Neuroscience Letters, 67 (1986) 263-268
263
Elsevier Scientific Publishers Ireland Ltd.
NSL 03974
C H A N G E S IN N E U R O N A L N U M B E R , D E N S I T Y A N D SIZE A C C O U N T FOR I N C R E A S E S IN V O L U M E OF S O N G - C O N T R O L NUCLEI D U R I N G S O N G D E V E L O P M E N T IN ZEBRA FINCHES
SARAH W. BOTTJER, ELIZABETH A. M I E S N E R and A R T H U R P. A R N O L D
Department of Psychology, and Laboratory of Neuroendocrinology, Brain Research Institute, University t?! CalifOrnia, Los Angeles, CA 90024 (U.S.A.) (Received September 16th, 1985; Revised version received December 12th, 1985; Accepted December 30th 1985)
Key words: neuron number
neuronal density
soma size- song-control nucleus - zebra finch
The caudal nucleus of the ventral hyperstriatum (HVc) and the robust nucleus of the archistriatum (RA) are two anatomically discrete brain regions that are known to be involved with song production in adult passerine birds. Both the HVc and RA increase greatly in volume during a restricted period of song development in male zebra finches, while brain regions not involved with song control show little or no increase in size. We report here that the increased volume of the HVc is attributable to an increase in the number of neurons during this period of song learning, whereas the growth of the RA is due to an increase in the somal size of neurons and a decrease in neuronal density. The pattern of results described is consistent with the idea that the HVc matures prior to the RA, and that the development of the RA may depend on the ingrowth of axons from the HVc and other song-control regions.
A specific neural circuit underlies the development of vocal learning in male passerine birds [4, 5, 8, 13, 14]. This neural circuit consists of an interconnected system of highly anatomically localized brain nuclei. Two telencephalic nuclei, the caudal nucleus of the ventral hyperstriatum (HVc) and the robust nucleus of the archistriatum (RA) are directly involved in song production in adult birds [11, 13]. Neurons in the HVc project directly onto RA neurons; neurons in the RA project monosynaptically onto the hypoglossal motor neurons that innervate the vocal muscles [13, 14]. We have previously reported that the HVc and RA increase dramatically in size during song development in male zebra finches, whereas brain regions not known to be involved with song exhibit little or no increase in volume during this time [4]. This result is interesting because it suggests that maturation of the HVc and RA is delayed, relative to other brain regions, until the time when song behavior is being learned. Male zebra finches begin to produce song-related vocalizations around 25 days of age; by 53 days most males have learned to produce the initial motor pattern of the stereotypical song they will sing throughout adulthood [5]. In this study, we have determined the somal area and the density of neurons in the HVc and RA of 25and 53-day male zebra finches; the gross nuclear volumes of the RA and HVc in these 0304-3940/86/$ 03.50 © 1986 Elsevier Scientitic Publishers Ireland Ltd.
264 birds have been reported previously [4]. The present results demonstrate that the increase in size of the HVc is due to an increase in the number of neurons in this region. In contrast, the increased size of the R A is attributable to an increase in the somal size of R A neurons and a decrease in their density. The increased number of HVc neurons m a y be required for control of learned song behavior, and the decreased density of RA neurons may indicate that they are receiving a greater afferent input from the HVc. The methods used to prepare the brain tissue used for this study have been described by Bottjer et al. [4]. Twelve male zebra finches, hatched in our breeding colony, were overdosed with anesthetic and perfused with saline followed by buffered formalin at either 23-29 days (mean=25.3; n - - 6 ) or 52-57 days (mean--53.5; n--6). Brains were frozen-sectioned in the transverse plane at 25 gm and alternate sections were stained with thionin. In order to measure soma size, sections were viewed at a final magnification of x 1000 and somata were traced with the aid of a camera lucida. The mean number of neurons traced for the 25-day birds was 97.0 and 101.3 in the R A and HVc, respectively; the corresponding numbers for the 53-day birds were 88.0 and 130.3. Approximately equal numbers of cells were traced on the left and right side of the brain for each bird. Neuronal density was measured by counting the number of neurons in which the nucleolus fell within an ocular grid which covered an area of 0.01 ram2; the grid counts were made in l0 strictly random locations throughout the central portion of each nucleus (the total area encompassed by 10 grids was 0.0025 mm3). Only the central portions of the HVc and R A were measured for both soma size and density because both of these brain regions have a uniform appearance when they are scanned throughout their extent at low power ( x 100-250) magnification. The criteria for tracing neuronal somata and counting neurons were as follows: neurons were easily distinguishable from glial cells because neurons contained a large, spherical nucleolus and usually had a large pale nucleus surrounded by dark cytoplasm. Glial cells contained several small, irregularly shaped nucleoli which were darkly stained. Aside from nucleoli, glial cells were lightly and evenly stained throughout the soma and never had a visible nucleus. These cytological criteria are consistent with those described by Gurney [7] for male zebra finches. Approximately 6% of the neurons in both the HVc and R A contained two nucleoli at each of the ages studied. For this reason, our density measurements may slightly overestimate the number of neurons. However, when neurons with two nucleoli were seen within a section, they were traced and counted as one neuron. The size of nucleoli did not change as a function of age. N o correction factor for split nucleoli was applied because nucleoli were presumed to be displaced rather than cut by the microtome knife (ref. 9, pp. 326--327). The distribution of cell size for each age group in the R A and HVc is shown in Figs. 1 and 2, respectively. There is a substantial difference in soma size between the two ages within the RA, but not within the HVc. This difference between the two nuclei is reflected in the mean somal areas shown in Table I. The difference in mean soma size was statistically significant in the R A (Fl,10 = 16.28, P = 0.003) but not in the HVc ( F < 1).
