Functional differentiation of small heat shock proteins in diapause-destined Artemia embryos



Encysted embryos of Artemia franciscana cease development and enter diapause, a state of metabolic suppression and enhanced stress tolerance. The development of diapause-destined Artemia embryos is characterized by the coordinated synthesis of the small heat shock proteins (sHsps) p26, ArHsp21 and ArHsp22, with the latter being stress inducible in adults. The amounts of sHsp mRNA and protein varied in Artemia cysts, suggesting transcriptional and translational regulation. By contrast to p26, knockdown of ArHsp21 by RNA interference had no effect on embryo development. ArHsp21 provided limited protection against stressors such as desiccation and freezing but not heat. ArHsp21 may have a non-essential or unidentified role in cysts. Injection of Artemia adults with amounts of ArHsp22 double-stranded RNA less than those used for other sHsps killed females and males, curtailing the analysis of ArHsp22 function in developing embryos and cysts. The results indicate that diapause-destined Artemia embryos synthesize varying amounts of sHsps as a result of differential gene expression and mRNA translation and also suggest that these sHsps have distinctive functions.


double-stranded RNA


green fluorescent protein


quantitative PCR


RNA interference


small heat shock protein


Diapause entails developmental delays associated with behavioural modification, morphological change, metabolic suppression and enhanced stress resistance, each ranging from slight to extreme in different organisms [1-5]. Diapause is especially common in insects [6, 7] and occurs in embryos, larvae, pupae and adults, although generally in only one developmental stage per species. Survival of environmental stress and synchronization of mating are achieved by diapause, a physiological/developmental process divided into at least three overlapping stages: initiation, maintenance and termination [2]. Initiation involves the modification of gene expression, protein synthesis and protein activity. Maintenance coincides with maximum metabolic suppression and stress tolerance and, at termination, metabolism and growth resume.

Embryos of the crustacean Artemia franciscana undertake one of two developmental routes [3, 8, 9]. As a result of ovoviviparous development, Artemia females release swimming nauplii. By contrast, Artemia embryos navigating the oviparous pathway are released from females as gastrulae enclosed in a protective shell [10]. These gastrulae are called cysts and they undergo diapause. During diapause in Artemia, cyst metabolism decreases beyond the point of easy detection [11, 12] and stress resistance increases greatly, whereby encysted embryos survive repeated freezing and thawing, heat, chemical exposure and years of anoxia [9, 12]. Metabolism resumes upon diapause termination and introduction to favourable temperature, aeration and moisture, the cyst shell ruptures, and a nauplius emerges [9, 13, 14]. During Artemia diapause, genes are both up- and downregulated [15]. Induced genes include p8, a stress responsive transcription cofactor [16, 17], artemin, a species-specific cysteine-enriched ferritin homologue that is a molecular chaperone [18, 19], several late embryogenesis abundant proteins [20] and three small heat shock proteins (sHsps) [21-24].

sHsp monomers, including those in Artemia, range in molecular mass from 15 to 42 kDa and they share an α-crystallin domain of ~ 90 residues, although otherwise they are not well conserved [25]. Monomers interact via the α-crystallin domain to form dimers, which are considered to be the building blocks of high molecular mass sHsp oligomers, a process influenced by the amino- and carboxyl-terminal regions of sHsps [26-31]. The amino-terminus assists in substrate binding, whereas the flexible carboxyl-terminus contributes to sHsp solubility [27-29, 32, 33]. When cells are stressed, sHsp oligomers undergo structural rearrangement and/or dissociation that is proposed to increase hydrophobicity and favour interaction with damaged proteins [26, 33-37]. Functioning in the absence of ATP, sHsps bind many different protein substrates, thereby contributing to cell survival during stress [36, 38]. ATP-independent function and substrate protection are important in diapause where ATP concentrations are often low and there is the potential for protein denaturation. Upon diapause termination and resumption of growth, the refolding of sHsp substrates requires ATP-dependent chaperones such as Hsp70 and Hsp90 [39-42].

Synthesis of the sHsps, ArHsp21, ArHsp22 and p26 is upregulated in diapause-destined Artemia embryos and ArHsp22 gene expression is induced by heat in Artemia adults but not in other life-history stages [15, 23, 24]. ArHsp21, ArHsp22 and p26 share sequence similarity (Fig. 1). Turbidimetric assays with citrate synthase and insulin show that ArHsp21 and ArHsp22 possess molecular chaperone activity in vitro [23, 24]. Similar to p26 [43], ArHsp21 and ArHsp22 may contribute to cyst biogenesis, maintenance of diapause and stress tolerance. To test this proposal, the mRNA and protein representing each sHsp were respectively measured by quantitative PCR (qPCR) and immunoprobing of western blots. RNA interference (RNAi) methodology, employing the injection of double-stranded RNA (dsRNA) into the egg sacs of mature females, yielded knockdown of ArHsp21 in cysts. ArHsp21 loss had no effect on embryo development and, at best, a modest effect on the stress tolerance of cysts. The injection of ArHsp22 dsRNA killed adult Artemia.

