We have studied cell lineage in the rat cerebral cortex using retroviral mediated gene transfer. By this method, a marker gene is inserted into dividing precursor cells such that their fate can be followed. We have applied this technique to two types of experiment. First, virus was used to label precursor cells of the cerebral cortex in situ during the period of neurogenesis. Second, cortical precursor cells were grown in dissociated cell culture, and virus was used to follow their development over the culture period. These experiments showed that the majority of precursor cells generate a single cell type - neurones, astrocytes, or oligodendrocytes. Moreover, this is true both in vivo and in dissociated cell culture. The only exception is a bipotential cell, which can generate both neurones and oligodendrocytes. These data suggest that the ventricular zone - the germinal layer of the embryonic cortex - is a mosaic of precursor cells of different restricted potentials.

Although precursor cells are restricted in terms of the cell types they generate, they seem not to be restricted in either the cortical laminae or cytoarchitectonie areas to which they can contribute. Both neuronal and grey matter astrocyte precursors contribute cells to multiple layers of both infra- and supragranular laminae. Moreover, in the hippocampal formation, neuronal precursors can contribute cells to more than one hippocampal field.

During development, cells take on a wide variety of different fates, providing the adult animal with the appropriate variety of different cell types in the appropriate proportions and positions. In the cerebral cortex, cell fate can be thought of as having three aspects. First, cells have to become one of the many types of neurones or glial cells of which the cortex is composed. Second, cortex is a layered structure, and the lamina to which a cell belongs (for a neurone at least) is an important part of its identity. Third, cortical cells belong to different cytoarchitectonie areas. Although the cortical mantle is a continuous sheet of tissue, it is divided vertically into areas that are anatomically and functionally distinct. Moreover, neocortical areas appear to be subdivided further into columns of cells, physiologically distinct from neighbouring columns (See Woolsey and Van der Loos, 1970; Hubei and Wiesel, 1977; and Mountcastle, 1978 inter alia).

All cortical neurones and macroglial cells are originally derived in the embryo from the ventricular zone. We have asked the question: how many types of precursor cell are there in the ventricular zone? Is there a single pluripotential precursor cell type that generates all the mature cortical cell types, as seems to be the case in the retina (Turner and Cepko, 1987; Wetts and Fraser, 1988; Holt et al. 1988; Turner et al. 1990), or are different cell types derived from different subpopulations of precursors? If the latter alternative is correct, in which aspect of cortical cell fate are precursors restricted? Are they restricted in the cell types they generate, or in the lamina or area to which they contribute?

We have approached these questions by using retroviral vectors as cell lineage markers (Price et al. 1987). By infecting precursor cells with virus, we introduce a marker gene into the infected cells. The gene is inherited undiluted by all the progeny of the infected cells, and its expression provides a histochemically detectable label of the marked clones. (For review see Price, 1987; Cepko, 1988; and Sanes, 1989.) We have employed this approach in two ways. First, we have infected embryonic cortical precursor cells in vivo and determined their normal fate. Second, we have infected precursor cells in dissociated cell cultures derived from embryonic cortex. In this way, we have been able to compare the behaviour of cells in vivo with that in dissociated culture, where the normal histogenetic interactions are disrupted.

In a series of experiments in vivo, we injected a retroviral vector into the cerebral vesicles of rat embryos in utero to label precursor cells of the telencephalic ventricular zone. The animals continue to develop normally for a variable specified perioc Then their cortices were analysed histochemically t determine which cell types the infected precursor cel had generated. Variations on this experiment include the type and amount of virus injected, and the age ( the animals at injection and at analysis.

In one set of experiments we injected BAG, retroviral vector that contains the lac-Z gene as marker (the gene that encodes E. coli galactosidase driven from the endogenous retroviral promoter (Prie et al. 1987). Sufficient virus was injected in one series c experiments to label about 20 clones per brain (Prie and Thurlow, 1988), and in a second series between on and six clones (Grove and Price, unpublished obsei valions). We got similar results with both groups e data. We injected virus at embryonic day 16 (E16 which is towards the middle of cortical neurogenesi; and analysed the brains on postnatal day 14 (P14).

