Ethyl 3-Aminobenzoate – DMF inhibitor

Adverse effects of serotonin depletion in developing zebrafish

Mark J. Airhart a, Deborah H. Lee a,⁎, Tracy D. Wilson b, Barney E. Miller a, Merry N. Miller b,
Richard G. Skalko a, Paul J. Monaco a
a Department of Anatomy and Cell Biology, Quillen College of Medicine, East Tennessee State University, Johnson City, TN 37614-1708, USA
b Department of Psychiatry, Quillen College of Medicine, East Tennessee State University, Johnson City, TN 37614-1708, USA

Abstract

In this study, p-chlorophenylalanine (pCPA), an inhibitor of tryptophan hydroxylase (the rate limiting en- zyme of serotonin synthesis), was used to reduce serotonin (5HT) levels during early development in zebra- fish embryos. One day old dechorionated embryos were treated with 25 μM pCPA for 24 h and subsequently rescued. Immunohistological studies using a 5HT antibody confirmed that 5HT neurons in the brain and spi- nal cord were depleted of transmitter by 2 days post fertilization (dpf). Twenty four hours after pCPA expo- sure embryos were unable to burst swim and were nearly paralyzed. Movement began to improve at 4 dpf, and by 7 dpf, larvae exhibited swimming activity. Rescued larvae continued to grow in rostrocaudal length over 5 days post-rescue, but their length was always 16–21% below controls. Surprisingly, both groups dis- played the same number of myotomes. To examine whether hypertonicity of myotomes in treated embryos played a role in their shorter rostrocaudal lengths, 1 dpf embryos were exposed to a combination of 25 μM pCPA and 0.6 mM of the sodium channel blocker ethyl 3-aminobenzoate methanesulfonate (MS-222). After a 24 hour exposure, the embryos exhibited the same rostrocaudal length as control embryos suggesting that myotome hypertonicity plays a major role in the decreased axial length of the treated larvae. In addition, pCPA treated 2 dpf embryos exhibited abnormal notochordal morphology that persisted throughout recov- ery. Reverse transcriptase polymerase chain reaction (RT-PCR) was performed to determine the relative levels of the serotonin 1A receptor (5HT1A) transcript and the serotonin transporter (SERT) transcript in the brain and spinal cord of control and treated embryos. Transcripts were present in both brain and spinal cord as early as 1 dpf and reached maximal concentrations by 3 dpf. Embryos treated with pCPA demonstrated a decrease in the concentration of 5HT1A transcript in both brain and spinal cord. While SERT transcript levels remained unaffected in brain, they were decreased in spinal cord. Five days subsequent to pCPA rescue, 5HT1A transcript concentrations remained decreased in brain while SERT transcript levels were elevated in both re- gions. These findings suggest that reduction of 5HT during early zebrafish development may have an adverse ef- fect on body length, notochordal morphology, locomotor behavior, and serotonin message-related expression.

1. Introduction

Serotonin (5HT), a monoamine neurotransmitter involved in a di- verse range of behaviors and physiological processes such as REM sleep, temperature regulation, and pain perception (Chase and Murphy, 1973; Jessell and Kelly, 1991; Popa et al., 2008), has also been found to play a regulatory role in numerous developmental events (Whitaker- Azmitia et al., 1996). During human gestation, abnormal serotonin levels are implicated in cardiovascular development, pulmonary hyper- tension and neuromuscular coordination (Chambers et al., 2006; Koren and Boucher, 2009). It has also been suggested that perturbation of se- rotonin during brain development may cause abnormal behaviors including depression, anxiety disorders, and autism (Lisboa et al., 2007; Lucki, 1998; Walsh et al., 2008). There is extensive literature ex- amining serotonin’s role in neuron proliferation, migration and differ- entiation, receptor ontogeny, terminal growth and synaptogenesis (Di Pino et al., 2004; Khozhai and Otellin, 2006; Lauder et al., 2000; Pronina et al., 2003; Shuey et al., 1993; van Kesteren and Spencer, 2003; Vitalis et al., 2007).

Studies using the tryptophan hydroxylase inhibitor, p-chloropheny- lalanine (pCPA) to deplete serotonin levels have demonstrated the importance of this neurotransmitter in the organization of locomotor pattern and modulation of motor output in the developing mammalian lumbar spinal cord (Nakajima et al., 1998; Pearlstein et al., 2005; Pﬂieger et al., 2002; Vinay et al., 2005). pCPA potently binds to tryptophan hydroxylase, the rate limiting enzyme in the serotonin synthesis pathway, and prevents the hydroxylation of tryptophan to 5-hydroxytryptophan, the immediate precursor of serotonin, resulting in the decrease of endogenous levels of serotonin in mammalian brain (Jéquier et al., 1967; Koe and Weissman, 1966). A study using pCPA to deplete serotonin over prenatal and early postnatal times in rats has shown retardation in raphe spinal axonal growth and terminal density in pCPA treated pups through postnatal day 10. This correlated with retardation in forelimb and hindlimb coordination for both swimming and walking, respectively. Recovery of motor activity did occur in pCPA treated rats by postnatal day 14. This implied that subse- quent to pCPA exposure, adequate transmitter levels returned in raphe neurons and re-established terminal growth and related function (Nakajima et al., 1998).

We have investigated the developing locomotor network in zebrafish to further elucidate the ontogeny and function of the serotonergic path- way. Swimming development in zebrafish has been well defined (Airhart et al., 2007; Brustein et al., 2003; Downes and Granato, 2006; Drapeau et al., 2002; Saint-Amant and Drapeau, 1998). Brieﬂy, the stages of swim- ming development include spontaneous coiling at approximately 17 h post fertilization (hpf), evoked coiling (21 hpf), burst swimming (27 hpf), and beat and glide swimming with swim bladder inﬂation at 4–5 days post fertilization (dpf). This last stage of swimming develop- ment is also referred to as spontaneous swimming activity which is char- acteristic of mature swimming. Recent studies have demonstrated a temporal correlation between the ontogeny of serotonin circuitry in the zebrafish spinal cord and the development of locomotor stages leading to mature beat and glide swimming (Drapeau et al., 2002; Saint-Amant and Drapeau, 1998). The development of the zebrafish serotonergic spi- nal cord circuitry consists of two well-defined pathways: (1) the raphe- spinal projection which develops within 3–4 dpf, and (2) a separate pop- ulation of serotonergic neurons in the spinal cord which are present as early as 1.0–1.5 dpf with growth cones observed by 2 dpf (Lillesaar et al., 2009; McLean and Fetcho, 2004a; Sallinen et al., 2009). This devel- opmental timing of serotonergic innervation closely correlates with the onset of mature swimming.