Figure 1.

Alignment of sHsps produced in diapause-destined Artemia embryos. The amino acid sequences of ArHsp21 (ABD19712), ArHsp22 (ABD19713) and p26 (AF031367) were aligned with clustalw ( The α-crystallin domains, containing a highly conserved arginine (shaded), are boxed, as is the IXI/V motif in the carboxyl terminal extension. Identical resides are indicated by an asterisk (*); conserved substitutions are indicated by a colon (:); semi-conserved substitutions are indicated by a dot (.). Adapted from Qiu Z and MacRae [24].


Quantitation of sHsps in Artemia cysts

Of the sHsps examined, mRNA encoding ArHsp22 was in the lowest amount, followed by ArHsp21 mRNA, which was six times more abundant than ArHsp22 mRNA (Fig. 2). p26 mRNA was most plentiful, with a copy number 9.8 times greater than ArHsp21 mRNA and 59.0 times greater than ArHsp22 mRNA (Fig. 2 and Tables 1 and 2). These findings suggest that genes for Artemia sHsp are differently transcribed. Immunoprobing of western blots demonstrated that ArHsp21 accounted for 1.2% of the soluble protein in Artemia cysts, whereas ArHsp22 and p26 corresponded to 0.39% and 6.99%, respectively (Figs 3 and 4). Overall, p26 was the most abundant of the sHsps in Artemia cysts in terms of both mRNA and protein, followed by ArHsp21 and ArHsp22. The amounts of sHsp mRNAs relative to one another were different from the corresponding values for proteins. p26 mRNA, the most abundant, appeared to be translated least efficiently, whereas mRNA encoding ArHsp22 was translated twice as efficiently as mRNA for ArHsp21 (i.e. the translation of mRNAs encoding ArHsp22 > ArHsp21 > p26) (Tables 1 and 2). This conclusion assumes the similar stability of all three sHsps and their mRNAs.

Table 1. Summary of sHsp mRNA and protein quantitation. The amount of each sHsp mRNA relative to tubulin and 18S rRNA and the percentage that each sHSP represented with respect to total soluble protein in cyst extracts is shown. Ratios of sHsp mRNA : 18S rRNA were multiplied by 60 to facilitate comparison with sHsp mRNA : tubulin mRNA
 sHsp mRNA : Tubulin mRNAsHsp mRNA : 18S rRNA × 60Mean sHsp mRNA ratiossHsp as a percentage of soluble protein
Table 2. Amounts of sHsp mRNA and protein relative to one another in Artemia cysts. Comparisons of sHsp mRNA and protein were performed using the data shown in Table 1. The mRNA : protein ratios were determined from the values given in the first two rows of Table 2 and they indicate the amount of each sHsp in cysts per arbitrary unit of sHsp mRNA. The final ratios provide a qualitative measure of how efficiently each sHsp mRNA is translated
sHspsArHsp21 : ArHsp22p26 : ArHsp22p26 : ArHsp21
mRNA : protein1.93.41.7
Figure 2.

Quantitation of sHsp mRNAs in Artemia cysts. sHsp transcripts in RNA extracts from cysts were converted to cDNA and amplified by qPCR to determine transcript copy numbers. mRNA copy numbers for each sHsp were compared with copy numbers of α-tubulin mRNA and 18S rRNA in three independently prepared RNA samples and the replicates were averaged. sHsp : 18S rRNA ratios were multiplied by 60 to facilitate comparison with the sHsp : tubulin mRNA ratios. 1, sHsp : tubulin; 2, sHsp : 18S rRNA; AU, arbitrary units. Error bars represent the SD.

Figure 3.

Quantitation of sHsps in Artemia cysts. To compare the amounts of (A) ArHsp21, (B) ArHsp22 and (C) p26 in cysts, purified sHsps and cyst extracts (A', B', C') were resolved by 12.5% SDS/PAGE and blotted to nitrocellulose. Proteins on blot A and A' were resolved in the same gel and blotted together, as was the case for each of the remaining sHsps. Individual membranes were probed separately with antibody to each sHsp followed by horseradish peroxidase-conjugated goat anti-(rabbit IgG) antibody. Immunoreactive proteins in three independently prepared samples were visualized by chemiluminescence and the inset blots show representative results. The fluorescence intensity of each band was determined using imagej and, for each sHsp, the intensity was normalized such that the highest intensity was equal to one. Immunofluorescence intensities were averaged for purified sHsps and cyst extracts, and each data set was curve-fitted with a linear regression. Error bars represent the SD. AU, arbitrary units.

