Introduction

Aquatic resources are the final recipients of terrestrial pollutants and effluent discharges from industrial and domestic sources, which exert toxic effects on aquatic organisms and species higher in the food chain. The threat of heavy metal pollution is well documented and remains a matter of concern (Jarup 2003; Verma and Singh 2005). Although often present in minute quantities, their non-biodegradable trait poses a risk to the environment and human health. Several techniques are available for accurate estimation of heavy metal concentrations in living systems (Breuil et al. 1998; Townsend et al. 1998; Tiemann et al. 1999). They require expensive equipment, considerable sample pre-treatment and cannot distinguish between bioavailable and non-available fractions. Therefore, it is essential to develop a reliable, efficient and cost-effective method which can monitor the presence of hazardous metals in the environment. Biosensors can act as an alternative to conventional analysis of heavy metals. Bio-available concentrations of contaminants can be directly measured by biosensors even when present at very low levels (Roberto et al. 2002). Researchers have developed metal-specific recombinant bacterial biosensors by fusing a metal regulatory protein gene promoter upstream of various reporter genes (Raja and Selvam 2011). However, being prokaryotic organisms, they may not indicate the toxic concentration ranges that trigger a response in eukaryotes. Therefore, attempts have been made to develop eukaryotic biosensors using yeasts, microalgae, ciliated protozoa and C. elegans, which facilitated the extrapolation of the results to higher eukaryotic organisms (Lagido et al. 2001; Gutierrez et al. 2015). In our study, we selected zebrafish, a common tropical eukaryotic multicellular vertebrate system, to develop a transgenic fish biosensor.

Zebrafish (Danio rerio) is widely used as a vertebrate model organism for research on toxicity (OECD 1992; Detrich et al. 1998), developmental biology (Driever et al. 1994), drug discovery, and number of other applications (Lele and Krone 1996), because zebrafish embryonic development is remarkably similar to that of humans. Zebrafish eggs are laid and fertilized externally, with large clutch size, rapid development and transparent embryos making it convenient to observe the movement and fate of individual cells during embryonic development (Kimmel et al. 1995). Carvan et al. (2000) gave a conceptual overview of the use of zebrafish for detecting detrimental effects of aquatic pollutants. Recently, transgenic zebrafish lines have been developed using stress promoters like heat shock protein (hsp) for detecting heavy metals (Blechinger et al. 2002; Seok et al. 2006, 2007; Wu et al. 2008; Lee et al. 2014).

Metallothioneins (MTs) are small, low molecular mass, cysteine-rich proteins that function as sensitive indicators of heavy metal contamination (Park et al. 2001). Metallothionein promoters containing metal-responsive elements (MREs) could be linked to fluorescent reporter genes to develop heavy metal-responsive biosensors. Although zebrafish has its own metallothionein gene, in the present study we used the metallothionein (MT-Ia1) promoter from green mussel, Perna viridis. This is mainly because being bivalve molluscs, green mussels have capability to accumulate high concentrations of heavy metals in their tissue as a result of filter feeding, and hence exhibit a high response towards heavy metals. Also, marine invertebrate metallothioneins are known to respond to a broad spectrum of heavy metals, Hg2+ > Cu2+ > Cd2+ > Zn2+ (Viarengo 1989). This property is particularly important, since a single biosensor potentially could be employed to detect several heavy metals simultaneously. Here, we report the use of MT-Ia1-DsRed2 transgene in a Tol2 transposon vector to develop a transgenic zebrafish heavy metal biosensor system.