Several studies have demonstrated that serotonin perturbation affects swimming behavior in zebrafish larvae (Airhart et al., 2007; Brustein et al., 2003; Njagi et al., 2010). For example, an injection of serotonin (100 μM) or the moderately selective serotonin agonist, quipazine (100 μM), into the pericardial sac of agarose-embedded zebrafish embryos and larvae only affected fictive spontaneous swim- ming activity in 4 dpf animals. Neither compound had any effect on younger animals (Brustein et al., 2003). Work from our laboratory using the serotonin reuptake inhibitor, ﬂuoxetine (PROZAC), to in- crease extracellular serotonin levels in zebrafish embryos and larvae confirmed the sensitive stage as coinciding with the final stage of swimming development, i.e., 4–5 dpf. A 24 hour bath exposure of ﬂu- oxetine (4.6 μM) to 4 or 5 dpf larvae resulted in a brief (2 h) episode of hyper-swimming activity followed 24 h later by a sustained de- crease in spontaneous swimming activity that lasted up to 14 days (Airhart et al., 2007).

The present study expands on our preliminary findings which showed that a 24 hour bath exposure of 25 μM pCPA at 1 dpf in zebra- fish embryos resulted in a nearly complete absence of burst swimming and a significant decrease in axial length at 2 dpf (Airhart et al., 2006). This shortening effect appeared to be due to simultaneous contractions of both sides of the axial musculature, resulting in shortening of the trunk with apparent notochordal damage. The notochord is required for the proper patterning of adjacent tissues in all developing verte- brates, serves as the major skeletal element for lower chordates, and provides an important element for locomotion in fish (Stemple et al., 1996). Somites, mesodermal segments on either side of the notochord, are formed when the paraxial mesoderm is divided into distinct blocks of tissue that develop sequentially in an anterior to posterior direction along the midline of the embryo with each new somite forming on the caudal side of existing somites until a total of thirty are formed in zebrafish embryos. As they form, somites undergo a series of develop- mental transitions to generate embryonic structures known as myo- tomes, the precursors of trunk musculature. In zebrafish, unlike some other vertebrates, somites are not transient structures; the first somite forms the first definitive myotome, etc., and this segmental muscle ar- rangement persists throughout adulthood. Somite formation and sub- sequent myotome differentiation are critical for the development of a functional locomotor system in zebrafish (Stemple et al., 1996; Kimmel et al., 1995).
This study also investigated the expression of two serotonin asso- ciated mRNAs, i.e., the 5HT1A receptor transcript and the SERT tran- script after pCPA treatment in 1 dpf zebrafish embryos using reverse transcriptase polymerase chain reaction (RT-PCR). Both of these sero- tonin related transcripts have been shown to be present in the em- bryonic and larval zebrafish central nervous system (Airhart et al., 2007; Norton et al., 2008; Wang et al., 2006). Their proteins are wide- spread, they appear early in development, and are the targets for many serotonergic drugs (Patel and Zhou, 2005; Zhou et al., 2000).

2. Materials and methods

2.1. Animals

Wild type adult zebrafish (Danio rerio) were purchased from Car- olina Biological Supply Co. (Burlington, NC), and standard procedures were used for mating, egg collection, and staging (Westerfield, 2000). Embryos were collected and pooled from same day natural matings, and reared as previously described (Airhart et al., 2007). In this study, embryos refer to zebrafish prior to hatching (0–3 dpf), while larvae refer to post-hatch animals (over 3 dpf). Developmental stages and motility were evaluated under an Olympus SZ stereo microscope (Shinjuku-ku, Tokyo, Japan) and only normal embryos were utilized in experiments. All experiments were reviewed and approved by the University Committee on Animal Care at East Tennessee State University under protocol #P080507 and were in accordance with the guidelines from the “Guide for the Care and Use of Laboratory An- imals” from the Institute of Laboratory Animal Resources, Commis- sion on Life Sciences, National Research Council, published by the National Academy Press, Washington, DC, 1996.

2.2. Chemical exposures

• 4-Chloro-dl-phenylalanine methyl ester hydrochloride (pCPA), α- Methyl-DL-tyrosine methyl ester hydrochloride (AMPT), and ethyl 3-aminobenzoate methanesulfonate (MS-222) were purchased from Sigma Chemical Co., St. Louis, MO. Stock solutions of pCPA (10 mM), AMPT (10 mM) and MS-222 (20 mM) were prepared in conditioned Instant Ocean (Aquarium Systems, Inc., Mentor, OH). Conditioned Instant Ocean consists of deionized water with 0.2 g/l Instant Ocean, 0.075 g/l NaHCO3, 0.008 g/l CaSO4, and 10 μg/l meth- ylene blue to inhibit fungal growth. Treatment solutions were dilut- ed from stock solutions with conditioned Instant Ocean. Standard bath exposures of pCPA, AMPT, MS-222 or a combination of pCPA/MS-222 began at 1 dpf and extended until 2 dpf at which time embryos were rescued by carefully pipetting incubation medi- um off and replacing with fresh changes of conditioned Instant Ocean and maintained until 7 dpf. Controls were maintained in con- ditioned Instant Ocean. Chemical exposures were conducted in 50 mm glass petri dishes containing 15 ml of solution and embryos were maintained in a light-timed incubator (14/10 h light/dark cycle) at 28.5 °C. Experiments were performed in triplicate with 10–20 animals per petri dish for each group.

For lowest observed effect concentration (LOEC) pCPA experi- ments, dechorionated embryos were exposed to 10, 25, 35, 50, and 100 μM pCPA beginning at 10 hpf and extending as long as 7 dpf. In- cubation medium was replaced daily with freshly prepared solutions diluted from a 10 mM stock solution. After 5 days of development, larvae were fed a diet of TetraMin baby fish food (Tetrawerke, Melle, Germany) and Artemia nauplii (Carolina Biological Supply Co., Burlington, NC). The two highest pCPA concentrations, 50 and 100 μM, were lethal by 48 hpf. Embryos exposed to 25 and 35 μM pCPA demonstrated no obvious abnormal effects on heart rate, circu- lation, or evoked coiling after 6 h of exposure (data not shown). After 24 h, treated embryos continued to demonstrate no observable effect on heart rate or circulation; however, they were nearly paralyzed, and there was a dose-dependent effect on embryo length: embryos exposed to 35 μM pCPA were on average 2.4 ± 0.2 mm in length, while embryos exposed to 25 μM pCPA were 2.6 ± 0.2 mm in length (Fig. 1). Animals exposed to 10 μM pCPA were no different from con- trols through 7 dpf (data not shown). 25 μM pCPA was thus chosen as the LOEC.For pCPA specificity experiments, embryos were exposed to 10, 25, 50, 100 μM and 1 mM of AMPT at 1 dpf, rescued at 2 dpf into con- ditioned Instant Ocean and subsequently evaluated through 5 dpf. MS-222 (0.6 mM) was used as described from the literature (Stehr et al., 2006).

