- Email: [email protected]
Arginine vasotocin effects on courtship behavior in male white perch (Morone americana)
Behavioural Brain Research 133 (2002) 177 /183 www.elsevier.com/locate/bbr
Research report
Arginine vasotocin effects on courtship behavior in male white perch (Morone americana) Stephen J. Salek, Craig V. Sullivan, John Godwin * Department of Zoology, North Carolina State University, Campus Box 7617, Raleigh, NC 27695, USA Received 14 August 2001; received in revised form 10 December 2001; accepted 10 December 2001
Abstract Arginine vasotocin (AVT) and its mammalian homologue, arginine vasopressin (AVP), have been shown to have widespread behavioral effects in vertebrates. AVT was evaluated for its effectiveness in stimulating an important courtship behavior termed ‘attending’ in male white perch, Morone americana. Attending consists of close and continuous following of the female with occasional contact in the abdominal area. We tested the behavioral effectiveness of AVT in stimulating attending when administered either intraperitoneally (IP) or intracerebroventricularly (ICV). We also tested IP injections of AVT alone and in combination with an AVP V1 receptor antagonist (Manning compound). None of the IP injections of either AVT or Manning compound produced consistent effects on attending behavior. In contrast, ICV injections of AVT did significantly increase attending behavior and at low dosages. Circulating levels of testosterone and 11-ketotestosterone were not affected approximately 80 min following injection by any of the treatments. The strong behavioral effects observed with ICV administration support a central site of action for AVT in stimulating attending behavior. This is a complex behavior that shows similarities to behaviors mediated by AVT and AVP in other vertebrates, providing further evidence of a conserved behavioral role for these peptides. # 2002 Published by Elsevier Science B.V. Keywords: Arginine vasotocin; Sexual behavior; Teleost; Androgen
1. Introduction The neuropeptides arginine vasopressin (AVP) and the homologous arginine vasotocin (AVT) in nonmammalian vertebrates have behavioral effects in a variety of vertebrate species (reviewed in Refs. [5,13,26]). These include effects on clasping and appetitive sexual behavior in newts (Taricha ) [24 /26,38], vocal and locomotor behavior in anuran amphibians [2,3,6,23,32,36], singing and sexual behavior in birds [4]17[22], and communicative, affiliative, and aggressive behavior in voles, mice, and hamsters (e.g. [1,7,8,16,40]). Similar findings have recently been published showing AVT effects on sexual, aggressive and communicative behaviors in teleost fishes [11,12,37]. The first demonstration of behavioral effects for neurohypophysial peptides was by Pickford [29], who showed that neurohypophysial extracts induced a
* Corresponding author. Tel.: 1-919-513-2396; fax: 1-919-5152698 E-mail address: [email protected] (J. Godwin).
‘spawning reflex’ in hypophysectomized killifish (Fundulus heteroclitus ). Subsequently, experiments with pure forms of specific neurohypophysial factors suggested AVT was responsible for this effect [30,39]. Isotocin (IST) could also induce this behavior but only at tenfold higher doses. In other closely related species of Cyprinodontids, AVT was shown to have similar effects [20,21,28]. Despite these studies and clear demonstrations of behavioral actions of AVT in other vertebrates, questions nevertheless remained regarding AVT effects on behavior in fishes until recently [11,12,37]. The spawning reflexes observed in Fundulus in response to injection with purified pituitary extracts were described by Pickford [29] as ‘violent S-shaped spasms’ during which ‘the fish showed no interest in each other but behaved independently’. The nature of this response and the large dosages necessary to elicit it in Fundulus and related species suggested that AVT could be acting peripherally through stimulation of smooth muscle [19,28]. Supporting this interpretation, both Peter [28] and Pickford et al. [31] found that AVT was no more effective when administered intracerebroventricularly
0166-4328/02/$ - see front matter # 2002 Published by Elsevier Science B.V. PII: S 0 1 6 6 - 4 3 2 8 ( 0 2 ) 0 0 0 0 3 - 7
178
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
(ICV) than intraperitoneally (IP). In contrast, Macey and coworkers [21] found that AVT-induced spawning reflexes in fish were essentially abolished by lesions of the preoptic nuclei, but not by lesions of comparable size in other brain regions, suggesting that AVT acted in the preoptic area. As noted above, Goodson and Bass [11,12] recently showed clear effects of AVT and IST on activity in neural circuits directly subserving vocalization behavior when these neuropeptides were administered directly into the preoptic area-anterior hypothalamus or midbrain tegmentum of the plainfin midshipman (Porichthys notatus ). The fish used in the studies of Goodson and Bass were anesthetized and the response measured was neuronal activity. Semsar and coworkers [37] recently showed that IP injections of AVT affected complex behaviors including sexual behavior, aggression, and feeding in the bluehead wrasse, Thalassoma bifasciatum . The primary courtship behavior of white perch (Morone americana ) leading to the spawning act is termed ‘attending’. Attending is close (within 30 cm) and continuous following of a female by a male and includes occasional contact of her abdominal area with the snout (described fully in [35]). In contrast to the spawning reflex of Cyprinodontid fishes, attending is a complex behavior which shows some strong similarities to appetitive sexual behaviors influenced by AVT in roughskin newts, frogs, and bluehead wrasses [3,37,38] and some similarities to affiliative behaviors affected by AVP in voles [42]. We therefore examined AVT effects on attending behavior of white perch in this study and compared the effectiveness of peripheral (IP) to central (ICV) administration of the peptide. We also measured plasma androgens since the androgen 11-ketotestosterone (11KT) influences the display of attending in white perch [34] and gonadal steroids critically affect the nature of AVT-induced reproductive behaviors in the rough-skinned newt [27]. We therefore asked whether the AVT and Manning treatments used here would influence plasma androgen levels (and potentially behavior by this route).