Figure 4.

Comparison of sHsps in Artemia cysts. The percentage of total soluble protein represented by each sHsp in Artemia cysts was determined by immunoprobing of western blots and calculated using the data shown in Fig. 3. The error bars represent the SD.

ArHsp21 knockdown in Artemia cysts

ArHsp21 cDNA and dsRNA yielded bands of the expected size in agarose gels with the dsRNA slightly larger than the cDNA (Fig. 5A). As determined by RT-PCR, the injection of Artemia females with dsRNA for ArHsp21 completely knocked down ArHsp21 mRNA in cysts (Fig. 5B). The lack of PCR products upon omission of reverse transcriptase demonstrated the absence of genomic DNA in RNA preparations. Immunoprobing of western blots showed that knockdown of ArHsp21 was complete and without effect on ArHsp22 (Fig. 5C).

Figure 5.

ArHsp21 knockdown by RNAi. (A) ArHsp21 dsRNA was visualized by electrophoresis in 1.2% agarose gels stained with SYBR®Safe DNA Gel Stain. Lane 1, Generuler 1-kb DNA size markers; 2, cDNA used to prepare ArHsp21 dsRNA; 3, ArHsp21 dsRNA. (B) RT-PCR was performed with RNA from cysts produced in females injected with either control solution (1 and 2) or ArHsp21 dsRNA (3 and 4). RNA samples were either incubated with reverse transcriptase before PCR (2 and 4) or this step was omitted (1 and 3). Electrophoresis was carried out in 1.2% agarose and gels were stained with SYBR®Safe. (C) Protein extracts from 55 cysts produced by females injected with either ArHsp21 dsRNA (1) or control solution (2). Proteins were resolved by 12.5% SDS/PAGE, blotted to nitrocellulose and reacted with antibody to ArHsp21 followed by horseradish peroxidase conjugated goat anti-(rabbit IgG) (upper lanes). The membranes were stripped and reacted with antibody to ArHsp22 followed by horseradish peroxidase conjugated goat anti-(rabbit IgG) (lower lanes). Immuno-reactive proteins were visualized by chemiluminescence.

Cysts lacking ArHsp21 exhibited normal behaviour

Post-injection mortality and behaviour of Artemia females individually receiving dsRNA for ArHsp21 and green fluorescent protein (GFP) or control solution were similar. Unlike p26 knockdown [43], the post-fertilization times to release from females of Artemia cysts with and without ArHsp21 were essentially identical. Accordingly, embryos developed at the same rate regardless of whether or not they contained ArHsp21 (not shown). By contrast to cysts deficient in p26, cysts either containing or lacking ArHsp21 failed to spontaneously terminate diapause after incubation in sea water for 90 days (not shown).

Stress tolerance of cysts lacking ArHsp21

Diapause of Artemia cysts collected from females in the laboratory was terminated by desiccation and freezing [43], which are both substantial stresses. Viability after diapause termination, as shown by hatching, was 54% and 47%, respectively, for cysts containing and lacking ArHsp21 (Fig. 6A), which are significantly different values (P = 0.032). Nauplii emerging from stressed cysts without ArHsp21 grew into adults and reproduced normally. Cysts not having ArHsp21, and for which diapause was terminated by desiccation and freezing, survived equally well ragardless of whether they were heated to 41 °C for 30 min or not heated (P = 0.935), indicating a minor role, if any, for ArHsp21 in the stress tolerance of encysted Artemia embryos (Fig. 6B).

Figure 6.

ArHsp21 knockdown had little effect on cyst stress tolerance. (A) Embryos from females injected with either control solution (1) or ArHsp21 dsRNA (2) were collected, diapause was terminated by desiccation and freezing and cyst hatching quantitated. (B) Diapause was terminated by desiccation and freezing of cysts lacking ArHsp21. Half of the cysts in a brood were kept at room temperature (1) and the other half heated at 41 °C for 30 min (2). The cysts were incubated in sea water and nauplii were counted and removed. Hatching (%) indicated viability. The experiment was performed in duplicate and error bars represent the SE.

