Evaluation of siRNA-mediated knockdown of heat shock protein 16.2 in adult Acropora cervicornis

Abstract

Acropora cervicornis is a critically endangered Caribbean coral species facing a precipitous decline due to disease and thermal bleaching. To combat the loss of ecologically important reef-building corals, novel molecular approaches can be used to understand mechanisms of resilience and ultimately provide a means to modulate coral performance. In this study, we tested a recently developed method for gene knockdown in adult corals that utilizes short interfering RNAs (siRNAs) and the RNA interference pathway. The Heat Shock Protein 16.2 (HSP16.2) gene in fragments of adult A. cervicornis was targeted for knockdown using a single siRNA construct. RT-qPCR analysis of HSP16.2 gene expression was performed between filtered seawater, no template control siRNA (NTC), and siRNA targeting HSP16.2 across four genotypes of A. cervicornis with known differences in thermal susceptibility (two heat-tolerant and two heat-susceptible genotypes). There was no significant difference in HSP16.2 gene expression due to the treatment of HSP16.2 siRNAs versus controls, but we did identify a single genotype-driven effect that corroborates evidence that A. cervicornis heat tolerance is driven largely by genotype. No difference in photophysiological performance or mortality was observed between experimental and control treatments. There were, however, some differences in mortality and HSP16.2 expression among genotypes. Our results reveal that siRNA-mediated knockdown was not successful for this species and gene and thus further optimization is needed prior to broader applicability of siRNA-mediated knockdown in corals. We provide an assessment of potential methodological improvements to support the development of future reverse-genetics studies in corals.

Introduction

Reef-building corals are foundational organisms responsible for creating complex habitats for diverse reef organisms, protecting coastlines from storm surges, and contributing billions of dollars annually in economic and cultural value (Beck et al., 2018; Souter et al., 2021; Costanza et al., 2014). Despite their economic and ecological importance, corals have experienced a precipitous decline due to anthropogenic stressors and their downstream effects (Hoegh-Guldberg et al., 2017; Hughes et al., 2017; Manzello et al., 2025). These anthropogenic influences manifest across global (ocean acidification and rising ocean temperatures) and local (pollutants and disease) scales and have been the main drivers to the loss of coral reef habitats throughout the tropics and subtropics (Eddy et al., 2021). Acute heat stress, in particular, has been found to manifest in the loss of symbiotic algae in the process known as coral bleaching (Douglas, 2003). Presently, the host-specific mechanisms that dictate coral bleaching and thermotolerance are under investigation but are not yet fully understood, especially at the molecular level (Cleves et al., 2020a).

To understand coral resilience and adaptation to thermal stress, novel approaches are needed in the face of increasingly prevalent marine heatwaves. Some strategies being applied include transcriptional studies of corals undergoing heat stress, the identification of coral genotypes, and the critical traits that underpin coral resilience to thermally extreme environments (Camp et al., 2019; Strader and Quigley, 2022; Dixon et al., 2020). Such studies have identified putative stress response pathways for corals contending with marine heatwaves and other stressors such as ocean acidification. Unfortunately, poor genetic annotation in corals, paired with a lack of knowledge on coral gene function, has limited the usability of results from transcriptional-based studies. Application of reverse-genetic approaches (i.e., those in which the phenotypic traits tied to a gene are elucidated by observing the effects on the organism after disrupting the expression of a single targeted gene; Ahringer, 2006) may fill these gene function knowledge gaps. Through this stepwise process of gene knockdown, the critical genes in a pathway, such as host-specific thermal response pathways vital to coral bleaching and thermal adaptation, can be determined.

While there exists a pressing need to determine gene function in corals, there has yet to be an established protocol for performing gene knockout or knockdown studies. The most well-known example of genetic manipulation of corals was performed utilizing the technique CRISPR/Cas9 to create gene knockout lines of Acropora millepora for Heat Shock Transcription Factor 1 (Cleves et al., 2020b). However, pairing the labor-intensive microinjector protocol and associated expensive specialized equipment with the slow growth of coral species, as this protocol is performed on coral embryos, limits the applicability of this technique. To contend with these limitations in scalability, other molecular techniques have begun to be explored. RNA interference (RNAi) is a technically simple gene knockdown technique in which the cell’s endogenous machinery (i.e. RNA interference pathway) is used to temporarily destroy the messenger RNA (mRNA) of a targeted gene (Novina and Sharp, 2004). To destroy the mRNA, one or many short interfering RNAs (siRNAs) are introduced to the organism via physical, biological, or chemical means. A commonly used chemical method is lipofection, in which the genetic payload is surrounded by liposomes brought into contact with a target cell’s membrane, wherein it fuses and delivers the surrounded payload into the cell (Felgner et al., 1987). In RNAi, this genetic payload can be one or many miRNAs that target a gene for knockdown (i.e., constructs). Therefore, siRNA constructs simply need to be mixed with a commercially available lipofection reagent, such as INTERFERin, and applied to target cells or tissues to perform RNAi (Jensen et al., 2014). It is noteworthy that this technique has certain temporal limitations for experimental design. Between delivery of siRNA payload and gene knockdown, there is a delay of approximately 24–48 h for the cell’s molecular machinery to effectively identify and destroy the target mRNA. As RNAi only works to destroy mRNA expression, the effects of the gene knockdown are transient and diminish after approximately one week (Raemdonck et al., 2008).