We identified four distinct types of clones. Thre were simple in composition: one was composed entirel of neurones; the second of grey matter astrocytes; an the third of white matter cells, probably glia or the: precursors. The fourth type of clone, however, include both neurones and white matter cells, each indis tinguishable from the cells in clones that wer composed of a single cell type (Price and Thurlov 1988).

The neuronal clones were small, mostly having on or two cells and rarely more than six cells. The majority of these cells were supragranular neurones as would be predicted from [3H]thymidine studies, which show that most infragranular neurones have already been generated at E16, the time of virus injection (Miller, 1986). In some of the labelled neurones, the X-gal staining was confined to the nucleus and to one or two small spots at the base of the principle dendrite. This pattern was so distinctive that we were able to use it as a marker for neurones. We have verified that this staining pattern is indeed unique to neurones by injecting nuclear stained cells with the fluorescent dye, Lucifer Yellow. This reveals the morphology of the whole cell, including its axon and dendrites, and consequently makes the cell easy to identify (Grove, Li, and Price, unpublished observations). Also, many nuclear-labelled cells stain with antibodies against the neuronal marker, MAP-2, or against the neurotransmitter, GABA, but they do not react with an antibody against the astrocyte marker, GFAP (Grove and Price, unpublished observations). However, not all neurones show this nuclear pattern of staining, and even in those that do it appears to be a transient phenomenon. In older animals, most neurones stain throughout their cytoplasm in a fashion that resembles a classical Golgi stained cell (Fig. 1).

Fig. 1.

A neuronal clone comprising two pyramidal neurones. Scale Bar, 50 μm.

Fig. 1.

A neuronal clone comprising two pyramidal neurones. Scale Bar, 50 μm.

Astrocyte clones are considerably larger than neuronal clones, typically comprising 30-40 cells. They are spread through all laminae of the cortical grey matter, often being found in little groups of two to four cells, as if a cell migrated into the cortex then divided once or twice more in situ. These cells are identifiable as astrocytes because of their morphology, because many have blood vessel or pial endfeet (Price and Thurlow, 1988), and because they stain with antibodies against GFAP (Grove and Price, unpublished observations). Astrocytes never have the restricted pattern of β- galactosidase staining seen in neurones.

The third type of clone is composed of cells that at post-natal day 14 (P14) are confined to the white matter. (With longer survival times these clones are found to have spread increasingly into the deepest layers of the grey matter.) At P14 these cells are difficult to identify as they are small and have few processes. They could be immature astrocytes or oligodendrocytes, or glial precursor cells. However, a small number have the morphology of oligodendrocytes. They have small round cell bodies and many long, club-ended processes running parallel with the white matter axons (Fig. 2). If the morphologically undifferentiated cells are glial precursors, then with longer survival times more clones of differentiated astrocytes or oligodendrocytes should appear. We are currently testing this prediction.

Fig. 2.

Part of a clone of oligodendrocytes. Three oligodendrocyte cell bodies are visible (arrows), as are many of their processes, which are running predominantly parallel with the axons of the external capsule. Scale bar, 50 μm.

Fig. 2.

Part of a clone of oligodendrocytes. Three oligodendrocyte cell bodies are visible (arrows), as are many of their processes, which are running predominantly parallel with the axons of the external capsule. Scale bar, 50 μm.

It is unlikely that any of the white matter cells are neurones. They do not look like neurones, and the neurones in this region - some of the earliest cells generated in the cortex (Raedler and Raedler, 1978) - would have been postmitotic when the viral injection was made at E16.

In the fourth type of clone, we found both neurones and white matter cells. The labelled white matter cells appeared immediately below the labelled neurones. This suggested that both groups of cells were derived from the same ventricular zone cell, the neurones having migrated out to the cortical plate while the glial cells migrated only as far as the white matter (Price and Thurlow, 1988). We were able to discount the possibility that these two cell types were derived from two neighbouring cells that had been labelled. The mixed clones constituted 9% of all clones, a frequency too great to be accounted for by a superimposition of clones, especially as this was the only type of mixed clone observed; no other combination of astrocytes, white matter glia, or neurones was found.

