The Use of Zebrafish Mutants to Identify Secondary Target Effects of Acetylcholine Esterase Inhibitors

Martine Behra*,1, Christelle Etard*,{dagger},1, Xavier Cousin{ddagger} and Uwe Strähle*,{dagger},2

* Institut de Genetique et de Biologie Moleculaire et Cellulaire, Illkirch, France; {dagger} Forschungszentrum Karlsruhe, Karlsruhe, Germany; and {ddagger} Unite Différenciation Cellulaire et Croissance – INRA, Montpellier, France

Received August 8, 2003; accepted October 20, 2003


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We are confronted with a large and steadily growing number of bioactive compounds, including drugs, pesticides, and industrial by-products. The assessment of target specificity and potential toxic effect on human health and the environment generates a strong demand for robust and cost-effective models with high predictive power. We investigated the potential of the zebrafish embryo as a whole organism, vertebrate model to assess the specificity of compounds that are known to inhibit acetylcholinesterase (AChE). Inhibitors of AChE are widely used as drugs and pesticides. By application of simple assays and comparison with the phenotype of embryos with genetic lesions in the ache gene, we demonstrate that only one of the AChE inhibitors (galanthamine) reproduces the phenotype of ache mutant embryos. The other compounds produced additional effects indicating secondary targets. Our work demonstrates the power of a genetic system for toxicological evaluations. The combination of genetics and transgenesis with the other experimental virtues of the zebrafish embryo, such as small size and low cost, offers a whole organism platform for medium to high throughput compound testing.

Key Words: zebrafish; acetylcholinesterase; inhibitors; animal model; secondary drug targets.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A powerful methodology in the development of bioactive chemicals has arisen from the synergy between structural information on the target molecules and combinatorial chemistry (Norin and Sundstrom, 2001Go). While the capability to synthesize novel compounds increased enormously in recent years, technologies to assess the biological activity and target specificities of these novel molecules did not match pace (Balls, 2002Go). A good drug is potent and highly specific with a well-defined target. Ideally, the phenotypic consequences of an inhibitory compound should be the same as that of a genetic mutation causing loss-of-function of the drug target. Simple genetic models such as yeast (Marton et al., 1998Go) or Caenorhabditis elegans (van Kessel et al., 1989Go) may help to identify primary and secondary targets, but they are too distant in evolutionary time to provide a comprehensive toxicological profile for vertebrates.

Besides associated ethical concerns, the use of mammals experimentally is expensive and labor intensive, limiting an application in large-scale screening programs. Embryos and larvae of the lower vertebrate zebrafish (Danio rerio) may offer a cheap and effective alternative. Zebrafish embryos/larvae are small and can be obtained in large numbers throughout the year at a fraction of the cost of a mouse or a rat. Zebrafish embryos develop outside of the mother, thereby allowing systematic compound screens (Peterson et al., 2000Go) from earliest life stages onwards. Moreover, a large number of mutations affecting signaling pathways and physiological processes relevant to human pathology and pharmacology are available (Barut and Zon, 2000Go; Driever et al., 1996Go; Haffter et al., 1996Go; Langheinrich et al., 2002Go; Stern and Zon, 2003Go).

For this study, we investigated the application of wild-type and mutant zebrafish embryos in drug testing by phenotypic comparisons with a mutation in the acetylcholinesterase (ache) gene. AChE catalyzes the hydrolysis of the neurotransmitter acetylcholine (ACh) and is crucial for cholinergic neurotransmission (Soreq and Seidman, 2001Go). AChE is the target of many toxins including snake venom, insecticides, and chemical weapons (Soreq and Seidman, 2001Go). The pharmacological use of AChE inhibitors includes treatment of the autoimmune disease myasthenia gravis, glaucoma, and Alzheimer’s disease (Soreq and Seidman, 2001Go).

We have previously identified a recessive mutation in the ache gene of zebrafish (Behra et al., 2002Go). Zebra fish AChE is highly related to that of mammals (Bertrand et al., 2001Go). A point mutation leading to replacement of a conserved serine at position 226 by an asparagine abolishes AChE enzymatic activity in homozygous mutant embryos. Mutants have impaired motility and develop a severe myopathy during the second day postfertilization (Behra et al., 2002Go). The myopathy is due to an overactivation of the muscle by accumulating ACh; genetic impairment of the nicotinic acetylcholine receptor (nAChR) on the muscle cell prevents the development of the myopathy in ache mutants (Behra et al., 2002Go).

The zebrafish presents a unique situation among vertebrates as the AChE is the only ACh-hydrolyzing enzyme in this organism. Homozygous ache mutant embryos and larvae lack detectable ACh-hydrolyzing activity and heterozygous adults show 50% of the ACh cleaving activity of wild-type siblings (Behra et al., 2002Go). Moreover, the zebrafish genome does not encode a functional butyrylcholine esterase, a related enzyme that can also hydrolyze ACh (Bertrand et al., 2001Go).

