over 3 years (Watson and Kelley 1992; Watson et al. 1993). Castrated males
maintain a masculine laryngeal weight and complement of all fast-twitch
muscle fibers. Castrated males stop calling even though their larynx is
capable of producing male-typical fast trills. The activational effects of
androgen must thus be on tissues other than the larynx, and the most likely
candidates are the androgen-target neurons in the vocal pathway.Two of the
major targets of androgens in the CNS are neurons of the laryngeal motor
nucleus (n. IX–X) and its major afferent nucleus, DTAM (Kelley 1980, 1981;
Pérez et al. 1996). The preoptic area is known to play an important role in
male reproductive behaviors in a variety of vertebrates (Kelley and Pfaff
1978; and see below in Section 3.1.3 on rodents). However, the POA
expresses estrogen receptors exclusively; because estrogen does not activate
advertisement calling, the POA is unlikely to be a site for this activational
effect (reviewed in Kelley 1996). Exactly what these activational effects are,
at the cellular and molecular levels, is presently not known.
Where does AVT act to mediate the expression of vocalizations? The
most likely target is, again, the CNS. Immunocytochemical studies reveal
that AVT is localized in many brain regions that are involved in the expression of vocal behaviors, including the pretrigeminal nucleus (DTAM of
Xenopus), the major input to vocal motor neurons (reviewed in Boyd 1997).
Interestingly, the concentration and the spatial distribution of AVT within
the brain are sexually distinct in bullfrogs (Boyd and Moore 1992; Boyd et
al. 1992; Boyd 1994b), corresponding to sex differences in vocal behavior
(i.e., advertisement calls are unique to males). Recent studies report that
the firing frequency of medullary neurons can be modified by direct application of AVT into brains (Rose et al. 1995).Together, these findings suggest
that AVT acts directly on the brain of male frogs by modifying the electrophysiological activity of neurons involved in the expression of vocalizations. Future studies will further identify how neuronal properties are
modified by the presence of AVT, and attempts should be made to understand how the modification of these higher-order neurons orchestrates the
expression of behavior.
3.1.3. Rodents
Many species of rodents produce ultrasonic vocalizations (≥20 kHz) in the
context of courtship prior to copulation (e.g., Floody 1979; Holman 1980;
Barfield and Thomas 1986). Following classic patterns of hormone induction of courtship vocalizations, male ultrasonic vocalizations are largely
androgen-dependent; in Mongolian gerbils (Meriones unguiculatus)
(Holman and Hutchison 1982), deer mice (Peromyscus maniculatus bairdi)
(Pomerantz et al. 1983), golden hamsters (Mesocricetus auratus) (Floody
1979), Long–Evans rats (Rattus norvegicus) (Barfield and Thomas 1986;
Matochik and Barfield 1991; Nyby et al. 1992), and house mice (Mus
musculus) (Nunez et al. 1978; Nunez and Tan 1984; Bean et al. 1986), cas294
A. Yamaguchi and D.B. Kelley
maintain a masculine laryngeal weight and complement of all fast-twitch
muscle fibers. Castrated males stop calling even though their larynx is
capable of producing male-typical fast trills. The activational effects of
androgen must thus be on tissues other than the larynx, and the most likely
candidates are the androgen-target neurons in the vocal pathway.Two of the
major targets of androgens in the CNS are neurons of the laryngeal motor
nucleus (n. IX–X) and its major afferent nucleus, DTAM (Kelley 1980, 1981;
Pérez et al. 1996). The preoptic area is known to play an important role in
male reproductive behaviors in a variety of vertebrates (Kelley and Pfaff
1978; and see below in Section 3.1.3 on rodents). However, the POA
expresses estrogen receptors exclusively; because estrogen does not activate
advertisement calling, the POA is unlikely to be a site for this activational
effect (reviewed in Kelley 1996). Exactly what these activational effects are,
at the cellular and molecular levels, is presently not known.
Where does AVT act to mediate the expression of vocalizations? The
most likely target is, again, the CNS. Immunocytochemical studies reveal
that AVT is localized in many brain regions that are involved in the expression of vocal behaviors, including the pretrigeminal nucleus (DTAM of
Xenopus), the major input to vocal motor neurons (reviewed in Boyd 1997).
Interestingly, the concentration and the spatial distribution of AVT within
the brain are sexually distinct in bullfrogs (Boyd and Moore 1992; Boyd et
al. 1992; Boyd 1994b), corresponding to sex differences in vocal behavior
(i.e., advertisement calls are unique to males). Recent studies report that
the firing frequency of medullary neurons can be modified by direct application of AVT into brains (Rose et al. 1995).Together, these findings suggest
that AVT acts directly on the brain of male frogs by modifying the electrophysiological activity of neurons involved in the expression of vocalizations. Future studies will further identify how neuronal properties are
modified by the presence of AVT, and attempts should be made to understand how the modification of these higher-order neurons orchestrates the
expression of behavior.
3.1.3. Rodents
Many species of rodents produce ultrasonic vocalizations (≥20 kHz) in the
context of courtship prior to copulation (e.g., Floody 1979; Holman 1980;
Barfield and Thomas 1986). Following classic patterns of hormone induction of courtship vocalizations, male ultrasonic vocalizations are largely
androgen-dependent; in Mongolian gerbils (Meriones unguiculatus)
(Holman and Hutchison 1982), deer mice (Peromyscus maniculatus bairdi)
(Pomerantz et al. 1983), golden hamsters (Mesocricetus auratus) (Floody
1979), Long–Evans rats (Rattus norvegicus) (Barfield and Thomas 1986;
Matochik and Barfield 1991; Nyby et al. 1992), and house mice (Mus
musculus) (Nunez et al. 1978; Nunez and Tan 1984; Bean et al. 1986), cas294
A. Yamaguchi and D.B. Kelley