Dr. Karten
began his talk by raising the problem of the evolutionary
origins of the neocortex in terms of comparative vertebrate
neuroanatomy. He pointed out that most brain regions
such as the cerebellum and spinal cord have a remarkably
similar structure in the brains of fish, amphibia, reptiles,
birds and mammals. In contrast, a clearly identifiable
neocortex is absent from nonmammalian species. In fact,
the presence or absence of the neocortex is as sure
a taxonomic quality of mammals as hair or mammary glands.
This raises the question of which structures in nonmammalian
brains gave rise to the neocortex.
Dr. Karten
went on to review historically the neuroanatomical literature
on homologies between the avian and mammalian forebrain.
The avian forebrain turns out to consist of a thin pallium
surrounding a very large mass of gray matter, in contrast
to the large cortex and smaller central gray matter
of mammalian brains. Early comparative neuroanatomists
assumed that the avian central gray matter mass were
homologous to the mammalian basal ganglia, because of
their similar position relative to the ventricles. Subsequent
studies, however, revealed a great heterogeneity of
these "striatum-like" structures. They can be roughly
divided into a basal and dorsal ventricular ridge. On
the basis of his own studies of the pigeon brain, using
staining patterns for acetyl cholinesterase, dopamine
and substance P, and the pattern of specific sensory
afferents, he concluded that in fact it is only the
basal ventricular ridge which is homologous to the mammalian
basal ganglia.
Similar studies
of connectivity patterns in the visual systems of birds
and mammals suggested that the dorsal ventricular ridge
(DVR) may bear homology to the primary input layers
(layer 4) of the extra striate regions of the mammalian
visual cortex. Another structure, the visual wulst was
found to be homologous to primary visual (striate) cortex.
This suggested that the cortex may therefore have originated
not from a single structure as suggested by Allman,
but from the integration of two separate structures.
Similar studies of somatosensory and auditory pathways
led to the same conclusion. A potential difficulty with
this view is that while the wulst is like the cortex
a laminated structure, the DVR is not. Dr. Karten then
digressed to recount another example where clearly homologous
neural structures are in one species highly laminated,
while in another closely related species they are not.
This example involves gustatory organs in fish. In the
catfish the nucleus which receives the vagal gustatory
afferents is crudely developed and non-laminated, while
in the goldfish, the same region is highly developed
and is fully laminated.
Dr. Karten
then went on to discuss the issue of how the DVR and
visual wulst develop. Birth dating studies using bromodeoxy
uridine suggested in mammals that early on there is
DVR equivalent, the subventricular zone (SVZ). A common
feature of cortical development is the "inside out"
pattern in which deep layer cells are born earliest
and superficial cells are born later. The deep layer
cells contain efferent neurons that project out of the
cortex, the middle layers contain recipient neurons
that receive thalamic inputs, and the superficial layers
contain, broadly speaking, interneurons that project
within the cortex. A similar birth dating pattern of
efferent, then recipient then interneuronal populations
was found in the avian brain, although here the separate
populations were located in separate regions rather
than in separate layers within the same region.
Dr. Karten
then speculated that the cell populations which occupy
different laminar positions in mammalian cortex and
different regional positions in the avian wulst and
DVR may in fact correspond to neuromeres, first described
by Ben Kalaine. He also raised the intriguing possibility
that recently described homeobox genes which have been
shown to label neuromeric structures in other parts
of the nervous system could potentially be used to test
his hypothesis of the dual origins of the mammalian
neocortex. Preliminary results suggest that in fact
there are HOX gene homologs which can be recognized
within the avian DVR.