RNAi has successfully achieved gene silencing in commercial plants for agricultural purposes (Small, 2007), in model organisms such as Drosophila for gene studies (Heigwer et al., 2018), and as an FDA-approved therapeutic for polyneuropathy in humans (Setten et al., 2019). Despite its widespread use in the biotechnology and biomedical sectors, it has only recently been applied to the study of cnidarians (but see Cleves, 2022). Importantly, while RNAi has been performed in several cnidarian and coral species, thus far it has been performed primarily on coral larvae as opposed to adult coral specimens (Dunn et al., 2007; Karabulut et al., 2019; Yuyama et al., 2021). Majerová and Drury (2022) successfully performed gene knockdown in adult Pocillopora acuta using RNAi with siRNAs targeting four predicted GFP-like genes and BI-1 (BAX inhibitor 1) to understand the role of thermal preconditioning in BI-1’s antioxidative response. Despite this, there has been no work in an adult Caribbean coral species using RNAi and siRNA to identify important genes and their function in response to heat stress. Therefore, in this study we attempted gene knockdown of Heat Shock Protein 16.2 (HSP16.2) in adult fragments of the critically endangered Caribbean coral, Acropora cervicornis, using RNAi. HSP16.2 is known to function as a molecular chaperone with affinity for unfolded proteins in C. elegans (Leroux et al., 1997), and while it is reliably upregulated in A. cervicornis exposed to acute heat stress (Parkinson et al., 2018), its causal role in thermotolerance remains unclear. c. utilizing RNAi is effective and this gene plays a functional role in thermal stress response, the knockdown cohort should exhibit lower thermotolerance as seen by faster bleaching and mortality rates. In addition to demonstrating the feasibility of this method for functional gene studies in adult corals, this study examined the role that HSP16.2 plays in coral thermotolerance during sustained thermal stress for two genets each of thermally tolerant (TT) and thermally susceptible (TS) Acropora cervicornis. Finally, we provide recommendations for optimization of RNAi in reverse-genetics studies, and implications for the potential roles of RNAi in coral restoration efforts.

Materials and methods

siRNA and primer design

BLOCK-iT™ RNAi Designer (ThermoFisher Scientific) was utilized to identify potential siRNAs targeting the A. cervicornis HSP16.2 gene from a transcribed RNA sequence (NCBI GASU01030017.1). Several of the top ten highest-ranked siRNAs were then run in BLAST (Altschul et al., 1990) using standard blastn parameters against the A. cervicornis genome to confirm on-target effects and no off-target binding (Locatelli et al., 2024). For the control non-targeting siRNA, a proprietary control scrambled Stealth siRNA was used (ThermoFisher Scientific), and similarly, the sequence was run in BLAST against the A. cervicornis genome to confirm no off-sequence binding would occur. Custom Silencer Select siRNAs (#10620312, ThermoFisher Scientific) were then ordered for two of the top ten highest-ranked siRNAs and the no template control siRNA (NTC). Unfortunately, one of the two selected siRNAs was unable to be synthesized by the manufacturer in time for the experiment due to supply chain delays and purification issues. This resulted in one siRNA construct for the experiment. The sequences of the siRNA and NTC siRNA used are as follows: HSP16.2 siRNA = CAUCUGGGAUCUUGAUGACUUUCUU; NTC siRNA = UUCCUCUCCACGCGCAGUACAUUUA (Stealth; Thermo Fisher Scientific). Primers for qPCR were designed using the OligoPerfect™ Primer Designer (ThermoFisher Scientific). The same HSP16.2 transcribed RNA sequence (GASU01030017.1) was used to design the qPCR primers and was also mapped to the genome to confirm binding specificity. For the endogenous control genes, candidate genes were chosen if they did not change in expression in response to thermal stress in A. cervicornis (Parkinson et al., 2018). The selected genes were MOB kinase activator-like protein 3 (GASU01031038.1), Actin-7 (GASU01085644.1), and Neuronal-specific septin (GTP-binding proteins) (GASU01084761.1). Control primers were tested using a dilution and melt curve analysis (Adams 2020). Only the Actin-7 housekeeping primers successfully amplified and were chosen as the housekeeping gene in this study. The sequence of the chosen qPCR control primer was for Actin-7 with the sequence (Forward) 5’-GATTCGCTCGAGAAAACGCC-3’ and (Reverse) 5’-GAAGCAGAGCTGAGAAGCCA-3’. The sequence of the chosen HSP16.2 qPCR primer was (Forward) 5’-TCGCTGTTGATGCCACTGAT-3’ and (Reverse) 5’-TTGGATGTTGACGCCGAGAA -3’.

Sample collection

Four A. cervicornis colonies were collected from the University of Miami’s Key Biscayne coral nursery off Virginia Key, FL (25.6763, -80.0987; depth = 9 m) in June 2022, representing two known thermally-tolerant (TT1, TT2) and two thermally-susceptible genotypes (TS1, TS2) as per the Key Biscayne coral nursery’s designation. The four genet colonies were then fragmented using a diamond band saw into ~2–3 cm long single-branch tips and allowed to acclimate in flow-through raceways in the Experimental Reef Laboratory (ERL) at the University of Miami’s Cooperative Institute for Marine and Atmospheric Studies (CIMAS) for one week at ambient temperatures of 27.5 °C. Once acclimated, fragments were assigned unique identification codes to track the genet and treatment throughout the experiment, and were affixed onto acrylic pucks using cyanoacrylate glue. A total of 14 fragments per genet were affixed to plugs for downstream experiments. Within each genet, three fragments were randomly marked to be treated with filtered seawater (seawater control), three were marked to be treated with the NTC siRNA (siRNA NTC), and eight were marked to be treated with the HSP16.2 siRNA (siRNA HSP16.2). One fragment had unexpected mortality during acclimation (prior to treatment), resulting in there being one less siRNA HSP16.2 treated fragment (n = 7 for genet TT2).