2. Materials and methods 2.1. Experimental animals We used white perch broodstock maintained at the North Carolina State University Pamlico Aquaculture Field Laboratory. They were transported to the Aquatic Research Facility on the North Carolina State University campus and maintained as described previously until testing [19]. We conducted these experiments during the breeding season for white perch in March, April, and May of 1998 and 1999. The white perch males used were 207.39/5.8 g and 237.99/2.3 mm (mean9/
SEM, N /94) in total length. All males were spermiating at the start of studies. Animal husbandry and experimental procedures followed the North Carolina State University Institutional Animal Care and Use Committee guidelines. 2.2. Injections We performed a series of experiments involving IP injections (experiments 1 /3) or ICV injections (experiment 4) to test behavioral effects of AVT and the AVP V1 receptor antagonist Manning compound. Experiment 1 compared attending behavior in males given either saline (0.9% NaCl) or 0.150 mg/g body weight AVT IP. Experiment 2 compared attending among males given saline, 0.150, or 1.0 mg/g AVT IP to test the repeatability of results from experiment 1 and test for a dose response. Experiment 3 tested the effect of a putative AVT receptor antagonist, the AVP V1 receptor antagonist Manning compound (2-Mercapto-beta,(CH2)5,O /Me /Tyr2, Arg8]-Vasopressin) on attending behavior when given IP both alone and in combination with AVT (1 mg/g body weight for both AVT and Manning compound; these doses were used by Propper and Dixon [32]). Experiment 4 tested the effects of AVT on attending behavior when delivered ICV. Injections were given under anesthesia (200 mg/l tricaine methanesulfonate in freshwater; Argent, Redmond, WA). Fish were anesthetized once for IP injections in experiments 1 and 2 and twice for experiment 3 which involved two separate injections. Fish were anesthetized once for ICV injections in experiment 4. We gave IP injections of peptides (AVT and Manning compound, Sigma, St. Louis, MO) with a 1-cc Luer LOK† syringe fitted with a 25 Ga needle (Becton Dickinson, Franklin Lakes, NJ) given in 100 /300 ml of 0.9% NaCl. ICV injections of peptides were made freehand into the third ventricle of the brain in 2 ml of 0.9% NaCl containing 10% Sheaffer ink (v/v) (Ft Madison, IA). Control animals were given 0.9% NaCl and ink only. Peptide solutions were freshly prepared and kept on ice. Syringes were filled just prior to injection. A small hole was made in the skull with a 20 Ga needle and then a 33 Ga needle fitted to a 10 ml Hamilton syringe (Reno, NV) was inserted into the hole to a depth of 8 mm. Lining up the syringe with the caudal side of the orbit and angling the syringe perpendicular to the head surface placed the needle in the middle of the brain approximately in the hypothalamus. Preliminary studies in which we sectioned each brain showed that ink was delivered to the ventricular system if no ink was found on the surface of the brain or in the braincase. The depth of the needle was controlled by placing a 200 ml pipette tip over the 33 Ga needle and trimming it so only 8 mm of the needle projected from the tip. We waited 10 s after inserting the needle, injected
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
2 ml over a period of 10 s, then waited 10 s more before removing the syringe. The hole was filled with Triple X antibiotic ointment (Food Lion, Salisbury, NC). Accuracy of injection was confirmed in all studies by opening the skull of each fish after behavioral trials and visually inspecting the brain. Samples with ink outside the brain were excluded from analyses. Both IP and ICV injections were completed rapidly (B/120 s) and fish were immediately returned to aerated water. Since the time out of water was brief, and in the interest of delivering injections rapidly, the gills of the fish were not irrigated during injections. 2.3. Behavior tests We evaluated the display of attending behavior using an ovulated female as the stimulus for male courtship behavior. To induce ovulation, several female perch were injected intramuscularly with human chorionic gonadotropin (hCG) (Schein Pharmaceuticals, Florham Park, NJ) at a dose of 330 IU/kg body weight 24 /48 h prior to behavioral testing. We assessed maturation of oocytes by anesthetizing females in quinaldine sulfate (50 mg/l) and subjecting them to ovarian biopsy [17]. Only females yielding ovulated eggs were used. Since males and females were similar in size, females were tagged with a silver streamer made of MylarTM tape (8 cm) inserted through the dorsal fin to mark them for identification during observations. Individual experimental males were captured from a holding tank with a hand net, given an injection (IP or ICV as described above), and placed into a holding tank for 15 min to recover from anesthesia. After this period, the experimental male was introduced into the test tank with the ovulated female. During the test period, attending was scored by one observer. Attending is defined as close (maximum 30 cm, but usually less) and continuous investigation of the female’s abdominal area which may include physical contact [34]. Male body sizes averaged approximately 24 cm, aiding in judging whether they were within 30 cm of the females abdominal area. Also, observers were blind to treatment to prevent bias in scoring this behavioral measure. The total time spent attending by a male was recorded by one observer over two 10 min intervals; the first interval began immediately after the male was introduced into the tank and the second began 50 min after the introduction. Following the second 10 min interval, the male was removed from the tank. Since males were given a 15 min recovery period after injection before being introduced to the test tank, these observation times began 15 and 65 min after injection. We performed behavior tests in 684-l rectangular tanks (1.9 /0.6 /0.6 m3), each illuminated with five 75 W red floodlights (Sylvania). Direct observations of courtship behavior were made through blinds located at
179
both ends of the tanks. Most testing occurred between 07:00 and 16:00 (65 males) and some occurred between 16:00 and 22:00 (29 males). There were no apparent differences in behavior between these two testing periods and behavioral tests were balanced across treatments for each experiment to prevent any consistent effect of testing time. All behavioral data presented and compared statistically in this paper are based on direct observation during behavioral testing. To assess the precision of these direct behavioral observations, attending was also rescored from videotapes and compared to direct observations using correlation analyses. For each behavioral test, a VHS video camera (Panasonic ag-187u, Seacaucus, NJ) equipped with a standard zoom lens (Panasonic Power Zoom /12) mated with a wide-angle lens (Panasonic Super Wide Pro 5050-0.5 /) and polarizing filter (Toshiba PL 49.0(s), Japan) was installed over each tank on a ball mount (Bogon, Italy). In our first experiment, all male scores were compared in this manner. In subsequent experiments, only four male’s scores per experiment were compared in this manner. Attending scores from direct observations and videotape analysis were highly and significantly correlated (n /15, R /0.94, P B/0.0001). 2.4. Blood sampling and hormone assays Immediately after behavior testing, fish were removed from the tanks, anesthetized in MS-222 (200 mg/ml), rinsed in fresh water, weighed, and bled by caudal puncture using a 3.0-cc syringe fitted with a 21-gauge heparinized needle. Approximately 1.5 ml blood was collected from each fish. Blood samples were stored on ice until being centrifuged (10 min at 10 000 /g ). The plasma was drawn off and stored at /80 8C until assay. Concentrations of 11KT and testosterone (T) were measured in duplicate 20 ml aliquots of blood plasma by highly specific RIAs. Briefly, we triple extracted samples in 3 ml of diethyl ether, dried the ether phase under a stream of nitrogen gas at 37 8C, and resuspended in 200 ml assay buffer for specific radioimmunoasays as described by Woods and Sullivan [41]. 2.5. Statistical analyses We log10 transformed data to reduce heteroschedasticity among treatment groups and performed statistical analyses using JMP statistical software (SAS, Cary, NC). We compared behavioral scores by either Student’s t-tests or one way analysis of variance (ANOVA) if there were more than two treatments. Individual means were compared following ANOVA using JMP’s ‘compare all pairs’ command to execute a Tukey / Kramer HSD comparison. Associations between attend-
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
180
ing scores from videotapes and direct observations were assessed by simple linear correlation (Pearson’s r).
3.2. Behavioral actions of AVT: effects of ICV administration
3. Results
In experiment 4, we injected males ICV with a saline control solution, 0.02 mg AVT, or 0.2 mg AVT. AVT significantly stimulated attending when administered at 0.02 mg/fish and 0.2 mg/fish (ANOVA, F2,11 /8.268, P B/0.01; Tukey /Kramer post hoc tests P B/0.05) (Fig. 1).
3.1. Behavioral actions of AVT: effects of IP administration
3.3. Effects of AVT: circulating androgen levels
IP injections of AVT and/or Manning compound in three separate experiments produced inconsistent results. In experiment 1, there was a significant increase in attending behavior by AVT-injected males relative to saline controls 65 min after injection (two-tailed t test, t /2.93, P /0.0013, 0.150 mg/g body weight), but not 15 min after injection (t /0.779 P /0.45, Table 1). We performed Experiment 2 identically to Experiment 1, but incorporated an additional dose of AVT. However, neither AVT dose (0.150 and 1.0 mg/g body weight) increased attending significantly relative to saline controls. In experiment 3, we found no significant differences among treatments where males received sequential injections spaced 15 min apart (saline/saline, saline/ AVT, Manning/saline, or Manning/AVT; AVT and Manning at 1.0 mg/g body weight (ANOVA, P /0.05 in all cases). The attending scores for males in all treatments in experiment 3 were at the high end of the range seen for males in experiments 1 and 2.
Table 1 Effect of IP AVT injections on courtship behavior Treatments attending
N
Attend early
Attend late
Total
Injected IP Experiment 1 Saline 0.150 mg/g
6 17.66.4 7 23.45.6
19.64.3 37.14.0*
18.64.0 30.74.8
Experiment 2 Saline 0.150 mg/g 1.0 mg/g
5 16.82.1 7 13.33.1 8 28.55.8
21.33.9 21.65.3 31.34.1
19.01.6 17.54.0 30.01.9
34.04.5 33.34.8 35.85.6 22.64.6
30.54.2 31.13.6 31.14.0 25.55.1
Experiment 3 Saline/Saline Saline/AVT Manning/Saline Manning/AVT
9 10 9 9
26.15.0 29.05.3 26.55.1 28.56.8
Attend early indicates attending scored for 10 min beginning 15 min post injection (meanSEM); attend late indicates attending scored for 10 min beginning 65 min post injection. Total attending indicates the average of attend early and attend late. Experiment 3: Manning and AVT each given at a dose of 1.0 mg/g body weight. * t -test comparing to Saline (0.9% NaCl) P B 0.001.
Regardless of the method of administration (IP or ICV), plasma levels of T and 11KT did not differ across treatments in samples taken approximately 80 min post injection (data not shown).
4. Discussion Numerous studies have identified a role for both AVP and AVT in stimulating sexual, affiliative and communicative behaviors in vertebrates (reviewed in Refs. [5,26,38]). Evidence for similar effects in fishes has been controversial because the original studies in killifishes examined a relatively simple behavior (the ‘spawning reflex’) that could have resulted from hormonal effects on peripheral tissues [28,31]. More recently, Goodson and Bass [10,12] provided clear evidence that AVT acts centrally to modulate activity in the neural circuits subserving vocal behavior in another teleost, the plainfin midshipman. This study complements the findings of Goodson and Bass by providing evidence that AVT acts centrally to stimulate sexual behavior in a freely-behaving teleost fish.
Fig. 1. The effect of ICV injection of saline or AVT on attending behavior in male white perch. Solid bars show attending scores (meanSEM) by treatment for the early time period (beginning at 15 min) while open bars depict the later time period (beginning at 65 min). Asterisk denotes significant difference; see text for details of statistical comparisons.
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
We found that IP administration of 0.150 mg/g body weight AVT produced a significant increase in attending behavior in male white perch in experiment 1 beginning at the later time point (65 min). However, we observed no effect at the same dose or even at the higher dose of 1.0 mg/g in experiment 2. The 1.0 mg/g dose used in experiment 2 is similar to that found to be behaviorally effective when delivered IP in several other species including two fishes (F. heteroclitus : [30]; T. bifasciatum : [37]; Rana catesbeianna: [2]; Bufo cognatus: [32]; Coturnix coturnix japonica, [4]). Similarly, Experiment 3 also failed to show treatment differences. Propper and Dixon [32] did find behavioral effects using Manning compound and a very similar sequential injection protocol in B. cognatus, but apparently did not anesthetize the animals. We cannot account for the very inconsistent effects with IP injections here, but the relative lack of effectiveness of peripheral administration is consistent with a central site of action. Administering AVT directly to the brain by ICV injection significantly increased attending behavior by 15 min after injection. Comparing the effectiveness of delivery, we found that ICV treatment produced the highest overall attending scores observed in these experiments and the effective dose injected into the brain was very small compared to the IP doses that were not effective (0.02 vs. 30 mg/fish). This effectiveness of ICV injections suggests that AVT is acting in the brain to mediate the effect on attending behavior. This finding is similar to previous findings in other species [4,25]. The ICV doses found to be effective in this study were comparable to those used in other species: P. notatus [13], Zonotrichia leucophrys gambelii [23], Taricha granulosa [25], but slightly higher than the minimum dose found to inhibit sexual behavior in male japanese quail (20 ng in this study vs. 2 ng in [4]). We did examine brains following behavioral testing to verify injection delivery into the brain, but did not assess potential damage caused by these injections histologically. While it is possible that such damage occurred, it appears unlikely that this would alter the conclusions drawn from this experiment as the damage would have had to increase attending behavior by males in both AVT-injected treatments relative to saline controls that would be presumed to have similar damage. Furthermore, the needle used had a small diameter (33 Ga) and the fish were active and did not exhibit obvious behavioral signs suggesting damage. The inhibition of sexual behaviors by AVT in male japanese quail stands in contrast to findings from males of other species such as those described here for white perch, rough-skinned newts, and other birds (reviewed in Ref. [3]). It may be relevant in this context that Castagna et al. studied castrated, T-implanted male quail that exhibited very high levels of sexual behavior. For example, control males in their experiments ob-
181
served females through a small window for approximately 85/90% of the duration of 5 min behavioral tests. In contrast, male white perch and male newts typically exhibit lower levels of sexual behavior in control treatments. Previous work has shown Manning compound to be effective inhibitor of sexual behavior when administered IP. For example, Propper and Dixon [32] found that Manning compound did reduce calling behavior induced by AVT injections in the toad B. cognatus when administered IP. Manning compound also effectively reduces courtship and territorial defense in bluehead wrasses when administered IP [37]. In contrast, we did not see a reduction in AVT-induced attending when Manning compound was administered IP 15 min prior to AVT or saline. All groups in experiment 3 exhibited high levels of attending, suggesting that attending should have been reduced if Manning compound were effective when delivered systemically. However, since IP injections were relatively ineffective in these experiments, we cannot conclude that Manning compound is not effective as an inhibitor in white perch. We found no evidence that the AVT and Manning compound treatments used in this study affected circulating levels of T and 11KT. There is therefore no evidence here that AVT-induced alterations in androgen levels could be having behavioral effects. However, we sampled at only a single time point and, in the case of ICV injections, approximately 1 h following the period when a significant effect on attending behavior was documented. This study was performed in a controlled laboratory setting, but several recent studies in fishes suggest AVT is an important mediator of reproductive behavior in the natural environment. The activity of the AVT system, as measured by levels of AVT mRNA or numbers of AVT immunoreactive cells, is correlated with behavioral phenotype in several species. Grober and Sumobe [15] found that when gobies (Trimma okinawae) undergo reversible sex change, there are correlated changes in the size of AVT producing cells in the forebrain. In the plainfin midshipman (P. notatus ), Foran and Bass [9] found that the smaller non-territorial (sneaker) males have smaller sized but greater numbers of AVT immunoreactive cells per gram body weight in the preoptic area of the brain than larger nest guarding males. In bluehead wrasses (T. bifasciatum ), the large territorial terminal phase males have greater hypothalamic AVT mRNA levels than females and non-territorial males, but these levels rise rapidly with female-to-male sex change [10,14]. Exogenous AVT can also induce courtship and aggressive behaviors in bluehead wrasses [37]. To our knowledge, this is the first demonstration that low doses of AVT administered centrally induce a behavioral response in a freely behaving fish that cannot be attributed to peripheral muscle activation (see also
182
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
Ref. [37]). Interestingly, attending behavior shows similarities to appetitive sexual and affiliative behaviors influenced by AVT/AVP in other species [3,37,38,42]. Studies in newts particularly show that AVT can increase responsiveness to somatosensory stimuli [33], which may be relevant here because of the frequent contact observed during the display of attending behavior in white perch. Our findings also build on the early work in killifish [29,30,39] along with recent work in the plainfin midshipman [11,12] and bluehead wrasse [37]. Taken together, these findings suggest AVT is an important mediator of sexual behavior across fishes and point to a conserved behavioral role for AVT and AVP in vertebrate animals.
Acknowledgements This work was supported, in part, by grants NA90AADS6062 and NA46ROG0087 to C.V.S from the National Sea Grant College Program (National Oceanic and Atmospheric Administration) to the North Carolina Sea Grant College Program. This work was also partially funded by a grant from the Sport Fishing Institute to C.V.S. and S.J.S. R.W. Clark and M.S. Hopper are acknowledged for producing the white perch used in these experiments. Dr A. McGinty and the staff of the North Carolina State University Pamlico Aquaculture Field Laboratory are appreciated for assistance with, and advice on, fish care. Dr J. Vandenbergh was generous in donating computer equipment used in the collection of behavioral data. Dr G. Weber provided helpful discussions throughout this project. This is a contribution of the W.M. Keck Center for Behavioral Biology at North Carolina State University.
References [1] Bluthe RM, Dantzer R. Role of the vomeronasal system in vasopressinergic modulation of social recognition in rats. Brain Res 1993;604:205 /10. [2] Boyd SK. Effect of vasotocin on locomotor activity in bullfrogs varies with developmental stage and sex. Horm Behav 1991;25:57 /69. [3] Boyd SK. Arginine vasotocin facilitation of advertisement calling and call phonotaxis in bullfrogs. Horm Behav 1994;28:232 /40 (DOI: 10.1006/hbeh.1994.1020). [4] Castagna C, Absil P, Foidart A, Balthazart J. Systemic and intracerebroventricular injections of vasotocin inhibit appetitive and consummatory components of male sexual behavior in japanese quail. Behav Neurosci 1998;112:233 /50. [5] De Vries GJ. Studying neurotransmitter systems to understand the development and function of sex differences in the brain: the case of vasopressin. In: Micevych PE, Hammer RP, editors. Neurobiological effects of sex steroid hormones. New York: Cambridge University Press, 1995:254 /80.
[6] Diakow C. A hormonal basis for breeding behavior in female frogs: vasotocin inhibits the release call of behavior Rana pipiens . Science 1978;199:1456 /7. [7] Ferris CF, Albers SM, Wesolowski SM, Goldman BD. Vasopressin injected into the hypothalamus triggers a stereotypic behavior in golden hamsters. Science 1984;224:521 /3. [8] Ferris CF, Melloni RH, Koppel G, Perry KW, Fuller RW, Delville Y. Vasopressin/serotonin interactions in the anterior hypothalamus control aggressive behavior in golden hamsters. J Neurosci 1997;17:4331 /40. [9] Foran CM, Bass AH. Preoptic GnRH and AVT: Axes for sexual plasticity in teleost fish. Gen Comp Endocrinol 1999;116:141 /52 (DOI:10.1006/gcen.1999.7357). [10] Godwin J, Sawby R, Warner RR, Crews D, Grober SG. Hypothalamic arginine vasotocin mRNA abundance variation across sexes and with sex change in a coral reef fish. Brain Behav Evol 2000;55:77 /84. [11] Goodson JL, Bass A. Forebrain peptides modulate sexually polymorphic vocal circuitry. Nature 2000a;403:769 /72. [12] Goodson JL, Bass A. Vasotocin innervation and modulation of vocal-acoustic circuitry in the teleost Porichthys notatus . J Comp Neurol 2000b;422:363 /79. [13] Goodson JL, Bass AH. Social behavior functions and related anatomical characteristics of vasotocin/vasopressin systems in vertebrates. Brain Res Rev 2001;35:246 /65. [14] Grammer M. The neurohormonal basis of individual behavioral variation in the bluehead wrasse, Thalassoma bifasciatum. Masters Thesis, North Carolina State University, Raleigh, NC; 1998. p. 44. [15] Grober MS, Sunobe T. Serial adult sex change involves rapid and reversible changes in forebrain neurochemistry. Neuroreport 1996;7:2945 /9. [16] Hennessey AC, Whitman DC, Albers HE. Microinjection of arginine /vasopressin into the periaqueductal gray stimulates flank marking in Syrian hamsters (Mesocricetus auratus ). Brain Res 1992;569:136 /40. [17] Kihlstrom JE, Danninge I. Neurohypophysial hormones and sexual behavior in males of the domestic fowl (Gallus domesticus L.) and the pigeon (Columba livia Gmel). Gen Comp Endocrinol 1972;18:115 /20. [18] King VW, Thomas P, Sullivan CV. Hormonal regulation of final maturation of striped bass oocytes in vitro. Gen Comp Endocrinol 1994;96:223 /33. [19] King WV, Berlinski DL, Sullivan CV. Involvement of gonadal steroids in the final oocyte maturation of white perch (Morone americana ) and white bass (M. chrysops ): in vivo and in vitro studies. Fish Phys Biochem 1995;14:489 /500. [20] Liley NR, Stacey NE. Hormones, pheromones and reproductive behavior. In: Hoar WS, Randall DJ, editors. Fish physiology, vol. 9B. New York: Academic Press, 1983:1 /63. [21] Macey MJ, Pickford GE, Peter RE. Forebrain localization of the spawning reflex response to exogenous neurohypophysial hormones in the killifish Fundulus heteroclitus . J Exp Zool 1974;190:269 /80. [22] Maney DL, Goode CT, Wingfield JC. Intraventricular infusion of arginine vasotocin induces singing in a female songbird. Neuroendocrinology 1997;9:487 /91. [23] Marler CA, Chu J, Wilczynski W. Arginine vasotocin injection increases the probability of calling in cricket frogs, but causes call changes characteristic of less aggressive males. Horm Behav 1995;29:554 /70 (DOI: 10.1006/hbeh.1995.1286). [24] Moore FL, Zoeller RT. Endocrine control of amphibian sexual behavior: evidence for a neurohormone /androgen interaction. Horm Behav 1979;13:203 /13. [25] Moore FL, Miller LJ. Arginine vasotocin induces sexual behavior of newts by acting on cells in the brain. Peptides 1983;4:97 /102.
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183 [26] Moore FL. Evolutionary precedents for behavioral actions of oxytocin and vasopressin. Ann Ny Acad Sci 1992;652:156 /65. [27] Moore FL, Wood RE, Boyd SK. Sex steroids and vasotocin interact in a female amphibian (Taricha granulosa ) to elicit female-like egg-laying behavior or male-like courtship. Horm Behav 1992;26(2):156 /66. [28] Peter RE. Preoptic nucleus in fishes-comparative discussion of function -activity relationships. Am Zool 1977;17:775 /85. [29] Pickford GE. Induction of a spawning reflex in a hypophysectomized killifish. Nature 1952;107:807 /8. [30] Pickford GE, Strecker EL. The spawning reflex response of the killifish Fundulus heteroclitus ; isotocin is relatively inactive in comparison with arginine vasotocin. Gen Comp Endocrinol 1977;32:132 /7. [31] Pickford GE, Knight WR, Knight JN. Where is the spawning reflex receptor for neurohypophyseal peptides in the killifish, Fundulus heteroclitus ? Rev Can Biol 1980;39:97 /105. [32] Propper RP, Dixon BD. Differential effects of arginine vasotocin and gonadotropin-releasing hormone on sexual behaviors in an anuran amphibian. Horm Behav 1997;32:99 /104 (DOI:10.1006/ hbeh.1997.1408). [33] Rose JD, Kinnaird JR, Moore FL. Neurophysiological effects of vasotocin and corticosterone on medullary neurons-implications for hormonal-control of amphibian courtship behavior. Neuroendocrinology 1995;62:406 /17. [34] Salek SJ, Sullivan CV, Godwin J. Courtship behavior of male white perch, Morone americana : evidence for control by androgens. Comp Biochem Physiol A 2001;130:731 /40.
183
[35] Salek SJ, Godwin J, Sullivan CV, Stacey NE. Courtship and Tank Spawning Behavior of Temperate Basses (genus Morone ). Trans Am Fish Soc 2001;130:833 /47. [36] Semsar KA, Klomberg KF, Marler C. Arginine vasotocin increases calling-site acquisition by nonresident male grey treefrogs. Anim Behav 1998;56:983 /7 (DOI: 10.1006/ anbe.1998.0863). [37] Semsar KA, Kandel FLM, Godwin J. Manipulation of the AVT system shifts social status and related courtship and aggressive behavior in the bluehead wrasse. Horm Behav 2001;40:21 /31. [38] Thompson RR, Moore FL. Vasotocin stimulates appetitive responses to the visual and pheromonal stimuli used by male roughskin newts during courtship. Horm Behav 2000;38:75 /85. [39] Wilhelmi AE, Pickford GE, Sawyer WH. Initiation of the spawning reflex response in fundulus by the administration of fish and mammalian neurohypophysial preparations and synthetic oxytocin. Endocrinology 1955;57:243 /52. [40] Winslow JT, Hastings N, Carter CS, Harbaugh CR, Insel RI. A role for central vasopressin in pair bonding in monogamous prairie voles. Nature 1993;365:545 /8. [41] Woods CWIII, Sullivan CV. Reproduction of striped bass, Morone saxatilis (Walbaum), broodstock: monitoring maturation and hormonal induction of spawning. J Aqua Fish Management 1993;24:213 /24. [42] Young L, Nilsen R, Waymire KG, MacGregor GR, Insel TR. Increased affiliative response to vasopressin in mice expressing the V1a receptor from a monogamous vole. Nature 1999;400:766 /8.
Research report
Arginine vasotocin effects on courtship behavior in male white perch (Morone americana) Stephen J. Salek, Craig V. Sullivan, John Godwin * Department of Zoology, North Carolina State University, Campus Box 7617, Raleigh, NC 27695, USA Received 14 August 2001; received in revised form 10 December 2001; accepted 10 December 2001
Abstract Arginine vasotocin (AVT) and its mammalian homologue, arginine vasopressin (AVP), have been shown to have widespread behavioral effects in vertebrates. AVT was evaluated for its effectiveness in stimulating an important courtship behavior termed ‘attending’ in male white perch, Morone americana. Attending consists of close and continuous following of the female with occasional contact in the abdominal area. We tested the behavioral effectiveness of AVT in stimulating attending when administered either intraperitoneally (IP) or intracerebroventricularly (ICV). We also tested IP injections of AVT alone and in combination with an AVP V1 receptor antagonist (Manning compound). None of the IP injections of either AVT or Manning compound produced consistent effects on attending behavior. In contrast, ICV injections of AVT did significantly increase attending behavior and at low dosages. Circulating levels of testosterone and 11-ketotestosterone were not affected approximately 80 min following injection by any of the treatments. The strong behavioral effects observed with ICV administration support a central site of action for AVT in stimulating attending behavior. This is a complex behavior that shows similarities to behaviors mediated by AVT and AVP in other vertebrates, providing further evidence of a conserved behavioral role for these peptides. # 2002 Published by Elsevier Science B.V. Keywords: Arginine vasotocin; Sexual behavior; Teleost; Androgen
1. Introduction The neuropeptides arginine vasopressin (AVP) and the homologous arginine vasotocin (AVT) in nonmammalian vertebrates have behavioral effects in a variety of vertebrate species (reviewed in Refs. [5,13,26]). These include effects on clasping and appetitive sexual behavior in newts (Taricha ) [24 /26,38], vocal and locomotor behavior in anuran amphibians [2,3,6,23,32,36], singing and sexual behavior in birds [4]17[22], and communicative, affiliative, and aggressive behavior in voles, mice, and hamsters (e.g. [1,7,8,16,40]). Similar findings have recently been published showing AVT effects on sexual, aggressive and communicative behaviors in teleost fishes [11,12,37]. The first demonstration of behavioral effects for neurohypophysial peptides was by Pickford [29], who showed that neurohypophysial extracts induced a
* Corresponding author. Tel.: 1-919-513-2396; fax: 1-919-5152698 E-mail address: [email protected] (J. Godwin).
‘spawning reflex’ in hypophysectomized killifish (Fundulus heteroclitus ). Subsequently, experiments with pure forms of specific neurohypophysial factors suggested AVT was responsible for this effect [30,39]. Isotocin (IST) could also induce this behavior but only at tenfold higher doses. In other closely related species of Cyprinodontids, AVT was shown to have similar effects [20,21,28]. Despite these studies and clear demonstrations of behavioral actions of AVT in other vertebrates, questions nevertheless remained regarding AVT effects on behavior in fishes until recently [11,12,37]. The spawning reflexes observed in Fundulus in response to injection with purified pituitary extracts were described by Pickford [29] as ‘violent S-shaped spasms’ during which ‘the fish showed no interest in each other but behaved independently’. The nature of this response and the large dosages necessary to elicit it in Fundulus and related species suggested that AVT could be acting peripherally through stimulation of smooth muscle [19,28]. Supporting this interpretation, both Peter [28] and Pickford et al. [31] found that AVT was no more effective when administered intracerebroventricularly
0166-4328/02/$ - see front matter # 2002 Published by Elsevier Science B.V. PII: S 0 1 6 6 - 4 3 2 8 ( 0 2 ) 0 0 0 0 3 - 7
178
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
(ICV) than intraperitoneally (IP). In contrast, Macey and coworkers [21] found that AVT-induced spawning reflexes in fish were essentially abolished by lesions of the preoptic nuclei, but not by lesions of comparable size in other brain regions, suggesting that AVT acted in the preoptic area. As noted above, Goodson and Bass [11,12] recently showed clear effects of AVT and IST on activity in neural circuits directly subserving vocalization behavior when these neuropeptides were administered directly into the preoptic area-anterior hypothalamus or midbrain tegmentum of the plainfin midshipman (Porichthys notatus ). The fish used in the studies of Goodson and Bass were anesthetized and the response measured was neuronal activity. Semsar and coworkers [37] recently showed that IP injections of AVT affected complex behaviors including sexual behavior, aggression, and feeding in the bluehead wrasse, Thalassoma bifasciatum . The primary courtship behavior of white perch (Morone americana ) leading to the spawning act is termed ‘attending’. Attending is close (within 30 cm) and continuous following of a female by a male and includes occasional contact of her abdominal area with the snout (described fully in [35]). In contrast to the spawning reflex of Cyprinodontid fishes, attending is a complex behavior which shows some strong similarities to appetitive sexual behaviors influenced by AVT in roughskin newts, frogs, and bluehead wrasses [3,37,38] and some similarities to affiliative behaviors affected by AVP in voles [42]. We therefore examined AVT effects on attending behavior of white perch in this study and compared the effectiveness of peripheral (IP) to central (ICV) administration of the peptide. We also measured plasma androgens since the androgen 11-ketotestosterone (11KT) influences the display of attending in white perch [34] and gonadal steroids critically affect the nature of AVT-induced reproductive behaviors in the rough-skinned newt [27]. We therefore asked whether the AVT and Manning treatments used here would influence plasma androgen levels (and potentially behavior by this route).
2. Materials and methods 2.1. Experimental animals We used white perch broodstock maintained at the North Carolina State University Pamlico Aquaculture Field Laboratory. They were transported to the Aquatic Research Facility on the North Carolina State University campus and maintained as described previously until testing [19]. We conducted these experiments during the breeding season for white perch in March, April, and May of 1998 and 1999. The white perch males used were 207.39/5.8 g and 237.99/2.3 mm (mean9/
SEM, N /94) in total length. All males were spermiating at the start of studies. Animal husbandry and experimental procedures followed the North Carolina State University Institutional Animal Care and Use Committee guidelines. 2.2. Injections We performed a series of experiments involving IP injections (experiments 1 /3) or ICV injections (experiment 4) to test behavioral effects of AVT and the AVP V1 receptor antagonist Manning compound. Experiment 1 compared attending behavior in males given either saline (0.9% NaCl) or 0.150 mg/g body weight AVT IP. Experiment 2 compared attending among males given saline, 0.150, or 1.0 mg/g AVT IP to test the repeatability of results from experiment 1 and test for a dose response. Experiment 3 tested the effect of a putative AVT receptor antagonist, the AVP V1 receptor antagonist Manning compound (2-Mercapto-beta,(CH2)5,O /Me /Tyr2, Arg8]-Vasopressin) on attending behavior when given IP both alone and in combination with AVT (1 mg/g body weight for both AVT and Manning compound; these doses were used by Propper and Dixon [32]). Experiment 4 tested the effects of AVT on attending behavior when delivered ICV. Injections were given under anesthesia (200 mg/l tricaine methanesulfonate in freshwater; Argent, Redmond, WA). Fish were anesthetized once for IP injections in experiments 1 and 2 and twice for experiment 3 which involved two separate injections. Fish were anesthetized once for ICV injections in experiment 4. We gave IP injections of peptides (AVT and Manning compound, Sigma, St. Louis, MO) with a 1-cc Luer LOK† syringe fitted with a 25 Ga needle (Becton Dickinson, Franklin Lakes, NJ) given in 100 /300 ml of 0.9% NaCl. ICV injections of peptides were made freehand into the third ventricle of the brain in 2 ml of 0.9% NaCl containing 10% Sheaffer ink (v/v) (Ft Madison, IA). Control animals were given 0.9% NaCl and ink only. Peptide solutions were freshly prepared and kept on ice. Syringes were filled just prior to injection. A small hole was made in the skull with a 20 Ga needle and then a 33 Ga needle fitted to a 10 ml Hamilton syringe (Reno, NV) was inserted into the hole to a depth of 8 mm. Lining up the syringe with the caudal side of the orbit and angling the syringe perpendicular to the head surface placed the needle in the middle of the brain approximately in the hypothalamus. Preliminary studies in which we sectioned each brain showed that ink was delivered to the ventricular system if no ink was found on the surface of the brain or in the braincase. The depth of the needle was controlled by placing a 200 ml pipette tip over the 33 Ga needle and trimming it so only 8 mm of the needle projected from the tip. We waited 10 s after inserting the needle, injected
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
2 ml over a period of 10 s, then waited 10 s more before removing the syringe. The hole was filled with Triple X antibiotic ointment (Food Lion, Salisbury, NC). Accuracy of injection was confirmed in all studies by opening the skull of each fish after behavioral trials and visually inspecting the brain. Samples with ink outside the brain were excluded from analyses. Both IP and ICV injections were completed rapidly (B/120 s) and fish were immediately returned to aerated water. Since the time out of water was brief, and in the interest of delivering injections rapidly, the gills of the fish were not irrigated during injections. 2.3. Behavior tests We evaluated the display of attending behavior using an ovulated female as the stimulus for male courtship behavior. To induce ovulation, several female perch were injected intramuscularly with human chorionic gonadotropin (hCG) (Schein Pharmaceuticals, Florham Park, NJ) at a dose of 330 IU/kg body weight 24 /48 h prior to behavioral testing. We assessed maturation of oocytes by anesthetizing females in quinaldine sulfate (50 mg/l) and subjecting them to ovarian biopsy [17]. Only females yielding ovulated eggs were used. Since males and females were similar in size, females were tagged with a silver streamer made of MylarTM tape (8 cm) inserted through the dorsal fin to mark them for identification during observations. Individual experimental males were captured from a holding tank with a hand net, given an injection (IP or ICV as described above), and placed into a holding tank for 15 min to recover from anesthesia. After this period, the experimental male was introduced into the test tank with the ovulated female. During the test period, attending was scored by one observer. Attending is defined as close (maximum 30 cm, but usually less) and continuous investigation of the female’s abdominal area which may include physical contact [34]. Male body sizes averaged approximately 24 cm, aiding in judging whether they were within 30 cm of the females abdominal area. Also, observers were blind to treatment to prevent bias in scoring this behavioral measure. The total time spent attending by a male was recorded by one observer over two 10 min intervals; the first interval began immediately after the male was introduced into the tank and the second began 50 min after the introduction. Following the second 10 min interval, the male was removed from the tank. Since males were given a 15 min recovery period after injection before being introduced to the test tank, these observation times began 15 and 65 min after injection. We performed behavior tests in 684-l rectangular tanks (1.9 /0.6 /0.6 m3), each illuminated with five 75 W red floodlights (Sylvania). Direct observations of courtship behavior were made through blinds located at
179
both ends of the tanks. Most testing occurred between 07:00 and 16:00 (65 males) and some occurred between 16:00 and 22:00 (29 males). There were no apparent differences in behavior between these two testing periods and behavioral tests were balanced across treatments for each experiment to prevent any consistent effect of testing time. All behavioral data presented and compared statistically in this paper are based on direct observation during behavioral testing. To assess the precision of these direct behavioral observations, attending was also rescored from videotapes and compared to direct observations using correlation analyses. For each behavioral test, a VHS video camera (Panasonic ag-187u, Seacaucus, NJ) equipped with a standard zoom lens (Panasonic Power Zoom /12) mated with a wide-angle lens (Panasonic Super Wide Pro 5050-0.5 /) and polarizing filter (Toshiba PL 49.0(s), Japan) was installed over each tank on a ball mount (Bogon, Italy). In our first experiment, all male scores were compared in this manner. In subsequent experiments, only four male’s scores per experiment were compared in this manner. Attending scores from direct observations and videotape analysis were highly and significantly correlated (n /15, R /0.94, P B/0.0001). 2.4. Blood sampling and hormone assays Immediately after behavior testing, fish were removed from the tanks, anesthetized in MS-222 (200 mg/ml), rinsed in fresh water, weighed, and bled by caudal puncture using a 3.0-cc syringe fitted with a 21-gauge heparinized needle. Approximately 1.5 ml blood was collected from each fish. Blood samples were stored on ice until being centrifuged (10 min at 10 000 /g ). The plasma was drawn off and stored at /80 8C until assay. Concentrations of 11KT and testosterone (T) were measured in duplicate 20 ml aliquots of blood plasma by highly specific RIAs. Briefly, we triple extracted samples in 3 ml of diethyl ether, dried the ether phase under a stream of nitrogen gas at 37 8C, and resuspended in 200 ml assay buffer for specific radioimmunoasays as described by Woods and Sullivan [41]. 2.5. Statistical analyses We log10 transformed data to reduce heteroschedasticity among treatment groups and performed statistical analyses using JMP statistical software (SAS, Cary, NC). We compared behavioral scores by either Student’s t-tests or one way analysis of variance (ANOVA) if there were more than two treatments. Individual means were compared following ANOVA using JMP’s ‘compare all pairs’ command to execute a Tukey / Kramer HSD comparison. Associations between attend-
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
180
ing scores from videotapes and direct observations were assessed by simple linear correlation (Pearson’s r).
3.2. Behavioral actions of AVT: effects of ICV administration
3. Results
In experiment 4, we injected males ICV with a saline control solution, 0.02 mg AVT, or 0.2 mg AVT. AVT significantly stimulated attending when administered at 0.02 mg/fish and 0.2 mg/fish (ANOVA, F2,11 /8.268, P B/0.01; Tukey /Kramer post hoc tests P B/0.05) (Fig. 1).
3.1. Behavioral actions of AVT: effects of IP administration
3.3. Effects of AVT: circulating androgen levels
IP injections of AVT and/or Manning compound in three separate experiments produced inconsistent results. In experiment 1, there was a significant increase in attending behavior by AVT-injected males relative to saline controls 65 min after injection (two-tailed t test, t /2.93, P /0.0013, 0.150 mg/g body weight), but not 15 min after injection (t /0.779 P /0.45, Table 1). We performed Experiment 2 identically to Experiment 1, but incorporated an additional dose of AVT. However, neither AVT dose (0.150 and 1.0 mg/g body weight) increased attending significantly relative to saline controls. In experiment 3, we found no significant differences among treatments where males received sequential injections spaced 15 min apart (saline/saline, saline/ AVT, Manning/saline, or Manning/AVT; AVT and Manning at 1.0 mg/g body weight (ANOVA, P /0.05 in all cases). The attending scores for males in all treatments in experiment 3 were at the high end of the range seen for males in experiments 1 and 2.
Table 1 Effect of IP AVT injections on courtship behavior Treatments attending
N
Attend early
Attend late
Total
Injected IP Experiment 1 Saline 0.150 mg/g
6 17.66.4 7 23.45.6
19.64.3 37.14.0*
18.64.0 30.74.8
Experiment 2 Saline 0.150 mg/g 1.0 mg/g
5 16.82.1 7 13.33.1 8 28.55.8
21.33.9 21.65.3 31.34.1
19.01.6 17.54.0 30.01.9
34.04.5 33.34.8 35.85.6 22.64.6
30.54.2 31.13.6 31.14.0 25.55.1
Experiment 3 Saline/Saline Saline/AVT Manning/Saline Manning/AVT
9 10 9 9
26.15.0 29.05.3 26.55.1 28.56.8
Attend early indicates attending scored for 10 min beginning 15 min post injection (meanSEM); attend late indicates attending scored for 10 min beginning 65 min post injection. Total attending indicates the average of attend early and attend late. Experiment 3: Manning and AVT each given at a dose of 1.0 mg/g body weight. * t -test comparing to Saline (0.9% NaCl) P B 0.001.
Regardless of the method of administration (IP or ICV), plasma levels of T and 11KT did not differ across treatments in samples taken approximately 80 min post injection (data not shown).
4. Discussion Numerous studies have identified a role for both AVP and AVT in stimulating sexual, affiliative and communicative behaviors in vertebrates (reviewed in Refs. [5,26,38]). Evidence for similar effects in fishes has been controversial because the original studies in killifishes examined a relatively simple behavior (the ‘spawning reflex’) that could have resulted from hormonal effects on peripheral tissues [28,31]. More recently, Goodson and Bass [10,12] provided clear evidence that AVT acts centrally to modulate activity in the neural circuits subserving vocal behavior in another teleost, the plainfin midshipman. This study complements the findings of Goodson and Bass by providing evidence that AVT acts centrally to stimulate sexual behavior in a freely-behaving teleost fish.
Fig. 1. The effect of ICV injection of saline or AVT on attending behavior in male white perch. Solid bars show attending scores (meanSEM) by treatment for the early time period (beginning at 15 min) while open bars depict the later time period (beginning at 65 min). Asterisk denotes significant difference; see text for details of statistical comparisons.
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
We found that IP administration of 0.150 mg/g body weight AVT produced a significant increase in attending behavior in male white perch in experiment 1 beginning at the later time point (65 min). However, we observed no effect at the same dose or even at the higher dose of 1.0 mg/g in experiment 2. The 1.0 mg/g dose used in experiment 2 is similar to that found to be behaviorally effective when delivered IP in several other species including two fishes (F. heteroclitus : [30]; T. bifasciatum : [37]; Rana catesbeianna: [2]; Bufo cognatus: [32]; Coturnix coturnix japonica, [4]). Similarly, Experiment 3 also failed to show treatment differences. Propper and Dixon [32] did find behavioral effects using Manning compound and a very similar sequential injection protocol in B. cognatus, but apparently did not anesthetize the animals. We cannot account for the very inconsistent effects with IP injections here, but the relative lack of effectiveness of peripheral administration is consistent with a central site of action. Administering AVT directly to the brain by ICV injection significantly increased attending behavior by 15 min after injection. Comparing the effectiveness of delivery, we found that ICV treatment produced the highest overall attending scores observed in these experiments and the effective dose injected into the brain was very small compared to the IP doses that were not effective (0.02 vs. 30 mg/fish). This effectiveness of ICV injections suggests that AVT is acting in the brain to mediate the effect on attending behavior. This finding is similar to previous findings in other species [4,25]. The ICV doses found to be effective in this study were comparable to those used in other species: P. notatus [13], Zonotrichia leucophrys gambelii [23], Taricha granulosa [25], but slightly higher than the minimum dose found to inhibit sexual behavior in male japanese quail (20 ng in this study vs. 2 ng in [4]). We did examine brains following behavioral testing to verify injection delivery into the brain, but did not assess potential damage caused by these injections histologically. While it is possible that such damage occurred, it appears unlikely that this would alter the conclusions drawn from this experiment as the damage would have had to increase attending behavior by males in both AVT-injected treatments relative to saline controls that would be presumed to have similar damage. Furthermore, the needle used had a small diameter (33 Ga) and the fish were active and did not exhibit obvious behavioral signs suggesting damage. The inhibition of sexual behaviors by AVT in male japanese quail stands in contrast to findings from males of other species such as those described here for white perch, rough-skinned newts, and other birds (reviewed in Ref. [3]). It may be relevant in this context that Castagna et al. studied castrated, T-implanted male quail that exhibited very high levels of sexual behavior. For example, control males in their experiments ob-
181
served females through a small window for approximately 85/90% of the duration of 5 min behavioral tests. In contrast, male white perch and male newts typically exhibit lower levels of sexual behavior in control treatments. Previous work has shown Manning compound to be effective inhibitor of sexual behavior when administered IP. For example, Propper and Dixon [32] found that Manning compound did reduce calling behavior induced by AVT injections in the toad B. cognatus when administered IP. Manning compound also effectively reduces courtship and territorial defense in bluehead wrasses when administered IP [37]. In contrast, we did not see a reduction in AVT-induced attending when Manning compound was administered IP 15 min prior to AVT or saline. All groups in experiment 3 exhibited high levels of attending, suggesting that attending should have been reduced if Manning compound were effective when delivered systemically. However, since IP injections were relatively ineffective in these experiments, we cannot conclude that Manning compound is not effective as an inhibitor in white perch. We found no evidence that the AVT and Manning compound treatments used in this study affected circulating levels of T and 11KT. There is therefore no evidence here that AVT-induced alterations in androgen levels could be having behavioral effects. However, we sampled at only a single time point and, in the case of ICV injections, approximately 1 h following the period when a significant effect on attending behavior was documented. This study was performed in a controlled laboratory setting, but several recent studies in fishes suggest AVT is an important mediator of reproductive behavior in the natural environment. The activity of the AVT system, as measured by levels of AVT mRNA or numbers of AVT immunoreactive cells, is correlated with behavioral phenotype in several species. Grober and Sumobe [15] found that when gobies (Trimma okinawae) undergo reversible sex change, there are correlated changes in the size of AVT producing cells in the forebrain. In the plainfin midshipman (P. notatus ), Foran and Bass [9] found that the smaller non-territorial (sneaker) males have smaller sized but greater numbers of AVT immunoreactive cells per gram body weight in the preoptic area of the brain than larger nest guarding males. In bluehead wrasses (T. bifasciatum ), the large territorial terminal phase males have greater hypothalamic AVT mRNA levels than females and non-territorial males, but these levels rise rapidly with female-to-male sex change [10,14]. Exogenous AVT can also induce courtship and aggressive behaviors in bluehead wrasses [37]. To our knowledge, this is the first demonstration that low doses of AVT administered centrally induce a behavioral response in a freely behaving fish that cannot be attributed to peripheral muscle activation (see also
182
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183
Ref. [37]). Interestingly, attending behavior shows similarities to appetitive sexual and affiliative behaviors influenced by AVT/AVP in other species [3,37,38,42]. Studies in newts particularly show that AVT can increase responsiveness to somatosensory stimuli [33], which may be relevant here because of the frequent contact observed during the display of attending behavior in white perch. Our findings also build on the early work in killifish [29,30,39] along with recent work in the plainfin midshipman [11,12] and bluehead wrasse [37]. Taken together, these findings suggest AVT is an important mediator of sexual behavior across fishes and point to a conserved behavioral role for AVT and AVP in vertebrate animals.
Acknowledgements This work was supported, in part, by grants NA90AADS6062 and NA46ROG0087 to C.V.S from the National Sea Grant College Program (National Oceanic and Atmospheric Administration) to the North Carolina Sea Grant College Program. This work was also partially funded by a grant from the Sport Fishing Institute to C.V.S. and S.J.S. R.W. Clark and M.S. Hopper are acknowledged for producing the white perch used in these experiments. Dr A. McGinty and the staff of the North Carolina State University Pamlico Aquaculture Field Laboratory are appreciated for assistance with, and advice on, fish care. Dr J. Vandenbergh was generous in donating computer equipment used in the collection of behavioral data. Dr G. Weber provided helpful discussions throughout this project. This is a contribution of the W.M. Keck Center for Behavioral Biology at North Carolina State University.
References [1] Bluthe RM, Dantzer R. Role of the vomeronasal system in vasopressinergic modulation of social recognition in rats. Brain Res 1993;604:205 /10. [2] Boyd SK. Effect of vasotocin on locomotor activity in bullfrogs varies with developmental stage and sex. Horm Behav 1991;25:57 /69. [3] Boyd SK. Arginine vasotocin facilitation of advertisement calling and call phonotaxis in bullfrogs. Horm Behav 1994;28:232 /40 (DOI: 10.1006/hbeh.1994.1020). [4] Castagna C, Absil P, Foidart A, Balthazart J. Systemic and intracerebroventricular injections of vasotocin inhibit appetitive and consummatory components of male sexual behavior in japanese quail. Behav Neurosci 1998;112:233 /50. [5] De Vries GJ. Studying neurotransmitter systems to understand the development and function of sex differences in the brain: the case of vasopressin. In: Micevych PE, Hammer RP, editors. Neurobiological effects of sex steroid hormones. New York: Cambridge University Press, 1995:254 /80.
[6] Diakow C. A hormonal basis for breeding behavior in female frogs: vasotocin inhibits the release call of behavior Rana pipiens . Science 1978;199:1456 /7. [7] Ferris CF, Albers SM, Wesolowski SM, Goldman BD. Vasopressin injected into the hypothalamus triggers a stereotypic behavior in golden hamsters. Science 1984;224:521 /3. [8] Ferris CF, Melloni RH, Koppel G, Perry KW, Fuller RW, Delville Y. Vasopressin/serotonin interactions in the anterior hypothalamus control aggressive behavior in golden hamsters. J Neurosci 1997;17:4331 /40. [9] Foran CM, Bass AH. Preoptic GnRH and AVT: Axes for sexual plasticity in teleost fish. Gen Comp Endocrinol 1999;116:141 /52 (DOI:10.1006/gcen.1999.7357). [10] Godwin J, Sawby R, Warner RR, Crews D, Grober SG. Hypothalamic arginine vasotocin mRNA abundance variation across sexes and with sex change in a coral reef fish. Brain Behav Evol 2000;55:77 /84. [11] Goodson JL, Bass A. Forebrain peptides modulate sexually polymorphic vocal circuitry. Nature 2000a;403:769 /72. [12] Goodson JL, Bass A. Vasotocin innervation and modulation of vocal-acoustic circuitry in the teleost Porichthys notatus . J Comp Neurol 2000b;422:363 /79. [13] Goodson JL, Bass AH. Social behavior functions and related anatomical characteristics of vasotocin/vasopressin systems in vertebrates. Brain Res Rev 2001;35:246 /65. [14] Grammer M. The neurohormonal basis of individual behavioral variation in the bluehead wrasse, Thalassoma bifasciatum. Masters Thesis, North Carolina State University, Raleigh, NC; 1998. p. 44. [15] Grober MS, Sunobe T. Serial adult sex change involves rapid and reversible changes in forebrain neurochemistry. Neuroreport 1996;7:2945 /9. [16] Hennessey AC, Whitman DC, Albers HE. Microinjection of arginine /vasopressin into the periaqueductal gray stimulates flank marking in Syrian hamsters (Mesocricetus auratus ). Brain Res 1992;569:136 /40. [17] Kihlstrom JE, Danninge I. Neurohypophysial hormones and sexual behavior in males of the domestic fowl (Gallus domesticus L.) and the pigeon (Columba livia Gmel). Gen Comp Endocrinol 1972;18:115 /20. [18] King VW, Thomas P, Sullivan CV. Hormonal regulation of final maturation of striped bass oocytes in vitro. Gen Comp Endocrinol 1994;96:223 /33. [19] King WV, Berlinski DL, Sullivan CV. Involvement of gonadal steroids in the final oocyte maturation of white perch (Morone americana ) and white bass (M. chrysops ): in vivo and in vitro studies. Fish Phys Biochem 1995;14:489 /500. [20] Liley NR, Stacey NE. Hormones, pheromones and reproductive behavior. In: Hoar WS, Randall DJ, editors. Fish physiology, vol. 9B. New York: Academic Press, 1983:1 /63. [21] Macey MJ, Pickford GE, Peter RE. Forebrain localization of the spawning reflex response to exogenous neurohypophysial hormones in the killifish Fundulus heteroclitus . J Exp Zool 1974;190:269 /80. [22] Maney DL, Goode CT, Wingfield JC. Intraventricular infusion of arginine vasotocin induces singing in a female songbird. Neuroendocrinology 1997;9:487 /91. [23] Marler CA, Chu J, Wilczynski W. Arginine vasotocin injection increases the probability of calling in cricket frogs, but causes call changes characteristic of less aggressive males. Horm Behav 1995;29:554 /70 (DOI: 10.1006/hbeh.1995.1286). [24] Moore FL, Zoeller RT. Endocrine control of amphibian sexual behavior: evidence for a neurohormone /androgen interaction. Horm Behav 1979;13:203 /13. [25] Moore FL, Miller LJ. Arginine vasotocin induces sexual behavior of newts by acting on cells in the brain. Peptides 1983;4:97 /102.
S.J. Salek et al. / Behavioural Brain Research 133 (2002) 177 /183 [26] Moore FL. Evolutionary precedents for behavioral actions of oxytocin and vasopressin. Ann Ny Acad Sci 1992;652:156 /65. [27] Moore FL, Wood RE, Boyd SK. Sex steroids and vasotocin interact in a female amphibian (Taricha granulosa ) to elicit female-like egg-laying behavior or male-like courtship. Horm Behav 1992;26(2):156 /66. [28] Peter RE. Preoptic nucleus in fishes-comparative discussion of function -activity relationships. Am Zool 1977;17:775 /85. [29] Pickford GE. Induction of a spawning reflex in a hypophysectomized killifish. Nature 1952;107:807 /8. [30] Pickford GE, Strecker EL. The spawning reflex response of the killifish Fundulus heteroclitus ; isotocin is relatively inactive in comparison with arginine vasotocin. Gen Comp Endocrinol 1977;32:132 /7. [31] Pickford GE, Knight WR, Knight JN. Where is the spawning reflex receptor for neurohypophyseal peptides in the killifish, Fundulus heteroclitus ? Rev Can Biol 1980;39:97 /105. [32] Propper RP, Dixon BD. Differential effects of arginine vasotocin and gonadotropin-releasing hormone on sexual behaviors in an anuran amphibian. Horm Behav 1997;32:99 /104 (DOI:10.1006/ hbeh.1997.1408). [33] Rose JD, Kinnaird JR, Moore FL. Neurophysiological effects of vasotocin and corticosterone on medullary neurons-implications for hormonal-control of amphibian courtship behavior. Neuroendocrinology 1995;62:406 /17. [34] Salek SJ, Sullivan CV, Godwin J. Courtship behavior of male white perch, Morone americana : evidence for control by androgens. Comp Biochem Physiol A 2001;130:731 /40.
183
[35] Salek SJ, Godwin J, Sullivan CV, Stacey NE. Courtship and Tank Spawning Behavior of Temperate Basses (genus Morone ). Trans Am Fish Soc 2001;130:833 /47. [36] Semsar KA, Klomberg KF, Marler C. Arginine vasotocin increases calling-site acquisition by nonresident male grey treefrogs. Anim Behav 1998;56:983 /7 (DOI: 10.1006/ anbe.1998.0863). [37] Semsar KA, Kandel FLM, Godwin J. Manipulation of the AVT system shifts social status and related courtship and aggressive behavior in the bluehead wrasse. Horm Behav 2001;40:21 /31. [38] Thompson RR, Moore FL. Vasotocin stimulates appetitive responses to the visual and pheromonal stimuli used by male roughskin newts during courtship. Horm Behav 2000;38:75 /85. [39] Wilhelmi AE, Pickford GE, Sawyer WH. Initiation of the spawning reflex response in fundulus by the administration of fish and mammalian neurohypophysial preparations and synthetic oxytocin. Endocrinology 1955;57:243 /52. [40] Winslow JT, Hastings N, Carter CS, Harbaugh CR, Insel RI. A role for central vasopressin in pair bonding in monogamous prairie voles. Nature 1993;365:545 /8. [41] Woods CWIII, Sullivan CV. Reproduction of striped bass, Morone saxatilis (Walbaum), broodstock: monitoring maturation and hormonal induction of spawning. J Aqua Fish Management 1993;24:213 /24. [42] Young L, Nilsen R, Waymire KG, MacGregor GR, Insel TR. Increased affiliative response to vasopressin in mice expressing the V1a receptor from a monogamous vole. Nature 1999;400:766 /8.