The phenotypic traits of the ache mutant provide criteria for a vertebrate that lacks ACh-hydrolyzing activity entirely and can thus serve as comparative standards for the specificity of AChE inhibitors. In this report, we used the zebrafish embryo as a whole organism model to test the specificity of chemicals. We chose four drugs known to be AChE inhibitors (Cousin et al., 1998Go; Giacobini, 1998Go) and compared their effects with the consequences of genetic elimination of AChE activity. Only one test compound (galanthamine, GAL) reproduced the ache mutant phenotype. The other three compounds (eserine, ESE; tacrine, TAC; and edrophonium, EDRO) showed effects in addition to those observed in ache mutants, demonstrating secondary target effects. In general terms, this work illustrates the power of the combination of zebrafish genetics with toxicological testing in the verification of drug target specificity and demonstrates the potential of the zebrafish embryo in the assessment of chemical toxicity.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Fish stocks and embryo production.
The wild-type zebrafish line wt ABO is a cross between fish purchased from a pet shop and the AB strain (University of Oregon, Eugene, Oregon) and was maintained as an inbred line for several years in the laboratory. Fish were bred and raised as described (Westerfield, 1993Go). The origin of the ache and the nic1 mutant and the -3.1ngn1:gfp transgenic zebrafish have been described (Behra et al., 2002Go; Blader et al., 2003Go; Sepich et al., 1998Go). To generate transgenic ache mutant zebrafish, ache+/- fish were mated with -3,1ngn1:gfp transgenic fish and the offspring were mated with an ache+/-;-3.1ngn1:gfp+/- genotype that was crossed to ache+/- animals. Animal care and experimentation were performed in compliance with French and European law (authorization B67-218-5 from February 23, 1999).

Inhibitor treatment.
Embryos were treated with the inhibitors GAL (1,2,3,4,6,7,7a,11c-octahydro-9-methoxy-2methylbenxzofuro[4,3,2-efg][2]benzazocin-6-ol), ESE (1' methylpyrrolidino [2':3':2:3] 1,3 dimethylindolin-5-yl N-methylcarbamate), TAC (1,2,3,4-tetrahydro-9-aminoacridine), and EDRO ([3-hydroxyphenyl] dimethylethylammonium bromide) diluted in 10% Hanks’ solution (Westerfield, 1993Go) from the 5-somite stage onwards. Solutions were replaced each day. Unless stated otherwise, 10-4 M ESE, 10-3 M GAL, 10-2 M EDRO, and 10-5 M TAC were used. These concentrations were determined in motility assays and by morphological inspection of embryos to phenocopy the ache mutant phenotype with respect to motility most closely. Higher concentrations of TAC and EDRO caused necrosis in the neural tube and were thus regarded to have a general toxic effect on the embryos.

Motility assays.
For video analysis, embryos or young larvae were embedded in low melting point agarose, as described previously (Behra et al., 2002Go; Westerfield, 1993Go). To allow recording of motility, the trunks and tails of embedded embryos were freed of agarose. Embryos were mounted under a dissecting microscope (Leica MZ FLIII, Leica, Bensheim, Germany) equipped with a CoolSnap video camera (Roper Scientific, Munich, Germany). Embryos were touched with a blunt metal rod at the body flank to trigger the startle response and swimming movements. Touch-evoked movements of embryos were monitored using the video camera recording 50 frames/s. Representative frames were overlaid to indicate the movement.

For measurement of the duration of swimming, individual embryos were placed in a 5-cm petri dish. After touching the embryos, movement was recorded with the video camera. The duration of the swimming movements was measured with a stopwatch from the recorded film. In these experiments, only the swimming movements were scored. However, most inhibitor-treated and ache mutant embryos exhibited the startle response with a very fast bend of the body axis in the 10 ms range followed by a slow straightening of the body axis over the next 200 ms. This movement was not scored as swimming movement since, with the exception of the initial bend, it does most likely not reflect an active movement. The data sets were subjected to statistical analysis with the Student’s t-test to assess the significance of the similarities and differences among wild-type, ache mutant, and inhibitor-treated embryos.

Measurement of AChE enzymatic activity.
Detection of AChE activity in situ was adapted from Karnovsky (Karnovsky and Roots, 1964Go). For IC50 measurements, AChE activity in supernatants of 4-day-old larvae was carried out in triplicate, as described previously (Behra et al., 2002Go; Bertrand et al., 2001Go; Ellman et al., 1961Go). Larvae were killed by incubation on ice for 5 min. Extract was prepared by passing embryos ten times through a 26-gauge syringe in LST buffer (20 mM Tris/HCl pH 7.0, 5 mM EDTA, 1% Triton X-100). Afterwards, extracts were centrifuged for 1 min at 10,000 rpm; supernatants were diluted in 100 mM phosphate buffer, pH 7.0, and 0.5 mM DTNB and preincubated with varying concentrations of inhibitors for 10 min (TAC, GAL, and EDRO) or 60 min (ESE). Acetylthiocholine was added to a final concentration of 1 mM before real-time determination of acetylthiocholine hydrolysis at 412 nm (Ellman et al., 1961Go). IC50 values were deduced from the plots of activity versus inhibitor concentration from three independent measurements at every inhibitor concentration.

Measurement of AChE activity in 2-day-old ESE-treated embryos was carried essentially as described above except that inhibitor was not added to the extract.

Confocal microscopy and birefringence.
Transgenic embryos were mounted in Aqua Polymount (Polysciences, Warrington, PA) and were analyzed in lateral views with a confocal microscope (Leica, Bensheim, Germany). Optical sections were taken through the entire embryo. Stacks of images were used in 3D reconstruction to generate lateral and transverse projections. Image processing was carried out with the programs TCSTK and TIMT (Behra et al., 2002Go); further details are available from Jean-Luc Vonesch (jlv{at}igbmc.u-strasbg.fr).