siRNA knockdown

Our method followed that of Majerova and Drury (2022), but with scaled-up working volumes utilized to fully cover adult coral fragments during incubation. Transfection mixtures were prepared for both HSP16.2 and NTC siRNA constructs concurrently for all genets to prevent batch effects. The volumes per reaction required adding 100 µl of reconstituted 10nM siRNA stock to 1250 µl- filtered and UV-sterilized seawater, vortexing for 10 s, then adding 80 µl INTERFERin (#101000028, Polyplus, Illkirch-Graffenstaden, France), vortexing 10 s, and waiting 10–20 minutes at room temperature. Once the solutions had been prepared, coral fragments were rapidly removed from flow-through raceways and placed into 50 ml conical tubes (Figure 1). Fragments then received either the seawater control, siRNA NTC, or siRNA HSP16.2 treatments. For the siRNA NTC and siRNA HSP16.2, the above treatment solutions were pipetted onto each fragment in their 50 ml tube, and allowed to sit for 2 min. Following this incubation, 23 ml of filtered, UV-sterilized seawater was added, covering the entire fragment. The submerged fragments were then incubated at ambient temperatures (27.5 °C) for 6 h, after which time corals were inspected for visible signs of stress (e.g., rapid tissue loss, retracted polyps) and then placed back into flow-through tanks at 27.5 °C for 48 h to allow for siRNA treatment to take effect. The seawater control fragments were put through the same physical process as the other two treatment groups, but with filtered seawater used in place of the siRNA mixtures.

Figure 1

Heat stress

Following the 48 h period to allow for sufficient knockdown of the target gene via the siRNA pathway, the three coral treatment groups were transferred from a 27.5 °C flow-through raceway to a thermal stress treatment raceway kept at 33 °C for 5 d. This temperature was based on previous literature that elicited HSP16.2 upregulation at similar temperatures (Parkinson et al., 2018), and trial experiments that showed A. cervicornis would undergo mortality within 5 d of thermal stress at 33 °C.

Immediately prior to the beginning of thermal stress, each coral fragment was sampled for gene expression analysis. A hammer and chisel were used to collect a ~1 cm piece of coral skeleton and tissue from the apical tip of the fragment. Tissue and skeleton were then placed into a 2 ml bead tube (Zymo Research) with 1 ml of cold TRIzol (#15596026, Thermo Fisher Scientific), placed on ice, and shaken vigorously by hand for 15 s to allow for TRIzol to infiltrate the tissue. Tissue samples were then immediately transferred to a -80 °C freezer until further processing. This sampling procedure was repeated for each sample at 4 h following initial exposure to the thermal stress treatment. The chisel was bleached and flame-sterilized with ethanol between samples to avoid cross contamination. The remaining live tissue on coral fragments were utilized for physiological assessment during thermal stress treatment.

Physiological assessment

Mortality for each fragment was determined by individually observing coral fragments in the morning and evening of each day and scoring as deceased in 0.5 d increments when loss of tissue was observed. Once a fragment was deceased, all mortality and IPAM measurements concluded. Mortality scores were assessed by a single observer throughout the week to control for any technician bias.

Imaging Pulse Amplitude Modulation (IPAM) fluorometry (Walz^®, IMAGING-PAM) was used to measure the photosynthetic efficiency (FV/FM) of Symbiodiniaceae within the coral tissue, as a means of quantifying bleaching and health throughout the thermal stress treatment. To establish baseline healthy FV/FM values for each coral fragment in all experimental treatments, measurements were taken the evening prior to siRNA treatment and 24 h following siRNA treatment. Once thermal stress began, FV/FM values were recorded for each fragment throughout the duration of the heat stress treatment nightly until mortality occurred. IPAM measurements involved following a modified protocol from Ralph et al. (2005), in which coral fragments were acclimated to the dark for 30 min and were subsequently pulsed with blue light-emitting diodes to map the chlorophyll a fluorescence yield (Ralph et al., 2005). The amount of light energy pulsed at the coral was a known value and the measured energy reflected back was then used to calculate the FV/FM, or amount of light used for photosynthesis by the coral’s symbiotic algae.

RNA extraction and gene expression analysis

Molecular workflow involved extracting RNA from coral tissue for gene expression analysis to assess the success of the knockdown. A Zymo Direct-zol RNA MiniPrep Extraction Kit protocol (#R2051, Zymo Research) with modifications described below was used to extract total RNA. Each bead tube containing coral tissue and TRIzol was bead-beaten twice for 1 min (6 m/s) on a FastPrep-24 (MP Biomedical), then allowed to rest at room temperature for 5 min. Beaten lysates were then spun down for 2 min at 16,000 × g to pellet debris. Equal volumes (700 µl) of the supernatant and 100% ethanol were transferred to a new tube and mixed thoroughly. A 700 µl aliquot was subsequently transferred to a Zymo-Spin Column in a collection tube and centrifuged for 30 s, a process that was then repeated for the remaining 700 µl. To elute RNA, 50 µl of DNase/RNase-free water heated to 60 °C was pipetted directly to the column, incubated for 2 min, and centrifuged for 30 s. The eluted RNA was then aliquoted on ice for quantification, with the remainder immediately transferred to a -80 °C freezer. RNA concentrations for each sample were then quantified using a Qubit Broad Range RNA kit (#Q10210, Invitrogen).

