequently exhibits a poor correlation with the expression abundance of the corresponding miRNA (Correia de Sousa et al., Reference Correia de Sousa, Gjorgjieva, Dolicka, Sobolewski and Foti2019). Third, a single miRNA often targets multiple mRNAs, and different miRNAs can concurrently regulate the same mRNA, forming complex regulatory networks (Esau, Reference Esau2008; Liu et al., Reference Liu, Shi, Yin, Su, Qi, Huang, Zhang and Yu2017). Fourth, miRNAs possess pleiotropic properties and intricate mutual interactions (Cannataro and Cione, Reference Cannataro and Cione2023). Collectively, these mechanisms account for the phenomenon whereby antagomir-279d did not alter miR-279d abundance but still exerted a regulatory effect on PaEno1 expression.
Based on the survival rate of P. akamusi larvae treated with miR-92a/279d antagomir and miR-92a/279d agomir after exposure to a sublethal dose of Cd2+, we found no evidence of their roles in Cd2+ resistance. This indicated that miR-92a and miR-279d do not directly enhance Cd tolerance. Instead, they may act as temporal regulators of PaEno1 within the broader stress-responsive network. Although agomir-induced overexpression of miR-92a/279d and antagomir-mediated inhibition of miR-92a persistently changed miRNA abundance during Cd2⁺ exposure, these alterations were not paralleled by consistent changes in PaEno1 expression. Instead, weak temporal decoupling was evident at several time points. Such subtle asynchrony is prevalent in in vivo regulatory networks, likely arising from transcriptional time lags, multilayered post-transcriptional modifications, and the involvement of additional co-regulatory molecules. These findings indicate that miR-92a and miR-279d modulate PaEno1 expression in a dynamic, time-dependent manner under heavy metal stress, with their regulatory interaction being transient rather than sustained. Under prolonged Cd2⁺ challenge, other uncharacterised miRNAs and post-transcriptional regulators may become dominant in controlling PaEno1 expression. This pattern reflects a layered, temporally sophisticated regulatory strategy employed by organisms to cope with environmental stress. This interpretation is supported by previous studies. A single gene can be targeted by multiple miRNAs, and its final expression is determined by integrated combinatorial regulation (Schmiedel et al., Reference Schmiedel, Klemm, Zheng, Sahay, Blüthgen, Marks and van Oudenaarden2015). Moreover, miRNA–target interactions often display distinct temporal and spatial specificity (van Wijk et al., Reference van Wijk, Zohar and Linial2022), reinforcing the context-dependent nature of miRNA-mediated regulation. Importantly, our whole-larva homogenate analysis cannot exclude the possibility that tissue-specific regulatory interactions between these miRNAs and PaEno1 are masked in specialised tissues responsible for Cd2⁺ detoxification.
The resistance of P. akamusi larvae to Cd2+ represents a complex biological process that likely involves a regulatory network and multiple genes. Several studies have indicated that insects expend substantial energy to mitigate the detrimental effects of Cd (Shen et al., Reference Shen, Hussain, Liu, Li, Yu, Zhao, Li and Yang2024). In response to physiological demands, the expression levels of Eno1 fluctuate according to the physiological state of the organism. Chironomus riparius populations can reduce Cd levels in their bodies through excretion (Doria et al., Reference Doria, Hannappel and Pfenninger2022). This process typically requires energy expenditure to facilitate excretion, and Eno1 can enhance ATP production during energy metabolism (Wang et al., Reference Wang, Li, Qiao, Ge, Ma and Wang2020). Previous research has demonstrated that elevated levels of PaEno1 may boost ATP production, potentially contributing to the resistance of this species to Cd2+ stress (Zheng et al., Reference Zheng, Li and Gao2020).
In this study, we observed the induction of PaHbV, PaHbVII, alpha-esterase, and COesterase when the expression of miR-279d was inhibited. Some studies have described this phenomenon as a compensatory effect, wherein the loss of function of one gene may be offset by another gene that possesses overlapping functions and expression patterns, thereby preserving cellular integrity (El-brolosy and Stainier, Reference El-brolosy and Stainier2017). Xu et al. (Reference Xu, Shi and Wang2016) demonstrated that organisms can activate effective compensatory mechanisms, including stress-related defence pathways, to facilitate metal detoxification. Chironomus haemoglobins (Hbs) serve as key environmental biomarkers and confer strong adaptability to extreme habitats in chironomid larvae (Ha and Choi, Reference Ha and Choi2008). With high oxygen affinity and polymorphism, Hbs facilitate oxygen transport and storage under polluted, anoxic conditions, sustaining aerobic metabolism and energy supply. They notably enhance larval tolerance to environmental stress and Cd2⁺ exposure. Additionally, Sun et al. (Reference Sun, Liu, Li, Zhao, Pan and Yan2021) demonstrated that copper (Cu) can induce the expression of the PaHb3, PaHb5, PaHb13, and PaHb15 genes, indicating a potential role for the Hb gene in the Cu tolerance of P. akamusi larvae. In this study, the overexpression of miR-279d significantly enhanced the expression of the HbV gene on day 3 of Cd2+ exposure. This increase is likely attributed to the elevated oxygen demand of P. akamusi larvae, which facilitates energy production and contributes to the tolerance of this insect to Cd2+ stress. Future investigations incorporating measurements of Cd accumulation, ATP content, and oxidative stress markers will help clarify the physiological consequences of this regulatory axis.