Birefringence was analyzed as described previously (Behra et al., 2002Go). Briefly, embryos were embedded in methylcellulose (Westerfield, 1993Go), mounted under a dissecting microscope (Leica MZ FLIII), and illuminated with polarized light. The parallel alignment of normal muscle fibers reflects polarized light. The degree of reflection was translated into a color code to reveal the differences in intensities among wild-type, mutant, and inhibitor-treated embryos.

Microinjection.
Microinjection of 10-2 M GAL into the yolk of one-cell stage embryos was carried out with a pressure-driven microinjector (Eppendorf AG, Hamburg, Germany), as described previously (Westerfield, 1993Go). The injection volume was measured by injection into an oil droplet on an electron microscope grid and the volume of a zebrafish embryo was estimated to be approximately 1000 nl.

Application of penetration enhancers.
In experiments that were intended to test the function of penetration enhancers, GAL was applied alone or together with the penetration enhancers from the onset of gastrulation (50% epiboly; see Westerfield, 1993Go for staging). In pilot experiments, the range of concentration of each penetration enhancer, which did not show toxicity, was tested. Dimethylsulfoxide applied at 2% (v/v) did not reduce the effective concentration of GAL and caused toxic effects when administered alone at 5% (v/v). Caprate (0.005%), hexadecyltrimethylammonium bromide, and cetyltrimethylammonium bromide gave a very weak enhancement of the GAL effect or were ineffective.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Inhibitors of AChE Phenocopy the Motility Defect of the ache Mutation
To validate the use of zebrafish embryos as a model for testing the specificity of AChE inhibitors, we chose four compounds known to be inhibitors of AChE. These include the drugs ESE, GAL, TAC, and EDRO (Cousin et al., 1998Go, and references therein).

We first assessed the effect of the compounds on motility of the embryo and young larva. Wild-type embryos show the first signs of motility at 18 h postfertilization (hpf). This initial beating of the tail is spontaneous but becomes touch-evoked by about 36 hpf. Motility in response to touch has two phases: an immediate strong bend away from the source of stimulation, the so-called startle response, followed by a longer phase of swimming movements that varies in duration. Starting from 27 hpf, spasmodic contractions of the body axis could be observed in ache mutants, while wild-type embryos performed smooth, wave-like movements of the trunk and tail (Behra et al., 2002Go). By 48–72 hpf, motility of the mutants was strongly impaired (Fig. 1A and BGo).



View larger version (130K):
[in this window]
[in a new window]
 
FIG. 1. Comparison of touch-evoked motility. Videomicrographs showing the touch-evoked movements of (A) wild-type, (B) ache mutant, and (C) 10-3 M GAL-treated embryos at 48 hpf. The heads of the animals were embedded in agarose and embryos were touched with a blunt metal rod. (A) Wild-type embryos reacted with a strong bend away from the source of stimulation and maintained swimming movements for several seconds. (B and C, middle panels) In contrast, ache mutant and GAL-treated embryos responded with a single bend of the tail away from the source of stimulation and returned slowly back to the starting position without swimming movements over the next 200 ms. (A, bottom) Wild-type embryos continued swimming for several seconds, while (B, bottom) mutants or (C, bottom) GAL-treated embryos were motionless. The individual panels represent overlays of selected video frames. The time of each sequence is indicated in the right bottom corner in ms. The middle and bottom panels of B and C represent 200 ms periods to document the slow straightening of the tail (middle panels) or the motionless state (bottom panels). In contrast, the middle and bottom panels of A represent 40 ms periods.

 
We determined the concentration of each of the four compounds at the point at which it affected motility in a manner similar to the ache mutation (Fig. 1Go and Table 1Go). Whole embryos were exposed to the compounds from the 5-somite stage onwards, the time at which ache starts to be expressed (Bertrand et al., 2001Go). Embryos bathed in 10-4 M ESE, 10-2 M EDRO, 10-3 M GAL, or 10-5 M TAC showed reduced motility at 27 and 48 hpf in a manner similar to ache mutants (Figs. 1CGo, 2Go and Table 1Go). The embryos under all treatment conditions had beating hearts, indicating that impaired motility was not due to a lethal effect of the compounds. Increasing the concentration by 10-fold did not elicit a stronger inhibition of motility and caused unspecific effects in the case of EDRO and TAC such as necrosis in the brain (data not shown).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Dose effect of AChE inhibitors on motility of 27h-old embryos
 


View larger version (8K):
[in this window]
[in a new window]
 
FIG. 2. Duration of swimming movements was reduced by the application of AChE inhibitors. Wild-type embryos were treated with 10-4 M ESE, 10-3 M GAL, 10-2 M EDRO, and 10-5 M TAC and the duration of the swimming response (seconds) was measured after a tactile stimulus to the body flank. Embryos were exposed to the compounds from the 5-somite stage onwards. Movement was recorded by video microscopy at 48 hpf and the duration of the swimming response was determined. The results were averaged from 48 to 70 embryos for each condition and standard deviations are indicated.