Following successful quantification, 100 ng of RNA per sample was reverse transcribed using the High Capacity cDNA reverse transcription kit (#4368813, ThermoFisher Scientific) using 0.5 µl of Ribolock RNase Inhibitor (#O0381, ThermoFisher Scientific) per reaction following the manufacturer’s protocol. Amplification of 7.5 ng of cDNA by qPCR with the PowerUp SYBR Green Master Mix (#A25742, Applied Biosystems) was performed in 10 µl reactions on an Applied Biosystems StepOne Plus Real-Time PCR System. Expression of the HSP16.2 gene was calculated relative to the seawater control coral within the same genet at time 0 with ΔΔCt method using the Actin-7 (GASU01085644.1) gene as the endogenous control. A knockdown threshold of 20% in gene expression of HSP16.2 in treated samples relative to NTC siRNA and seawater control samples was determined in order to match a similar threshold set in GFP knockdown seen in the Majerova and Drury 2022 study.

Statistical analysis

Statistical analysis was conducted in R (R Core Team, 2021). The effect of treatment group (seawater controls, siRNA NTC, and siRNA HSP16.2) on HSP16.2 gene expression was analyzed using a mixed model controlling for genet as a random effect due to the clonal nature of coral fragments (Table 1) using the package lme4 (Bates et al., 2015). A one-way ANOVA was performed on each genet separately to analyze the effect of treatment on HSP16.2 gene expression with homogeneity of variance, normality of residuals, and Shapiro-Wilk tests performed to check the assumptions of the ANOVA with Bonferroni correction for multiple comparisons applied (Table 2).

Statistical analysis of IPAM data was first calculated using the Walz^® software to generate photosynthetic performance (FV/FM) metrics for each sample over time. These values were then analyzed for significant differences among genet and treatment through time using one-way ANOVA. Mortality data were quantified as the number of days from the onset of the high temperature thermal stress to complete mortality. These data were analyzed for significant differences among treatments within genet using survivorship analyses (Cox regression for survival analysis) in the package survival (Therneau, 2021), with survminer (Kassambara et al., 2021) used for visualization.

Results

Gene expression response

Gene expression analysis using RT-qPCR to quantify HSP16.2 activity revealed that 26% (8/31) of HSP16.2 siRNA-treated samples showed ≥ 20% decreased expression following heat exposure (TT1, TT2, TS1, and TS2) when normalized to the siRNA NTC and seawater control samples regardless of genet (Figure 2). HSP16.2 gene expression within control groups (both siRNA NTC and seawater) was variable, and one-way ANOVA revealed no significant difference in gene expression between treatment groups (One-way ANOVA, p(Treatment) = 0.6650; Figure 3A). Genet-level effects of this knockdown were observed, with genet TT1 representing 5 of the 8 knockdown samples (Figure 2A) and having higher, but not statistically significantly higher HSP16.2 expression compared to genets TT2, TS1, and TS2 (Figure 3B, Tables 1,  2). When expression of HSP16.2 by treatment was analyzed within each genet separately, TT1 and TS1 had elevated but not significant expression for the NTC siRNA treatment as opposed to the seawater controls and siRNA HSP16.2 groups (Figure 4). No significant difference in gene expression among treatments within genet was detected apart from gene expression in the NTC siRNA treatment group relative to the siRNA HSP16.2-treated samples within the TT2 genet (Wilcoxon Rank Sum, p = 0.0330; Figure 4B).

Figure 2

Figure 3

Figure 4

Physiological response

All treatment groups reached deceased status in roughly the same amount of time (seawater control = 3.25 ± 0.23 d, NTC siRNA = 3.54 ± 0.22, HSP16.2 siRNA = 3.23 ± 0.15), with mortality of most fragments occurring rapidly between day 2 and 5, yielding no significant differences according to a Cox regression for survival analysis (Table 3). Over the five-day thermal stress experiment, none of the siRNA HSP16.2-treated corals were significantly more likely to reach a survival probability of zero before the seawater control or siRNA NTC groups (Figure 5). However, genet TS2 did show a global log-rank p-value below alpha (p = 0.019) for survivorship due to the siRNA NTC treatment having a greater survival probability than the seawater control and siRNA HSP16.2 samples. There was no difference in FV/FM measurements between treatment groups throughout the experiment, with all three groups seeing a similar decline in FV/FM during sustained thermal stress (ANOVA, p = 0.798; Figure 6), but a significant difference by genet was observed (p = 0.002; Table 4). Further analysis with a Tukey’s Honestly Significant Difference (HSD) test revealed this significant difference in genet was explained by genet TS2’s interaction with TT1 (p = 0.043) and TT2 (p = 0.0003; Table 5).

Figure 5

Figure 6

Discussion

Study results

Majerova and Drury (2022) successfully knocked down the anti-apoptotic protein Bax-Inhibitor-1 in Pocillopora acuta using the same siRNA approach. Our RNA interference (RNAi) application did not result in a significant effect in knocking down the activity of the HSP16.2 gene across all four genets of adult A. cervicornis. There was, however, some evidence of partial RNAi, as 26% (8/31) of treated coral fragments showed HSP16.2 knockdown of at least 20% reduction in comparison to controls within the same genet. Bleaching rate and survival, as measured by IPAM and mortality respectively, was statistically indistinguishable for all treatment groups across genets, so bleaching rate and mortality were unaffected by the siRNA treatment. This lack of phenotypic response was not surprising, as the HSP16.2 knockdown was muted in its efficacy, and all samples were shown to be expressing HSP16.2 to some degree. This level of HSP16.2 expression seen in all samples may have reached a potential minimum threshold of gene expression for a viable thermal stress response, leading to the lack of differentiation in phenotypic feedback. It is also possible that HSP16.2 expression may not have an effect on bleaching rate and survival, as the function of HSP16.2 is not fully understood.

