вход по аккаунту


Morphological differences between minicolumns in human and nonhuman primate cortex.

код для вставкиСкачать
Morphological Differences Between Minicolumns in
Human and Nonhuman Primate Cortex
Daniel P. Buxhoeveden,1,2* Andrew E. Switala,1 Emil Roy,1 Mark Litaker,3 and Manuel F. Casanova1
Department of Psychiatry, Medical College of Georgia, Augusta, Georgia 30904
Department of Anthropology, University of South Carolina, Columbia, South Carolina 29208
Department of Biostatistics, Medical College of Georgia, Augusta, Georgia 30904
brain evolution; planum temporale; language
Our study performed a quantitative investigation of minicolumns in the planum temporale
(PT) of human, chimpanzee, and rhesus monkey brains.
This analysis distinguished minicolumns in the human
cortex from those of the other nonhuman primates. Human cell columns are larger, contain more neuropil
space, and pack more cells into the core area of the
The search for anatomical differences between the
brains of humans and nonhuman primates, especially the African apes, has proven limited. This
applies especially to areas of the brain long associated with language, despite the uniqueness of the
human form of language among primates (Deacon,
1997). MR imaging studies report identical asymmetries of the planum temporale (PT) in both humans
and the African great apes, though not in Old or
New World monkeys (Gannon et al., 1998; Hopkins
et al., 1998).
The use of MR-imaging has also found the human
frontal lobe to be no larger than expected in an ape
brain our size, despite some long-held assumptions
to the contrary (Semendeferi et al., 1997). It is important to note that decades ago, data countered the
notion that humans had relatively more frontal lobe
(Brodmann, 1912; Brummelkamp, 1939; Holloway,
1968; LeBoucq, 1928.). Arguably, information regarding cortical volume alone is limited and difficult
to interpret (Deacon, 1997). This assertion does not
diminish the importance of macroscopic studies,
since microscopic studies also involve difficulties of
their own. Rather, basing comparisons of primate
brains on macroanatomical structures alone necessarily overlooks the proliferation of functional units
in large regions of the cortex; each has the potential
for different pathways and intracortical connections.
Macroscopic examination of the brain at the gross
anatomical level, therefore, often fails to adequately
detect organizational differences in cortical tissue
contained within these units. While it provides information at one level of inquiry (e.g., is there more
or less temporal lobe?), it cannot determine what is
occurring beneath the tissue. As Holloway (1968)
column than those of the other primates tested. Because
the minicolumn is a basic anatomical and functional
unit of the cortex, this strong evidence showed reorganization in this area of the human brain. The relationship between the minicolumn and cortical volume is also
discussed. Am J Phys Anthropol 115:361–371, 2001.
2001 Wiley-Liss, Inc.
asserted, “A comparison of mass is not a comparison
of equal units.”
A brief list of some of the more recent comparisons
between human and nonhuman primate brains includes studies of morphology and size (Conroy et al.,
1998, 2000; Falk et al., 1989, 1999; Holloway, 1968,
1973, 1974; Jerison, 1977, 1985; Radinsky, 1972,
1973, 1975, 1979; Rilling and Insel, 1999; Semendeferi and Damasio, 2000; Semendeferi et al., 1997,
1998), metabolism (Armstrong, 1990), histology
(Holloway, 1968; Heilbroner and Holloway, 1989),
gyral folding (Armstrong et al., 1991; Zilles et al.,
1988, 1989), cerebral blood flow (Falk et al., 1988;
Falk, 1993; Tobias and Falk, 1988), and cerebral
asymmetries (Corballis et al., 2000; Falk, 1978; Gazzaniga, 2000a,b; Holloway and De La Costelareymondie, 1982; LeMay and Geschwind, 1975; LeMay,
1976, 1985; Yeni-Komshian and Benson, 1976).
To discern differences between humans and primate brains, research must explore the internal operations of the cortex in ways that best interface
anatomy with physiology. To do this, we propose
using the cortical minicolumn as a template. We
have found significant differences between the morphology of cell columns in the human planum temporale and those of the chimpanzee or monkey (Fig.
1). Researchers have generally assumed that minicolumns share the same morphology among species
(Carpenter et al., 1976; Mountcastle, 1997). How*Correspondence to: Daniel Buxhoeveden, Ph.D., Downtown VA
Medical Center, 116-A, Psychiatry Service, 3B-121, Augusta, GA
30904. E-mail: [email protected]
Received 19 September 2000; accepted 6 May 2001.
Fig. 1.