 
To quantify the impairment of motility, we measured the duration of swimming movements after touching wild-type, ache mutant, or inhibitor-treated embryos (Fig. 2Go and Table 1Go). Wild-type embryos carried out swimming movements for an average of 2.8 s after being stimulated. In contrast, ache mutants showed on average only a very short phase of swimming. Both GAL and ESE reduced the average duration of swimming by 95% (Fig. 2Go), while TAC and EDRO were slightly less effective in reducing the average swimming episode (86 and 70%, respectively; Fig. 2Go and Table 2Go). The data sets were pairwise subjected to the Student’s t-test. In comparison to that of wild-type embryos, the duration of swimming of ache mutant and inhibitor-treated embryos was significantly reduced (p < 0.05, Table 2Go). The difference between ache mutant and ESE-/GAL-treated embryos is not significant (p > 0.05, Table 2Go). The lower efficiency of TAC and EDRO in reducing the duration of swimming is, however, significant (p < 0.05, Table 2Go). Although ache mutant and inhibitor-treated embryos showed strong reductions in swimming duration, they responded in many instances with one fast bend of the body axis. Moreover, the inhibitor-treated embryos, like ache mutants, had beating hearts, demonstrating that impaired motility was not due to a general toxic effect of the drugs (Table 2Go). Thus, zebrafish embryos treated with inhibitors of AChE have impaired motility similar to that of embryos that carry a loss-of-function mutation in the ache gene.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Reduction of duration of swimming relative to control/wild type (CO/WT) in ache mutant, 10-4 M ESE–, 10-3 M GAL–, 10-5 M TAC–, and 10-2 M EDRO–treated embryos
 
Inhibitors Differ in Their Ability to Induce Muscle Degeneration
Lack of AChE activity caused a progressive myopathy in homozygous ache mutant embryos. Mutant muscle fibers formed vacuoles and myofibrils were not aligned in parallel as in wild-type embryos (Behra et al., 2002Go). As a consequence, the reflection of polarized light (or birefringence) from the flank of mutant embryos was severely reduced at 48–72 hpf (compare Fig. 3A and BGo). We tested whether the AChE inhibitors would cause similar muscle defects. Embryos were exposed to the inhibitors from the 5-somite stage and were analyzed under polarized light at 72 hpf. GAL caused severe muscle defects in wild-type embryos (Fig. 3DGo) that were not increased by absence of the ache gene (Fig. 3EGo). Development of a myopathy in ache mutants required the nicotinic acetylcholine receptor (nAChR), since mutations in the {alpha}1 subunit of nAChR (nic1) suppressed the myopathy of ache mutants (Fig. 3J–LGo and Behra et al., 2002Go, for details). In contrast to wild-type embryos, GAL-treated, nic1 mutant embryos that lack functional nAChR (Sepich et al., 1998Go) did not develop a myopathy (Fig. 3FGo compared with C and D). Thus, the GAL-induced muscle degeneration depended on a functional nAChR similar to the myopathy caused by a lack of ache gene activity. In comparison to GAL (Fig. 3DGo), EDRO and TAC induced a weaker reduction in birefringence (data not shown) in agreement with their slightly weaker effect on motility (Fig. 2Go and Table 2Go).



View larger version (85K):
[in this window]
[in a new window]
 
FIG. 3. GAL but not ESE caused a myopathy similar to that observed in ache mutants. (A, D, and G) Wild-type, (B, E, and H) ache mutant, and (C, F, and I) nAChR (nic1) mutant embryos (72 hpf) were either (A to C) untreated or treated with (D to F) 10-3 M GAL or (G to I) 10-4 M ESE. Lack of ache caused defects in the axial musculature, which are visualized by illuminating embryos with polarized light. (A, C, F, and G to I) Strong reflection of polarized light, translated to red/yellow colors by the spectral representation, indicated a normal parallel alignment of myofibrils. (B, represented by a blue/green color code) In contrast, myofibrillar arrangement was disrupted in ache mutants. This defect was phenocopied by treatment of wild-type embryos with (D) GAL but not with (G) ESE. (F) The GAL-induced myopathy was suppressed by lack of a functional nAChR (nic1). (H) Application of ESE to ache mutant embryos suppressed the myopathy of ache mutants. The figure provides a view onto the axial musculature at the hindgut extension; dorsal is up and anterior is to the left. The scale bar represents 50 µm. The rainbow scale indicates the quantitative differences in reflected polarized light represented by the color code. (J to L) Schematic summary of the epistatic relationships of AChE and AChR (J) in wild-type embryos, (K) in the development of a myopathy in ache mutants, and (L) its suppression in ache;nic1 double mutants. ACh released from the nerve terminal accumulated in the neuromuscular cleft of ache mutants and led to overactivation of the muscle by interaction with the nAChR on the muscle membrane. As a consequence, the myofibers were damaged. Overactivation of the muscle was prevented in ache;nic1 double mutants in which the nAChR was missing to transmit the effect to the muscle cells. As a consequence, the myopathy did not develop. (H) The phenotype of the double mutant is chemocopied by treatment of ache mutants with ESE.

 
Curiously, ESE, even though it impaired motility strongly (Fig. 2Go), did not cause a significant reduction in birefringence (Fig. 3GGo). ESE may not inhibit AChE at all in the zebrafish or may have additional activities. To exclude the first possibility, we performed histochemical and spectrophotometric measurements of AChE activity in extracts or intact embryos treated with ESE. Embryos exposed to 10-5 ESE showed a reduction of AChE histochemical activity, and exposure to 10-4 M ESE almost completely abolished the histochemical reaction (Figs. 4A–4DGo), comparable with that seen in the ache mutant (Fig. 4DGo). Similarly, when measured by the colorimetric Elman assay, the AChE activity was reduced by 72 and 91% in extracts from embryos treated with 10-5 and 10-4 M ESE, respectively. Hence, ESE was able to inhibit zebrafish AChE efficiently. The lack of a myopathy is thus likely due to a secondary target effect of ESE. The lack of a myopathy in embryos that lack a functional nAChR suggests that the nAChR may be a possible secondary target of ESE. If this notion is correct, the myopathy of ache mutant embryos should be suppressed by treatment with ESE. Indeed, treatment of ache mutants with ESE almost completely abolished the myopathy (Fig. 3HGo) observed in untreated ache mutants (Fig. 3BGo). These results show that ESE has a secondary effect, most likely caused by an antagonistic interaction with the nAChR. In vitro studies suggested that ESE can act as an antagonist of the Torpedo nAChR (Kawai et al., 1999Go).