In a series of tissue culture studies, we attempted to reproduce in culture the in vivo experiment described above. Cells were dissociated from E16 embryonic cortex, and plated onto a support monolayer of cortical astrocytes. They were infected with a titre of virus that labelled between three and eight precursor cells of the several hundred thousand cells plated per coverslip. After 12 days in vitro, the cultures were stained with an anti-/3-galactosidase serum together with a pair of cell type-specific monoclonal antibodies. By using different combinations of antibodies, we were able to identify all the major cell types in the cultures and the clones in which they occurred. We used antibodies against either neurofilament (Wood and Anderton, 1981) or the microtubule associated protein, MAP2 (Binder et al. 1986) to recognise neurones; oligodendrocytes were identified with anti-galactocerebroside (Ranscht et al. 1982) or 04 (Sommer and Schachner, 1981); and astrocytes were stained with GFAP (Bignami et al. 1972).

We found four different types of clones, just as in the in vivo studies (Williams et al. 1991). Three of these were composed of a single cell type - neurones, astrocytes, or oligodendrocytes. The fourth type of clone included two cell types, neurones and oligodendrocytes. So, there was a close correspondence between the in vivo and culture experiments. In both there were neuronal clones and astrocyte clones. In culture there were clones of oligodendrocytes; in vivo there were clones that included oligodendrocytes but also included cells that we think are glial precursors. Finally, in both experiments we saw evidence of a bipotential precursor that generated neurones and oligodendrocytes in culture, neurones and white matter cells in vivo.

We have drawn two conclusions from this result. First, the four types of precursor cells we see represent stable phenotypes. As these precursor cells are neuroepithelial cells of the ventricular zone, it might have been supposed that their developmental potential would be restricted by cell-cell interactions in the neuroepithelial cell layer. Thus in the less restricted tissue culture environment, cells might have shown a broader potential. This was not observed, which suggests that the restricted patterns of behaviour are stable, cell-autonomous phenotypes.

Second, although the majority of both neurones and oligodendrocytes at E16 are generated from separate precursor cell populations, roughly 20% of the precursor cells that give rise to neurones, also generate oligodendrocytes.

Just as the question can be posed, is a precursor cell committed to the production of one cell type; so it can be asked, is it committed to the production of cells for one area or one lamina?

Laminar distribution of clones

Our experiments show that both the astrocyte and neuronal precursors can contribute cells to multiple laminae. Many astrocyte clones contribute cells to all laminae from I to VI (Price and Thurlow, 1988). However, astrocytes may not have a lamina identity in the sense that neurones do, so this result is probably not surprising. A narrower distribution is found for neuronal clones in most of our experiments simply because most infragranular neurones have already been generated at the time of labelling at E16 (Miller, 1986). Even injections at E14 give clones of mostly supra-granular neurones, although the clones are larger and include more infragranular neurones than clones in animals injected at E16. Despite this bias, we have seen no evidence of any simple restriction in the laminae to which neuronal precursors can contribute. These data do not exclude the possibility that there are subpopulations among the neuronal precursors that preferentially send cells to one lamina, or to a particular combination of laminae. However, we can conclude that there are precursor cells that can contribute to both infragranular and supragranular laminae (Fig. 3), and that within supragranular cortex, precursors can contribute to more than one lamina.

Fig. 3.

Camera lucida drawings of three neuronal clones. The labelled neurones are represented by filled circles, and their positions are indicated in relation to cortical laminae (labelled with roman numerals: EC is external capsule). Each clone is distributed across several cortical laminae. Scale bar, 500 μm.

Fig. 3.

Camera lucida drawings of three neuronal clones. The labelled neurones are represented by filled circles, and their positions are indicated in relation to cortical laminae (labelled with roman numerals: EC is external capsule). Each clone is distributed across several cortical laminae. Scale bar, 500 μm.

