siRNA transfection in larvae of the barnacle Amphibalanus amphitrite

Barnacles are typical marine fouling organisms and have been used as model organisms in biofouling and antifouling research for many years (Holm, 2012). So far, all the reported functional studie of genes, signal molecules and pathways in barnacle larval settlement have solely relied on chemical inhibitors, activators or analogues (Clare et al., 1995; He et al., 2012; Zhang et al., 2012, 2013). However, these chemicals, which are often claimed to be highly specific and active, might have unexpected or unnoticed effects and thus raise serious concerns about the reliability of previous results. Furthermore, not all genes and signal molecules have specific inhibitors, activators or analogues. These limitations in conventional functional analysis of genes and pathways hampered progress in molecular mechanism studies of barnacle settlement and development.

RNA interference (RNAi) is a powerful tool for studying gene function (Premsrirut et al., 2011) and is more specific to the target gene than chemical treatments. The manipulation of RNA in conventional model organisms is well developed. Lipofection (Zhang et al., 2014), viral infection (Ben-Shoshan et al., 2014) and electroporation (Xu et al., 2014) are widely applied to cultured cells. Microinjection of small interfering RNA (siRNA) into the animal tissue of interest has been successfully used in insects such as Drosophila (Zhang et al., 2006), Acyrthosiphon pisum (Jaubert-Possamai et al., 2007), Aedes aegypti (Isoe et al., 2007) and Thermobia domestica (Ohde et al., 2009), Crustacea such as Daphnia magna (Kato et al., 2011), Artemia franciscana (Copf et al., 2004) and Parhyale hawaiensis (Liubicich et al., 2009), the arachnids Tetranychus urticae (Khila and Grbić, 2007) and Cupiennius salei (Schoppmeier and Damen, 2001) and the oligochaete Enchytraeus japonensis (Takeo et al., 2010). Environmental exposure to siRNA solution (environmental RNAi) was also reported to be effective in the flatworms Caenorhabditis elegans (Timmons and Fire, 1998) and Macrostomum lignano (De Mulder et al., 2009), and the rotifer Brachionus manjavacas (Snell et al., 2011). Electroporation is successfully used to silence genes in the cnidarian Hydra (Lohmann et al., 1999). However, RNAi has not yet been successfully applied to barnacles.

In this study, we sought to develop a method to introduce siRNA into Amphibalanus amphitrite (Darwin 1854) larvae. The p38 mitogen-activated protein kinase (p38 MAPK), which regulates larval settlement in A. amphitrite through dual phosphorylation on the activated sites (He et al., 2012) was chosen as the target gene. The designed double-stranded siRNA against p38 MAPK was synthesized chemically and labelled with the red fluorescent dye ROX at both 5′ ends. By means of two commercial lipofection reagents, larvae were fed with siRNA at either nauplius VI or cyprid stage, or both. The efficiency of transfection was tested using western blots against p38 MAPK and phosphorylated p38 MAPK (pp38 MAPK) as well as real-time quantitative PCR and settlement bioassays.

To transfect siRNA into the animals, larvae were fed with siRNA at nauplius VI stage (nauplius transfection), early cyprid stage (cyprid transfection), or both (double transfection). The concentration of lipofectin and RNAfectin was set as 4 μl ml⁻¹ and 1 μl ml⁻¹, respectively, based on a previous concentration test (supplementary material Fig. S1). After 6 h, both lipofectin and RNAfectin effectively brought siRNA into the larvae (Fig. 1). For nauplius VI larvae, the red fluorescence was mainly concentrated in guts, but widely spread throughout the whole bodies (Fig. 1A,B). The red fluorescent signals were detected almost everywhere in the larval bodies in transfected cyprids (Fig. 1D,E). Double transfection allowed siRNA to diffuse into the larvae remarkably well, with strong fluorescent signals in guts, limps, antennules and near the ventral part of the carapace (Fig. 1G,H). There was no visual difference in signal distribution or intensity between lipofectin and RNAfectin treatments. In the blank controls, no obvious red fluorescent signal was detected (Fig. 1C,F,I).

Generally, siRNA diffuses first into outer cells in multicellular organisms by forming liposomes, followed by systemic RNAi, which means gene-specific silencing information is transmitted from a primary group of cells into a secondary group of cells (Winston et al., 2002). In the present study, both nauplii and cyprids have a thick carapace. However, the red fluorescent signals were found to attach to the surface of the larval body and concentrated in guts, revealing that siRNA might diffuse into the surface cells and also gut cells after uptake. Then, systemic RNAi may spread to other cells of barnacle larvae, because weaker ROX fluorescent signals were detected in other tissues.

