Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats.

In the autoradiograms of young rats injected with thymidine-H3 many of the granule cells of the dentate gyrus were found labeled. The number of labeled cells declined rapidly with increased age at the time of injection. Histological studies showed the presence in young rats of a large germinal matrix of mitotic cells in the ependymal and subependymal layers of the third and lateral ventricles. The areal extent and cell population of this germinal pool declined rapidly from birth on with a transient rise with a peak at about 15 days. During this latter period the number of "undifferentiated" cells near the granular layer of the dentate gyrus showed a rapid rise with a subsequent decline. The decline in the number of "undifferentiated" cells was accompanied by a rise in the number of differentiated granule cells. Cell counts in homologous parts of the dentate gyrus indicated a six-fold increase in the number of differentiated granule cells from birth to three months. We postulated that un­ differentiated cells migrate postnatally from the forebrain ventricles to the hippo­ campus where they become differentiated. The possible functional significance of delayed hippocampal neurogenesis is discussed with reference to our finding of incor­ poration of testosterone-H3 by cells of the hippocampus, implicating that they may function as receptors of gonadal hormones. It is commonly held that neurons in the granulel cells of the hippocampus. This central nervous system of higher verte­ finding was subsequently confirmed in brates are formed during embryonic devel­ normal adult rats and adult cats after intraperitioneal or intraventricular injec­ opment and that neurogenesis does not tion of thymidine-H (Altman, '63a). Since occur postnatally. This belief is based on there is good evidence that thymidine, a the absence of neurons with mitotic figures specific precursor of chromosomal DNA, is in the brains of adult birds and mammals, utilized exclusively by the nuclei of cells in general, and the absence of signs of that are preparing for multiplication regenerative neuronal proliferation follow­ (Hughes, '59; Hughes, et aI., '58; Leblond, ing brain lesions or trauma in particular. et aI., '59; Taylor, et al., '57), these results This conclusion has not been seriously suggested the possibility of neurogenesis in questioned until recently, even though some forebrain structures in adult mam­ some investigators argued for the exist­ mals. That the autoradiographic "labeling" ance in the mature brain of "indifferent of cell nuclei in the brain following injeC­ cells" (Schaper, 1897) or "medulloblasts" tion of thymidine-H is associated with (Bailey and Cushing, '25) which can dif­ cellular proliferation was supported by our ferentiate into neurons (for a history, see observation of the labeling of a good pro­ Globus and Kuhlenbeck, '44; Jones, '32; portion of those glia cells that were in­ Kershman, '38), while others claimed to duced to multiply in regions of experimen­ have observed mitotic neurons in young tal brain lesions or in areas structurally mammals (for a history, see Kjellgren, '44). and functionally connected with the trau­ In pilot studies employing fine-resolu­ matized sites (Altman, '62a). The same tion autoradiography, we have recently conclusion could also be drawn from the observed (Altman, '62b) that following finding that neuroglia and microglia cells intracranial injection of thymidine-H" into young adult rats there was an accumula­ 1 This study was supported by the U. S. Atomic Energy Commission, and supplJementary aid was re­ tion of reduced silver grains over the nu­ ceived from the John A. Hartford Foundation. We clei of a few neurons in the neocortex and, wish to thank Elizabeth Altman, William J. Anderson and Louise Wasserman for their assistance in various more commonly and consistently, over the phases of this program. J. COMPo NEUR., 124: 319-336. 31~