There are several potential reasons for the muted efficacy of the RNAi protocol in this study. It is possible that the construct efficacy was low and resulted in less than the 20% reduction of HSP16.2 expression compared to control that was used by Majerova and Drury (2022) or in no measurable phenotypic or physiological effect. During this study, we were unable to quantify construct uptake efficacy and its effects on protein expression due to widespread supply chain issues that persisted from the Covid-19 pandemic. For example, we were unable to secure fluorescently labeled constructs that could have been used to measure construct uptake or Western blot antibodies that could have been used to measure protein expression (Kurien and Scofield, 2006; Medarova et al., 2007). Thus, these metrics could not be incorporated in this study to complement the RT-qPCR and phenotypic expression metrics to assess siRNA knockdown success. The expression of HSP16.2 may not have a phenotypic effect on resilience to stressors in corals as measured here or the phenotypic or physiological effect may require a longer time to manifest relative to knockdown and monitoring times. Additionally, the temporal dynamics of HSP16.2 could have played a role in the physiological outcomes, as all samples had baseline HSP16.2 expression after 24 h following thermal stress onset regardless of treatment. The expression of heat shock proteins has been shown to vary over the day-night cycle depending on coral species (Han et al., 2024), as the dynamics of HSP16.2 have not been fully determined in A. cervicornis, the effect of the temporal dynamics of HSP16.2 therefore cannot be ruled out. One thermally tolerant genet (TT1) showed significantly higher HSP16.2 expression compared to the thermally susceptible genets (TS2 and TS1) and the second thermotolerant genet (TT2) when exposed to thermal stress. This may mean that genet-level differences in baseline HSP16.2 expression or other genes in the stress response pathway may have played a role as well. Therefore, future studies should examine the entire molecular pathway of HSP16.2 and other heat shock proteins, then target upstream transcription factors to modulate overall expression of stressor-related proteins. Another factor to consider is the temperature selected during the course of this experiment. The chosen temperature of 33 °C was determined to cause mortality in all A. cervicornis samples by the conclusion of the week-long experiment. It is possible that a less-severe thermal condition may have allowed for physiological and genetic differences in treatment to manifest over the course of the experiment. This underpins the major drawback of RNAi in that it has a transient effect, necessitating the need for balancing the time of both experimental treatment and knockdown longevity.

When expression of HSP16.2 was analyzed by treatment within each genet, genets TT1 and TS1 had elevated expression of HSP16.2 for the siRNA NTC but depressed expression for the seawater control and siRNA HSP16.2 treatments (Figure 4). The lower HSP16.2 expression in genets TT1 and TS1 with the siRNA HSP16.2 treatment may represent some effect of the treatment on depressing HSP16.2 expression. While this pattern was not statistically significant, it may indicate that expression of HSP16.2 can be elevated in response to the transfection reagents, which are cytotoxic in high enough concentrations (Dass, 2002), although it is worth noting no visible signs of toxicity were seen between treatment groups. This elevation in expression, due to reagent-induced cytotoxicity, may have been observed in this study, as HSP16.2 is a stress response gene for thermal and oxidative stress that has not had its functionality fully confirmed in invertebrates (Strayer et al., 2003). This pattern, however, does not hold for genets TS2 and TT2, with genet TT2 showing a strong reduction in HSP16.2 expression with the siRNA NTC treatment. Thus, the stress response of HSP16.2 to siRNA transfection appears to differ substantially among genets.

Future steps and implications for coral restoration

Expanding the methodological toolkit of molecular ecologists will be necessary for conducting a comprehensive examination of coral molecular thermotolerance. In the future, gene knockdowns utilizing RNAi can be paired with RNA-Seq to ascertain not just the effect of a single gene, but how the modulation of a single gene may affect the response of the entire transcriptome. As whole-transcriptome gene expression techniques continue to decrease in cost and genomic references improve, the study of coral thermotolerance may advance rapidly, lending to a combination of methods such as the one presented in this study. While this study was unsuccessful in establishing a link between gene knockdown of HSP16.2 and physiological outcomes, its lack of success highlights that the methods still need optimization to be applicable across different corals and gene targets. Therefore, we have identified some next steps for improving the outcome of future siRNA knockdown studies in adult scleractinian corals, which may lead to consistent results as seen in the previous siRNA knockdown in adult P. acuta (Majerová and Drury, 2022).

Future studies should: (1) Incorporate additional constructs to target HSP16.2 and other genes of interest. RNAi workflows involving siRNAs generally involve the use of several constructs per gene target to contend with differential construct efficacy and the potential for off-target effects of single constructs in high concentrations (Hannus et al., 2014; Neumeier and Meister, 2021). In this study, while we identified numerous potential HSP16.2 constructs and chose two for the experiment, we could only use one due to time constraints, supply chain delays, and manufacturing issues. Future studies should therefore use siRNA pools of multiple constructs, or several independent constructs to increase chances of successful knockdown.

(2) Use other metrics for evaluating knockdown success from construct uptake, such as using fluorescently labeled constructs, to protein expression, such as proteomics or Western blot. Studies have demonstrated the feasibility of using a mixture of methodologies to verify siRNA uptake and localization in experimental organisms, these approaches have even been applied in vivo in murine models (Medarova et al., 2007).