Materials and methods

Construction of biosensor plasmid

The metallothionein Ia1 (MT-Ia1) gene promoter (1190 bp) was amplified from genomic DNA of green mussel, P. viridis using Pvir-MTI-F (TCCTTCCTCAGCA TGAAAC) and Pvir-MTI-R (AAGTGGCTGTATGTCTCAGTTG) primers. A 25-µl PCR mix consisting of 1 µl of template (75 ng), 1 µl each of forward and reverse primers (10 pmol), 2 µl of dNTPs (2.5 mM each), 2.5 µl of 10 × Taq buffer (with 1.5 mM MgCl2) and 0.25 µl (1 U) of Taq DNA polymerase (Thermoscientific, USA) was prepared in nuclease-free water. PCR conditions included initial denaturation at 95 °C for 5 min; followed by 35 cycles of 94 °C for 30 s, 52 °C for 30 s and 72 °C for 1 min; and final extension at 72 °C for 8 min. The amplicon was T/A cloned and confirmed by sequencing. The promoter was then amplified from the clone using linker primers having BglII and EcoRV sites in the above-mentioned forward and reverse primers, respectively. This fragment was directionally cloned into the pTol2 vector (a gift from Dr. S.C. Ekker, Mayo Clinic, USA). The DsRed2 reporter gene (922 bp) was PCR-amplified from pFRM-DsRed2 plasmid (kind gift from Dr. S.C. Ekker, Mayo Clinic, USA) using linker primers (DsRed2-F: AAAGATATCAGTTCAGCCGGAATTCACC and DsRed2-R: AAAAAGCTTACAGA GTGAGCCGATCCGAG) with EcoRV and HindIII RE sites in the forward and reverse primers, respectively. PCR conditions were the same as above except that the annealing temperature was set at 55 °C. The amplicon was gel-purified, digested with EcoRV and HindIII restriction enzymes and cloned downstream to the MT-Ia1 promoter in the similarly digested pTol2-MT-Ia1 plasmid. The construct with the promoter-reporter cassette was transformed into E. coli DH5α cells. Positive clones were confirmed by colony PCR and sequencing (Sambrook and Russell 2001). The biosensor construct was named pTol2-MT-Ia1-DsRed2 (Fig. 1).

Fig. 1
figure 1

Schematic representation of the biosensor construct in mini Tol2 transposon vector. The construct comprises the metallothionein Ia1 (MT-Ia1) promoter from green mussel, P. viridis, the DsRed2 reporter gene, and the mini Tol2 terminal inverted repeat sequences required for transposition. (Color figure online)

Full size image

Zebrafish maintenance

Wild-type zebrafish were maintained at 28 °C in aquaculture system (63 mg/l Red Sea Coral Pro Salt in RO-purified water) on a 14 h light and 10 h dark photoperiod cycle. Newly hatched fish larvae were fed Paramecium. Mature fish were fed thrice daily on freshly hatched brine shrimp. Adult zebrafish were bred in small tanks with 1:2 female-male ratio to obtain embryos (Westerfield 2000). The care and treatment of fish used in this study were in accordance with the guidelines of the CPCSEA [(Committee for the Purpose of Control and Supervision of Experiments on Animals), Ministry of Environment and Forests (Animal Welfare Division), Govt of India] on the care and use of animals in scientific research, and the work plan was approved by appropriate committees.

Microinjection of biosensor construct and screening for transgenics

Tol2 transposase mRNA was synthesized from the pDB600 plasmid (kind gift from Dr. S. C. Ekker, Mayo Clinic, USA) using the mMESSAGE mMACHINE Capped RNA In vitro Transcription kit (Ambion, USA) and stored at −80 °C. Zebrafish embryos at the one/two cell stage were microinjected at the blastoderm/yolk interface with 3 nL of solution containing 8.3 ng/µl of pTol2-MT-Ia1-DsRed2 plasmid and 100 ng/µl of Tol2 transposase mRNA. Injected embryos were reared till maturity in an aquaculture system with a few drops of methylene blue as an anti-fungal agent. After grow-out, the mature fish were mated with wild-type counterparts, and pools of 6 embryos were screened for the presence of the transgene by PCR amplification of the MT-Ia1 promoter and DsRed2 gene from genomic DNA. Transgenic parents were labelled and housed separately.