The distribution of clones between cortical areas

In addition to the lamina spread, there is a considerable degree of tangential dispersion of clones in the cerebral cortex (see Fig. 3). This is true for both neuronal and astrocyte clones. The exact extent of this dispersion has been difficult to quantify, but is at least several hundred micrometers (Luskin et al. 1988; Price and Thurlow, 1988; Walsh and Cepko, 1988). This is clearly different from clones in the retina which are so narrowly distributed that even in the mature animal a clone is a discrete column of cells (Turner and Cepko, 1987). However, the rodent cortex may not be so different from chick optic tectum as was earlier thought to be the case. The first studies of clonal spread in the optic tectum showed that the predominant distribution of clones was narrowly radial (Gray et al. 1988). However, it is now clear that some tectal neurones also undergo considerable tangential spread (Gray and Sanes, 1991).

Given the spread in the cortex, the question arises - are clones being distributed across the anatomical and physiological boundaries that divide the cortex? We have studied the distribution of clones in relation to two types of boundaries. (1) The boundaries between barrels in the somatosensory cortex (Moore and Price, unpublished data). (2) The boundaries in the hippocampal formation between the subiculum and Ammon’s horn, and between the CA fields of the hippocampus proper (Grove and Price, unpublished observations).

In both cases we find that clones include cells that fall on both sides of a boundary. So by E16, the day of labelling in these experiments, these borders are not clonal restriction boundaries, such as exist between rhombomeres in the hindbrain (Fraser et al. 1990). This is not to say that clonal restriction boundaries do not occur at other sites in the telencephalon.

These findings can be interpreted in two ways. It could be that the boundaries we are considering have not been formed at E16, the time of viral injection. This interpretation is consistent with a large body of data which shows that the barrel boundaries do not appear until after birth in the rodent, and are strongly influenced by afferent input (Van der Loos and Woolsey, 1973). It is also consistent with the observation that heterotopic and heterochronic grafts between cortical regions from late embryos regulate, in that they take on some of the connectional characteristics of the host region (Stanfield and O’Leary, 1985; O’Leary and Stanfield, 1989). However, similar grafting experiments have suggested that limbic cortex becomes committed between E12 and E14 to the expression of a limbic system-specific marker, and does not regulate if grafted to sensorimotor cortical regions after this time (Barbe and Levitt, 1991). A hypothesis that is consistent with all these data is that large sectors of cortex (i.e. limbic and sensorimotor regions) become delineated before the period of neurogenesis, but divisions within these regions (e.g. subiculum, CA fields) arise later. We have not studied the borders between limbic and sensorimotor cortex for lineage restriction.

A second possibility is that individual area delineations do exist in the cortex prior to neurogenesis, but unlike borders between rhombomeres, these boundaries are not clonal compartment boundaries. There are several feasible ways in which such boundaries could be maintained even though postmitotic cells are free to cross them. For example, a cell’s area identity could be specified in the ventricular zone, but in a manner that could be regulated by the identity of surrounding cells. Consequently, although nominally specified, a neurone that migrated into a neighbouring area would be influenced to switch its fate appropriately. In this manner, a boundary would be maintained between regions even though individual cells were free to cross it.

The observations of Balaban et al. (1988) in the chick forebrain would be consistent with this type of model. They grafted pieces of quail forebrain isochronically and isotopically into a chick host at neural fold stages. They observed considerable mixing of postmitotic neurones of host and graft origins. However, there was much less mixing across the border of the graft in the ventricular zone.

We conclude that the boundaries we have studied are not clonal compartment boundaries at the stages we have examined. Thus precursor cells can contribute to more than one cortical area. This means either that they are not specified in this respect, or that such a specification is plastic and can be changed when cells migrate into neighbouring regions.

Our experiments show that there is considerable heterogeneity among ventricular zone cells. Combining our in vivo and in vitro observations, it appears that neurones, grey matter astrocytes, and oligodendrocytes each have a dedicated population of precursor cells.