2.3. Morphological and behavioral assessments

Zebrafish embryos and larvae were examined with an Olympus SZ stereo microscope and an Olympus VANOX microscope (Shibuyaku, Tokyo, Japan).
Digital photographs were obtained with a Sony HQX DSC-F717 camera (Tokyo, Japan) using a MM99 universal microscope adapter (Martin Microscope Co., Easley, SC). Zebrafish were lightly anesthe- tized with 0.3 mM MS-222 for imaging as necessary. For morpholog- ical and behavioral assessments, control and treated embryos and larvae from each developmental stage were pooled from same day natural matings of approximately 90 fish (equal sex ratio). Tested an- imals were removed from the samples to prevent recounting. Myotomes were counted and rostrocaudal lengths of embryos (1– 3 dpf) and larvae (4–7 dpf) were measured using an American Optical (Buffalo, NY) filar micrometer 10× eyepiece fitted on an Olympus SZ stereo microscope (total magnification 100×). Older larvae (N 4 dpf) were lightly anesthetized with 0.3 mM MS-222 in conditioned Instant Ocean to prevent movement during measurement. Standard body length measurements to the nearest tenth of a millimeter were taken from the most anterior mouth region to the juncture of the body and caudal fin. Representative myotome measurements from the head (myotome 5), trunk (myotome 15), and tail (myotome 23) regions were taken, measuring from the beginning to the end of each myotome, and expressed to the nearest micrometer.

Fig. 1. Rostrocaudal length in zebrafish embryos after pCPA exposure. (A) Lateral image of a control 2 dpf zebrafish embryo. (B) Lateral image of a 2 dpf embryo exposed to 25 μM pCPA from 1 to 2 dpf. (C) Lateral image of a 2 dpf embryo exposed to 35 μM pCPA from 1 to 2 dpf. The rostrocaudal length of the head regions are indicated by arrowed lines. Solid vertical lines represent the most rostral point on the head, and oblique dotted lines indicate the beginning of the first myotome. The rostrocaudal length of the head is equal in both control and treated embryos demonstrating that total body length differ- ences are due to differences in trunk lengths. Scale bar= 0.5 mm.

The number of gridlines crossed in 3 min after response to mechan- ical stimuli was used as a measurement of motor development for 2– 4 dpf embryos. Dechorionated 1 dpf zebrafish embryos were treated with 25 μM pCPA for 24 h and rescued into conditioned Instant Ocean at 2 dpf. Individual 2, 3, and 4 dpf embryos were placed in a small glass petri dish (35 ×10 mm) containing conditioned Instant Ocean atop a 35 mm grid of sixteen 7.0 mm squares resting on heated micro- scope stage that maintained the temperature at 28.5 ±0.5 °C. Preceding hatching (3 dpf) and swim bladder inﬂation (4–5 dpf), dechorionated embryos lie on their side at the bottom of the petri dish and show little volitional movement. When larvae are touched, they will respond by moving along the bottom of the petri dish, sometimes very quickly. This stage of swimming development is often referred to as burst swim- ming. After a 3 min acclimation period, each embryo was gently stimu- lated with a tungsten probe up to three times to illicit movement and the number of gridlines crossed in 3 min was recorded. Motor behavior of individual 5–7 dpf zebrafish larvae was accessed similarly to that of aforementioned 2–4 dpf embryos, but without the need for manual stimulation. At these ages, zebrafish larvae exhibit the final stage of loco- motor development, referred to as beat and glide, or spontaneous swim- ming activity characterized by volitional movement after inﬂation of the swim bladder. Movement was recorded with a Sony MPEG-Movie HQX DSC-F717 camera (Tokyo, Japan) mounted on the microscope with a MM99 universal microscope adapter (Martin Microscope Co., Easley, SC). Recordings were viewed on Dell Optiplex GX 280 computer (Round Rock, TX), and the number of gridlines crossed in 3 min was recorded.

2.4. Immunohistochemistry

Manually dechorionated 2 dpf zebrafish embryos were deeply anesthetized in 0.76 mM MS-222 and immersion fixed in 4% parafor- maldehyde in phosphate buffered saline (0.01 M PBS; 0.9% NaCl; pH 7.4; Sigma) overnight at 4 °C. Fixed embryos were washed thorough- ly in PBS and dehydrated through a graded methanol/PBS series (1:3, 1:1, 3:1), and then passed through three changes of absolute metha- nol and held in the last methanol change at −20 °C. Embryos were re-hydrated through a graded methanol/PBS series (3:1, 1:1, 1:3) containing 0.5% Triton-X-100 (PBSTx; Sigma). To facilitate reagent penetration, tissue permeabilization was achieved by sequentially passing embryos through deionized H20 at room temperature for 60 min, 100% acetone at −20 °C for 20 min, and collagenase (1 mg/ml; Sigma) at room temperature for 30 min. Endogenous per- oxidase activity was quenched by incubating the embryos in 0.3% H- 2O2/absolute methanol for 30 min at room temperature. After thorough washes in PBSTx, embryos were transferred to a Sylgard dish containing PBSTx, and the yolk sac and the epidermis overlying the brain and trunk were removed to allow better penetration of blocking and antibody solutions. To block nonspecific staining, em- bryos were incubated with avidin D (Vector Laboratories, Burlingame, CA) at 4 drops/ml in 10% Normal Goat Serum (NGS)/PBSTx overnight at 4 °C. After thorough washes in PBSTx, embryos were incubated in primary antiserum (rabbit polyclonal anti-5HT; ImmunoStar, Hud- son, WI) diluted 1:500 in 10% NGS/PBSTx with biotin blocking re- agent (Vector Labs; 4 drops/ml) for 2–3 days at 4 °C. After all-day washes (6–8 h) in PBSTx, embryos were incubated overnight at 4 °C in goat anti-rabbit secondary antibody (1:200; Vectastain Elite ABC kit, Vector Labs). After thorough PBSTx washes, embryos were incubated with Vectastain Elite ABC reagent (Vector Labs) in PBSTx for 40 min at room temperature, washed in PBSTx, and then trans- ferred into multi-well plates for staining (2–3 embryos per well). Em- bryos were presoaked in 3, 3′-diaminobenzidine (DAB) buffer (DAB substrate kit for peroxidase; Vector Labs) for 15 min at room temper- ature. DAB presoak was removed, and fresh DAB/H2O2 solution was added to the wells following the manufacturer’s instructions. Reac- tions were monitored under a dissecting microscope and stopped by adding PBS directly to the wells. After thorough washing in PBS, embryos were dehydrated through increasing concentrations of etha- nol, cleared in methyl salicylate, and mounted in Permount (Fisher Scientific, Pittsburg, PA).