Artemia adults died upon injection with ArHsp22 dsRNA

ArHsp22 cDNA and dsRNA yielded bands of the expected size upon electrophoresis in agarose gels with the dsRNA slightly larger than the cDNA (Fig. 7A, inset). The initial survival of Artemia females injected individually with approximately equal amounts of dsRNAs specific to either ArHsp22, ArHsp21, p26 or GFP, as well as control solution, was similar. Swimming, feeding and tail extension returned simultaneously after injection. However, compared to injection with control solution, more females receiving ArHsp22 dsRNA were dead 1 day after injection (Fig. 7A). Stationary animals with immobile appendages were considered dead. By day 2 after injection, 28% of females receiving dsRNA specific to ArHsp22 were viable, whereas 84% of females receiving control solution survived. All females injected with ArHsp22 dsRNA died within 4 days of injection, whereas 84% of females administered control solution were viable (Fig. 7A). Reduction of dsRNA for ArHsp22 to 50% and 25% of the amount used in the original injection resulted in the death of adults. The survival of females injected with 12.5% of the normal dose of dsRNA was the same as for females injected with control solution but ArHsp22 was not reduced in cysts produced by these animals (not shown). Exposure to ArHsp22 dsRNA killed all adult males by 2 days post-injection, whereas approximately half of the males injected with control solution were viable after 4 days (Fig. 7A). When adult Artemia were near death, their appendages assumed a feathery appearance. Aberrant appendages were observed 1 day after injection with dsRNA for ArHsp22, and these animals usually died within 12 h (Fig. 7B). The egg sacs of Artemia females typically emptied near to or soon after death, with this occurring regardless of whether or not females were injected with dsRNA.

Figure 7.

Injection of Artemia adults with ArHsp22 dsRNA caused death. (A) dsRNA specific to ArHsp22 was prepared (inset). Lane 1, Generuler 1-kb DNA size markers; 2, cDNA used to generate ArHsp22 dsRNA; 3, ArHsp22 dsRNA. Electrophoresis was carried out in 1.2% agarose and RNA was visualized with SYBR®Safe. Marker size (bp) is given on the left. Females and males were injected with control solution (blue and green, respectively) and dsRNA specific to ArHsp22 (red and purple, respectively). Animals were incubated in sea water and viability was determined daily. The experiment was performed in duplicate with ~ 10 animals per group. Error bars represent the SE. (B) Light micrographs of adult Artemia females showing a healthy female (left) and an animal about to die (right). a, appendage; g, gut; h, head; es, egg sac containing cysts; o, oocytes. Scale bar = 2.5 mm.


The differential accumulation of molecular chaperones, including the sHsps, Hsp60, Hsp70 and Hsp90, characterizes diapause in many organisms [1, 4, 44, 45]. The ATP-independent sHsps are the first line of defence against irreversible protein denaturation induced by stress, as occurs during diapause and other dormancies. In support of this idea, the upregulation of sHsps among animals displaying profound metabolic suppression and/or tolerance to extreme stresses such as desiccation, anoxia and heat, is common. Examples include the brine shrimp A. franciscana [21-24], the rotifer Brachionus plicatilis [46, 47], the land snail Sphincterochila zonata [48] and the flesh fly Sarcophaga crassipalpis [45]. Despite the many examples of sHsp upregulation in diapause animals, the number of cases where the function of these proteins has been examined in vivo by RNAi methodology is small. Knockdown of Hsp23 in S. crassipalpis limits cold tolerance during diapause [45] whereas, in Artemia, elimination of the diapause-specific sHsp, p26 affects stress tolerance, embryo development and diapause termination [43]. RNAi dependent knockdown of the sHsps Hsp22 and Hsp23 in nondiapause Drospophila melanogaster slows recovery from chill coma [49].

Three sHsps are synthesized at comparable times in diapause-destined Artemia embryos and they accumulate maximally in cysts. These sHsps include p26, an abundant protein investigated in detail [21, 22, 43, 50-53], and the more recently discovered ArHsp21 and ArHsp22 [15, 23, 24]. mRNA encoding p26 is more abundant than mRNA encoding ArHsp21 and ArHsp22, perhaps as a consequence of differential transcription rates as a result of variation in upstream regulatory elements within the genes encoding each sHsp. This proposal is speculative because the regulatory region of only the p26 gene is sequenced [54]. The relative amounts of each sHsp mRNA differ from the relative amounts of each sHsp, indicating either differential translation of the mRNAs or variation in protein stability and/or degradation. For example, p26 mRNA appears to be translated least efficiently and ArHsp22 may be the most stable of the three sHsps, although it is anticipated that all of the sHsps are relatively stable. The cellular locations of Artemia sHsps under different physiological conditions are partially elucidated, with p26 and ArHsp22 entering the cyst nuclei, whereas ArHsp21 does not [8, 24, 55, 56]. Disparities in accumulation and intracellular distribution may explain some of the functional distinctions exhibited in the present study, as well in previous studies investigating ArHsp21, ArHsp22 and p26 [43].