View larger version (145K):
[in this window]
[in a new window]
 
FIG. 4. In situ determination of AChE activity in embryos treated with ESE. Determination of AChE activity in (A) untreated; (B) 10-5 M ESE-treated, wild-type embryos; (C) 10-4 M ESE-treated, wild-type embryos; and (D) ache mutant embryos. The figure provides a view onto the axial musculature at the hindgut extension; dorsal is up and anterior is to the left. The scale bar represents 30 µm.

 
Secondary Effects of Inhibitors in the Nervous System
To assess a possible effect of the AChE inhibitors on the developing nervous system, we used a transgenic line that expresses the green fluorescent protein (GFP) gene under control of the regulatory sequences of the zebrafish neurogenin1 gene (-3.1ngn1:gfp; Blader et al., 2003Go). The neurogenin1 gene is a neural determination gene that controls differentiation of Rohon Beard sensory neurons involved in the perception of touch (Korzh and Strahle, 2002Go). The transgene was strongly expressed in Rohon Beard neurons (Fig. 5AGo). Also, a few scattered interneurons located immediately below the dorsal longitudinal fascicle (dlf) were marked by weak GFP expression (Fig. 5AGo). The overall pattern of GFP expression was unchanged in ache-/- embryos (n = 20) at 27 hpf (Fig. 5BGo), suggesting that a lack of ache activity did not affect the expression of the transgene and the pattern of neurogenesis at this stage. When embryos were exposed to either EDRO (Fig. 5CGo) or TAC (Fig. 5DGo) from the 5-somite stage to 27 hpf, the pattern of GFP-expressing neurons was abnormal (69% treated embryos [n = 19] and 70% treated embryos [n = 37], respectively). Ectopic, GFP-expressing neurons were noted at the ventral aspects of the neural tube in the drug-treated embryos, suggesting that the two drugs caused aberrant neurogenesis or cell migration (Figs. 5CGo and 5DGo). An abnormal pattern of neurogenesis was not observed in the ache mutants (Fig. 5BGo, n = 20). This effect was not the result of inhibition of AChE activity and likely reflects a secondary target effect.



View larger version (97K):
[in this window]
[in a new window]
 
FIG. 5. Edrophonium and tacrine altered the pattern of neurogenesis in the spinal cord. (A) Wild-type, (B) ache mutant, (C) 10-2 M EDRO-, (D) 10-5 M TAC-, (E) 10-3 M GAL-, and (F) 10-4 M ESE-treated embryos carrying the 3.1ngn1:gfp transgene. This transgene expressed GFP in Rohon-Beard sensory neurons (rb) and weakly in some interneurons (i) of the dorsal spinal cord. It also marked the dorsal longitudinal fascicles (dlf) at the dorsolateral aspect of the neural tube. The pattern of GFP expression was indistinguishable in wild-type and ache mutant embryos. In contrast, (C and D, arrows) both EDRO- and TAC-treated embryos showed ectopic GFP expression in ventral neural tube regions. Arrows in E and F show examples of cells that are only slightly ventrally displaced in GAL- and ESE-treated embryos. These deviations were a variation of the normal body plan, as they occured with a similar frequency in wild-type embryos also. Embryos are 27 h old and are shown in lateral views; dorsal is up and anterior is to the left. The scale bar represents 15 µm.

 
These drastic changes in neuronal pattern were also not observed in embryos treated with ESE or GAL (Figs. 5FGo and 5EGo; n = 35 and n = 33, respectively). Occasionally, ESE- or GAL-treated embryos that had slightly displaced cells were noted (Fig. 5F and EGo, arrows; 23 and 21%, respectively). However, these cells were also observed in untreated embryos at a similar frequency.

The Inhibitors Block Zebrafish AChE Effectively at Micro- to Nanomolar Ranges
When assessed by their dose-dependent effect on motility (Fig. 1Go and Table 1Go) or by in situ analysis of AChE activity in the case of ESE (Fig. 4B and CGo), the effective blocking concentration of each of the four inhibitors was high in comparison to reported IC50 values of AChEs from other species (Ariel et al., 1998Go; Harvey 1995Go; Shafferman et al., 1992Go; Stein and Lewis, 1969Go). Three reasons could account for this: (1) the inhibitors have lower affinities to the AChE of zebrafish than to that of other vertebrates; (2) the zebrafish embryo may take up these chemicals from the environment rather inefficiently, hence requiring higher doses of inhibitor; and/or (3) the compounds are metabolized in the embryo.

To assess the first possibility, the IC50 of the four inhibitors was determined in extracts of zebrafish larvae. The extracts were preincubated with varying concentrations of the inhibitors before AChE activity was measured spectrophotometrically (Ellman et al., 1961Go). ESE and TAC inhibited zebrafish AChE half-maximally at 125 ± 15 and 114 ± 23 nM, respectively (Fig. 6Go). The respective IC50 values of GAL and EDRO were 2800 ± 21 µM and 1300 ± 100 nM. These IC50 values are in the range of those reported for AChEs of other species (Ariel et al., 1998Go; Harvey, 1995Go; Shafferman et al., 1992Go; Stein and Lewis, 1969Go). Thus, the zebrafish AChE is as sensitive to the inhibitors as the AChE of other vertebrates.