265
- "-"
25 D A Y S
30
.......... 5 3 D A Y S
N
....... t~
E o
20
o co "~ --'1 .Q
10
x..
'o 60
120 soma
180 size
240
300
( u m 2)
Fig. 1. The distribution of soma size of individual neurons in RA for 25-and 53-day birds. Values plotted tire mean percentages of cells within a given size class _+ S.E.M. Cell size is plotted along the abscissa as bins of 30 lain: width.
o~ v (D u
25 D A Y S
30
AYS
r.t)
E o
20
CO
"S C
o "" ""I A21
10
,,...,
"ID I
I
I
60
t
I
120 soma
I
I
180 size
I
I
240
( u m 2)
Fig. 2. The distribution of soma size of individual neurons in HVc for 25- and 53-day birds, plotted as in Fig. I. T h e d e n s i t y o f n e u r o n s w i t h i n e a c h b r a i n r e g i o n is g i v e n in T a b l e I. T h e r e is a s u b s t a n t i a l d i f f e r e n c e in n e u r o n (F~ ~ 0 = 4 7 . 6 8 , P = 0 . 0 0 0 1 )
density between
with younger
25- a n d 5 3 - d a y b i r d s w i t h i n t h e R A
birds having
approximately
twice as many
266 TABLE I MEANS _+S.D.s FOR 25- A N D 53-DAY MALES Volume a (mm 3)
Somal area (tam 2)
Neuron densityb (mm 3)
Neuron number
RA 25 days 53 days
0.228+0.030 0.402__.0.101
112.24+24.29 157.42 + 12.73
129,132+ 19,796 66,400+ 10,164
29,081 +2472 26,085__+4535
HVc 25 days 53 days
0.408+0.085 0.521 +0.124
98.57+10.71 98.16_+ 7.31
122,868+11,276 147,532-t- 13,924
49,698+ 9272 76,485_+ 17,928
~'These data represent gross nuclear volumes and are taken from Bottjer et al. [4]; values correspond to the summed areas for left and right sides in the RA and HVc, respectively. bNeuron density is given as number of neurons per mm 3.
neurons per unit area as older birds. There is also a difference in density between 25- and 53-day birds within the HVc (Fl,10 = 11.37, P=0.007), but in this case the density of neurons is higher in the older birds. The total number of neurons was estimated by multiplying the neuron density times the gross nuclear volume, which was reported for this same data set in a prior study [4]. These data are shown in Table I. Both the total volume of the HVc and the density of neurons within the HVc increase between 25 and 53 days, resulting in a significant increase in neuron number between these two ages (Fl,10 = 10.57, P = 0.009). Although the volume of the R A increasessubstantially between 25 and 53 days, the density of neurons is so much less in older birds that the estimated number of neurons is not significantly different (Fl.lo = 2.02, P = 0.18). An interesting correlate of the increased density of neurons in the HVc is an apparent shift in the spatial arrangement of neurons. Neurons of 53-day birds tend to cluster in discrete groups of approximately 4-8 neurons. These clusters generally include two or more large neurons and several smaller neurons. In contrast, neurons of 25day birds appear to be much more randomly distributed, with little or no clustering apparent. An observer attempted to judge the HVc of each bird (age unknown to observer) as possessing either a high or a low degree of neuron clustering. O f the 12 birds in this study, all 6 birds in the 53-day group were judged as 'clusterers' and 5 out of 6 birds in the 25-day group were judged as 'non-clusterers'. It is tempting to speculate that this change in the distribution of cell density represents the formation of functional units within the HVc that are related to song learning. The total volumes of both the HVc and R A increase substantially during song development, with the increase in size of the HVc preceding that of R A [4]. The present results show that the increased size of the HVc reflects an increase in neuronal number, raising the possibility that ongoing neuronal proliferation and migration into the HVc are contributing to this increase. N o t t e b o h m et al. have already demonstrated that neurogenesis occurs in the HVc of adult songbirds, but does not result in a net increase in t h e number of neurons within the HVc during adulthood [6, 12,
267