LC50 studies

To obtain 24-h LC50 values for various heavy metals, dechorionated 48-h old wild-type zebrafish embryos were exposed to 0, 0.1, 0.2, 0.3, 0.4 or 0.5 ppm (parts per million) of Hg2+and Cu2+; 0, 5, 6, 7, 8, 9 or 10 ppm of Cd2+; 0, 5, 7, 9, 11, 13 or 15 ppm of Pb2+ ions or 0, 10, 12, 14, 16, 18 or 20 ppm of Zn2+ ions at 28.5 °C for 24 h. These concentrations were determined from the results of earlier dose-ranging studies for each metal ion (data not shown). Unexposed controls were maintained alongside. Triplicates were run with 10 embryos in each Petri plate. Surviving individuals were counted 24-h post-treatment. The LC50 values obtained were used to decide the concentration range to be used for induction studies.

Induction of transgenics with heavy metals and fluorescence microscopy

Confirmed transgenics were bred with wild counterparts to obtain F1 progeny, which were dechorionated and exposed to different heavy metals at 48 h post-fertilization (hpf) to confirm and quantify reporter expression. The embryos were exposed to 0.05, 0.1, 0.3 or 0.5 ppm of Hg2+, Cd2+, Cu2+ or Pb2+ ions; or 5, 10, 15 or 20 ppm of Zn2+ ions for 8 h. Unexposed transgenic embryos and unexposed wild-type embryos constituted the two negative control sets. Duplicates were run with ten embryos in each treatment group. 8 h post-exposure, DsRed2 expression was observed under a Zeiss Axioscope 2 compound microscope with a rhodamine filter set (excitation and emission wavelengths of 546 ± 10 and 570 ± 10 nm, respectively). A Zeiss Axiocam digital camera was used for imaging the fluorescent reporter expression in the transgenic zebrafish. Constant exposure settings were used throughout the imaging of experimental larvae. The fluorescence images were used to quantify the red fluorescence in the control as well as experimental groups by ImageJ software. The relative fluorescence intensity in the larvae after metal exposure was calculated by normalizing to the background fluorescence in the un-induced controls.

Real-time PCR

Quantitative RT-PCR (qRT-PCR) was performed to confirm expression of the reporter gene in zebrafish larvae. Ten embryos from each treatment group (as described above) were pooled and stored in RNAlater solution (Qiagen, NL) at −20 °C until total RNA was isolated using RNeasy mini kit (Qiagen, NL). First-strand cDNA was synthesized from 1 µg of total RNA and oligo(dT) primer using First Strand cDNA Synthesis kit (Thermoscientific, USA). Real-time PCR was performed on ABI 7500 Real-time PCR machine using Quantifast SYBR Green PCR mix (Qiagen, NL). Primers qDsR-F (CTACCTGGTGGAGTTCAAGTCC) and qDsR-R (CGCTACAGGAACAGGTGGTG) were designed to amplify a 165-bp fragment of the DsRed2 gene, while qGapdh-F (GTGGAGTCTACTGGTGTCTTC) and qGapdh-R primers (GTGCAGGAGGCA TTGCTTACA) were designed to amplify a 173-bp fragment of the GAPDH gene that served as reference for baseline-level gene expression. Gene Runner software v. 3.0 was the primer design software used. A 15-µl PCR mix comprising 7.5 µl of 2 × SYBR Green master mix (Thermoscientific, USA), 1 µl of cDNA, 1 µl each of forward and reverse primers (0.3 µM) and 4.5 µl of nuclease-free water was prepared. The PCR program included an initial denaturation at 95 °C for 5 min; with 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Experimental duplicates were run. The comparative Ct method was used to estimate the relative expression of DsRed2 mRNA. Fold-change in DsRed2 expression was calculated by the \(2^{{ - {\Delta \Delta }C_{\text{t}} }}\) method (Schmittgen and Livak 2008).