In addition, there is a bipotential cell, which we call the N-O cell, that can give rise to both neurones and oligodendrocytes. We presume that this N-O cell is the precursor of both the neuronal and oligodendrocyte precursors. In this sense, it would be an earlier cell ontogenetically than the two more restricted precursors to which it gives rise. However, it is also possible that the N-O cell represents an independent, parallel lineage by which oligodendrocytes and neurones are generated. A third possibility is that the N-O cell is not a separate precursor cell type at all. Rather what we have observed is the propensity of some precursor cells to switch between the neurone- and oligodendrocyteproducing phenotypes. More work is needed to clarify the relationships between these different precursor cell types.

The oligodendrocyte precursor we have observed is probably an O-2A like cell. The O-2A cell was first defined in rat optic nerve cultures as a bipotential cell, capable of generating both oligodendrocytes and a variety of astrocyte, the type 2 astrocyte (see Raff, 1989, for review). The cell type that is produced is determined by whether or not the culture includes certain factors found in fetal calf serum (FCS). In the absence of such factors, the cells generate oligodendrocytes; in the presence of FCS, type 2 astrocytes are generated (Raff et al. 1983). Although they were originally described in the optic nerve, cells with the O-2A phenotype have since been described in a variety of regions of the CNS, including the cerebral cortex (Williams et al. 1985; Dubois-Dalcq, 1987; Ingraham and McCarthy, 1989; Grinspan et al. 1990; Vaysse and Goldman, 1990). As the cultures described above did not contain FCS, any O-2A cells present would be expected to generate oligodendrocytes. Consequently, we presume that the cortical oligodendrocyte precursor is indeed an O-2A cell, or a pre-O-2A cell. In line with this supposition, in the presence of FCS our cultures give rise to fewer clones of oligodendrocytes and more astrocytes, but we have yet to demonstrate that these astrocytes are type 2 cells serologically. In fact, judging by morphological criteria, the classification of cortical astrocytes might turn out to be more complex than it is in cultures of optic nerve. Clones of astrocytes in our cultures (grown in either the presence or absence of FCS) are composed of a much greater variety of morphological types than are similar cultures derived from the optic nerve.

If we accept that the oligodendrocyte precursor is an 0-2A cell, the other outstanding question at this time is, do the in vivo white matter clones include astrocytes as well as oligodendrocytes? This question has enhanced significance as there is no evidence so far that the O-2A cell gives rise to more than oligodendrocytes in the in vivo optic nerve. We are currently studying this question.

All the data presented here relate to the period of development after neurogenesis has begun, and as we have described, precursor cells show a restricted potential at these stages. However, a question that is begged by these results is, how are these restricted precursor cells derived? We would expect that before neurogenesis begins, there would be a population of multipotential precursors from which the later precursor cells are generated. Certainly, lineage studies of earlier stages of chick CNS have demonstrated cells with broader potentials in both tectum (Gray et al. 1988; Galileo et al. 1990) and spinal cord (Leber et al. 1990) than we have observed in the cerebral cortex. However, it is not clear in these CNS regions when or how (or indeed, if) precursor cells of more restricted potential arise. In vivo cell lineage studies of younger rodent embryos are technically difficult, but clearly they are required to unify these disparate observations.

We would like to thank Drs L. I. Binder, A. Matus, and M. C. Raff for kindly providing us with antibodies. We would also like to acknowledge the financial support of the Multiple Sclerosis Society.