Esterases are a diverse enzyme family that hydrolyse ester compounds, among which carboxylesterases exhibit broad substrate specificity and are linked to insecticide resistance, while the alpha-esterase gene encodes an isoform participating in xenobiotic detoxification, lipid metabolism and reproductive regulation. As key glycoproteins, cholinesterases regulate cholinergic nerve signal transmission; their inhibition leads to acetylcholine accumulation and neurotoxicity. Owing to their high sensitivity to organophosphorus inhibition, carboxylesterase and cholinesterase have become major research targets in insects exposed to organophosphorus insecticides (Cruse et al., Reference Cruse, Moural and Zhu2023; Oksana et al., Reference Lockridge, Quinn and Radić2023). Recent research has demonstrated that metal exposure can induce the activity of both enzymes. For example, high activity of carboxylesterases has been observed in female ground beetles along a gradient of Zn2+ pollution (Gao et al., Reference Gao, Zhao, Yang, Zhang, Bai, Hu and Hu2014), insecticide resistance in Aedes albopictus (Grigoraki et al., Reference Grigoraki, Pipini, Labbé, Chaskopoulou, Weill and Vontas2017), and Oxya chinensis exposed to Cd2+ (Zhang et al., Reference Zhang, Song, Wu, Yang, Zhang, Li, Ma and Guo2014). In the present study, we explored the expression induction of the alpha-esterase gene under late-stage Cd2⁺ stress following miR-279d agomir treatment. We also determined the changes in carboxylesterase activity on days 9 and 12 of Cd2⁺ exposure after intervention with miR-279d agomir and miR-92. Compared with our findings, elevated acetylcholinesterase activity has also been reported in the soil–cotton plant–cotton bollworm system under Cd2+ stress (Shen et al., Reference Shen, Hussain, Liu, Li, Yu, Zhao, Li and Yang2024).
However, the interpretation of the stimulation of carboxylesterases and cholinesterase activity by metals remains unclear, as metal ions may not be necessary cofactors for either enzyme. Some studies suggested that Cd may indirectly induce the activity of carboxylesterases and promote the production of metabolites from carboxylesterase substrates (Chang et al., Reference Chang, Xu, Wang, Zhu, Ru, Xu, Chen and Li2024). Although the roles of both enzymes in lipid metabolism have been recognised for many years (Reuter and Geus, Reference Reuter and Geus1987; Ross et al., Reference Ross, Streit and Herring2010), few studies have explored their functions in the context of environmental stress regarding lipid metabolism and energy supply. This raises the speculation that the induced activity of carboxylesterases and cholinesterases may play a role in lipid metabolism to support energy supply when organisms are subjected to metal stress, potentially challenging traditional understandings of the mechanisms by which both enzymes respond to environmental stressors.
Collectively, the coordinated expression of Hb, esterase, and energy metabolism-related genes enables P. akamusi larvae to mount a dynamic adaptive response against Cd2⁺ toxicity following miR-92a and miR-279d agomir treatment. The distinct upregulation of key genes on day 9 represents a stage-specific molecular buffering strategy, which triggers physiological compensation to rebalance metabolic homeostasis and mitigate Cd-induced damage, consistent with the temporal regulatory characteristics of miRNAs under persistent environmental stress. Functionally, P. akamusi may enhance oxygen supply and energy production via Hb genes, while esterase-mediated lipid metabolism optimises energy allocation, jointly improving larval tolerance to Cd2⁺ stress. Notably, the proposed compensatory model remains preliminary due to the lack of network correlation and multivariate analyses; further multi-omics integration and statistical modelling are warranted to validate these regulatory interactions in future research.
Conclusions
This study revealed that PaEno1 acts as a putative target of miR-92a and miR-279d with coordinated expression patterns under Cd2⁺ stress, while direct physical binding remains unvalidated due to the absence of dual-luciferase reporter and site mutagenesis assays, suggesting potential indirect regulation. Neither miRNA significantly altered larval survival upon Cd exposure, indicating they are not key determinants of Cd tolerance. Given the ineffective inhibition of miR-279d by its antagomir, its regulatory role is mainly inferred from agomir overexpression data. Collectively, miR-92a and miR-279d function as regulatory modulators rather than direct contributors to Cd tolerance, and they are not the exclusive upstream regulators of PaEno1, supporting the existence of a complex miRNA–gene network that orchestrates the larval stress response instead of a single linear pathway. Several limitations should be acknowledged herein: the lack of molecular validation for miRNA–target interaction, the poor knockdown efficiency of miR-279d antagomir, and whole-larva RNA sampling that may mask tissue-specific regulation in critical functional tissues such as the gut and fat body. Future research is recommended to verify direct binding relationships, optimise antagomir delivery systems, and adopt tissue-specific expression profiling to further dissect the spatiotemporal regulatory mechanism underlying Cd2⁺ stress adaptation in P. akamusi larvae.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S000748532610128X.
Acknowledgements
The authors thank Professor Daichou Zhang of Prefectural University of Kumamoto in Japan for his guidance in the writing process.
Data availability statement
The datasets used and/or analysed during the current study can be obtained from the corresponding author upon reasonable request.
Ethics and approvals
Not applicable.
Financial support
This work was financially supported by the Research Project Supported by Shanxi Scholarship Council (2023-030), National Natural Science Foundation of China (42477031 and 42307303).
Competing interests
The authors declare no competing interests.