RNAi was employed to compare ArHsp21 and ArHsp22 functions in vivo and to determine whether the elimination of these sHsps affected Artemia embryos in the same way as the loss of p26 [43]. The survival of cysts containing ArHsp21 after diapause termination by desiccation and freezing was slightly higher than for cysts lacking this sHsp, although the difference was minor. By contrast, cysts lacking ArHsp21 survived equally well whether heated at 41 °C or not heated. The observed discrepancy in stress tolerance may have been a result of the cysts containing and lacking ArHsp21 originating from different females. As another possibility, exposure to desiccation and freezing may be more stressful for Artemia cysts than exposure to heating, or cell parameters may be affected differently by dissimilar stresses, thus leading to the protection of cysts by ArHsp21 during one type of stress but not another. The elimination of ArHsp21 had no apparent effect on embryo development or diapause maintenance and termination. The results indicate that ArHsp21 plays either a minor or non-essential role in diapause not only in relation to stress tolerance, but also during embryo development and diapause termination. The function(s) of ArHsp21 after knockdown may be assumed by other sHsps, two of which occur in diapause-destined Artemia embryos. In this context, the amount of ArHsp21 in Artemia cysts remains constant when p26 is knocked down, demonstrating that the increased synthesis of ArHsp21 does not compensate for lost p26. Such results indicate either that ArHsp21 has function(s) different from those exhibited by p26 [43] or that regulation of the ArHsp21 and p26 genes varies.

ArHsp22 varies from other known diapause-related sHsps in Artemia because its synthesis is induced by heat in adults [24]. Additionally, the injection of adults with ArHsp22 dsRNA, in amounts that are 50% and 75% lower than those used for p26 and ArHsp21, caused death. Reducing ArHsp22 dsRNA to 12.5% of that normally used neither killed adult females, nor reduced ArHsp22 in cysts produced by these females. The handling and injection of adults may constitute stresses that induce the synthesis of ArHsp22 required for protection. Consequently, when ArHsp22 dsRNA is injected into adults, the ArHsp22 mRNA produced in response to handling is destroyed, resulting in a loss of the protein, reduced stress tolerance and death. During dsRNA injection, adult Artemia are placed on a cold agar plate for ~ 5 min and excess sea water is removed. Additionally, the injection itself may produce a stress response. Stressors other than heat elicit the expression of several heat shock protein genes, including cold in the insects D. melanogaster [57] and S. crassipalpis [58], desiccation in snails [48] and handling in flatfish [59]. To our knowledge, the results reported in the present study comprise the first evidence indicating that the short-term handling of an organism associated with the injection of dsRNA leads to a stress response, which may have consequences for experiments where heat shock proteins are studied by RNAi. The death of females made it impossible to determine the contribution of ArHsp22 to stress tolerance and cyst development.

To summarize, experiments carried out previously in vitro demonstrated that ArHsp21 and ArHsp22 are molecular chaperones [23, 24], whereas the findings obtained in the present study reveal that, unlike the situation for p26, knockdown of ArHsp21 has little or no effect on cyst stress tolerance. The same may be true for ArHsp22, which is present in cysts in lower amounts than ArHsp21. The results suggest less prominent roles during stress tolerance and development for ArHsp21 and ArHsp22 compared to p26.

Materials and methods

Culture of Artemia

In total, 6 g of dry Artemia cysts from the Great Salt Lake, Utah, USA (INVE Aquaculture, Inc., Ogden, UT, USA) were incubated at room temperature with aeration for 36–48 h in twice-filtered and autoclaved sea water from Halifax Harbor (termed sea water throughout the present study). Empty shells produced as a result of cyst hatching were removed from the surface of the seawater after cultures stood without agitation for 30 min. Nauplii were harvested and grown to adults in sea water at room temperature with constant aeration and daily feeding with Isochrysis sp. (clone synonym T-Iso) obtained from The Provasoli-Guillard National Center for Culture of Marine Phytoplankton (West Boothbay Harbor, ME, USA). Females were observed for up to five rounds of reproduction and nauplii arising from fertilized females, either by ovoviviparous or oviparous development, were raised to adults in small weigh boats containing seawater.