Vertical cell columns in lamina III in the PT of primate cortex. Viewed at 100⫻.
ever, no one has consistently examined this topic,
despite indications of differences across species and
between different cortical regions of the brain (Buxhoeveden et al., 1996; Peters and Sethares, 1996;
Preuss et al., 1999; Schlaug et al., 1995; Seldon,
1981a,b). Recent work demonstrates that although
researchers have long assumed that humans and
primates have very similar primary visual cortexes,
distinctive morphological differences exist between
human and chimpanzee V1 in lamina 4 (Preuss et
al., 1999).
The minicolumn is a basic anatomical and physiological unit of many areas of the brain (Favorov and
Kelly, 1994a– c; Mountcastle, 1997). However, this
may not hold true for the entire cortex (Jones, 2000;
Swindale, 2000). The minicolumn interfaces cellular
organization with the larger macrocolumn or module (Fig. 2). The physiological basis for the minicolumn has been supported by experiments using microelectrodes, nerve regeneration, optical density,
and metabolic activity (Favorov and Whitsel,
1988a,b; Favorov and Diamond, 1990; Kaas et al.,
1981; Kohn et al., 1997; Lee and Whitsel, 1992;
Mountcastle, 1957, 1978, 1997; Tommerdahl et al.,
1993; Yuste et al., 1992). Because it is a fundamental anatomical and physiological unit of the cortex,
changes in the morphology of the column may be
expected to reflect alterations in function. Columns
are easy to recognize with the aid of a low-power
microscope in many areas of the cortex. However,
morphological differences of the kind described below are not discernible to the naked eye and require
quantified computerized imaging.
The size of minicolumns based on in vivo physiological studies, primarily in the somatosensory and
visual cortex, seems to range between 30 – 60 ␮m
(Favorov and Whitsel, 1988a,b; Favorov and Diamond, 1990; Kaas et al., 1981; Peters and Yilmaz,
1993; Peters and Walsh, 1972). The smaller columns
are found in the primary visual cortex. A mean size
of 56 ␮m, based on apical dendrite clusters, has been
suggested across many different species including
rabbits, rats, cats, monkeys, and humans (Mountcastle, 1997). A metabolic (2DG) examination of columns in the somatosensory cortexes of cats and
monkeys revealed column widths between 30 –50
␮m (Tommerdahl et al., 1993). It is also important to
recognize that cell column size can vary between
cortical areas (Buxhoeveden et al., 1996; Schlaug et
al., 1995).
The use of the minicolumn in clinical neurology
and comparative studies is just beginning to make
an impact. Recent studies reveal modular organization in the posterior portion of area 22 in humans
(Galuske, 2000), and the structure of minicolumns
has been studied in Alzheimer’s disease (Buldyrev et
al, 2000).
We examined the planum temporale in the left
hemispheres in a population of brains at the Yakovlev-Haleem Collection in Washington, DC. We compared nine human, eight chimpanzee, and seven
rhesus monkey brains. The human samples were
adult except for one 16-year-old. The chimpanzees
averaged 8.6 years, with two of them at 13 years and
one 4-year-old. We considered the lower ages important, since the average age of the chimpanzee brains
was generally younger than that of the humans. The
rhesus monkey brains were listed as either adult or
young adult. We focused on cell columns in lamina
III, because the pyramidal cells and the vertical
organization are most pronounced in this layer. At
higher magnifications, anatomical columns were often very difficult to discern in lamina IV and were
essentially nonexistent in layer II. Our present
method works best for vertically oriented columns,
which are found most consistently in lamina III. A
comparison of laminae in the human cortex demonstrates that layer III in Tpt is the most conservative,
having undergone the least amount of change from
the fetal radial cell column (Buxhoeveden et al.,
1996, 2000). In the future, we will be examining
other laminae for comparative differences across
Potential problems include shrinkage from tissue
preparation, Z-axis artifact, and the plane of cut, all
of which may cause distortion. For more information, the reader is referred to Buxhoeveden et al.
rior axis in coronal section. Correction factors in this
case are based on measurements of atrophy (disease) and shrinkage (processing) in each brain, i.e.,
(V1/V2)1/3 where V1 and V2 are the tissue volume
before and after processing, respectively. In the cerebral cortex, our computerized methods detect and
identify cells unidirectionally from pia to white matter or vice versa, making unnecessary corrections for
shrinkage in this dimension.
Because the width of the column cortex measured
and the thickness of section are constant, shrinkage
of the cortex will result in overcounting of neurons.