(3) Increase the transfection efficiency of the constructs through methodological refinement; in particular, optimizing construct concentrations and incubation times. Troubleshooting transfection reagent and siRNA construct concentrations through dilution experiments is a step that can improve transfection efficiency and minimize issues with cytotoxicity of reagents (Jarvis 2023, Kargaard et al., 2019). We did not observe any visible signs of stress in siRNA-treated coral fragments in this study, as all samples showed exposed polyps during the 6-h incubation period. Thus, either the incubation time could be further extended, or increased concentrations of siRNA could perhaps be used without negative impacts on coral health. Similarly, adjusting incubation times of the siRNA payload and the transfection reagent could influence transfection efficiency as seen with Lipofectamine2000 (ThermoFisher), a similar transfection reagent that works based on the liposome–DNA complex formation (Dalby et al., 2004).

(4) Optimize methods to minimize physiological stress responses that may act as a potential barrier to the penetration of siRNA reagents or mask the effects of gene knockdown. In corals, including A. cervicornis and P. acuta, mucus secretion is a common biological stress response that may act as a potential barrier to the penetration of siRNA reagents (Majerová and Drury, 2022). Acroporids have been found to continuously release several liters of mucus per day per m2 of reef area when submerged, and roughly double in mucus production when exposed to air (Wild et al., 2004, Wild et al., 2005). As the siRNA treatment involved removing fragments from water to directly apply the transfection mixture, along with a 2 min incubation exposed to air, excessive mucus production in this species may have reduced the penetration of lipofection reagents and payload. Given that mucus can act as a physical barrier to reagent diffusion, the heavy mucus secretion noted by the authors during siRNA treatment of coral samples may have hindered or prevented transfection. Optimizing methods or developing an automated or closed-system for treatment could be potential solutions to minimize this issue.

(5) Subject corals to a gradient of stress to assess the effect of siRNA knockdown and allow for detection of phenotypic and physiological changes that may be masked at high stress levels that induce mortality.

Optimization of these methods for successful repeatable knockdown of gene expression in scleractinians may be useful for coral conservation and restoration efforts. The identification, and importantly, verification of genes associated with stress tolerance and resilience from functional gene studies may be applied to the conservation and restoration of coral reef ecosystems in several ways pending thorough methodological refinement. The assessment of thermal resilience of nursery corals for outplanting can be done by checking for the upregulation of a few target genes, with genotype being found to play a significant role in determining growth rates of outplanted corals (Henley et al., 2022) and gene expression (Young et al., 2024) in both Pacific and Atlantic species respectively. Assisted evolution of corals through breeding and movement (i.e., assisted gene flow; Quigley et al., 2019) can increase thermal resilience, and the identification of candidate genes will make the process more efficient and scalable, given an understanding of the verified identity of specific adaptive genes. Lastly, direct genetic engineering of corals to increase fitness must be preceded by work that determines which genes require upregulation (Knowlton and Leray, 2015).

Conclusions

This study represents an attempt at further establishing a gene knockdown technique in adult corals for the study of scleractinian coral resilience at the genetic level. Although we failed to generate a successful knockdown of HSP16.2 in this study, we identified steps for optimization that future studies may consider: (1) incorporate additional siRNA constructs, (2) use metrics to qualify siRNA uptake success such as fluorescently labeled siRNAs, (3) optimize construct concentrations and incubation times through stepwise experimentation, (4) minimize physiological stress responses that may act as a potential barrier to the penetration of siRNA reagents, and (5) test corals in a stress gradient to detect subtle indications of knockdown. Such optimizations and consideration of inter-genotype variation in gene expression could contribute to establishment of effective gene knockdown protocols to understand coral resilience and adaptation to stress, and will advance our understanding of gene function in corals. This knowledge could then be applied to coral conservation and restoration to maintain sustainable healthy coral ecosystems in an era of changing climate.

References

• 1

  AdamsG. (2020). A beginner’s guide to RT-PCR, qPCR and RT-qPCR. The Biochemist, 42(3), 48–53.

  • Google Scholar

• 2

  AhringerJ. (2006). Reverse genetics. WormBook: The Online Review of C. elegans Biology. WormBook.

  • Google Scholar

• 3

  AltschulS. F.GishW.MillerW.MyersE. W.LipmanD. J. (1990). Basic local alignment search tool. J. Mol. Biol.215, 403–410. doi: 10.1016/s0022-2836(05)80360-2. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  BatesD.MächlerM.BolkerB.WalkerS. (2015). Fitting linear mixed-effects models using lme4. J. Stat. Software67, 1–58. doi: 10.18637/jss.v067.i01

  • CrossRef
  • Google Scholar

• 5

  BeckM. W.LosadaI. J.MenéndezP.RegueroB. G.Díaz-SimalP.FernándezF. (2018). The global flood protection savings provided by coral reefs. Nat. Commun.9 (1), 2186. doi: 10.1038/s41467-018-04568-z. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  CampE. F.EdmondsonJ.DohenyA.RumneyJ.GrimaA. J.HueteA.et al (2019). Mangrove lagoons of the Great Barrier Reef support coral populations persisting under extreme environmental conditions. Mar. Ecol. Prog. Ser.625, 1–14. doi: 10.3354/meps13073. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  ClevesP. A. (2022). “ A need for reverse genetics to study coral biology and inform conservation efforts,” in Coral Reef Conservation and Restoration in the Omics Age, edited by van OppenM. J. H.Aranda LastraM.. (Cham: Springer International Publishing), 167–178.