Whole mounts were photographed using a Sony HQX DSC-F717 camera with a MM99 universal microscope adapter mounted on an Olympus VANOX microscope. Images were compiled using Adobe Photoshop 7.0 (San Jose, CA). The specificity of the reaction was ver- ified by the use of two controls: omission of the primary antibody or by replacement of the primary antibody with non-immune serum. No staining was observed in either control. Serotonin immunohisto- chemistry experiments were repeated three times using 10 embryos each time.

2.5. Reverse transcriptase polymerase chain reaction (RT-PCR) experiments

2.5.1. 5HT1A receptor transcript and SERT transcript developmental experiment

Zebrafish embryos/larvae from 1 to 7 dpf were manually dechor- ionated and deeply anesthetized in 0.76 mM MS-222 in conditioned Instant Ocean. The epidermis of each embryo/larva was removed to eliminate contaminating sources of serotonergic cells/neurons, and in the case of 4–7 dpf larvae, the gills, swim bladder, and intestinal tract were also removed. Embryos and larvae were sectioned into brain (head) and spinal cord (tail) segments so that each could be processed separately. Anesthetized animals were placed in a slide- well containing conditioned Instant Ocean. Each embryo/larva was positioned and sectioned by iridectomy scissors placed at the rostral border of the yolk sac and swim bladder with the blades angled to- ward the first myotome. After the cut was made, the tail was inspected to insure that the caudal rhombencelphalon was not at- tached. Five heads or five tails were placed in tubes containing 200 μl of RNAlater (Ambion, Austin, TX) and held at 4 °C until all de- velopmental groups were accumulated. Total RNA was extracted from tissues and treated with DNase I using RNAqueous-Micro (Ambion) following the manufacturer’s instructions for solid tissue isolation. RNA integrity and concentration were verified as previously described (Airhart et al., 2007). β-actin was used as an internal con- trol. cDNA for SERT and 5-HT1A receptor was PCR amplified with primers developed in our laboratory based on genomic data for the zebrafish and produced single bands at the expected molecular weights. The β-actin zebrafish primers, amplification conditions and sequences were previously described (Airhart et al., 2007). Concen- tration of each sample was determined by running 1 μl of PCR product on a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA) with a the epidermis and intestinal tract were removed from experimental and control embryos. Procedures for the serotonin depletion study experiments were identical to those described above.

2.6. Statistical analysis

Morphological, behavioral, and RT-PCR data are expressed as the means±SD. Tukey’s Honestly Significance Difference Test was per- formed for post-ANOVA pair-wise comparisons. p values of b 0.05 were considered significant.

3. Results

3.1. Effects of serotonin depletion on movement and length

This study examined the effect of serotonin depletion on early move- ment (burst swimming) and rostrocaudal length using the LOEC (25 μM) of pCPA. One day post fertilization zebrafish embryos were bath exposed to 25 μM pCPA, rescued into conditioned Instant Ocean at 2 dpf and allowed to develop through 7 dpf. At 1 dpf, control zebrafish embryos exhibited evoked coiling (i.e., alternating bilateral trunk ﬂexion with no forward movement). When manually stimulated at 2 dpf, con- trol embryos exhibited burst swimming and were able to travel the di- ameter of a petri dish (5.0 cm). Control animals continued to progress in swimming development through 7 dpf. pCPA treated embryos showed significantly less movement than controls on every day exam- ined (Fig. 2; p values b 0.05). At 2 dpf, treated embryos appeared to quiv- er on their sides with little or no positional displacement through 4 dpf. Beginning at 4 dpf, treated larvae began to show minimal improvement in swimming activity. By 6–7 dpf, i.e., 4–5 days after rescue; larvae appeared to demonstrate some spontaneous or volitional swimming ac- tivity, but significantly less than controls (Fig. 2; p values b 0.05).

Embryos treated with pCPA were significantly shorter along their rostrocaudal axis as compared to controls at 2 dpf, and these differences were apparent through 7 dpf (Fig. 3; p values b 0.01). Both treated and control 2 dpf embryos exhibited the same number of myotomes (n =30); however, representative myotomes from the head (myotome 5), trunk (myotome 15), and tail (myotome 23) regions of treated em- bryos were shorter than controls (Fig. 4; p values b 0.05). The notochord from control 2 dpf embryos consisted of large clear, vacuolated disk shaped cells with a small rim of cytoplasm. Cells were surrounded by a thin connective tissue sheath and were aligned one adjacent to the other with no apparent intervening intercellular spaces (Fig. 5B). In pCPA treated embryos, the notochord showed damaged regions along its length. These damaged regions exhibited fragmentation of vacuoles within the notochordal cells with an apparent in-growth of the fibrous connective tissue sheath as shown in the boxed area in Fig. 5D.

Fig. 3. Changes in rostrocaudal length of control and pCPA exposed embryos/larvae. Embryos were exposed to 25 μM pCPA at 1 dpf and rescued 24 h later. Rostrocaudal lengths increased in rescued larvae, but did not return to control lengths even 6 days after rescue (p valuesb 0.01). Each data point represents the mean±S.D. of 30 embryos/larvae.

3.2. Serotonin immunochemistry in control and pCPA treated embryos

Immunohistochemistry was utilized to qualitatively examine pCPA’s inhibition of serotonin synthesis in CNS neurons. Embryos at 1 dpf were treated with 25 μM pCPA and rescued 24 h later. The treated embryos and 2 dpf control embryos were processed for immunohistochemistry using a serotonin antibody. Control embryos exhibited serotonin local- ization in the same nuclear groups (regions) within the brain and spinal cord (Fig. 6A and C) and as described in numerous other investigations (Bellipanni et al., 2002; Lillesaar et al., 2009; McLean and Fetcho, 2004a; Sallinen et al., 2009). No reaction product was observed in any of the nuclear groups within the brain and spinal cord of pCPA treated embryos thus demonstrating that pCPA caused a complete reduction of serotonin concentration in these neurons (Fig. 6B and D).