Quantitation of sHsp mRNAs in Artemia cysts

RNA was extracted with Trizol (Invitrogen, Burlington, ON, Canada) from rehydrated Artemia cysts (INVE Aquaculture, Inc.). cDNA was generated from 0.1 μg of RNA using the SuperScript® III First-Strand Synthesis Kit (Invitrogen) with oligo dT20 primers. qPCR, using 0.5 μL of cDNA, was conducted with the QuantiTect® SYBR® Green PCR Kit (Qiagen, Mississauga, ON, Canada) in a Rotor-Gene RG-3000 (Corbett Research, Sydney, Australia) with primers for α-tubulin, 18S rRNA, ArHsp21, ArHsp22 and p26 (Table 3) at 10 μm. To quantify cDNA, a standard curve was prepared using known copy numbers of DNA fragments representing either each protein of interest or 18S rRNA. DNA fragments were generated by PCR using 0.5 μL of cDNA with Platinum PCR SuperMix (Qiagen) and 10 μm forward and reverse primers for ArHsp21, ArHsp22, p26, α-tubulin and 18S rRNA (Table 3). The concentration of PCR products for each template was determined by measuring absorbance at 260 nm, and copy number was calculated based on PCR product length and a base pair mass of 650 Da ( The DNA fragment preparations were diluted in a ten-fold series with TE buffer (10 mm Tris, 1 mm EDTA, pH 8.0) and 0.5 μL of each dilution was used as template with the QuantiTect® SYBR® Green PCR Kit (Qiagen). The standard curve of the resulting Ct values was fitted and the copy number of each cDNA was determined using rotor-gene 6 software (Corbett Research). The copy numbers of ArHsp21, ArHsp22 and p26 were normalized against α-tubulin and 18S rRNA and these values were plotted. Because the ratio of sHsp : 18S rRNA was so much lower that the ratio of sHsp : tubulin, the former was multiplied by 60 to facilitate comparison and graphing of the two ratios. Melting curve analysis was performed for each experiment. mRNA for each sHsp was quantitated in three independently prepared samples from cysts.

Table 3. Primers used for RT-PCR, qPCR and dsRNA synthesis
cDNA amplifiedPrimers (5′- to 3′)
  1. a

    From Zhao et al. [61].

  2. All primers were obtained from Integrated DNA Technologies (Coralville, IA, USA). The T7 promoter is shaded.

dsRNA production

Quantitation of sHsps in Artemia cysts

Full-length cDNAs encoding ArHsp21, ArHsp22 and p26 were cloned into either His-tagged prokaryotic expression vector pRSET A or pRSET C (Invitrogen) [23, 24, 60]. The cDNA inserts were sequenced and the recombinant plasmids were transformed into Escherichia coli BL21(DE3) pLysS (Invitrogen). sHsp synthesis was induced with 1 mm isopropyl thio-β-D-galactoside for 6 h at 37 °C, and each recombinant sHsp was purified from cell free extracts of E. coli on BD TALON metal affinity resin (BD Biosciences, Mississauga, ON, Canada) with wash buffer (40 mm HEPES, 300 mm NaCl, 15 mm imidazole, pH 7.4) and elution buffer (40 mm HEPES, 300 mm NaCl, 150 mm imidazole, pH 7.4). Protein extracts were prepared from commercially obtained rehydrated Artemia cysts [8]. Protein samples were concentrated and desalted with Microcon Ultracel YM-10 centrifugal filter devices (Millipore EMD, Billerica, MA, USA). Protein concentrations of purified sHsps and cyst extracts were determined by the Coomassie Protein Assay using BSA as standard (ThermoScientific, Rockford, IL, USA). Varying but known amounts of cyst extract protein and purified sHsps were resolved by 12.5% SDS/PAGE and transferred to nitrocellulose membranes. The PiNK Plus Prestained Protein Ladder (FroggaBio, Toronto, ON, Canada) was used as a molecular mass marker. Each sHsp was quantitated in three independently prepared protein samples.