Southern blotting

Southern blot hybridization was carried out to confirm the genomic integration of the biosensor cassette following standard protocol (Sambrook and Russell 2001). A 165-bp region of the DsRed2 gene sequence within the biosensor cassette was used as the probe. Probe labelling was done using the PCR DIG probe synthesis kit (Roche Applied Science, Switzerland) following the manufacturer’s instructions. Briefly, genomic DNA was isolated from F1 embryos (offspring of transgenic × wild-type), and the presence of the transgene was confirmed by PCR as mentioned above. The genomic DNA was partially digested with HindIII which had a single site at one end of the promoter-reporter cassette. The restriction enzyme-digested fragments were separated on 1 % agarose gels and transferred to a positively charged nylon membrane (Pall Life Sciences, USA) using alkaline upward capillary transfer. The membrane was probed with DIG-labelled probe, and the hybridization was detected using the DIG-Nucleic Acid detection kit (Roche Applied Science, Switzerland) following the manufacturer’s instructions. The results were documented using a Syngene Gel documentation system.

Statistical analysis

Real-time PCR and relative fluorescent intensity data were analyzed by one-way ANOVA, and Tukey’s test was done to determine statistical significance of changes observed in reporter expression among different treatment groups.

Results

Microinjection of biosensor construct and screening for transgenics

The biosensor construct pTol2-MT-Ia1-DsRed2 was microinjected into one/two cell stage zebrafish embryos along with in vitro transcribed Tol2 transposase mRNA. In total, 1117 embryos were injected during the course of this study, and about 60 % survival was obtained at 72 h post-injection. However, only 87 fish (8 %) survived to maturity, of which 32 produced offspring (F1) when mated with wild-type mates. Genomic DNA was isolated from pools of 48 h-old F1 embryos, and the transgenics were confirmed by PCR amplification of MT-Ia1 and DsRed2 regions of the transgene cassette. Four males and five females of the 32 microinjected broodstock were confirmed as transgenics, and the integration rate was calculated to be 28 %. The transgenic parents were mated with their wild counterparts to obtain F1 embryos for further studies.

LC50 of heavy metals for zebrafish embryos

All the heavy metals caused a dose-dependent increase in mortality of zebrafish embryos. The LC50 at 24 h was estimated as 0.4 ± 0.01, 0.5, 8.5 ± 0.21, 8.75 ± 0.35, 17 ± 0.28 ppm for Hg2+, Cu2+, Cd2+, Pb2+ and Zn2+ ions, respectively. The sub-lethal concentrations used for the subsequent induction studies were decided based on these values.

Reporter detection in transgenic larvae by fluorescence microscopy

Heavy metal-induced DsRed2 reporter expression recorded by fluorescence microscopy could be seen in almost 90 % of the transgenic embryos in a mosaic pattern detected largely in dorsal and ventral retina, gills and the skin epithelium (Fig. 2). The relative fluorescence intensity in the experimental groups as quantified by ImageJ software revealed highest RFP expression in Hg2+-induced larvae (Fig. 3). The fluorescence intensity also increased with increasing metal concentration. Low levels of red fluorescence were observed in egg yolk and eyes of some unexposed transgenic larvae, and those exposed to Zn2+ and Pb2+. This indicates some leakage from the promoter because no fluorescence was recorded in the wild-type controls. However, this leaky expression was never observed in body parts other than eyes and yolk sac, which could be because of the transparency of these body parts. The larvae were phenotypically normal at all metal concentrations used for induction, except at 0.5 ppm for Hg2+, where no animals survived. No other mortality was recorded during the experiment.