Fig. 4. Schematic representation of a 2 dpf zebrafish embryo with myotomes shown as chevrons. Rostrocaudal lengths (mm) and representative myotome lengths (μm) in 2 dpf control and pCPA treated zebrafish embryos are indicated. Measurements (mean±S.D.) were made after bath exposure of dechorionated 1 dpf embryos to 25 μM pCPA and subsequent rescue at 2 dpf. Rostrocaudal lengths of treated embryos were shorter than control lengths; representative myotomes of treated embryos were also shorter as compared to their representative controls (p valuesb 0.05; n= 10).

Fig. 5. Treatment with 25 μM pCPA at 24 hpf leads to shorter body length and notochordal damage by 2 dpf. (A) Lateral image of a control 2 dpf embryo. (B) Enlargement of caudal portion of control embryo showing the notochord (N). The highlighted box indicates a portion of a normal notochord which is in comparison to the same region of notochord in a pCPA treated embryo shown in D. (C) Lateral image of a 2 dpf embryo exposed to 25 μM pCPA from 1 to 2 dpf. (D) Enlargement of caudal portion of treated embryo showing notochordal damage. Scale bar= 0.5 mm (A, C). Scale bar= 0.05 mm (B, D).

3.3. Effects of pCPA/MS-222

The anesthetic MS-222 is a voltage-gated Na+ channel blocker that has been used by others to eliminate rostrocaudal shortening due to myotome hypertonicity in zebrafish embryos (Hirata et al., 2004, 2005; Stehr et al., 2006). To address the possibility that the rostrocaudal shortening we observed in pCPA treated embryos was caused by myo- tome hypertonicity, 1 dpf embryos were exposed to 0.6 mM MS-222, 25 μM pCPA, or a combination of 25 μM pCPA and 0.6 mM MS-222, res- cued at 2 dpf and then followed for one additional day. Rostrocaudal lengths of MS-222 treated embryos were the same as controls on both days, while pCPA treatment resulted in embryos that were shorter than controls and the MS-222 group (Table 1; p b 0.05). The combina- tion treatment of pCPA/MS-222 appeared to mitigate rostrocaudal shortening in 2 dpf embryos, but by 3 dpf combination treated embryos were shorter as compared to 2 dpf embryos and controls (Table 1; p values b 0.05).

Fig. 6. Serotonin antibody staining in a 2 dpf embryo. (A) Dorsal image of a 2 dpf control zebrafish head showing serotonin labeled cells in the arrowhead population, AP; hypo- thalamus, HY; superior raphe, SR; and inferior raphe, IR. (B) Dorsal image of the head of a 2 dpf zebrafish exposed to 25 μM pCPA at 1 dpf showing no serotonin labeling. (C) Lateral image of the spinal cord of a 2 dpf control zebrafish showing ventromedial, VM serotonin labeled cells. (D) Lateral image of the spinal cord of a 2 dpf zebrafish exposed to 25 μM pCPA at 1 dpf showing no serotonin labeled cells. The epidermis overlying the brain and trunk were removed to allow better penetration of blocking and antibody solutions. Scale bars: A, B= 95 μm; C, D= 50 μm.

3.4. Speciﬁcity of pCPA

• The possibility that depleted dopamine, rather than serotonin, was responsible for the observed results was considered. Embryos were treated with alpha-methyl-DL-tyrosine methyl ester hydro- chloride (AMPT), a specific inhibitor of tyrosine hydroxylase, the rate-limiting enzyme of dopamine synthesis (Corrodi and Hanson, 1966). One dpf embryos were exposed to 10, 25, 50, 100 μM and 1 mM AMPT in a dose response experiment, rescued at 2 dpf and evaluated through 5 dpf. At the four lowest concentrations, axial length and burst swimming (as described in Section 3.2 above) appeared no different from controls at 2 dpf. At the highest concen- tration of AMPT (1 mM), axial lengths were not different from controls; but, at 2 dpf, movement as assessed by burst swimming was reduced. These embryos could move only 1.0–1.5 cm rather than the 5.0 cm diameter of the petri dish that control embryos could move. While diminished, burst swimming did occur. By 3 dpf, treated AMPT embryos exhibited normal axial length and movement comparable to control larvae, and by 5 dpf, AMPT trea- ted embryos were not distinguishable from controls with respect to axial length and volitional movement (data not shown).

3.5. Changes in the developmental expression of 5HT1A receptor and SERT transcripts in brain and spinal cord

In a previous study using whole embryos we demonstrated by RT- PCR that mRNA expression for two serotonin associated proteins, the 5HT1A receptor and SERT (the serotonin transporter) appeared early in development (Airhart et al., 2007). In the present study we were able to demonstrate that these transcripts appear early (1 dpf) in iso- lated regions of the CNS (brain and spinal cord). The data suggest that both brain and spinal cord exhibit similar temporal expression of 5HT1A receptor and SERT transcripts (Fig. 7).

3.6. Changes in the developmental expression of 5HT1A receptor and SERT transcripts in brain and spinal cord after pCPA exposure

RT-PCR was also utilized to determine if there were differences in transcript levels for 5HT1A receptor and SERT in 2 dpf embryos after a 24 hour exposure to 25 μM pCPA. Twenty four hours after pCPA expo- sure, 2 dpf embryos showed a decrease in 5HT1A receptor transcripts in both brain and spinal cord (Fig. 8A) as compared to controls (brain, p b 0.01; spinal cord, p b 0.05). SERT transcript levels remained unaffected in brain, but were decreased in spinal cord (Fig. 8C; p b 0.01).

4. Discussion

This study investigated the impact of serotonin depletion in embry- onic and larval zebrafish by using the tryptophan hydroxylase inhibiter pCPA to reduce serotonin synthesis. We examined the effects of seroto- nin reduction on locomotor behavior, rostrocaudal length and noto- chordal morphology, and the expression of two mRNA transcripts associated with serotonin action, i.e., the 5HT1A receptor and SERT.
pCPA treated embryos rescued at 2 dpf demonstrated minimal im- provement in mobility over the post-rescue period. One and 2 days after rescue, embryos remained nearly paralyzed and exhibited no burst swimming. By 6–7 dpf, i.e., 4–5 days after rescue; embryos demonstrated spontaneous or volitional swimming activity that plateaued well below control animals at the end of the experiment (7 dpf). It was surprising significantly higher in both brain and spinal cord of rescued larvae as compared to controls (Fig. 8D; brain, p b 0.01; spinal cord, p b 0.01).