15]. The increase in somal area and the decrease in density of neurons in the RA suggest that these cells are undergoing fundamental maturational changes between 25 and 53 days, as song behavior is beginning to acquire its adult form. The decrease in density may indicate that the dendritic arbor of these neurons is increasing, and/or that terminals of neurons from other brain regions are growing into the RA during this time. The two major sources of afferent input to the RA are from the HVc and the magnocellular nucleus of the anterior neostriatum ( M A N ) [13, 14]. Konishi and Akutagawa [10] have reported that injection of tritiated proline into the HVc results in label throughout the RA in 35-day male zebra finches, but results in little or no label in 25-day males. We have detected axonal projections from the M A N to the RA in males as young as 25 days. In addition, we have reported previously that the number of neurons in the M A N declines dramatically between 25 and 53 days of age: there are approximately twice as m a n y neurons in the M A N in 25-day male zebra finches as there are in 53-day males [4]. This pattern of findings may indicate that the loss of M A N neurons is triggered by the increase in number of HVc neurons, possibly bccausc HVc neurons compete effectively against M A N neurons for synaptic contacts with RA neurons. Alternatively, the loss of M A N neurons may be attributable to some other event, but may allow HVc axons to grow into the RA, where they may induce secondary growth via some trophic influence. In any case, these maturational changes all correlate with the occurrence of song learning, and each may reflect specific events responsible for learned changes in this complex behavior. A final issue pertains to the role of hormones in the development of the song-control system. Both the HVc and M A N contain a high proportion of androgen-accumulating cells in adult male zebra finches [1, 2]. An important question is therefore whether the ability of cells in the HVc and M A N to concentrate androgens during song development affects the maturation of a monosynaptically related nucleus such as the RA. M A N cells in 25-day males accumulate dihydrotestosterone much less effectively than MAN cells in 60-day birds [3]. Thus, it seems possible that increased accumulation of androgens in the M A N and/or HVc may be importantly involved in the regulation of neuronal number and the outgrowth of axons from these regions, as well as the ability of HVc and M A N neurons to make and receive synaptic contacts with other brain regions. This research was supported by USPHS Grants NS 18392 to S.W.B. and NS19645 to A.P.A., and by a grant from the McKnight Foundation to S.W.B. We thank Dale Sengelaub for his comments. [ Arnold, A.P., Quantitative analysis of sex differences in hormone accumulation in the zebra linch brain: methodology and theoretical issues, J. Comp. Neurol., 189 (1980) 4 2 1 4 3 6 . 2 Arnold, A.P., N o n e b o h m , F. and Pfaff, D.W., H o r m o n e concentrating cells in vocal control and other areas of the brain of the zebra finch, J. Com p. Neurol., 165 (1976) 487 512. 3 Bottjer, S.W. and Arnold, A.P., Development of androgen accumulation in a forebrain nucleus that mediates song learning in zebra finches, Soc. Neurosci. Abstr., 15 (1985) 159. 4 Bottjer, S.W., Glaessner, S.L. and Arnold, A.P., Ontogeny of brain nuclei controlling song learning and behavior in zebra [inches, J. Neurosci,. 5 ([985) 1556 1563.
268 5 Bottjer, S.W., Miesner, E.A. and Arnold, A.P., Forebrain lesions disrupt development but not maintenance of song in passerine birds, Science, 224 (1984) 901-903. 6 Goldman, S.A. and Nottebohm, F., Neuronal production, migration, and differentiation in a vocal control nucleus of the adult female canary brain, Proc. Natl. Acad. Sci. USA, 80 (1983) 2390-2394. 7 Gurney, M.E., Hormonal control of cell form and number in the zebra finch song system, J. Neurosci.. 1 (1981) 658-673. 8 Kelley, D.B. and Nottebohm, F., Projections of a telencephalic auditory nucleus Field L in the canary, J. Comp. Neurol., 183 (1979) 455-469. 9 Konigsmark, B.W., Methods for the counting of neurons. In W.J.H. Nauta and S.O.E. Ebbeson (Eds.), Contemporary Research Methods in Neuroanatomy, Springer-Verlag, New York, 1970. 10 Konishi, M. and Akutagawa, E., Neuronal growth, atrophy and death in a sexually dimorphic song nucleus in the zebra finch brain, Nature (London), 315 (1985) 145-147. I 1 McCasland, J.S, and Konishi, M., Interaction betwen auditory and motor activities in an avian song control nucleus, Proc. Natl. Acad. Sci. USA, 78 (1981) 7815-7819. 12 Nottebohm, F., Birdsong as a model in which to study brain processes related to learning, Condor. 86 (1984) 227-236. 13 Nottebohm, F., Stokes, T.M. and Leonard, C.M., Central control of song in the canary, J. Comp. Neurol., 165 (1976) 457468. 14 Nottebohm, F., Kelley, D.B. and Paton, J.S., Connections of vocal control nuclei in the canary telencephalon, J. Comp. Neurol., 207 (1982) 344-357. 15 Paton, J.A. and Nottebohm, F., Neurons generated in the adult brain are recruited into functional circuits, Science, 225 (1984) 1046-1048.