Western blot analysis showed that both p38 and pp38 MAPK were slightly reduced when lipofectin was applied, and sharply dropped in RNAfectin treatments (Fig. 2) during double transfection. Transfection of control siRNA did not affect the levels of p38 and pp38 MAPK (Fig. 2). Nauplius transfection did not significantly affect the levels of p38 or pp38 (supplementary material Fig. S2A,B). For cyprid transfection, 0.4 and 0.8 μg ml⁻¹ of siRNA decreased the level of p38 MAPK when either lipofectin or RNAfectin was used. However, the level of pp38 MAPK showed no apparent changes under the same conditions (supplementary material Figure S2C,D).

Further examination using real-time quantitative PCR revealed that double transfection using either lipofectin or RNAfectin significantly suppressed the mRNA level of p38 MAPK, compared with reagent controls or control siRNA treatments (Fig. 3). These results suggested that the downregulation of the protein level of p38 MAPK in these treatments resulted from the decreased mRNA level of p38 MAPK.

Similarly, settlement bioassay results revealed that double transfection of 0.4 and 0.8 μg ml⁻¹ siRNA using RNAfectin significantly inhibited the larval settlement (Fig. 4B), compared with the reagent control. However, no effects were observed when lipofectin was used (Fig. 4A). Nauplius transfection and cyprid transfection had no effects on settlement and mortality, regardless of the transfection reagent (supplementary material Fig. S4).

The p38 MAPK executes biological functions through the dual phosphorylation of p38 protein molecules at Thr180 and Tyr182 (He et al., 2012). In barnacle larvae, p38 MAPK molecules are stored in the cells after being translated from the mRNA. In this state, p38 MAPK molecules do not have any functions. Once the larvae are stimulated, p38 MAPK will be phosphorylated by MAP kinase kinase 3 (MKK3), which then promotes larval settlement (Zhang et al., 2013). Using double transfection, the levels of p38 MAPK were decreased in lipofectin treatments and in treatments with 0.4 μl ml⁻¹ siRNA and RNAfectin. However, the levels of pp38 MAPK were not changed, because the level of residual p38 MAPK molecules was still high enough to be phosphorylated, ensuring larval settlement rates in these treatments as high as that in the control. This viewpoint can be partially supported by the fact that only a part of p38 MAPK molecules needs to be phosphorylated to fully execute biological functions in cyprids under normal conditions, as the level of pp38 MAPK can be still elevated in response to crude extracts of adult barnacles, compared with that in the control (He et al., 2012).

Several possibilities might explain why double transfection was more efficient in gene knockdown. Firstly, double transfection enlarged the interaction area between liposome and larvae. Nauplius VI larvae moult once before transforming to cyprids. The body surfaces of both nauplii and the newly transformed cyprids were exposed to siRNA in double transfection, meaning that more surface cells interacted with siRNA directly. Moreover, nauplii ingested siRNA into their guts, but cyprid is a non-feeding stage, so less siRNA could be ingested into guts in cyprid transfection only. In double transfection, siRNA firstly got into the nauplius guts. After transforming to cyprids, larvae still retained part of these siRNA molecules in their guts, as a high intensity of ROX was detected there, providing another opportunity for cyprids in double transfections to obtain more siRNA than those transfected only at the cyprid stage. Secondly, double transfection extended the time for siRNA uptake. The singly transfected larvae were only exposed to siRNA solution for 12–24 h, but exposure was for 36 h in the double transfection. In the nauplius transfection, the level of p38 MAPK returned to a normal level 24 h after the initiation of transfection (supplementary material Fig. S2E). The second transfection in double transfections is essential to maintain a low level of p38 MAPK. Thirdly, doubly transfected larvae had a longer time for degradation of the residual p38 MAPK molecules. The p38 MAPK is synthesized and stored in cytoplasm. Knockdown of p38 MAPK minimized the production of mRNA and blocked the synthesis of new p38 MAPK protein, but did not affect the phosphorylation process on p38 MAPK. Longer treatment time in double transfections ensured a longer time to break down the residual p38 MAPK, resulting in reduced levels of pp38 MAPK, and hence, settlement rate. In conclusion, the present study developed a method for siRNA transfection of barnacle larvae, which would be helpful for future functional genomics and biofouling mechanism studies on barnacles.

The procedures for adult barnacle collection and larvae culture are described in Zhang et al. (2013). Nauplii developed to the VI stage after 3 days and transformed into cyprids after 4 days. To collect nauplius VI larvae and cyprids, the culture system was filtered through a series of different pore-sized meshes (355, 280, 180, 154 and 110 μm) 1–2 times. Larvae on the 355 and 280 μm pore meshes were mainly nauplius VI larvae, with cyprids mainly retained by 154 μm pore mesh.