Fig. 7. Developmental expression of 5HT1A receptor and SERT transcripts in the brain and spinal cord in zebrafish. One-day time intervals from 1 to 7 days are shown. In each case, percentages are given relative to β-actin levels. Data represent means±S.D. of 15 embryos/larvae.

A similar set of experiments was performed to determine if transcript levels recovered in rescued pCPA treated embryos. Rescued 2 dpf embry- os were transferred into conditioned Instant Ocean and allowed to re- cover for five days. A recovery period of five days was chosen because at this time improved swimming was observed (Fig. 2). After five days of recovery (7 dpf), 5HT1A receptor transcript levels remained below control levels in brain (Fig. 8B, p b 0.05) but spinal cord values were not statistically distinguishable from controls. SERT transcript levels were to observe no recovery in notochordal morphology or axial body length after rescue. In fact, at 6 and 7 dpf, when swimming had improved, the degree of body shortening was near maximum levels. Therefore, multiple toxic mechanisms may be associated with serotonin depletion. The swim- ming behavior that we observed in rescued fish may be explained by de novo synthesis of tryptophan hydroxylase and concurrent metabolic breakdown of the pCPA-tryptophan hydroxylase complex. Restoration of tryptophan hydroxylase activity could then lead to a re-establishment of normal serotonin levels in serotonergic neurons and their terminals. However, the continual decrease in rostrocaudal length suggests that se- rotonin circuitry may be abnormal in rescued animals.

Tryptophan hydroxylase and tyrosine hydroxylase are aromatic amino acid hydroxylases that share a similar primary structure (Candy and Collet, 2005). Tyrosine hydroxylase is the rate-limiting enzyme in catecholamine synthesis, and one of its products, dopamine, plays a role in zebrafish locomotor development (Bretaud et al., 2004). To evaluate whether the phenotypic effect of pCPA was due to inhibition of dopamine synthesis as opposed to serotonin, we exposed embryos to the tyrosine hydroxylase inhibitor, AMPT. While rescued embryos initially demon- strated reduced movements, their swimming ability eventually recovered and was indistinguishable from controls; their axial length was also equal to control embryos. We therefore conclude that the effects observed were not due to reduction in dopamine.

Fig. 8. 5HT1A receptor and SERT transcript expression in brain and spinal cord of 2 dpf and 7 dpf zebrafish after pCPA exposure at 1 dpf and rescue at 2 dpf. (A). 5HT1A receptor transcript expression in brain and spinal cord of control and 25 μM pCPA treated 2 dpf embryos. Twenty four hours after pCPA exposure, embryos demonstrated a decrease in 5HT1A receptor transcripts in both brain and spinal cord as compared to controls (p valuesb 0.01). (B). 5HT1A receptor transcript expression in brain and spinal cord of control and pCPA treated 7 dpf larvae. After five days of recovery (at 7 dpf), 5HT1A receptor transcript levels remained below control levels in brain (p b 0.05). 5HT1A receptor levels in spinal cord at 7 dpf were not different than control levels. (C). SERT transcript expression in brain and spinal cord of 25 μM pCPA treated 2 dpf larvae. SERT transcript levels were not dif- ferent from controls in brain after pCPA exposure, but were decreased in spinal cord as compared to controls (p b 0.01). (D). SERT transcript expression in brain and spinal cord of control and pCPA treated 7 dpf larvae. SERT transcript expression was higher in both brain and spinal cord of rescued larvae as compared to controls (p valuesb 0.01). Values are given relative to β-actin levels. Data represent means±S.D. of 15 larvae. Asterisks indicate significant differences (*p valuesb 0.01; **p valuesb 0.05).

Embryos exposed to pCPA exhibited mobility impairment and were shorter in overall body length as compared to untreated con- trols. Shorter lengths were due to smaller myotome size; however, myotome numbers were not affected. Our observations suggest that the sensory component of movement is not affected by serotonin de- pletion because small contractions of the myotomes were observed after mechanical stimulation. Notochords of treated embryos showed damaged regions along their lengths, suggesting abnormal muscle contractions as a result of myotome hypertonicity. Using the sodium channel blocker, MS-222, we tested the possibility that hypertonicity of the myotomes, i.e., sustained contractions on each side of the trunk, resulted in reduced axial length and abnormal morphology of the notochord, the later perhaps due to mechanical compression. The effects we observed on rostrocaudal length were temporary; at 3 dpf the beneficial effect of MS-222 was no longer seen. We attribute this to the fact that MS-222 was rapidly metabolized.

There are two sources of serotonergic innervation to the embryonic zebrafish spinal cord: intrinsic serotonergic neurons of the spinal cord itself and the inferior raphe nucleus (Lillesaar et al., 2009; McLean and Fetcho, 2004a, b). The serotonergic neurons of the inferior raphe nucle- us are present between 2 and 3 dpf, and their descending axons appear nearly simultaneously with initial neuron differentiation. These axons along with dopaminergic axons have been observed in a ventral path- way in the zebrafish hindbrain that extends to the beginning of the spi- nal cord by 2–3 dpf (Lillesaar et al., 2009; McLean and Fetcho, 2004b). The intrinsic serotonergic neurons of the zebrafish spinal cord differen- tiate at an early age (1–2 dpf). These neurons are localized throughout the entire length of the zebrafish spinal cord and are referred to as ven- tromedial neurons (Fig. 6C). They form a bilateral column of nerve con- nections by 24–32 hpf (McLean and Fetcho, 2004a; Sallinen et al., 2009). Their axons have unipolar processes that project ventrolaterally and then dorsally into the motor column. Axons descend a short dis- tance (1–2 myotome segments) and are associated with growth cones that lie adjacent to soma and dendrites of primary and secondary motor neurons (McLean and Fetcho, 2004a). Our immunohistochemical data demonstrate that serotonin is depleted in the inferior raphe nuclei and in intrinsic spinal neurons of embryos treated with pCPA. We sug- gest that the depletion of serotonin within intrinsic neurons of the spi- nal cord may be responsible for the nearly complete paralysis pCPA treated embryos at 2 dpf.