Protein-containing membranes were incubated in 5% Carnation low fat milk (Nestlé, Vevey, Switzerland) solution for 1 h at room temperature followed by exposure for 20 min at room temperature to antibody raised against ArHsp21 [23], p26 [22] and the ArHsp22 peptide N'-PWEEENEGEFRSGI-C' (amino acids 177–190) (Pacific Immunology Corp, Ramona, CA, USA). The antibodies to ArHsp21, p26 and ArHsp22 were diluted 1 : 5000, 1 : 10 000 and 1 : 5000, respectively, in 10 mm Tris containing 140 mm NaCl (pH 7.4) (Tris-NaCl). The membranes were washed three times for 5 min in Tris-NaCl and 0.1% Tween 20 (pH 7.4) (Tris-NaCl-Tween) and three times for 5 min in 10 mm TRIS containing 1 m NaCl, 0.5% Tween 20 (pH 7.4). The membranes were incubated for 20 min at room temperature in horseradish peroxidase-conjugated goat anti-(rabbit IgG) (Sigma-Aldrich, Oakville, ON, Canada) diluted 1 : 10 000 in Tris-NaCl. The membranes were washed as described above and antibody-reactive proteins were visualized with ECL Plus Western Blotting Detection Reagents (GE Healthcare, Baie d'Urfe, QC, Canada) and a DNR Bio-Imaging Systems MF-ChemiBIS 3.2 gel documentation system (Montreal Biotech, Montreal, QC, Canada).

imagej ( was used to determine the intensity of fluorescent bands corresponding to sHsps in samples of cyst extracts and purified sHsps. For each membrane containing cyst extract and purified sHsp, the intensity of fluorescent bands was normalized such that the strongest band was equal to one. Curves, fitted with a linear regression, were generated for the mean staining intensities of purified sHsp and the antibody reactive bands in cyst extracts to calculate the percentage that a particular sHsp represented of the total soluble protein in cyst extracts. For example, the intensity of ArHsp22 immunostaining with 100 μg of cyst extract corresponds to 0.39 μg of purified ArHsp22, or 0.39% of total protein. The experiment was performed in triplicate with independently prepared cyst extracts.

Preparation of dsRNA for ArHsp21 and ArHsp22

ArHsp21 and ArHsp22 cDNAs cloned in pRSET A (Invitrogen) were harvested from overnight cultures of E. coli BL21(DE3) pLysS (Invitrogen) using a miniprep kit (Sigma-Aldrich). Forward and reverse primers for ArHsp21 and ArHsp22 were used to amplify the cDNA by PCR and to attach T7 promoters to the 5′ ends (Table 3). PCR was performed using Platinum Taq DNA polymerase (Invitrogen) in accordance with the manufacturer's suggested concentration of 1 μL of cDNA: 5 min at 94 °C followed by 30 cycles of 94 °C for 30 s, 57 °C and 62 °C for 30 s for ArHsp21 and ArHsp22, respectively, and 72 °C for 1 min, followed by 10 min at 72 °C. The PCR products were resolved in 2% agarose gels in 0.5 × TAE (20 mm Tris, 10 mm acetic acid, 0.5 mm EDTA, pH 8.5) at 80 V and stained with SYBR®Safe DNA Gel Stain (Invitrogen) at twice the manufacturer's recommended concentration for 1 h before visualization with a Bio-Imaging Systems MF-ChemiBIS 3.2 gel documentation system (DNR Bio-Imaging Systems Ltd, Jerusalem, Israel). dsRNAs were generated from PCR products with the MEGAscript® RNAi kit (Ambion Applied Biosystems, Austin, TX, USA) in accordance with the manufacturer's instructions, including extension of incubation with RNA polymerase to 3 h. dsRNA was purified with the RNAeasy kit (Qiagen) and synthesis was confirmed by electrophoresis.

dsRNA specific to GFP was generated using forward and reverse primers (Table 3) [61]. The GFP cDNA used as template was cloned in the vector pEGFP-N1 (Clontech, Mountain View, CA, USA). The conditions described above were used to produce dsRNA for GFP, except that annealing was performed at 60 °C and MgCl2 was 2.25 mm.

Injection of Artemia adults with dsRNA

dsRNA specific to ArHsp21, ArHsp22 and GFP, as well as elution solution from the dsRNA kit, was mixed separately in a 1 : 1 ratio (v/v) with 0.5% phenol red in Dulbecco's NaCl/Pi (Sigma-Aldrich). The mixture of elution solution and phenol red was termed control solution. dsRNAs and control solution were introduced into the egg sacs of mature Artemia females or the unprotected area near the eye of males using a Nanoject II Microinjector (Drummond Scientific Co., Broomall, PA, USA) [43]. The amount of dsRNA injected for ArHsp21, ArHsp22 and GFP was 80 ng in 250 nL of elution buffer from the MEGAscript® RNAi kit (Ambion Applied Biosystems). Animals were immobilized on cooled 3% agar and gently blotted with Kimwipes (Kimberly-Clark Corp., Irving, TX, USA). Females destined to produce nauplii and cysts were differentiated by the absence or presence of a shell gland and by oocyte colour [22]. Adults were monitored for at least 2 h after injection and those animals exhibiting straight tails, normal swimming and feeding, and dye retention were used in the experiments. Fertilization (marked by fusion of lateral egg sacs), embryo development and morphological changes in animals were observed post-injection with an SZ61 stereomicroscope (Olympus, Tokyo, Japan).