Therefore, correction factors are necessary in these
two dimensions, i.e., (V2/V2)2/3. Corsellis et al. (1975)
argued that for nitrocellulose-embedded tissue,
shrinkage considerations may be omitted for cases
where cell count per unit area (defined by natural
boundaries) is the only parameter considered. The
material in this study limits possible confounds by
defining areas based on anatomical boundaries and
by embedding the material in plastic. Since brain
weights are available for human and most of the
monkey brains before and after fixation, appropriate
correction factors can be applied. These corrections
factors are minimal. The brains available for our
study were fixed in 10% formalin for several months.
As the distortion due to fixation stabilizes after no
more than approximately 3 weeks (Quester and
Schroder, 1997), all brains would have reached their
stable, postfixation state when studied. Thus, the
duration of fixation can be ruled out as a potential
Unfortunately, processing information was available only for the human brains and four monkeys,
and not for the chimpanzees. However, there is no
reason the chimpanzee brains should display
shrinkage outside the range of the monkeys or humans, especially since all brains were handled and
processed according to standardized procedures, a
feature of the Yakovlev-Haleem collection. The
amount of shrinkage in human brains ranged from a
low of 5% to a high of 35.8%. The mean was 28.7%.
For the monkeys the range was much tighter, with a
low of 24.7% and a high of 34.7%. The mean was
29.4% and thus essentially identical to the humans
and totally within their range.
Fig. 2. The minicolumn is a bridge between macroanatomy
and microanatomy. It is also a bridge between anatomy and
To eliminate considerations of shrinkage factors
in two dimensions, it is necessary to assess all neurons on a particular section delimited by anatomical
boundaries (e.g., Tpt) rather than on an arbitrary
grid. However, as the same section thickness is used
in all cases where this neuronal population is
counted or identified, shrinkage must still be corrected for depth of section, i.e., the anterior-poste-
Z-axis artifact
The potential for Z-axis overlap of cells from other
minicolumns or minicolumns dissected by sectioning
is a problem present in this kind of cytoarchitectural
analysis. We have dealt with it as follows: 1) We
omitted from analysis lines of cells that display a
lack of continuity, such as large empty spaces in the
center of the minicolumn, indicative of Z-axis artifact, and which do not match the general pattern of
unequivocally clear minicolumns. 2) We used a
threshold level to eliminate incomplete cellular
minicolumns. From an analysis of thousands of
minicolumns, the mean cell number for each lamina
becomes the standard to determine the threshold
level. Generally, we will not consider minicolumns
with fewer than 10 cells in them complete. This level
is set by the operator and can be changed at any
time before or after data has been acquired. 3) The
35-␮m depth of section was approximately equal to
the radial size of a single cell column in the typical
human adult cortex (taking into account shrinkage
factors; see below). Though this does not eliminate
the problem, it minimizes the potential for too much
Z-axis overlap within one section.
Plane of cut
Quite possibly, the orientation of the cut can affect
some of the morphological parameters we use to
describe minicolumns. This can be considered a random error with bias (e.g., where length of minicolumns tends to be reduced). Since results tend to
occur in one direction, no simple way exists to average its effects. Measuring segments of minicolumns
helps avoid this error. Barring any undue manipulation, to the extent bias remains due to the plane of
cut, it would be a constant variable with an effect
equal or similar in both populations (i.e., study and
In the past, the relationship of the VI/white matter border to the pial surface has guided our determination of orientation (Buxhoeveden et al., 1996).
Others have used similar visual approaches to measure minicolumns (Kohn et al., 1997; Tommerdahl et
al., 1993). With whole brains available for study, we
can control for orientation, selecting the optimal
plane of cut for each cortical area to be studied.
The materials were Nissl-stained, celloidin-embedded, and cut in the coronal or sagittal plane at
35-␮m thickness. Since we found no differences between the two planes of cut, we could see no theoretical reason why analyses of cell columns would be
affected unless they were uniformly ovoid in shape
in one direction, which the literature does not support (Calvin, 1998; Favorov and Kelly, 1994a,b;
Mountcastle, 1997).
Method of visualization
Our method is a modification of an earlier version
described elsewhere (Buxhoeveden et al., 2000). A
region of interest (ROI) must be isolated and transferred to the computer imaging system. The ROI is
obtained from a microscopic field of view at a chosen
magnification. Magnifications of 100⫻ are chosen to
resolve both individual perikarya and the complete
laminar depth. Usually, the ROI consists of the entire field of view as seen in the photomicrograph.
Sometimes, however, partial sections are used due
to artifacts or other distortions.