A variety of studies have also shown that perturbations in serotonin levels can alter the RNA messages of serotonin receptor subtypes and SERT, the serotonin transporter protein (Airhart et al., 2007; Lauder et al., 2000; Rattray et al., 1996; Whitaker-Azmitia and Azmitia, 1989). Studies have also shown that the 5HT1A receptor plays an important role in the development of neural structures in a variety of animals in- cluding zebrafish (Norton et al., 2008; Patel and Zhou, 2005). In the pre- sent study we followed the ontogeny of the 5HT1A receptor and SERT transcripts using RT-PCR. Significant levels of both transcripts appear in brain and spinal cord as early as 1 dpf, and by 3 dpf, transcript levels approached maximal concentrations in both CNS structures. The con- centrations of 5HT1A receptor and SERT transcripts through 7 dpf are nearly all localized to the developing brain and spinal cord of zebrafish embryos. Exceptions to this include the small number of neuroepithelial cells of the gill filament primorida that express serotonin at 5 dpf (Jonz and Nurse, 2005), and enteric neurons and enterochromaffin cells of the gastrointestinal epithelium in 5 dpf larvae (Njagi et al., 2010). In our studies, gills, swim bladder, and intestinal tract were removed to elim- inate contaminating sources of serotonergic cells/neurons. Therefore, we assume that the 5HT1A receptor and SERT transcript values observed in our studies relate to brain and spinal cord. The presence of transcripts in the developing spinal cord is almost certainly due to a relatively large contingent of serotonergic neurons contained throughout its length. Using an antibody for serotonin, others have shown that by 32 hpf two to three neuron cell bodies per zebrafish spinal segment are pre- sent with axons visible by 2 dpf (McLean and Fetcho, 2004a). A more re- cent investigation also using an antibody to serotonin observed these neurons and axons as early as 24 hpf (Sallinen et al., 2009). Early differ- entiation of serotonergic neurons in the spinal cord makes them a likely target for pCPA’s abnormal effects in 2 dpf treated embryos.

Data in the present study indicate that 24 h after pCPA exposure (at 2 dpf), 5HT1A transcript levels in both brain and spinal cord were re- duced. Five days after recovery (at 7 dpf), 5HT1A receptor transcript levels remained below control levels in brain. While transcript levels appeared to increase in spinal cord, they were not statistically distin- guishable from control levels. Twenty four hours after pCPA exposure (at 2 dpf), SERT transcript levels in brain did not appear different than controls, while in spinal cord, SERT transcript levels were reduced. Five days after recovery (at 7 dpf), SERT transcript levels in both brain and spinal cord were increased when compared to their controls. These findings suggest that dynamic changes in steady-state levels of both transcripts can occur during recovery. The RT-PCR data presented in this study demonstrates that concentration of 5HT1A receptor and SERT transcripts in the brain and spinal cord are affected by pCPA treat- ment — thus the proteins are likely functional in embryonic zebrafish. These findings necessitate the need to further investigate serotonin’s role in neurodevelopment.

Conﬂict of interest statement

The authors declare that there are no conﬂicts of interest.

Acknowledgements

It is with great sadness that we report the loss of two authors and members of the Zebrafish Neurobehavioral Laboratory at East Ten- nessee State University. Dr. Mark J. Airhart and Dr. Barney E. Miller died unexpectedly during the final stages of the preparation of this manuscript. We dedicate this research paper to their memory and to their love of science.