  • Google Scholar

• 8

  ClevesP. A.ShumakerA.LeeJ.PutnamH. M.BhattacharyaD. (2020). Unknown to known: Advancing knowledge of coral gene function. Trends Genet.36 (2), 93–104. doi: 10.1016/j.tig.2019.11.001. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  ClevesP. A.TinocoA. I.BradfordJ.PerrinD.BayL. K.PringleJ. R. (2020). Reduced thermal tolerance in a coral carrying CRISPR-induced mutations in the gene for a heat-shock transcription factor. Proc. Natl. Acad. Sci.117 (46), 28899–28905. doi: 10.1073/pnas.1920779117. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  CostanzaR.De GrootR.SuttonP.Van der PloegS.AndersonS. J.KubiszewskiI.et al (2014). Changes in the global value of ecosystem services. Glob. Environ. Change26, 152–158. doi: 10.1016/j.gloenvcha.2014.04.002. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  DalbyB.CatesS.HarrisA.OhkiE. C.TilkinsM. L.PriceP. J.et al (2004). Advanced transfection with Lipofectamine 2000 reagent: primary neurons, siRNA, and high-throughput applications. Methods33, 95–103. doi: 10.1016/j.ymeth.2003.11.023. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  DassC. R. (2002). Cytotoxicity issues pertinent to lipoplex-mediated gene therapy in-vivo. J. Pharm. Pharmacol.54, 593–601. doi: 10.1211/0022357021778817. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  DixonG.AbbottE.MatzM. (2020). Meta-analysis of the coral environmental stress response: Acropora corals show opposing responses depending on stress intensity. Mol. Ecol.29, 2855–2870. doi: 10.1111/mec.15535. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  DouglasA. E. (2003). Coral bleaching–how and why? Mar. pollut. Bull.46, 385–392.

  • Google Scholar

• 15

  DunnS. R.PhillipsW. S.GreenD. R.WeisV. M. (2007). Knockdown of actin and caspase gene expression by RNA interference in the symbiotic anemone Aiptasia pallida. Biol. Bull.212 (3), 250–258. doi: 10.2307/25066607. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  EddyT. D.LamV. W. Y.ReygondeauG.Cisneros-MontemayorA. M.GreerK.PalomaresM. L. D.et al (2021). Global decline in capacity of coral reefs to provide ecosystem services. One Earth4, 1278–1285. doi: 10.1016/j.oneear.2021.08.016. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  FelgnerP. L.GadekT. R.HolmM.RomanR.ChanH. W.WenzM.et al (1987). Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. U.S.A.84, 7413–7417. doi: 10.1073/pnas.84.21.7413. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  HanT.LiaoX.GuoZ.ChenJ. Y.HeC.LuZ. (2024). Deciphering temporal gene expression dynamics in multiple coral species exposed to heat stress: Implications for predicting resilience. Sci. Total Environ.912, 169021. doi: 10.1016/j.scitotenv.2023.169021. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19

  HannusM.BeitzingerM.EngelmannJ. C.WeickertM. T.SpangR.HannusS.et al (2014). siPools: highly complex but accurately defined siRNA pools eliminate off-target effects. Nucleic Acids Res.42, 8049–8061. doi: 10.1093/nar/gku480. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  HeigwerF.PortF.BoutrosM. (2018). RNA interference (RNAi) screening in Drosophila. Genetics208, 853–874. doi: 10.1534/genetics.117.300077. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  HenleyE. M.BouwmeesterJ.JuryC. P.ToonenR. J.QuinnM.LagerC. V. A.et al (2022). Growth and survival among Hawaiian corals outplanted from tanks to an ocean nursery are driven by individual genotype and species differences rather than preconditioning to thermal stress. PeerJ10, e13112. doi: 10.7717/peerj.13112. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  Hoegh-GuldbergO.PoloczanskaE. S.SkirvingW.DoveS. (2017). Coral reef ecosystems under climate change and ocean acidification. Front. Mar. Sci.4, 252954. doi: 10.3389/fmars.2017.00158. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23

  HughesT. P.BarnesM. L.BellwoodD. R.CinnerJ. E.CummingG. S.JacksonJ. B.et al (2017). Coral reefs in the anthropocene. Nature546 (7656), 82–90. doi: 10.1038/nature22901. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  JensenK.AndersonJ. A.GlassE. J. (2014). Comparison of small interfering RNA (siRNA) delivery into bovine monocyte-derived macrophages by transfection and electroporation. Veterinary Immunol. Immunopathology158, 224–232. doi: 10.1016/j.vetimm.2014.02.002. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  KarabulutA.HeS.ChenC. Y.McKinneyS. A.GibsonM. C. (2019). Electroporation of short hairpin RNAs for rapid and efficient gene knockdown in the starlet sea anemone, Nematostella vectensis. Dev. Biol.448 (1), 7–15. doi: 10.1016/j.ydbio.2019.01.005. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26

  KargaardA.SluijterJ. P. G.KlumpermanB. (2019). Polymeric siRNA gene delivery – transfection efficiency versus cytotoxicity. J. Controlled Release316, 263–291. doi: 10.1016/j.jconrel.2019.10.046. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  KassambaraA.KosinskiM.BiecekP. (2021). survminer: drawing survival curves using “ggplot2”, R package version 0.4.9.

  • Google Scholar

• 28

  KnowltonN.LerayM. (2015). “ Exploring coral reefs using the tools of molecular genetics,” in Coral reefs in the anthropocene, edited by BirkelandC.. (Dordrecht: Springer).