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 6. Efficient inhibition of zebrafish AChE by GAL, ESE, TAC, and EDRO in vitro. Extracts from 4-day-old embryos were incubated with increasing concentrations of ESE, GAL, EDRO, and TAC, and AChE activity was measured using the Ellman assay. IC50 values were in the nano- to micromolar range.

 
To exclude the possibility that the drugs are metabolized in the embryo, GAL was injected into the yolk cell of one-cell stage embryos. We injected approximately 1 nl of 10-2 M GAL per embryo resulting in an intraembryonic concentration of about 10-5 M, assuming free diffusion within the embryo and no reduction due to growth of the embryo and diffusion out of the embryo until the 72 h stage. All injected embryos (n = 50) showed impaired motility and reduced reflection of polarized light at 72 hpf in comparison to mock-injected controls (Fig. 7A and BGo). Assuming no losses due to dilution or diffusion between the injection and examination of the effect 72 h later, the estimated intraembryonic concentration was at least 10-fold lower than that required in bath application to "chemocopy" the ache mutant phenotype. Thus, intraembryonic degradation of the drugs is unlikely to account for the relative insensitivity of whole embryos.



View larger version (113K):
[in this window]
[in a new window]
 
FIG. 7. Microinjection of GAL or coapplication of a penetration enhancer reduced the concentration required to "chemocopy" the myopathy of ache mutants. (A and B) Birefringence of (A) uninjected embryos or (B) embryos injected with GAL (10-2 M) giving an approximate intraembryonic concentration of 10-5 M GAL. (C to H) Birefringence of embryos treated with (C and D) 10-4, (E and F) 10-5, and (G and H) 10-6 M GAL in the (C, E, and G) absence or (D, F, and H) presence of 0.05% (v/v) Tween 80 (Tw80). The penetration enhancer Tween 80 reduced the GAL concentration that was required to induce a myopathy by 10- to 100-fold. Embryos are oriented anterior left and dorsal up. Views onto the trunk at the level of the yolk extension are shown. Color coding of the reflection of polarized light is as described in Figure 3Go.

 
Taken together, these results suggest that GAL penetrates the zebrafish embryo ineffectively. As microinjection is not the most convenient way to deliver compounds to a large number of embryos, we next assessed whether the drug effect could be improved by coapplying penetration enhancers. Dimethylsulfoxide, caprate, and cetyltrimethylammonium bromide applied at various concentrations (see Experimental Protocol) did not increase the sensitivity to GAL efficiently (data not shown). However, when GAL was applied together with 0.05 or 0.01% (v/v) Tween 80, a significant effect could be noted. While 10-5 or 10-6 M GAL administered alone was ineffective (100% embryos; n = 30; Fig. 7E and GGo), the same concentrations of GAL coapplied with 0.05% Tween 80 impaired motility and reduced the birefringence (60% embryos; n = 30; Fig. 7F and HGo). Administration of Tween 80 (0.01 or 0.05%) alone did not affect development or viability in any detectable way (data not shown). Thus, impaired penetration is a likely explanation for the relative insensitivity of whole embryos, and the barrier can be overcome by applying the penetration enhancer Tween 80.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We assessed the potential of wild-type and mutant zebrafish embryos as models for determining the specificity of inhibitors of AChE. Using this class of inhibitors, our work was also aimed at evaluating, in general, the use of wild-type and mutant zebrafish embryos in testing the toxic or teratological effect of chemicals. Among the four compounds tested, only GAL "chemocopied" the ache mutant phenotype. Like genetic loss of ache function, GAL treatment impaired motility, caused a myopathy, but did not affect the differentiation of neurons in the spinal cord. Moreover, the myopathy in ache mutants (Behra et al., 2002Go) and GAL-treated embryos (this study) both required an intact nAChR. The other inhibitors had additional effects not seen in ache mutants. Both TAC and EDRO caused impaired neuronal development that was not detected in ache mutants. ESE even cured the muscle defects of the ache mutants. As a similar suppression of the muscle phenotype of ache mutants was observed when the AChR was genetically removed (Behra et al., 2002Go), this suggests that ESE acted as an antagonist of the AChR. This notion is in agreement with a previous in vitro study (Kawai et al., 1999Go) in which it was shown that ESE acted as an antagonist of the nAChR. In summary, our work demonstrates the power of wild-type and mutant zebrafish embryos to assess the target specificity of AChE inhibitors.

We noticed a drawback in that the concentrations required to "chemocopy" the ache mutant phenotype in intact embryos were higher in comparison to the reported in vitro inhibitory concentration for AChEs from other species (Ariel et al., 1998Go; Harvey, 1995Go; Shafferman et al., 1992Go; Stein and Lewis, 1969Go). This does not appear to be due to a low affinity of the inhibitors to zebrafish AChE or degradation of the compounds in the embryos. A likely explanation is the inefficient uptake of the compounds. While this relative insensitivity of zebrafish embryos may be a hindrance in primary screening programs of compound libraries where the available amounts of chemical are rather limited, it may not present an obstacle for second-phase toxicological assessment of candidates or the analysis of the toxicological mechanism. Interestingly, not all penetration enhancers were equally effective, suggesting that the type of penetration enhancer may have to be taken into consideration for each compound class to be tested.