References

Airhart MJ, Miller B, Lee D, Wilson T, Miller M, Skalko RG. The effects of serotonin de- pletion on locomotor development in zebrafish. Birth Defects Res A Clin Mol Tera- tol 2006;76:334.
Airhart MJ, Lee DH, Wilson TD, Miller BE, Miller MN, Skalko RG. Movement disorders and neurochemical changes in zebrafish larvae after bath exposure to ﬂuoxetine (PROZAC). Neurotoxicol Teratol 2007;29:652–64.
Bellipanni G, Rink E, Bally-Cuif L. Cloning of two tryptophan hydroxylase genes expressed in the diencephalon of the developing zebrafish brain. Mech Dev 2002;119S:S215–20.
Bretaud S, Lee S, Guo S. Sensitivity of zebrafish to environmental toxins implicated in Parkinson’s disease. Neurotoxicol Teratol 2004;26:857–64.
Brustein E, Saint-Amant L, Buss RR, Chong M, McDearmid JR, Drapeau P. Steps during the development of the zebrafish locomotor network. J Physiol 2003;97:77–86.
Candy J, Collet C. Two tyrosine hydroxylase genes in teleosts. Biochim Biophys Acta 2005;1727:35–44.
Chambers CD, Hernandez-Diaz S, Van Marter LJ, Werler MM, Louik C, Jones KL, et al. Se- lective serotonin-reuptake inhibitors and risk of persistent pulmonary hyperten- sion of the newborn. N Engl J Med 2006;354:579–87.
Chase TN, Murphy DL. Serotonin and central nervous system function. Annu Rev Phar- macol 1973;13:181–97.
Corrodi H, Hanson LCF. Central effects of an inhibitor of tyrosine hydroxylation. Psy- chopharmacologia 1966;10:116–25.
Di Pino G, Moessner R, Lesch KP, Lauder JM, Persico AM. Roles for serotonin in neuro- development: more than just neural transmission. Curr Neuropharmacol 2004;2: 403–17.
Downes GB, Granato M. Supraspinal input is dispensable to generate glycine-mediated locomotive behaviors in the zebrafish embryo. J Neurobiol 2006;66:437–51.
Drapeau P, Saint-Amant L, Buss RR, Chong M, McDearmid JR, Brustein E. Development of the locomotor network in zebrafish. Prog Neurobiol 2002;68:85-111.
Hirata H, Saint-Amant L, Waterbury J, Cui W, Zhou W, Li Q, et al. accordion, a zebrafish behavioral mutant, has a muscle relaxation defect due to a mutation in the ATPase Ca2+ pump SERCA1. Development 2004;131:5457–68.
Hirata H, Saint-Amant L, Downes GB, Cui WW, Zhou W, Granato M, et al. Zebrafish bandoneon mutants display behavioral defects due to a mutation in the glycine receptor β-subunit. Proc Natl Acad Sci USA 2005;102:8345–50.
Jessell TM, Kelly DD. Pain and analgesia. In: Kandel ER, Schwartz JH, Jessell TM, editors.
Principles of neural science. New York: Elsevier; 1991. p. 385–99.
Jéquier E, Lovenberg W, Sjoerdsma A. Tryptophan hydroxylase inhibition: the mecha- nism by which p-chlorophenylalanine depletes rat brain serotonin. Mol Pharmacol 1967;3:274–8.
Jonz MJ, Nurse CA. Development of oxygen sensing in the gills of zebrafish. J Exp Biol 2005;208:1537–49.
Khozhai LI, Otellin VA. Formation of the neocortex in mice developing in conditions of prenatal serotonin deficiency. Neurosci Behav Physiol 2006;36:513–7.
Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic de- velopment of the zebrafish. Dev Dyn 1995;203:253–310.
Koe BK, Weissman A. p-chlorophenylalanine: a specific depletor of brain serotonin. J Pharmacol Exp Ther 1966;154:499–516.
Koren G, Boucher N. Adverse effects in neonates exposed to SSRIs and SNRI in late ges- tation. Can J Clin Pharmacol 2009;16:e66–7.
Lauder JM, Liu J, Grayson DR. In utero exposure to serotonergic drugs alters neonatal expression of 5-HT1A receptor transcripts: a quantitative RT-PCR study. Int J Dev Neurosci 2000;18:171–6.
Lillesaar C, Stigloher C, Tannhäuser B, Wullimann MF, Bally-Cuif L. Axonal projections originating from raphe serotonergic neurons in the developing and adult zebrafish, Danio rerio, using transgenics to visualize raphe-specific pet1 expression. J Comp Neurol 2009;512:158–82.
Lisboa SFS, Oliveira PE, Costa LC, Venâncio EJ, Moreira EG. Behavioral evaluation of male and female mice pups exposed to ﬂuoxetine during pregnancy and lactation. Pharmacology 2007;80:49–56.
Lucki I. The spectrum of behaviors inﬂuenced by serotonin. Biol Psychiatry 1998;44:151–62. McLean DL, Fetcho JR. Ontogeny and innervation patterns of dopaminergic, noradren- ergic, and serotonergic neurons in larval zebrafish. J Comp Neurol 2004a;480:
38–56.
McLean DL, Fetcho JR. Relationship of tyrosine hydroxylase and serotonin immunoreactivity to sensorimotor circuitry in larval zebrafish. J Comp Neurol 2004b;480:57–71.
Nakajima K, Matsuyama K, Mori S. Prenatal administration of para-chlorophenylala- nine results in suppression of serotonergic system and disturbance of swimming movements in newborn rats. Neurosci Res 1998;31:155–69.
Njagi J, Ball M, Best M, Wallace KN, Andreescu S. Electrochemical quantification of se- rotonin in the live embryonic zebrafish intestine. Anal Chem 2010;82:1822–30.
Norton WHJ, Folchert A, Bally-Cuif L. Comparative analysis of serotonin receptor (HTR1A/HTR1B families) and transporter (slc6a4a/b) gene expression in the zebra- fish brain. J Comp Neurol 2008;511:521–42.
Patel TD, Zhou FC. Ontogeny of 5-HT1A receptor expression in the developing hippo- campus. Brain Res Dev Brain Res 2005;157:42–57.
Pearlstein E, Mabrouk FB, Pﬂieger JF, Vinay L. Serotonin refines the locomotor-related alternations in the in vitro neonatal rat spinal cord. Eur J Neurosci 2005;21: 1338–46.
Pﬂieger JF, Clarac F, Vinay L. Postural modifications and neuronal excitability changes induced by short-term serotonin depletion during neonatal development in the rat. J Neurosci 2002;22:5108–17.
Popa D, Léna C, Alexandre C, Adrien J. Lasting syndrome of depression produced by re- duction in serotonin uptake during postnatal development: evidence from sleep, stress, and behavior. J Neurosci 2008;28:3546–54.
Pronina T, Ugrumov M, Adamskaya E, Kuznetsova T, Shishkina I, Babichev V, et al. Inﬂu- ence of serotonin on the development and migration of gonadotropin-releasing hormone neurones in rat foetuses. J Neuroendocrinol 2003;15:549–58.
Rattray M, Baldessari S, Gobbi M, Mennini T, Samanin R, Bendotti C. p-Chlorophenyla- lanine changes serotonin transporter mRNA levels and expression of the gene product. J Neurochem 1996;67:463–72.
Saint-Amant L, Drapeau P. Time course of the development of motor behaviors in the zebrafish embryo. J Neurobiol 1998;37:622–32.
Sallinen V, Sundvik M, Reenilä I, Peitsaro N, Khrustalyov D, Anichtchik O, et al. Hyper- serotonergic phenotype after monoamine oxidase inhibition in larval zebrafish. J Neurochem 2009;109:403–15.
Shuey DL, Sadler TW, Tamir H, Lauder JM. Serotonin and morphogenesis. Anat Embryol (Berl) 1993;187:75–85.
Stehr CM, Linbo TL, Incardona JP, Scholz NL. The developmental neurotoxicity of fipro- nil: notochord degeneration and locomotor defects in zebrafish embryos and lar- vae. Toxicol Sci 2006;92:270–8.
Stemple DL, Solnica-Krezel L, Zwartkruis F, Neuhauss SCF, Schier AF, Malicki J, et al. Mutations affecting development of the notochord in zebrafish. Development 1996;123:117–28.
van Kesteren RE, Spencer GE. The role of neurotransmitters in neurite outgrowth and synapse formation. Rev Neurosci 2003;14:217–31.
Vinay L, Ben-Mabrouk F, Brocard F, Clarac F, Jean-Xavier C, Pearlstein E, et al. Perinatal development of the motor systems involved in postural control. Neural Plast 2005;12:131–9.
Vitalis T, Cases O, Passemard S, Callebert J, Parnavelas JG. Embryonic depletion of sero- tonin affects cortical development. Eur J Neurosci 2007;26:331–44.
Walsh CA, Morrow EM, Rubenstein JLR. Autism and brain development. Cell 2008;135: 396–400.
Wang Y, Takai R, Yoshioka H, Shirabe K. Characterization and expression of serotonin transporter genes in zebrafish. Tohoku J Exp Med 2006;208:267–74.
Westerfield M. The zebrafish book: a guide for the laboratory use of zebrafish (Danio rerio). 4th ed. Eugene, OR: University of Oregon Press; 2000.
Whitaker-Azmitia PM, Azmitia EC. Stimulation of astroglial serotonin receptors pro- duces culture media which regulates growth of serotonergic neurons. Brain Res 1989;497:80–5.
Whitaker-Azmitia PM, Druse M, Walker P, Lauder JM. Serotonin as a developmental signal. Behav Brain Res 1996;73:19–29.
Zhou FC, Sari Y, Zhang JK. Expression of serotonin transporter protein in developing rat brain. Dev Brain Res 2000;119:33–45.