  • Google Scholar

• 29

  KurienB. T.ScofieldR. H. (2006). Western blotting. Methods38, 283–293. doi: 10.1016/j.ymeth.2005.11.007. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  LerouxM. R.MelkiR.GordonB.BatelierG.CandidoE. P. (1997). Structure-function studies on small heat shock protein oligomeric assembly and interaction with unfolded polypeptides. J. Biol. Chem.272, 24646–24656. doi: 10.1074/jbc.272.39.24646. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  LocatelliN. S.KitchenS. A.StankiewiczK. H.OsborneC. C.DellaertZ.ElderH.et al. (2024). Chromosome-level genome assemblies and genetic maps reveal heterochiasmy and macrosynteny in endangered Atlantic Acropora. BMC genomics, 25(1), 1119.

  • Google Scholar

• 32

  MajerováE.DruryC. (2022). Thermal preconditioning in a reef-building coral alleviates oxidative damage through a BI-1-mediated antioxidant response. Front. Mar. Sci.9, 971332.

  • Google Scholar

• 33

  ManzelloD. P.CunningR.KarpR. F.BakerA. C.BartelsE.BonhagR.et al (2025). Heat-driven functional extinction of Caribbean Acropora corals from Florida’s Coral Reef. Science390 (6771), 361–366. doi: 10.1126/science.adx7825. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  MedarovaZ.PhamW.FarrarC.PetkovaV.MooreA. (2007). In vivo imaging of siRNA delivery and silencing in tumors. Nat. Med.13 (3, 372–377. doi: 10.1038/nm1486. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  NeumeierJ.MeisterG. (2021). siRNA specificity: RNAi mechanisms and strategies to reduce off-target effects. Front. Plant Sci.11. doi: 10.3389/fpls.2020.526455. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36

  NovinaC. D.SharpP. A. (2004). The RNAi revolution. Nature430, 161–164. doi: 10.1038/430161a. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  ParkinsonJ. E.BartelsE.Devlin-DuranteM. K.LusticC.NedimyerK.SchopmeyerS.et al. (2018). Extensive transcriptional variation poses a challenge to thermal stress biomarker development for endangered corals. Mol. Ecol.27, 1103–1119. doi: 10.7287/peerj.preprints.3158v1

  • CrossRef
  • Google Scholar

• 38

  QuigleyK. M.BayL. K.van OppenM. J. H. (2019). The active spread of adaptive variation for reef resilience. Ecol. Evol.9, 11122–11135. doi: 10.1002/ece3.5616. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  RalphP. J.SchreiberU.GademannR.KühlM.LarkumA. W. (2004). Coral photobiology studied with a new imaging pulse amplitude modulated fluorometer 1. Journal of Phycology41(2), 335–342.

  • Google Scholar

• 40

  R Core Team (2021). R: A language and environment for statistical computing (Vienna, Austria: R Foundation for Statistical Computing). Available online at: https://www.R-project.org/ (Accessed July 15, 2022).

  • Google Scholar

• 41

  RaemdonckK.VandenbrouckeR. E.DemeesterJ.SandersN. N.De SmedtS. C. (2008). Maintaining the silence: reflections on long-term RNAi. Drug Discov. Today13, 917–931. doi: 10.1016/j.drudis.2008.06.008. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  SettenR. L.RossiJ. J.HanS. (2019). The current state and future directions of RNAi-based therapeutics. Nat. Rev. Drug Discov.18, 421–446. doi: 10.1038/s41573-019-0017-4. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  SmallI. (2007). RNAi for revealing and engineering plant gene functions. Curr. Opin. Biotechnol.18, 148–153. doi: 10.1016/j.copbio.2007.01.012. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  SouterD.PlanesS.WicquartJ.LoganM.OburaD.StaubF. (2021). Status of coral reefs of the world: 2020 report. Eds. SouterD.PlanesS.WicquartJ.LoganM.OburaD.StaubF. ( Global Coral Reef Monitoring Network (GCRMN) and International Coral Reef Initiative (ICRI).

  • Google Scholar

• 45

  StraderM. E.QuigleyK. M. (2022). The role of gene expression and symbiosis in reef-building coral acquired heat tolerance. Nat. Commun.13, 4513. doi: 10.1038/s41467-022-32217-z. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  StrayerA.WuZ.ChristenY.LinkC. D.LuoY. (2003). Expression of the small heat-shock protein Hsp16–2 in Caenorhabditis elegans is suppressed by Ginkgo biloba extract EGb 761. FASEB J.17, 2305–2307.

  • Google Scholar

• 47

  TherneauT. M. (2021). survival: a package for survival analysis in R, R package version 3.2-13.

  • Google Scholar

• 48

  WildC.HolgerW.MarkusH. (2005). Influence of coral mucus on nutrient fluxes in carbonate sands. Mar. Ecol. Prog. Ser.287, 87–98. doi: 10.3354/meps287087. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  WildC.HuettelM.KlueterA.KrembS. G.RasheedM. Y.JørgensenB. B. (2004). Coral mucus functions as an energy carrier and particle trap in the reef ecosystem. Nature428 (6978), 66–70. doi: 10.1038/nature02344. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  YoungB. D.WilliamsD. E.BrightA. J.PetersonA.Traylor-KnowlesN.RosalesS. M. (2024). Genet identity and season drive gene expression in outplanted Acropora palmata at different reef sites. Sci. Rep.14, 29444. doi: 10.1038/s41598-024-80479-y. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  YuyamaI.HiguchiT.HidakaM. (2021). Application of RNA interference technology to acroporid juvenile corals. Front. Mar. Sci.8, 688876. doi: 10.3389/fmars.2021.688876. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar