paternal behavior is associated with central...

Behavioral Neuroscience2001. Vol. 115. No. 6. 1341-1348

Copyright 2001 by the American Psychological Association, Inc.0735-7044/D1/$5.00 DOI: 10.1037//0735-7044.115.6.134I

Paternal Behavior Is Associated With Central Neurohormone ReceptorBinding Patterns in Meadow Voles (Microtus pennsylvanicus)

Karen J. Parker, Lisa F. Kinney, Katie M. Phillips, and Theresa M. LeeUniversity of Michigan

Paternal and nonpaternal voles (microtus) have different arginine-vasopressin (AVP) and oxytocin (OT)receptor patterns in the extended amygdala, a neural pathway associated with parental behavior. Usingreceptor autoradiography, the authors examined whether AVP and OT receptor patterns were associatedwith facultative paternal behavior in either sexually and parentally inexperienced or experienced meadowvoles (Microtus pennsylvanicus). Experienced, in contrast to inexperienced, males had less AVP bindingin the lateral septum (LS), more AVP binding in the anterior olfactory nucleus (AON), and more OTbinding in the AON, bed nucleus of the stria terminalis, LS, and lateral amygdala. Thus, specific AVPreceptor patterns, which co-occur with paternal care in consistently paternal voles, also may be associatedwith paternal care (when present) in typically nonpaternal species. This study also demonstrated apossible relationship between OT receptor patterns and paternal state in male mammals.

Paternal care, in its broadest sense is defined as "all nongameticinvestments in offspring following fertilization" (Wittenberger,1981). In rodents, the presence of paternal care is often associatedwith social living and harsh breeding conditions (Kleiman &Malcolm, 1981) and may be either a consistent or facultativefeature of a given social system. For example, Peromyscus cali-fomicus (California mice) live in habitats with patchy resourcebases and engage in consistent biparental care to increase offspringsurvivorship in this resource-poor environment (Gubernick,Wright, & Brown, 1993). In contrast, typically nonparental rodentsmay also engage in parental care, but, rather than consistentlyoccurring, facultative paternal behavior is usually expressed situ-ationally to offset fitness costs associated with unpredictable socialor seasonal changes: decreased access to mates or compromisedoffspring survivorship because of a patchy resource base, highpredation risk, or increased offspring energy needs (Kleiman,1977). Such facultative care has been reported in typically nonpa-ternal Peromyscus maniculatus (deer mice; Mihok, 1979), Mar-mota caligata (hoary marmots; Barash, 1975), and Peromyscus

Karen J. Parker, Lisa F. Kinney, Katie M. Phillips, and Theresa M. Lee,Department of Psychology and Reproductive Sciences Program, Universityof Michigan.

Karen J. Parker is now at the Department of Psychiatry and BehavioralSciences, Stanford University School of Medicine.

This research was supported by National Institutes of Health Reproduc-tive Sciences Program Training Gram 5T32HD07048 to Karen J. Parker,and by National Institute of Child Health and Human Development GrantHD24575 and National Science Foundation Grant IBN9808814 to TheresaM. Lee. We thank Tammy Jechura for expert computer assistance, IsraelLiberzon for the receptor binding protocol, and Bob Thompson for thegenerous use of laboratory facilities and equipment.

Correspondence concerning this article should be addressed to TheresaM. Lee, Department of Psychology, University of Michigan, 525 EastUniversity Avenue, Ann Arbor, Michigan 48109-1109. Electronic mailmay be sent to [email protected].

leucopus (white-footed mice; Schug, Vessey, & Underwood,1992) during colder months.

To date, both the evolutionary rationale for paternal effortexpenditure (Kleiman & Malcolm, 1981; Trivers, 1972) and thesocial and environmental regulation of paternal behavior expres-sion (reviewed by Dewsbury, 1985) have been well elucidated forboth consistently and facultatively paternal species. However, theneural pathways that mediate the expression of paternal behaviorhave been studied primarily in consistently paternal species (re-viewed by Wang, Young, De Vries, & Insel, 1998). Thus, it is notyet known whether the same neurobiology regulates paternal be-havior when it is facultatively, rather than consistently, initiated.Nevertheless, our understanding of the neurobiology that promotespaternal behavior has been greatly advanced by studies conductedin Microtus, an ideal genus in which to investigate the underlyingneural circuits associated with the presence and absence of pairbonding and paternal behavior in closely related rodent species.

Initial studies demonstrated that sexually and parentally inex-perienced monogamous Microtus ochrogaster (prairie voles) andnonmonogamous Microtus montanus (montane voles) showedconsiderable differences in arginine-vasopressin (AVP) and oxy-tocin (OT) receptor patterns in the extended amygdala (Insel &Shapiro, 1992; Insel, Wang, & Ferris, 1994). The extended amyg-dala neural pathway is composed of the accessory olfactory nu-cleus (AON), amygdala, bed nucleus of the stria terminalis(BNST), lateral septum (LS), and medial preoptic area of thehypothalamus (MPOA) and is important for the initiation of af-filiative, mating, and parenting behaviors in several rodent species(reviewed by Newman, 1999). Because monogamous species oftenconcomitantly exhibit paternal care, partner preferences, andstranger-directed aggression and nonmonogamous species do not,it has been suggested that receptor differences in this pathwayreflect species differences in evolved patterns of social organiza-tion (Wang et al., 1998).

Since these first studies were published, a working model of theunderlying neurobiology of male prairie vole social and affiliative

1341

1342 PARKER, KINNEY, PHILLIPS, AND LEE

behaviors has emerged. Behavioral research has shown that maleprairie voles develop selective partner preferences after 24 hr ofmated cohabitation with a female (Insel, Preston, & Winslow,1995), and these behavioral changes are associated with enhancedpaternal responsiveness (Bamshad, Novak & De Vries, 1994).Several pharmacological experiments have implicated a role forAVP (Winslow, Hastings, Carter, Harbaugh, & Insel, 1993) andOT (Cho, De Vries, Williams, & Carter, 1999) in male partnerpreference onset. Although a functional role for OT has not yetbeen investigated in regulating paternal behavior in voles, centraladministration of AVP into either the lateral ventricle or the LSenhances paternal behavior expression (Wang, Ferris, & De Vries,1994).

Several lesion studies (bilateral bulbectomy in Kirkpatrick, Wil-liams, Slotnick, & Carter, 1994; corticomedial and medial nucleiof the amygdala [MeAmyg] in Kirkpatrick, Carter, Newman, &Insel, 1994) in combination with a number of early gene (c-fos)expression (indexed by immunoreactivity [Fos-ir]), AVP-ir fiberdensity, and AVP messenger RNA (mRNA) peptide expressionstudies have also been helpful in indicating a likely role forspecific neural pathways in regulating paternal care in prairievoles. The importance of the extended amygdala is supported byincreased Fos-ir in the MeAmyg (Kirkpatrick, Kim, & Insel, 1994)and MPOA (Wang, Hulihan, & Insel, 1997) after male interactionswith pups. After mating and cohabiting with a female, males showincreased AVP mRNA expression in BNST cell bodies and de-creased AVP-ir fibers in the LS and lateral habenular nucleus(Bamshad et al., 1994; Wang, Smith, Major, & De Vries, 1994, butalso see Lonstein & De Vries, 1999), and these differences areassociated with increased paternal responsiveness (Bamshad et al.,1994).

The extended amygdala also contains significant numbers ofAVP receptors (AON-amygdala-BNST-LS; Insel et al., 1994;Wang, Young, Liu, & Insel, 1997; Wang, Liu, Young, & Insel,2000). Because paternal behavior is associated with increasedsynthesis and release of AVP (Bamshad et al., 1994), and centralAVP receptor binding studies have found no detectable differencesin receptor number or location between sexually inexperiencedmales and sires (Wang et al., 2000), it does not appear that achange in AVP receptor number is critical for the expression ofconsistent paternal behavior in prairie voles.

Although these studies delineate neural pathways involved inregulating consistent paternal behavior, exactly how neural sys-tems act to facilitate parental behavior may vary by whether aspecies exhibits parental behavior in a consistent or facultativemanner. For example, unlike male prairie voles, most femalerodents are neophobic as nulliparous adults (e.g., rats) and onlybecome maternal peripartum (reviewed by Rosenblatt, 1990). Thetransition from neophobia to maternal behavior onset occurs con-comitantly with greater OT receptor numbers in the extendedamygdala late in pregnancy (Insel, 1986; Insel & Shapiro, 1992).For montane voles, which only exhibit maternal behavior postpar-tum, these behavioral changes are associated with greater numbersof OT receptors in the lateral amygdala (LatAmyg) compared withnulliparous females (Insel & Shapiro, 1992).

As stated earlier, exactly how (and which) neural systems act toregulate facultative paternal behavior has yet to be determined.However, behavioral and pharmacological research conducted in

Microtus pennsylvanicus (meadow voles), a typically nonpaternalmicrotine species, provides a foundation by which to begin inves-tigating the neural underpinnings of facultative paternal behaviorregulation.

In some laboratory populations, meadow vole sires share nestswith a mate rather than establishing separate nest sites (Storey,Bradbury, & Joyce, 1994), drive off intruding males (Storey, 1996;Storey, French, & Payne, 1995), demonstrate appreciable care foryoung (Dewsbury, 1983; Hartung & Dewsbury, 1979; Storey etal., 1994; Storey & Joyce, 1995; Storey & Snow, 1987; Wilson,1982), and mate with an unfamiliar female without diminishingcare for pups (Storey & Snow, 1987). Additionally, biparental careincreases pup growth rates (Storey & Snow, 1987). More recently,it has been shown that sexually and parentally inexperienced adultmeadow voles also exhibit appreciable paternal care (Parker &Lee, in press; Storey & Joyce, 1995), and housing under winter,short day (SD) lengths enhances this effect (Parker & Lee, inpress).

Similar to prairie voles (Wang, Ferris, & De Vries, 1994),research in our laboratory has demonstrated a role for AVP inregulating meadow vole paternal state; central administration ofAVP suppresses pup-directed aggression in previously aggressivemales and promotes paternal behavior in previously pup-unresponsive males, and AVP antagonists block the onset ofpaternal behavior expression (Parker & Lee, 2001). However,administration of AVP does not induce paternal behavior in ag-gressive males, nor does it enhance paternal behavior expression inalready paternal males. This finding suggests that other neuralchanges are involved in regulating paternal behavior expression,for example, changes in AVP (and/or OT) receptor number.

The goal of these experiments was to determine whether AVPand/or OT receptor binding density or location differed betweensexually and parentally inexperienced (hereafter inexperienced)males and sexually and parentally experienced (hereafter experi-enced) males. After behavioral testing and central receptor auto-radiography, inexperienced and experienced meadow voles wereassessed for differences in AVP (Experiment 1) and OT (Experi-ment 2) receptor patterns in various neuroanatomical areas in theextended amygdala neural pathway.

General Method

Subjects

Subjects (Microtus pennsylvanicus), derived from wild-caught volesindigenous to northwestern Pennsylvania and southwestern New York,were bom to breeding pairs in an established colony at the University ofMichigan. Weanling meadow vole pups were removed from the dam andsire at 19 days of age and housed in same-sex sibling dyads in either longday (LD; 14 hr light/day) or SD (10 hr light/day) conditions. (LD and SDvoles were evenly distributed across conditions, and independent t testsshowed that LD and SD voles within each group showed no differences inregional receptor binding. Subsequently, all groups were collapsed acrossphotoperiod to simplify additional analyses.) Subjects (total autoradiogra-phy n - 34; n = 24 for Experiment 1, n - 10 for Experiment 2) werehoused in 26.67 X 21.59 x 13.97-cm polypropylene cages on pine shavingbedding with food (Purina Mouse Chow No. 5015, Ralston-Purina, St.Louis, MO) and water available ad libitum. Subject rooms were maintainedat 21 ± 2°C with low ambient noise conditions. Subjects remained sohoused until the beginning of the experimental procedure (at 11-13 weeks

AVP AND OT RECEPTOR BINDING IN MALE MEADOW VOLES 1343

of age). At this time, males were assigned to specific testing conditions. InExperiment 1, we examined whether the following groups showed similaror different regional AVP receptor patterns: (a) inexperienced, paternalmales (n = 6) compared with experienced, paternal males (« = 6), (b)inexperienced, aggressive males (n = 6) compared with inexperienced,pup-unresponsive males (n = 6), and (c) all paternal males (n = 12)compared with all nonpaternal males (« = 12). In Experiment 2, OTreceptor patterns were assessed in inexperienced, pup-unresponsive males(n = 5) and experienced, paternal males (n = 5).

Paternal Behavior Testing

As previously described (Parker & Lee, 2001, in press), the testing tosort males into groups was as follows. Inexperienced males (that receivedno social manipulation) were tested with an alien pup between 11 and 13weeks of age. Experienced males were tested for paternal behavior withone of their own pups after 72 hr of postpartum pup exposure, after pairingwith a female at 11 weeks of age, and cohabitation with their matethroughout pregnancy and delivery of the litter. Other than these differ-ences, paternal behavior testing was identical for inexperienced and expe-rienced males. Each male was placed in a novel polypro-pylene 48.26 X 26.67 x 20.32-cm cage with fresh bedding. Males wereallowed to become familiar with the new environment for 5 min, and thena 2-5 day-old pup was introduced to the opposite end of the cage from themale. Each test was carried out during the lighted phase of the light cycleand was videotaped for 10 min with a Panasonic camera and wide-anglelens on a time-lapse VCR. Based on the experimenter's rating, each male'sbehavior was scored categorically as aggressive (rough handling or charg-ing pup, resulting in pup vocalization or injury), unresponsive (briefinvestigatory sniffing or no contacting or interacting with pup), or paternal(grooming, huddling, and/or retrieving).

Receptor Autoradiography

After behavioral testing, subjects were anesthetized with halothane anddecapitated. Brains were quickly removed, frozen on dry ice, and stored at-70 °C until sectioned. Brain slices of 20 /nm were cut in a cryostat at19 ± 1 °C, slide mounted (Colorfrost/Plus slides, Fisher Scientific, Atlanta,GA), and stored at —70 °C. Coronal sections were cut through four brainareas from rostral to caudal (see later discussion). Brain slices were thawedfor 30 min at room temperature before commencement of the bindingassay. AVP receptor autoradiography was performed using a highly selec-tive linear AVP V,a receptor antagonist, [125I]-phenylacetyl-D-Tyr(Me)-Phe-Gln-Asn-Arg-Pro-Arg-Tyr-NH2 (New England Nuclear, Wellesley,MA; 2,200 Ci/mmol, 0.1 nmol). Slides were incubated in buffer (50 mmolTris, 10 mmol magnesium chloride, 0.1% bovine serum albumin [BSA])for 60 min, washed three times (8 min each) in buffer (50 mmol Tris, 100mmol choline chloride, 0.1% BSA, 0.01% Triton X-100), and brieflydipped in double-distilled water. All buffer chemicals were obtained fromSigma Chemical (St. Louis, MO). Nonspecific binding was determinedusing l-/xM concentration of AVP, Cys-Tyr-Phe-Gln-Asn-Cys-Pro-Arg-Gly-NH2 (Peninsula Laboratories, San Carlos, CA). The OT binding pro-tocol was identical to the AVP binding protocol, except a highly selective[125I]-ornithine vasotocin analog, d(CH2)5[Tyr(Me)2, Thr4, Orn8, [125I]Tyr9-NH2] (New England Nuclear; 2,200 Ci/mmol, 0.1 nM), was used todetermine specific binding, and nonspecific binding was assessed usingl-ju,mol concentration of OT, Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Leu-Gly-NH2

(Peninsula Laboratories). Slides were dried under a stream of cool air,placed in film cassettes, and apposed to BioMax MR X-ray film (Amer-sham, Buckinghamshire, England) for 48 hr (AVP) or 24 hr (OT). Inpreliminary trials, X-ray films were developed after different lengths ofexposure for both vasopressin and OT (e.g., 24 hr, 48 hr, and 72 hr) todetermine a suitable exposure time that was in the linear response range of

the film. Film optical densities were measured using Image J for the PC, acomputerized imaging system from the National Institutes of Health.

Data Analysis

For each subject, nine sections per area per side of brain were analyzed(i.e., six specific binding sections and three nonspecific binding sectionsper area per side of the brain) by an experimenter who was unaware ofsubjects' behavioral data, hi a few cases, tissue sections were damagedduring autoradiographic processing and were excluded from analysis bythe mutual consent of two experimenters (each unaware of subjects'experimental condition). Specific binding and nonspecific binding wereaveraged separately for each side of the brain, nonspecific binding wassubtracted from specific binding for each side of the brain to yield totalbinding for left and right structural areas, and finally, left and right wereaveraged together to represent a single data point for each vole/area. Brainregions for image analysis were based on previous studies demonstratingan association between behavior and AVP or OT receptor binding con-ducted in various rodent species (Insel, 1986, 1990; Insel & Shapiro, 1992;Insel et al., 1994) and according to the areas that showed the most intensereceptor binding in these experiments. For Experiment 1 (i.e., AVP recep-tor binding), the following brain regions that had considerable AVP bind-ing were examined: AON, LS (anterior LS at the level of the vertical limbof the diagonal band; lateral, medial, and central measurements of the LSat the level of the anterior commissure), diagonal band, BNST, MeAmyg,and central amygdala (CeAmyg). For Experiment 2 (i.e., OT receptorbinding), the AON, BNST, LS at the level of the vertical limb of thediagonal band, LatAmyg, and CeAmyg were examined. Data were ana-lyzed separately for AVP and OT receptor autoradiography groups and arepresented separately.

Experiment 1

AVP Statistical Analyses

Using independent t tests, we assessed whether different types of pater-nal males (e.g., inexperienced, paternal vs. experienced, paternal) anddifferent types of nonpaternal males (e.g., inexperienced, aggressive vs.inexperienced, pup-unresponsive) differed from each other. Because nosignificant differences in regional AVP receptor binding were found withineither behavioral group, groups were collapsed within each behavioralcategory to simplify subsequent statistical analyses. Second, independent ttests were used to determine whether nonpaternal meadow voles andpaternal meadow voles differed in AVP receptor binding for each neuro-anatomical area.

Results

Paternal males showed more AVP receptor binding than non-paternal males in the AON, ;(20) = 2.192, p = .040, and showedless AVP binding in the LS at the level of the diagonal band,f(21) = -2.890, p = .009, and the lateral, f(15) = -2.583, p =.021; central, ;(15) = -2.501, p = .024; and medial septum,r(15) = —2.697, p = .017, at the level of the anterior commissurethan did nonparental males (see Figures 1 and 2). No differenceswere found in the diagonal band (p = .771), BNST (p = .852),MeAmyg (p = .968), or CeAmyg (p = .408; see Figures 1 and 3).

Experiment 2

OT Statistical Analyses

For OT receptor binding, independent t tests were used to determinewhether inexperienced, pup-unresponsive males differed from experi-


Paternal Males Non-paternal Males

A B

AON AON

D

DB>

BNST

G ll

Figure 1. Photomicrographs of arginine-vasopressin receptor optical densities in extended amygdala structuresof paternal and nonpaternal male meadow voles. A and B: anterior olfactory nucleus (AON); C and D: anteriorlateral septum (S) and diagonal band (DB); E and F: lateral, central, and medial septum (S) and bed nucleus ofthe stria terminalis (BNST) at the level of the anterior commissure; G and H: central and medial amygdala (A).

enced, paternal males in OT receptor distribution for each of the followingneuroanatomical areas: AON, BNST, LS at the level of diagonal band,LatAmyg, and CeAmyg.

Results

Experienced, paternal males showed greater OT binding thaninexperienced, pup-unresponsive males in the AON, r(8) = 2.389,p = .044; LS, r(8) = 7.550, p < .001; BNST, r(8) = 4.219, p =.003; and LatAmyg, f(8) = 2.865, p = .028 (see Figures 4 and 5).Although not statistically significant, OT receptor binding waselevated in the CeAmyg of experienced males (p = .095), and itshould be noted that two-tailed tests and low sample sizes mayhave masked a significant different in OT binding for experiencedmales, relative to inexperienced males, in this area.

General Discussion

Data from Experiment 1 indicate that AVP receptor patterns inthe extended amygdala are associated with paternal state inmeadow voles. Specifically, paternal males demonstrate more re-ceptor binding in the AON but less AVP receptor binding invarious areas of the LS than nonpaternal males. In Experiment 2,sexually and parentally inexperienced and experienced males showregional differences in OT receptor distribution; paternal, experi-enced males have more OT receptor binding in the AON, LS,BNST, and LatAmyg and show a similar trend in the CeAmygcompared with nonpaternal, inexperienced males.

Whether or not activity in meadow vole AVP and OT neuralpathways functions to generate the onset of paternal behaviorremains unknown and awaits further investigation by other exper-imental techniques. Nonetheless, findings from experiments in


(/>cCD

|".e

o 0.5

|°-4

0.3

as

o 0.2D3(D

LUCO

0.1

I

A

Figure 2. Paternal meadow voles (filled bars) have higher arginine-vasopressin (AVP) binding in the anterior olfactory nucleus (AON) andlower AVP binding in the anterior lateral septum (ANTSEPTUM), lateralseptum (LATSEPTUM), central septum (CENTSEPTUM), and medialseptum (MEDSEPTUM) compared with nonpaternal males (open bars).Asterisks indicate a minimum significant difference (p < .05) betweenpaternal and nonpaternal males.

other rodent species coupled with pharmacological and behavioralresearch in our laboratory suggest several intriguing possibilitiesfor the functional significance of these observed differences inAVP and OT receptor patterns in relation to meadow vole paternalstate.

In the present studies, paternal behavior onset is marked bysubstantially lower AVP receptor binding in the lateral septumcompared with nonpaternal males. This finding is similar to thoseobserved in previous studies that have shown that characteristi-cally paternal microtine species (i.e., prairie voles) show fewerAVP receptors in the LS compared with characteristically nonpa-ternal species (i.e., montane voles) in both sexually and parentallyinexperienced and experienced states (Insel et al., 1994; Wang etal., 2000). In other nonpaternal rodents, it has been suggested thatseptal AVP may be involved in the acquisition and storage ofinformation related to situations that invoke fear and anxiety (rats;Everts & Koolhaas, 1999), and higher levels of septal AVP (asindexed by AVP-ir) covary with increased aggressive behavior(mice; Compaan, Buijs, Pool, de Ruiter, & Koolhaas, 1993).Because the location of AVP receptor sites demarcates whereendogenous peptides can exert their effects, perhaps the low septalAVP receptor binding that was observed for parental males reflectsthe diminished ability of AVP to induce aggressive and fearfulbehaviors in response to neonates, as was exhibited by aggressiveand pup-unresponsive males with high septal AVP receptor bind-ing. In this respect, it might be hypothesized that high numbers ofAVP receptors in the septum exert inhibitory control over paternalbehavior expression. Thus, down-regulation of AVP receptors inthis site may exert a permissive effect on males and may allow

them to respond to the specific social and environmental cues thatinduce paternal onset.

Paternal males also showed higher AVP receptor binding in theAON. Exactly why AVP receptor binding differs in this neuroana-tomical area is unclear. However, because injections of AVP intothe AON enhance social recognition in rats (Dluzen, Muraoka,Engelmann, & Landgraf, 1998), and bilateral bulbectomy reducespaternal behavior in prairie voles (Kirkpatrick, Williams, et al.,1994), it is possible that AVP receptors in the AON may play asignificant role in recognizing pup odors during paternal states.

In addition to the observed differences in regional AVP receptorbinding, sexually and parentally experienced males also showhigher OT receptor binding in the AON, LatAmyg, CeAmyg,BNST, and LS compared with inexperienced males. These struc-tures are part of the extended amygdala neural pathway that hasbeen previously implicated in regulating maternal behavior onsetin mammalian rodents (Fleming & Walsh, 1994; Numan & Shee-han, 1997). In several species, greater OT receptor binding in theBNST (female rats; Insel, 1986) and the LatAmyg (female mon-tane voles; Insel & Shapiro, 1992) is associated with the onset ofmaternal behavior. Unfortunately, one limitation of our study isthat we cannot distinguish between whether OT receptor bindingdifferences in inexperienced and experienced voles are truly due topaternal state or whether they are produced by differences in themating and cohabiting experience of sires compared with virginmales. Nonetheless, the possibility exists, and should be consid-ered, that the observed differences in OT receptor binding inspecific neuroanatomical sites may act to facilitate the onset ofparental behavior in males.

Previous pharmacological experiments in inexperiencedmeadow voles have demonstrated a dose- and behavioral state-

.£•'tocu>^ 0.603O

"o.o 0.5

^Q.

S 0.4&a.^ 0.3

"co.I 0.2O)2?if 0.1LU

^P n n

_

-

-

T T

1rf1ii A M

fFl

Figure 3. Paternal males (filled bars) do not differ from nonpaternalmales (open bars) in arginine-vasopressin (AVP) receptor binding in thediagonal band (DIAGBAND), bed nucleus of the stria terminalis (BNST),central amygdala (CENAMYGDALA), or medial amygdala (MED-AMYGDALA).


Paternal Males Non-paternal Males

is

1AON

BNST BNST

< A

Figure 4. Photomicrographs of oxytocin receptor optical densities inextended amygdala structures of paternal and nonpaternal male meadowvoles. A and B: anterior olfactory nucleus (AON); C and D: lateral septum(S) and bed nucleus of the stria terminalis (BNST) at the level of theanterior commissure; E and F: lateral and central amygdala (A).

dependent effect of centrally administered AVP on aggressive andpaternal behaviors (Parker & Lee, 2001). For baseline aggressivemales, low (1 ng AVP), but not high (3 ng AVP), doses of AVPsuppressed pup-directed aggression, but these same doses of AVPfailed to induce paternal behavior. In pup-unresponsive males,3-ng, but not 1 -ng, doses of AVP promoted paternal behavior, andAVP antagonists administered to this group before 24 hr of un-mated cohabitation with a female completely inhibited the reliableonset of paternal behavior. However, AVP injections did notenhance paternal behavior in already paternal voles. Previously, itwas hypothesized that these results suggested that the receptorsystems of these different baseline behavioral groups were notidentical, and this possibility might account for the observeddifferential responses to AVP injection. Experiment 1 partiallysupports this assertion; inexperienced, paternal males do differfrom inexperienced, aggressive and pup-unresponsive males inolfactory and septal AVP receptor binding, but the latter twogroups do not differ from each other. Thus, it is not clear whyAVP induces paternal behavior in pup-unresponsive, but notpup-aggressive, males. One possibility is that centrally admin-istered AVP is acting not on AVP receptors but on OT recep-tors. This is supported by research that has shown that at lowphysiological doses exogenously administered AVP and OTprimarily bind to their own receptors, but at high pharmacolog-ical doses AVP and OT exert behavioral effects by binding notonly to their own but also to each other's receptor sites (Bar-beris & Tribollet, 1996). Thus, the possibility that some of these

behavioral effects were mediated through regional OT receptorscannot be ruled out. Unfortunately, no data are available onwhether centrally administered OT induces changes in paternalstate in meadow voles or any other rodent species. Nor is itknown whether inexperienced, aggressive, pup-unresponsive,or paternal males differ in regional OT receptor binding. How-ever, because differences in both AVP and OT receptor bindingexist between sexually and parentally inexperienced and expe-rienced males, determining whether meadow voles respond tocentral administration of OT and whether they demonstratedifferences in regional OT receptor binding in all of the afore-mentioned inexperienced behavioral states merits furtherinvestigation.

In conclusion, these studies have demonstrated that specificcentral AVP and OT receptor patterns co-occur with paternalbehavior in meadow voles. In particular, the presence of pater-nal behavior is associated with lower AVP receptor binding inthe LS, an area of the brain associated with aggression (Com-paan et al., 1993) and higher OT receptor binding in theextended amygdala in areas associated with maternal care(Fleming & Walsh, 1994; Numan & Sheehan, 1997). Althoughthese studies mark a first step in identifying putative neuralsystems that regulate facultative paternal behavior, whether ornot activity in meadow vole AVP and OT neural pathwaysfunctions to generate the onset of paternal behavior, and if sounder which specific socioecological circumstances, remainsunknown.

tn

0.5 rCO.o"5.2 0.4o"5.

£ 0.3

OH 0.2o

2 0.1tuop

£ 0.0

1JL

Figure 5. Paternal, sexually, and parentally experienced meadow voles(filled bars) have greater oxytocin (OT) receptor binding in the anteriorolfactory nucleus (AON), lateral septum (SEPTUM), bed nucleus of thestria terminalis (BNST), lateral amygdala (LATAMYGDALA), and centralamygdala (CENAMYGDALA) compared with nonpaternal sexually andparentally inexperienced males (open bars). Asterisks indicate a minimumsignificant difference (p < .05) between paternal and nonpaternal males;dagger indicates a nonsignificant difference (p < .10) between paternaland nonpaternal males.


References

Barashad, M., Novak, M. A., & De Vries, G. J. (1994). Cohabitation altersvasopressin innervation and paternal behavior in prairie voles (Microtusochrogaster). Physiology & Behavior, 56, 751-758.

Barash, D. P. (1975). Ecology of paternal behavior in the hoary marmot(Marmota caligata): An evolutionary interpretation. Journal of Mam-malogy, 56, 613-618.

Barberis, C., & Tribollet, E. (1996). Vasopressin and oxytocin receptors inthe central nervous system. Critical Reviews in Neurobiology, 10, 119-154.

Cho, M. M., De Vries, A. C., Williams, J. R., & Carter, C. S. (1999). Theeffect of oxytocin and vasopressin on partner preferences in male andfemale prairie voles (Microtus ochrogaster). Behavioral Neuroscience,113, 1071-1079.

Compaan, J. C., Buijs, R. M., Pool, C. W., de Ruiter, A. J., & Koolhaas,J. M. (1993). Differential lateral septal vasopressin innervation in ag-gressive and nonaggressive male mice. Brain Research Bulletin, 30,1-6.

Dewsbury, D. A. (1983). A comparative study of rodent social behavior ina seminatural enclosure. Aggressive Behavior, 9, 207-215.

Dewsbury, D. A. (1985). Parental behavior in rodents. Journal of AmericanZoology, 25, 841-852.

Dluzen, D. E., Muraoka, S., Engelmann, M., & Landgraf, R. (1998). Theeffects of infusion of arginine vasopressin, oxytocin, or their antagonistsinto the olfactory bulb upon social recognition responses in male rats.Peptides, 19, 999-1005.

Evens, H. G. J., & Koolhaas, J. M. (1999). Differential modulation oflateral septal vasopressin receptor blockade in spatial learning, socialrecognition, and anxiety-related behaviors in rats. Behavioural BrainResearch, 99, 7-16.

Fleming, A. S., & Walsh, C. (1994). Neuropsychology of maternal behav-ior in the rat: c-fos expression during mother-litter interactions. Psycho-neuroendocrinology, 19, 429-443.

Gubernick, D. J., Wright, S. L., & Brown, R. E. (1993). The loss of father'spresence for offspring survival in the monogamous California mouse,Peromyscus californicus. Animal Behaviour, 46, 539-546.

Hartung, T. G., & Dewsbury, D. A. (1979). Paternal behavior in six speciesof muroid rodents. Behavioral and Neural Biology, 26, 466-478.

Insel, T. R. (1986). Postpartum increases in brain oxytocin binding. Neu-roendocrinology, 44, 515-518.

Insel, T. R. (1990). Regional changes in brain oxytocin receptors post-partum: Time-course and relationship to maternal behaviour. Journal ofNeuroendocrinology, 2, 539-545.

Insel, T. R., Preston, S., & Winslow, J. T. (1995). Mating in the monog-amous male: Behavioral consequences. Physiology & Behavior, 57,615-627.

Insel, T. R., & Shapiro, L. E. (1992). Oxytocin receptor distribution reflectssocial organization in monogamous and polygamous voles. Proceedingsof the National Academy of Sciences, USA, 89, 5981-5985.

Insel, T. R., Wang, Z. X., & Ferris, C. F. (1994). Patterns of brainvasopressin receptor distribution associated with social organization inmicrotine rodents. Journal of Neuroscience, 14, 5381-5392.

Kirkpatrick, B., Carter, C. S., Newman, S. W., & Insel, T. R. (1994).Axon-sparing lesions of the medial nucleus of the amygdala decreaseaffiliative behaviors in the prairie vole (M. ochrogaster): Behavioral andanatomical specificity. Behavioral Neuroscience, 108, 501-513.

Kirkpatrick, B., Kim, J. W., & Insel, T. R. (1994). Limbic system fosexpression associated with paternal behavior. Brain Research, 658,112-118.

Kirkpatrick, B., Williams, J. R., Slotnick, B. M., & Carter, C. S. (1994).Olfactory bulbectomy decreases social behavior in male prairie voles(M. ochrogaster). Physiology & Behavior, 55, 885-889.

Kleiman, D. G. (1977). Monogamy in mammals. Quarterly Review ofBiology, 52, 39-69.

Kleiman, D. G., & Malcolm, J. R. (1981). The evolution of male paternalinvestment in mammals. In D. J. Gubernick & P. H. Klopfer (Eds.),Parental care in mammals (pp. 347-387). New York: Plenum.

Lonstein, J. S., & De Vries, G. J. (1999). Sex differences in the parentalbehaviour of adult virgin prairie voles: Independence from gonadalhormones and vasopressin. Journal of Neuroendocrinology, 11, 441-449.

Mihok, S. (1979). Behavioral structure and demography of subarcticClethrionomys gapperi and Peromyscus maniculatus. Canadian Journalof Zoology, 57, 1520-1535.

Newman, S. W. (1999). The medial extended amygdala in male reproduc-tive behavior: A node in the mammalian social behavior network. In J. F.McGinty (Ed.), Annals of the New York Academy of Sciences: Vol. 877.Advancing from the ventral striatum to the extended amygdala: Impli-cations for neuropsychiatry and drug use (pp. 242-257). New York:New York Academy of Sciences.

Numan, M., & Sheehan, T. P. (1997). Neuroanatomical circuitry formammalian maternal behavior. In C. S. Carter, I. I. Lederhendler, & B.Kirkpatrick (Eds.), The integrative neurobiology of affiliation (Vol. 807,pp. 101-125). New York: New York Academy of Sciences.

Parker, K. J., & Lee, T. M. (2001). Central vasopressin administrationregulates the onset of facultative paternal behavior in Microtus pennsyl-vanicus (meadow voles). Hormones and Behavior, 39, 285-294.

Parker, K. J., & Lee, T. M. (in press). Social and environmental factorsinfluence the suppression of pup-directed aggression and developmentof paternal behavior in captive meadow voles (Microtus pennsylvani-cus). Journal of Comparative Psychology.

Rosenblatt, J. S. (1990). Landmarks in the physiological study of maternalbehavior with special reference to the rat. In N. A. Krasnegor (Ed.),Mammalian parenting: Biochemical, neurobiological, and behavioraldeterminants (pp. 40-60). New York: Oxford University Press.

Schug, M. D., Vessey, S. H., & Underwood, E. M. (1992). Paternalbehavior in a natural population of white-footed mice (Peromyscusleucopus). American Midland Naturalist, 127, 373-380.

Storey, A. E. (1996). Behavioral interactions increase pregnancy blockingby unfamiliar male meadow voles. Physiology & Behavior, 60, 1093-1098.

Storey, A. E., Bradbury, C. G., & Joyce, T. L. (1994). Nest attendance inmale meadow voles: The role of the female in regulating male interac-tions with pups. Animal Behaviour, 47, 1037-1046.

Storey, A. E., French, R. J., & Payne, R. (1995). Sperm competition andmate guarding in meadow voles (Microtus pennsylvanicus). Ethology,101, 265-279.

Storey, A. E., & Joyce, T. L. (1995). Pup contact promotes paternalresponsiveness in male meadow voles. Animal Behaviour, 49, 1-10.

Storey, A. E., & Snow, D. T. (1987). Male identity and enclosure sizeaffect paternal attendance of meadow voles, Microtus pennsylvanicus.Animal Behavior, 35, 411-419.

Trivers, R. (1972). Parental investment and sexual selection. In B. Camp-bell (Ed.), Sexual selection and the descent of man 1871-1971 (pp.136-179). Chicago: Aldine.

Wang, Z., Ferris, C. F., & De Vries, G. J. (1994). Role of septal vasopressininnervation in paternal behavior in prairie voles (Microtus ochrogaster).Proceedings of the National Academy of Sciences, USA, 91, 400-404.

Wang, Z., Hulihan, T. J., & Insel, T. R. (1997). Sexual and social experi-ence is associated with different patterns of behavior and neural activa-tion in male prairie voles. Brain Research, 767, 321-332.

Wang, Z. X., Liu, Y., Young, L. J., & Insel, T. R. (2000). Hypothalamicvasopressin gene expression increases in both males and females post-partum in a biparental rodent. Journal of Neuroendocrinology, 12,111-120.


Wang, Z., Smith, W., Major, D. E., & De Vries, G. J. (1994). Sex andspecies differences in the effects of cohabitation on vasopressin mes-senger RNA expression in the bed nucleus of the stria terminalis inprairie voles (Microtus ochrogaster) and meadow voles (Microtus penn-sylvanicus). Brain Research, 650, 212-218.

Wang, Z., Young, L. J., De Vries, G. J., & Insel, T. R. (1998). Voles andvasopressin: A review of molecular, cellular, and behavioral studies ofpair bonding and paternal behaviors. In I. J. A. Urban, J. P. H. Burbach,& D. De Wied (Eds.), Voles and vasopressin (pp. 483-499). New York:Elsevier.

Wang, Z., Young, L. J., Liu, Y., & Insel, T. R. (1997). Species differencesin vasopressin receptor binding are evident early in development: Com-

parative anatomic studies in prairie and montane voles. Journal ofComparative Neurology, 378, 535-546.

Wilson, S. C. (1982). Parent-young contact in prairie and meadow voles.Journal of Mammalogy, 63, 300-305.

Winslow, J. T., Hastings, N., Carter, C. S., Harbaugh, C. R., & Insel, T. R.(1993, October 7). A role for central vasopressin in pair bonding inmonogamous prairie voles. Nature, 365, 545-548.

Wittenberger, J. F. (1981). Animal social behavior. Boston: Duxbury Press.

Received November 6, 2000Revision received April 13, 2001

Accepted July 10, 2001 •

730 Plot Street, H.E., MMfalnston, DC 20002-4242

Washington. Tf. 20002-1

l Neucxwciera

kutracMotu to Pubtbhrn

paternal behavior is associated with central...

Documents