The effective in vivo dose to impair motility appears to be linked to the chemical structure of the AChE inhibitors. EDRO was the least efficient (10-2 M) at inhibiting motility in bath applications even though its in vitro IC50 is in the micromolar range. ESE "chemocopied" the ache mutant phenotype less effectively than TAC, but both compounds inhibited the AChE activity with an IC50 around 100 nM. Other compounds were previously applied to zebrafish embryos at micro- and nanomolar concentrations and caused significant effects (Peterson et al., 2000Go, 2001Go). We show that, at least in the case of GAL, direct injection of the compound or the coadministration of the penetration enhancer Tween 80 can reduce the effective dose in the embryos, suggesting that the uptake of the AChE inhibitors may represent a barrier for the AChE inhibitors.

Our work illustrates the potential of zebrafish in the assessment of toxicological and teratological effects of compounds; zebrafish mutants and mutant phenotypes act as blueprints of drug effect for assessing target specificity and gaining insight into the toxic mechanism in a whole organism setup. Ideally, if an inhibitor is specific it should reproduce the phenotype of a loss-of-function mutation faithfully with no additional effects. However, reality can be more complex than this theoretical concept. Complications may be imposed by the complexity of the responses of intact organisms. Firstly, the phenotypes of a loss-of-function mutation may not reflect the effect of an inhibitor that blocks its target only partially, as submaximal inhibition may have different phenotypic consequences. Secondly, detoxifying pathways may break down a chemical to metabolites that act unspecifically, even though the administered inhibitor is perfectly specific for its target. Another complication may arise from the fact that genetically altered organisms can have varying phenotypes depending on the specific lesion and the genetic background. We have not observed variation of the ache mutant phenotypes used here, even when crossed into distantly related strains such as the WIK mapping strain (M.B. and U.S., unpublished). However, this does not mean that this may not present a problem in other genetic backgrounds and for other genetic lesions in the zebrafish.

Despite these caveats, we believe that, with the increasing number of cloned mutant loci, the zebrafish will provide a simple, cost-effective platform to assess the effects of chemicals on many developmental and physiological processes. The potential of the zebrafish is even further increased by the steadily growing number of transgenic lines that permit monitoring of many physiological processes in the living animal. The small size, transparency, and extrauterine development of the zebrafish embryo allow the application of simple assays, which can easily be modified to satisfy medium- and possibly also high-throughput applications. In summary, even though the concentrations required to achieve an effect may be relatively high for certain chemicals, the zebrafish embryo can provide a valuable, whole organism assay to satisfy the demand for efficient toxicological test systems by regulators and industry.


    ACKNOWLEDGMENTS
 
We are grateful to N. Fischer, D. Biellmann, C. Vialle, J. L. Vonesch, D. Hentsch, and M. Boeglin for technical assistance, artwork, and animal care. We thank S. Rastegar and C. S. Lam for comments and discussion and D. Dembele for help with the statistical analysis. U.S. was supported by CNRS/INSERM/ULP/HUS, AFM, ARC, Volkswagen Stiftung, and AICR.


    NOTES
 
1 Authors contributed equally to this work. Back

2 To whom correspondence should be addressed at Institut de Genetique et de Biologie Moleculaire et Cellulaire, 1 rue Laurent Fries, BP10142, 67404 Illkirch, France. Fax: +33 3 88 65 32 01. E-mail: uwe{at}igbmc.u-strasbg.fr. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Ariel, N., Ordentlich, A., Barak, D., Bino, T., Velan, B., and Shafferman, A. (1998). The "aromatic patch" of three proximal residues in the human acetylcholinesterase active centre allows for versatile interaction modes with inhibitors. Biochem. J. 335(Pt 1), 95–102.[ISI][Medline]

Balls, M. (2002). Alternatives to animal experimentation yesterday, today, and tomorrow: Some thoughts on leaving ECVAM. Altex 19, 225–227.[ISI][Medline]

Barut, B. A., and Zon, L. I. (2000). Realizing the potential of zebra fish as a model for human disease. Physiol. Genomics 2, 49–51.[Abstract/Free Full Text]

Behra, M., Cousin, X., Bertrand, C., Vonesch, J. L., Biellmann, D., Chatonnet, A., and Strahle, U. (2002). Acetylcholinesterase is required for neuronal and muscular development in the zebra fish embryo. Nat. Neurosci. 5, 111–118.[CrossRef][ISI][Medline]

Bertrand, C., Chatonnet, A., Takke, C., Yan, Y. L., Postlethwait, J., Toutant, J. P., and Cousin, X. (2001). Zebra fish acetylcholinesterase is encoded by a single gene localized on linkage group 7. Gene structure and polymorphism: Molecular forms and expression pattern during development. J. Biol. Chem. 276, 464–474.[Abstract/Free Full Text]

Blader, P., Plessy, C., and Strahle, U. (2003). Multiple regulatory elements with spatially and temporally distinct activities control neurogenin1 expression in primary neurons of the zebra fish embryo. Mech. Dev. 120, 211–218.[CrossRef][ISI][Medline]

Cousin, X., Hotelier, T., Giles, K., Toutant, J. P., and Chatonnet, A. (1998). aCHEdb: the database system for ESTHER, the {alpha}/ß fold family of proteins and the cholinesterase gene server. Nucleic Acids Res. 26, 226–228.[Abstract/Free Full Text]