Two transfection reagents, lipofectin and RNAfectin, were purchased from Tiangen Biotechnology Company (Beijing, China). To test possible adverse effects of these two reagents, cyprids were treated with a series of dilutions of lipofectin or RNAfectin (1, 2, 4 and 8 μl ml⁻¹ for lipofectin; 0.5, 1, 2 and 4 μl ml⁻¹ for RNAfectin). For each treatment, 20 newly formed cyprids (<4 h) were introduced into 1 ml dilution of reagents in filtered seawater (FSW). The experiments were conducted in 24-well polystyrene plates (Becton Dickinson Labware). All the plates were kept at room temperature in the dark. After 24 h, the numbers of dead, settled and swimming larvae were counted.

The siRNA for A. amphitrite p38 MAPK (KC287236) spans nucleotides 472–490 and the target sequence is 5′-GACTGCGAGCTCAAGATCC-3′ (Hu et al., 2007). A BLAST search of the NCBI database confirmed this segment to be specific for A. amphitrite p38 MAPK. The sense and antisense siRNA sequences were synthesized by Genscript Company (Nanjing, China), separately. To enable visual observation of the transfection process, the two siRNA strands were labelled with red fluorescent dye ROX at the 5′ ends and then purified using high-performance liquid chromatography (HPLC). Finally, these two strands were annealed to form double-strand siRNA. The non-sense siRNA (control siRNA; 5′-ROX-UUCUCCGAAGGUGUCACGUtt-3′) (Hu et al., 2007) was synthesized simultaneously. This sequence of control siRNA was compared with our transcriptome database of A. amphitrite and no hits were found.

Based on the results of preliminary tests, the concentration of lipofectin and RNAfectin was set at 4 μl ml⁻¹ and 1 μl ml⁻¹, respectively, and two concentrations of siRNA (0.4 and 0.8 μg ml⁻¹) were tested. For a 1 ml culture volume, a suitable amount of siRNA and 4 μl lipofectin or 1 μl RNAfectin was diluted in 250 μl Milli-Q water, respectively, and then mixed well by vortexing. The mixture was incubated at room temperature for 20 min to form siRNA–liposome complexes. At the same time, reagent and blank controls were also prepared, using the same quantity of transfection reagent or FSW instead of transfection mixture, respectively.

Three groups of transfection: cyprid transfection, nauplius transfection and double transfection were conducted. For the cyprid transfection, 20 newly formed cyprids were placed into one polystyrene well containing 0.5 ml FSW and fed with siRNA–liposome complex. For the nauplius transfection, around 100 nauplius VI larvae were introduced into 2 ml FSW with 1×10⁶ cells ml⁻¹ of algae, and then incubated with siRNA–liposome complex. After 12 h, transformed cyprids were collected for real-time quantitative PCR, western blot and settlement bioassays. Cyprids transformed from the nauplius tranfection were moved to new wells and transfected with siRNA for a second time (as in the cyprid only transfection), for ‘double’ transfection. All experiments were repeated at least three times. During the transfection process, larvae were placed under a fluorescent microscope to observe the diffusion of siRNA into larval bodies.

Unsettled cyprids in all the treatments were collected for western blot analysis. The protocols for total protein extraction, protein concentration determination and western blot are described in Zhang et al. (2013). Antibodies against pp38 MAPK (Cell Signaling Technology, USA) and p38 MAPK (raised by He et al., 2012) were used to determine the level of pp38 MAPK and p38 MAPK, respectively. Actin (Cell Signaling Technology, USA) was detected as a loading control. The intensity of each band was determined using ImageJ software.

To confirm the positive results in the double transfection, total RNA from these samples was extracted for real-time qPCR according to the protocol described previously (Zhang et al., 2012). Primers used are shown in supplementary material Table S1. The gene cytochrome b (cyt b) was selected as the internal reference (Bacchetti De Gregoris et al., 2009). The final relative fold changes were calculated using the 2(–ΔΔC_t) method (Livak and Schmittgen, 2001).

To test the effects of siRNA transfection against p38 MAPK on the phenotypic changes of larvae, settlement bioassay were conducted for each transfection treatment. Twenty cyprids from each treatment were placed in a polystyrene well with (for cyprid and double transfection) or without (nauplius transfection) transfection mixture. After 24 h, the numbers of swimming, settled and dead cyprids were counted under a dissecting microscope, respectively. Mortality was calculated as the ratio of dead cyprids to all larvae in the well; and the settlement rate was represented as the percentage of settled cyprids to all larvae alive (including settled and swimming larvae).

Data were analysed using SPSS 11.5 software. Means of transfection treatments were compared with the reagent control using one-way analysis, followed by stepwise multiple comparisons using the Student–Newman–Keuls (SNK) method to identify sample means that are significantly different. P<0.05 was set as the significance criterion.