Fig. 2
figure 2

Fluorescence imaging of F1 transgenic zebrafish (D. rerio) larvae (48 h post-fertilization) exposed for 8 h to Hg2+, Cd2+, Cu2+, Pb2+ or Zn2+ at doses that gave maximum fluorescence. Fluorescence expression is observed only in the yolk sac of un-induced transgenic control, while it is present in other body parts in the induced larvae. BL bright field, FL fluorescence. Scale bar 200 µm

Full size image
Fig. 3
figure 3

Quantification of DsRed2 fluorescence in F1 transgenic zebrafish larvae exposed to Hg2+, Cd2+, Cu2+, Pb2+ or Zn2+ (a–e) at different doses. The mean fluorescence intensity values normalized to background fluorescence from un-induced controls for each experimental group were plotted. Data represents mean ± SEM. (ns not significant; **p < 0.01; ***p < 0.001)

Full size image

Reporter detection in transgenic larvae by real-time PCR

Heavy metal-induced larvae from each plate were pooled, and expression of the DsRed2 reporter gene in all the treatments and controls was confirmed by real-time PCR (Fig. 4). While the DsRed2 transcript was absent in the wild-type controls, it was detected in the unexposed transgenic controls, showing leaky expression from the promoter. The difference in mean fold-change in reporter expression calculated with respect to the unexposed transgenic controls was significant in larvae induced with Hg2+, Cd2+ and Cu2+ ions, and the fold-change difference increased with the concentration of inducing ions. Eightfold increase was observed on exposure to 0.3 ppm Hg2+ (p < 0.001), while no larvae survived the 0.5 ppm exposure. In case of Cd2+ (p < 0.001) and Cu2+ (p < 0.01) ions, sixfold and twofold significant up-regulation, respectively, was noted at 0.5 ppm. No increase in reporter expression was observed on induction with lead and zinc. The heavy metals induced promoter function in the order Hg2+ > Cd2+ > Cu2+ at the highest sub-lethal concentration tested.

Fig. 4
figure 4

Real-time PCR analysis of DsRed2 mRNA expression in F1 transgenic zebrafish larvae exposed to Hg2+, Cd2+, Cu2+, Pb2+ or Zn2+ (a–e) at different doses. Data represents mean ± SEM. ns expression not significantly different from control; **expression significantly higher than control (p < 0.01); ***expression highly significant above the control (p < 0.001)

Full size image

Confirmation of transgene integration

The transgenic fish identified by PCR screening were mated with wild-type fish, and 15-day old F1 embryos were pooled, with 10 embryos in each pool. Assuming one integration event per transgenic individual, on mating with wild-type counterparts, 50 % of the embryos would be expected to carry the transgene. It was expected that each integration event would be visible as one hybridized band since there was no other restriction site for HindIII in the transgene. If the transgene construct is extra-chromosomal, then a single band of ~4.7 kb is expected. However, six bands of sizes 6, 3.8, 2.8, 2.5, 2.4 and 1.0 kb were detected in pooled F1 embryos from a single cross (transgenic × wild), indicating at least six integration events in the transgenic parent (Fig. 5). Since the DNA was from a pool of embryos and not from an individual sample, the results could not be used for copy number estimation.

Fig. 5
figure 5

Southern blotting using DIG-labelled dsRed probe confirms integration. Lane 1 Generuler 1 Kb ladder (Thermoscientific, USA); lane 2 pooled F1 embryos from a single transgenic × wild-type cross shows six bands of sizes 6, 3.8, 2.8, 2.5, 2.4 and 1.0 kb, indicating six integration events