Balaban
,
E.
,
Teillet
,
M-A.
and
Le Douarin
,
N.
(
1988
).
Application of the quail-chick chimera system to the study of brain development and behavior
.
Science
241
,
1339
1345
.
Barbe
,
M. F.
and
Levitt
,
P.
(
1991
).
The early commitment of fetal neurons to the limbic cortex
.
J. Neurosci
.
11
,
519
533
.
Bignami
,
A.
,
Eng
,
L. F.
,
Dahl
,
D.
and
Uyeda
,
C. T.
(
1972
).
Localisation of the glial fibrillary acidic protein in astrocytes by immunofluorescence
.
Brain Res
.
43
,
429
435
.
Binder
,
L. L
,
Frankfurter
,
A.
and REBHUN, L. 1
. (
1986
).
Differential localization of MAP-2 and tau in mammalian neurons in situ
.
Ann. N.Y. Acad. Sci
.
466
,
145
166
.
Cepko
,
C.
(
1988
).
Retrovirus vectors and their applications in neurobiology
.
Neuron
1
,
345
353
.
Dubois-Dalcq
,
M.
(
1987
).
Characterisation of a slowly proliferative cell along the oligodendrocyte pathway
.
EMBO J
.
6
,
2587
2595
.
Fraser
,
S.
,
Keynes
,
R.
and
Lumsden
,
A.
(
1990
).
Segmentation in the chick embryo hindbrain is defined by cell lineage restrictions
.
Nature
344
,
431
435
.
Galileo
,
D. S.
,
Gray
,
G. E.
,
Owens
,
G. C.
,
Majors
,
J.
and
Sanes
,
J. R.
(
1990
).
Neurons and glia arise from a common progenitor in chicken optic tectum: demonstration with two retroviruses and cell type-specific antibodies
.
Proc. natn. Acad. Sci. U.S.A
.
87
,
458
462
.
Gray
,
G. E.
,
Glover
,
J. C.
,
Majors
,
J.
and
Sanes
,
J. R.
(
1988
).
Radial arrangement of clonally related cells in the chicken optic tectum: lineage analysis with a recombinant retrovirus
.
Proc, natn. Acad. Sci. U.S.A
.
85
,
7356
7360
.
Gray
,
G. E.
and
Sanes
,
J. R.
(
1991
).
Migratory paths and phenotypic choices of clonally related cells in the avian optic tectum
.
Neuron
6
,
211
225
.
Grinspan
,
J. B.
,
Stern
,
J. L.
,
Pustilnik
,
S. M.
and
Pleasure
,
D.
(
1990
).
Cerebral white matter contains PDGF-responsive precursors to O2A cells
.
J. Neurosci
.
10
,
1866
1873
.
Holt
,
C. E.
,
Bertsch
,
T. W.
,
Ellis
,
H.
and
Harris
,
W. A.
(
1988
).
Cellular determination in the Xenopus retina is independent of lineage and birth date
.
Neuron
1
,
15
26
.
Hubel
,
D. H.
and
Wiesel
,
T. N.
(
1977
).
Functional architecture of macque monkey visual cortex
.
Proc. R. Soc. Lond. B
.
198
,
1
—59.
Ingraham
,
C. A.
and
Mccarthy
,
K. D.
(
1989
).
Plasticity of process-bearing glial cell cutures from neonatal rat cerebral cortical tissue
.
J. Neurosci
.
9
,
63
69
.
Leber
,
S. M.
,
Breedlove
,
S. M.
and
Sanes
,
J. R.
(
1990
).
Lineage, arrangement, and death of clonally related motoneurons in chick spinal cord
.
J. Neurosci
.
10
,
2451
2462
.
Luskin
,
M. B.
,
Pearlman
,
A. L.
and
Sanes
,
J. R.
(
1988
).
Cell lineage in the cerebral cortex of the mouse studied in vivo and in vitro with a recombinant retrovirus
.
Neuron
1
,
635
647
.
Miller
,
M. W.
(
1986
).
Effects of alcohol on the generation and migration of cerebral cortical neurons
.
Science
233
,
1308
1311
.
Mountcastle
,
V. B.
(
1978
).
An organising principle for cerebral function: the unit module and the distribution system
. In
The Mindful Brain
(eds
Edelman
,
G. M.
and
Mountcastle
,
V. B.
)
MIT Press
:
Cambridge, USA
, pp.
7
50
.
O’leary
,
D. D. M.
and
Stanfield
,
B. B.
(
1989
).
Selective elimination of axons extended by developing cortical neurons is dependent on regional locale: experiments utilizing fetal cortical transplants
.
J. Neurosci
.
9
,
2230
2246
.
Price
,
J.
(
1987
).
Retroviruses and the study of cell lineage
.
Development
101
,
409
419
.
Price
,
J.
and
Thurlow
,
L.
(
1988
).
Cell lineage in the rat cerebral cortex: a study using retroviral-mediated gene transfer
.
Development
104
,
473
482
.
Price
,
J.
,
Turner
,
D.
and
Cepko
,
C.
(
1987
).
Lineage analysis in the vertebrate nervous system by retrovirus-mediated gene transfer
.
Proc. natn. Acad. Sci. U.S.A
.
84
,
156
160
.
Raedler
,
E.
and
Raedler
,
A.
(
1978
).
Autoradiographic study of early neurogenesis in rat neocortex
.
Anat. Embryol
.
154
,
267
312
.
Raff
,
M. C.
(
1989
).
Glial cell diversification in the rat optic nerve
.
Science
243
,
1450
1455
.
Raff
,
M. C.
,
Miller
,
R. H.
and
Noble
,
M.
(
1983
).
A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on the culture medium
.
Nature
303
,
390
396
.
Ranscht
,
B.
,
Clapshaw
,
P. A.
,
Price
,
J.
,
Nobel
,
M.
and
Seifert
,
W.
(
1982
).
Development of oligodendrocytes and Schwann cells studied with a monoclonal antibody against galactocerebroside
.
Proc. natn. Acad. Sci. U.S.A
.
79
,
2709
2713
.
Sanes
,
J. R.
(
1989
).
Analysing cell lineage with a recombinant retrovirus
.
TINS
12
,
21
28
.
Sommer
,
I.
and
Schachner
,
M.
(
1981
).
Monoclonal antibodies (O1 and 04) to oligodendrocyte cell surfaces: an immunocytological study in the central nervous system
.
Devi Biol
.
83
,
311
327
.
Stanfield
,
B. B.
and
O’leary
,
D. D. M.
(
1985
).
Fetal occipital cortical neurones transplanted to the rostral cortex can extend and maintain a pyramidal tract axon
.
Nature
313
,
135
137
.
Turner
,
D.
and
Cepko
,
C.
(
1987
).
Cell lineage in the rat retina: a common progenitor for neurons and glia persists late in development
.
Nature
328
,
131
136
.
Turner
,
D. L.
,
Snyder
,
E. Y.
and
Cepko
,
C. L.
(
1990
).
Lineageindependent determination of cell type in the embryonic mouse retina
.
Neuron
4
,
833
845
.
Van Der Loos
,
H.
and
Woolsey
,
T. A.
(
1973
).
Somatosensory cortex: structural alterations following early injury to sense organs
.
Science
179
,
395
397
.
Vaysse
,
P. J.-J.
and
Goldman
,
J. E.
(
1990
).
A clonal analysis of glial lineages in neonatal forebrain development in vitro
.
Neuron
5
,
227
235
.
Walsh
,
C.
and
Cepko
,
C. L.
(
1988
).
Clonally related cortical cells show several migration patterns
.
Science
241
,
1342
1345
.
Wetts
,
R.
and
Fraser
,
S. E.
(
1988
).
Multipotential precursors can give rise to all major cell types of the frog retina
.
Science
239
,
1142
1145
.
Williams
,
B. P.
,
Abney
,
E. R.
and
Raff
,
M. C.
(
1985
).
Macroglia cell development in embryonic rat brain: studies using monoclonal antibodies, fluorescence activated cell sorting, and cell culture
.
Devi Biol
.
112
,
126
134
.
Williams
,
B. P.
,
Read
,
J.
and
Price
,
J.
(
1991
).
The generation of neurons and oligodendrocytes from a common precursor cell
.
Neuron In press
.
Wood
,
J. D.
and
Anderton
,
B. H.
(
1981
).
Monoclonal antibodies to mammalian neurofilaments
.
Biosci. Rep
.
1
,
263
268
.
Woolsey
,
T. A.
and
Van Der Loos
,
H.
(
1970
).
The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex
.
Brain Res
.
17
,
205
242
.