Analysis of ArHsp21 mRNA knockdown in Artemia cysts

Approximately 60 hydrated cysts released from Artemia females individually receiving either dsRNA specific to ArHsp21 and GFP or control solution were incubated in sea water at room temperature for 10 days, homogenized with a microfuge pestle (Fisher Scientific, Ottawa, ON, Canada) in 600 μL of RLT reagent from the RNAeasy kit (Quiagen) and passed through a 20-gauge needle ten times. The mixture was centrifuged for 1 min at 20 g and the supernatant transferred to a fresh tube. cDNA was made with the Superscript® III First-Strand Synthesis Kit (Invitrogen) using oligo-dT20 primers. In total, 4 μL of the cDNA was amplified using forward and reverse primers for ArHsp21 (Table 3) and Taq polymerase (Fermentas, Glen Burnie, MD, USA): 5 min at 94 °C, 40 cycles of 94 °C for 30 s, 60 °C for 30 s and 72 °C for 1 min, followed by 10 min at 72 °C. PCR products were resolved in 2% agarose gels in 0.5 × TAE buffer at 100 V, stained with SYBR®Safe DNA Gel Stain and visualized in a Bio-Imaging Systems MF-ChemiBIS 3.2 gel documentation system.

Analysis of ArHsp21 knockdown in Artemia cysts

Protein extracts were prepared from 55 cysts produced by females injected separately with dsRNA specific to ArHsp21 and GFP or with control solution. The cysts were incubated in sea water at room temperature for at least 10 days after release from females, transferred to ice cold distilled water and collected by centrifugation for 1 min at 20 g. The supernatants were discarded and each tube received 10 μL of 2 × sample buffer for SDS/PAGE (250 mm Tris, 280 mm SDS, 40% (v/v) glycerol, 0.2% (w/v) bromophenol blue, pH 6.8) and 10 μL of ice cold Pipes buffer (100 mm Pipes, 1 mm MgCl2, 1 mm EGTA, pH 7.4) after which the cysts were homogenized using a microfuge pestle. The homogenate was placed in a boiling water bath for 5 min and all tubes were centrifuged at 4 °C for 10 min at 8600 g. In total, 15 μL of each sample was resolved by 12.5% SDS/PAGE and transferred to nitrocellulose membranes at 100 mA overnight at room temperature. PageRuler® Plus Pre-stained Protein Ladder (ThermoScientific) was used as molecular mass marker. The blots were incubated with antibody to ArHsp21. Occasionally, antibodies were removed from nitrocellulose membranes by incubation for 30 min at 50 °C in membrane stripping buffer (62.5 mm Tris-HCl, pH 6.7, 100 mm β-mercaptoethanol, 2% SDS). Membranes were washed twice for 10 min in Tris-NaCl-Tween, rinsed with distilled H2O, washed in Tris-NaCl several times and re-probed.

Determination of cyst stress tolerance

Cysts released from females injected with dsRNA specific to either ArHsp21 or GFP or with control solution were stored in sea water for 10 days. To break diapause, the cysts were desiccated for 14 days over Drierite (Drierite, Nashville, TN, USA) followed by incubation at −20 °C for ~ 8 weeks. Hatching efficiency after diapause breakage, used to indicate viability, was determined by incubating cysts at room temperature for at least 1 week in sea water with nauplii counted and removed. To test heat tolerance, diapause was terminated by desiccating and freezing broods of cysts lacking ArHsp21 from the same female. Half of the cysts from a brood were held at room temperature and the other half was heated at 41 °C for 30 min, after which cysts were transferred to sea water and hatching was assessed. The experiments were carried out in duplicate and a two-tailed t-test was performed.

Effect of ArHsp22 dsRNA on adult Artemia

Groups of 12 adult Artemia injected with either ArHsp22 or GFP dsRNA, or control solution, were monitored with a microscope over 4 days to determine survival. Females were also injected with a two-fold serial dilution of ArHsp22 dsRNA in control solution such that animals received 50%, 25% and 12.5% of the normal dose of dsRNA. The animals were photographed with an Infinity 1–1 camera (Lumenera, Ottawa, ON, Canada) mounted on an SZ61 stereomicroscope. The experiment was performed in triplicate.


This work was supported by a Natural Sciences and Engineering Research Council of Canada Discovery Grant (RGPIN/7661-2011) to THM.