Full size image

Discussion

Use of zebrafish, a small tropical fish, has revolutionized the field of developmental biology and soon became the favourite vertebrate model. Owing to its transparent body and easy maintenance, it has been used extensively for developing transgenic models expressing reporter proteins for environmental toxicity testing. Several researchers worldwide have developed transgenic zebrafish lines expressing fluorescent proteins under the control of promoter elements such as estrogen response elements (ERE; Chen et al. 2010, Lee et al. 2012a, b), aryl hydrocarbon response elements (AHREs; Mattingly et al. 2001), heat-shock protein (hsp) promoters (Blechinger et al. 2002; Seok et al. 2007; Wu et al. 2008), glycoprotein hormone promoters (Ji et al. 2012; Cheng et al. 2014), DNA-damage inducible promoters (Gireesh-Babu et al. 2012) and tissue-specific promoters (Wan et al. 2006; Sun et al. 2010; Almeida et al. 2014) for aquatic pollution monitoring. So far, most of the transgenic zebrafish biosensors developed for heavy metal detection are based on hsp promoter elements that are triggered by several other stressors (Blechinger et al. 2002). In most cell types even prior to stress, HSPs constitute 1–2 % of total protein, suggesting a strong constitutive basal expression from the hsp promoter (Stephanou and Latchman 2011). In this scenario, it would be difficult to quantify the heavy metal levels using reporter expression, as it indicates a cumulative effect rather than a heavy metal-specific effect. MTs are reported to be specifically induced in response to heavy metal treatment, whereas the synthesis of hsp70 appears to be a general stress response (Misra et al. 1989). Cioci et al. (2000) compared the responsiveness of transgenic Caenorhabditis elegans containing hsp-regulated versus MTl-regulated transgenes, and reported that the MTl reporter system is a more sensitive monitor for metal exposure. Also, metallothionein gene promoters contain metal-response elements (MREs) that are responsive to heavy metals, and hence are a better choice for developing biosensors for heavy metals. Recently, Liu et al. (2016) have generated a mt:egfp transgenic zebrafish as a biosensor with a zebrafish MT promoter that is sensitive to environmental concentrations of zinc and cadmium.

Here, we report the use of a broad-spectrum metallothionein (MT-Ia1) promoter from a marine invertebrate, P. viridis, for developing a zebrafish biosensor system. Viarengo (1989) reported that marine invertebrate metallothioneins are responsive to heavy metals in the order of Hg2+ > Cu2+ > Cd2+ > Zn2+. Khoo and Patel (1999) reported the promoter sequence of the P. viridis MT-Ia1 gene. The green mussel from which it is derived is itself used as a bioindicator of heavy metals (Putri et al. 2012). The MT-Ia1 promoter consists of three active protein-binding (AP1) sites and two metal-responsive elements (MRE) that regulate its expression in response to heavy metals (Khoo and Patel 1999). The DsRed2 fluorescent reporter used here is a variant of the original red fluorescent protein isolated from a coral, Discosoma sp. Unlike GFP, it has higher signal-to-noise ratio and is resistant to photobleaching (Baird et al. 2000, Clay and Ramakrishnan 2005). However, the wild-type protein has disadvantages of slow maturation and insolubility. Through a combination of random and site-directed mutagenesis, Bevis and Glick (2002) developed a variant called DsRed2 with improved solubility, enhanced fluorescence and speed of maturation. An asparagine-to-glutamine substitution at position 42 resulted in 10–15 times faster maturation of DsRed, but also increased the level of green emission. This green emission was further suppressed by additional amino acid substitutions, and a reduction of net charge near the N terminus of the protein resulted in enhanced solubility.

Use of the Tol2 transposon vector in this study resulted in a germ-line integration rate of 28 %, which is higher than the 1.8 % obtained by Wu et al. (2008) and Lee et al. (2011). In these studies, transgenesis was performed by microinjecting a linearized naked plasmid vector that integrates into the host genome less efficiently. The Tol2 transposon vector has been used effectively before and is known to integrate randomly into the vertebrate genome (Kawakami et al. 1998; Balciunas et al. 2006; Zou et al. 2006). The high germ-line integration rates (60–70 %) reported earlier for the Tol2 transposon vector could be due to the use of a constitutive promoter that helped in screening the embryos for GFP fluorescence before being reared to maturity (Urasaki et al. 2006). Since an inducible promoter was used here, all embryos were reared, and this could have resulted in the lower integration rate of 28 %. Although Tol2 has been observed to create single-copy insertions most often (Kawakami 2007), at least six integration events were detected in one transgenic individual in this study. Using a similar vector, Urasaki et al. (2006) found the total number of insertions transmitted by a single founder fish varies from 1 to 15.

The mosaic pattern of reporter expression observed after induction with heavy metals could be due to pigmentation obscuring the fluorescence, integration of head-to-tail concatemers, position effects at the integration site or variation in copy number, as was observed for GFP expression in the F1 embryos of transgenic Nile tilapia (Fujimura and Kocher 2011). Although no red fluorescence was detected in wild-type controls, some leaky expression of DsRed2 was seen in un-induced transgenic embryos by real-time PCR. This leaky expression could be due to the basal expression of this promoter in the absence of inducing heavy metals. Background GFP fluorescence was observed in transgenic C. elegans containing the MT2 promoter fused to the GFP gene, which was 10-fold to 15-fold lower than the post-induction values (Ma et al. 2009). MTs are closely associated with homeostasis of essential metal ions and hence some level of constitutive expression is expected. MT expression is also known to be controlled by developmentally regulated pathways (Chen et al. 2004). Since, in the present study zebrafish larvae (48 hpf) were used for heavy metal exposure assay, there is a possibility of basal expression from the MT-Ia1 promoter. However, under the fluorescence microscope, this expression was visible as fluorescence in the eyes and yolk sac which was easily distinguishable from metal-induced fluorescence in other body parts. This pattern of background fluorescence in the control was found to be consistent across different larvae and experimental groups. Hence, this fluorescence is not expected to interfere in any way with the heavy metal detection assay.

Kusik et al. (2008) used an electrophile-responsive element (EPRE) and luciferase-green fluorescence protein (LUC-GFP) fusion reporter to detect Hg2+ ions ranging from 0.1 to 1 µM. Human and zebrafish hsp70 promoters have been employed by a number of workers to detect Cd2+ (0.2–125 µM), CuSO4 (1–1.5 µM) and As3+ (10–300 µM), and reporter response was found to be dose-dependent (Blechinger et al. 2002; Seok et al. 2006, 2007). Wu et al. (2008) developed a heat shock-inducible gfp transgenic zebrafish line by cloning the zebrafish hsp27 promoter in a pEGFP-1 vector. The transgenic line has been maintained for six generations. Induction was carried out using cadmium (1.35–135 µM) and sodium arsenate (64–640 µM). It was selectively inducible by arsenic in a dose-dependent manner. In the present work too, the transgenic larvae showed a dose dependent induction pattern for the heavy metals studied. According to the Environmental (Protection) Rules of India (1986), the water quality standards for effluent discharge of mercury, cadmium and copper are 0.01, 1–2, and 3 ppm, respectively. The minimum detection limit by the zebrafish biosensor developed in the present study is 0.1 ppm for mercury and 0.5 ppm for cadmium and copper as evident by real-time PCR. However, it is to be noted that the transgenic larvae used here are mosaic for the transgene. Hence, the exact detection range of the heavy metals by this biosensor may be determined only once a true transgenic line is established. At that point, these transgenic zebrafish might be used to detect the presence of heavy metals in freshwater samples. For this, the adult transgenic zebrafish biosensors could be exposed to effluent water samples in the laboratory and the expression of red fluorescence protein be studied by fluorescence microscopy. However, it is even possible to visualize the red fluorescence in daylight or under ultra-violet light (385 nm) provided it expresses in large quantities, alleviating the need for fluorescence microscopy. If this becomes a reality, the day will come where these zebrafish can be introduced into cages near effluent treatment plants to test effluent water samples prior to their discharge in receiving water bodies.

Conclusion

The present transgenic zebrafish biosensor for sensing heavy metal toxicity employs the P. viridis MT-Ia1 metallothionein promoter fragment known to respond to a broad spectrum of heavy metals. These fish are expected to report presence of heavy metals in the bio-available fraction at levels toxic to eukaryotes and can be used as a primary screening method for water samples from areas likely to be polluted by the heavy metals. Quantification can be thus limited to positive samples to minimize the cost of expensive analytical techniques.