Driever, W., Solnica-Krezel, L., Schier, A. F., Neuhauss, S. C., Malicki, J., Stemple, D. L., Stainier, D. Y., Zwartkruis, F., Abdelilah, S., Rangini, Z., et al. (1996). A genetic screen for mutations affecting embryogenesis in zebra fish. Development 123, 37–46.[Abstract/Free Full Text]

Ellman, G. L., Courtney, K. D., Andres, V., and Featherstone, R. M. (1961). A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7, 88–95.[CrossRef][ISI][Medline]

Giacobini, E. (1998). Invited review. Cholinesterase inhibitors for Alzheimer’s disease therapy: From tacrine to future applications. Neurochem. Int. 32, 413–419.[CrossRef][ISI][Medline]

Harvey, A. L. (1995). The pharmacology of galanthamine and its analogues. Pharmacol. Ther. 68, 113–128.[CrossRef][ISI][Medline]

Haffter, P., Granato, M., Brand, M., Mullins, M. C., Hammerschmidt, M., Kane, D. A., Odenthal, J., van Eeden, F. J., Jiang, Y. J., Heisenberg, C. P., Kelsh, R. N., Furutani-Seiki, M., Vogelsang, E., Beuchle, D., Schach, U., Fabian, C., and Nusslein-Volhard, C. (1996). The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio. Development 123, 1–36.

Karnovsky, M. J., and Roots, L. (1964). A direct colouring thiocholine method for cholinesterases. J. Histochem. Cytochem. 12, 219–221.[ISI][Medline]

Kawai, H., Carlson, B. J., Okita, D. K., and Raftery, M. A. (1999). Eserine and other tertiary amine interactions with Torpedo acetylcholine receptor postsynaptic membrane vesicles. Biochemistry 38, 134–141.[CrossRef][ISI][Medline]

Korzh, V., and Strahle, U. (2002). Proneural, prosensory, antiglial: The many faces of neurogenins. Trends Neurosci. 25, 603–605.[CrossRef][ISI][Medline]

Langheinrich, U., Hennen, E., Stott, G., and Vacun, G. (2002). Zebra fish as a model organism for the identification and characterization of drugs and genes affecting p53 signaling. Curr. Biol. 12, 2023–2028.[CrossRef][ISI][Medline]

Marton, M. J., DeRisi, J. L., Bennett, H. A., Iyer, V. R., Meyer, M. R., Roberts, C. J., Stoughton, R., Burchard, J., Slade, D., Dai, H., et al. (1998). Drug target validation and identification of secondary drug target effects using DNA microarrays. Nat. Med. 4, 1293–1301.[CrossRef][ISI][Medline]

Norin, M., and Sundstrom, M. (2001). Protein models in drug discovery. Curr. Opin. Drug Discov. Dev. 4, 284–290.[Medline]

Peterson, R. T., Link, B. A., Dowling, J. E., and Schreiber, S. L. (2000). Small molecule developmental screens reveal the logic and timing of vertebrate development. Proc. Natl. Acad. Sci. U.S.A. 97, 12965–12969.[Abstract/Free Full Text]

Peterson, R. T., Mably, J. D., Chen, J. N., and Fishman, M. C. (2001). Convergence of distinct pathways to heart patterning revealed by the small molecule concentramide and the mutation heart-and-soul. Curr. Biol. 11, 1481–1491.[CrossRef][ISI][Medline]

Sepich, D. S., Wegner, J., O’Shea, S., and Westerfield, M. (1998). An altered intron inhibits synthesis of the acetylcholine receptor {alpha}-subunit in the paralyzed zebra fish mutant nic1. Genetics 148, 361–372.[Abstract/Free Full Text]

Shafferman, A., Velan, B., Ordentlich, A., Kronman, C., Grosfeld, H., Leitner, M., Flashner, Y., Cohen, S., Barak, D., and Ariel, N. (1992). Substrate inhibition of acetylcholinesterase: Residues affecting signal transduction from the surface to the catalytic center. EMBO J. 11, 3561–3568.[Abstract]

Soreq, H., and Seidman, S. (2001). Acetylcholinesterase—new roles for an old actor. Nat. Rev. Neurosci. 2, 294–302.[CrossRef][ISI][Medline]

Stein, H. H., and Lewis, G. J. (1969). Noncompetitive inhibition of acetylcholinesterase by eserine. Biochem. Pharmacol. 18, 1679–1684.[CrossRef][ISI][Medline]

Stern, H. M., and Zon, L. I. (2003). Opinion: Cancer genetics and drug discovery in the zebra fish. Nat. Rev. Cancer 3, 533–539.[CrossRef][ISI][Medline]

van Kessel, W., Brocades Zaalberg, R., and Seinen, W. (1989). Testing enironmental pollutants on soil organisms: A simple assay to investigate the toxicity of environmental pollutants on soil organisms, using CDCl2 and nematodes. Ecotoxicol. Environ. Saf. 18, 181–190.[ISI][Medline]

Westerfield, M. (1993). The Zebra Fish Book. University of Oregon Press, Eugene, Oregon.





This Article
Abstract
FREE Full Text (PDF)
All Versions of this Article:
77/2/325    most recent
kfh020v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (3)
Disclaimer
Request Permissions
Google Scholar
Articles by Behra, M.
Articles by Strähle, U.
PubMed
PubMed Citation
Articles by Behra, M.
Articles by Strähle, U.