The first stage of the column detection routine
divides a region of interest into overlapping horizontal strips. Cell concentration v is defined for each
strip by
␯ 共x兲 ⬅
⫺8共x⫺x i 兲 2 /d 2
sum being taken over all cells in that strip. The
characteristic scale d is the width of a box that,
placed at random in the field, would enclose one cell
on average. The relative maxima and minima of v
mark the locations of the centers of aggregation of
cell columns and the space between them, respectively. Note that the full width of the Gaussian in
the definition of v is such that cells within less than
1/2 of each other contribute to the same relative
maximum of v and are not resolved into distinct
columns. In order that the primary contribution to v
would come from a region containing, on average,
one cell, the height of the horizontal strips is approximately 2d. The method defines minicolumns as vertical clusters of large neurons delimited on either
side by cell-sparse areas. Imaginary lines through
the sparse areas partition a field into polygonal regions. We refer to such a polygon together with the
totality of small and large neurons contained within
it as a minicolumn segment, yielding several descriptive statistics (Fig. 3).
This study reports on five measures: column
width, neuropil space, grey level ratio, core grey level
ratio, and linearity (Fig. 4). Column width (CW)
describes the horizontal width of a single column, or
equivalently, the distance between adjacent columns. The peripheral neuropil space (NS) is the
width of the cell-poor space on both sides of the
column. NS is found by subtracting the average
width of the column core from CW. We define the
column core as that part of the column that contains
90% of the cell bodies. So that the width of the main
aggregate of cells is insensitive to “outlier” points,
up to 10% of the cells in the column may lie in the
region designated as peripheral neuropil space.
Neuropil traditionally consists of neuronal processes
with high synaptic content and glia (Haug, 1986).
Because our method does not count cells, we accounted for cell density by measuring the amount of
area occupied by Nissl-stained segments, which we
call the grey level ratio (GLR). This measure is based
on the GLI method developed elsewhere and
adapted specifically for use within minicolumns
(Schleicher et al., 1986; Zilles et al., 1982). Before
column detection, the GLR is computed from the
original image by thresholding. We use the total
area of Nissl-stained objects, or pixels below the
threshold, divided by the image area, to estimate the
cellular density within the minicolumn. While this
measurement does not provide us with cell numbers,
it estimates the amount of space dedicated to cell
soma within a column. This measure includes neurons and glia and does not discriminate between
potential differences in ratios between them. An
analysis of cell size within our samples found no
significant difference between the human and nonhuman primate population, which means that density differences would be due to the population of
Fig. 3. A: Large pyramidal cells, used by the column detection routine to identify areas of high cell concentration, are highlighted
in black. B: White lines depict the Euclidean minimum spanning trees of the cell column segments, used by the analysis routine to
describe minicolumn structure. C: Faint gray lines surrounding minimum spanning trees depict the boundaries of the areas of high
cell concentration, used by the analysis routine to compute column width and peripheral neuropil space.
cells within the column and not the size of the cells.
The core grey level ratio (CGLR) is simply the GLR of
the core region of the columns (described above).
Linearity (LIN) is the similarity of the minicolumn
to a straight line. It is computed from the minimum
spanning tree of the column (Fig. 3B), which is the
set of lines linking the cells with the smallest total
length of all such sets. The diameter of the minimum
spanning tree is the longest path from one cell to
another. In Figure 3B, one can see the diameters by
tracing the paths from the uppermost cell in each
column to the lowermost cell. The straight-line distance between these two cells, divided by the length
of the diameter, is LIN. As the cells of the minicolumn approach a linear arrangement, LIN approaches 1; as they deviate from linearity, LIN approaches zero.
Statistical methods
Mixed-model analysis of variance (ANOVA) was
used to evaluate differences in CW, NS, and GLI
between hemispheres across species, taking into account correlation among repeated measurements
made on the same experimental units, while allowing unequal numbers of observations on the different units. A rank transformation was used, due to
nonnormality of the distributions of the sample
data. The statistical test, which evaluates equality
Fig. 4. Schematic representation of lamina IIIb cell columns.
Shaded area represents the neuropil space (so-called by Seldon,
1981) in the periphery of cell columns. This area contains mostly
synapses, dendrites, and unmyelinated axons. The unshaded section is the region in which the majority of the cell soma reside.
Some of the processes found in this part of the column are represented here.
TABLE 1. Mean values for cell column parameters (mean ⫾ sd)1
CW (␮m)
51.0 ⫾ 5.5
36.6 ⫾ 1.7
36.4 ⫾ 3.4
16.9 ⫾ 2.8
26.0 ⫾ 4.0
26.0 ⫾ 4.7
34.3 ⫾ 5.7
46.6 ⫾ 4.8
44.0 ⫾ 5.9
NS (␮m)
21.6 ⫾ 3.8
17.3 ⫾ 1.0
15.8 ⫾ 2.8
0.80 ⫾ 0.01
0.85 ⫾ 0.01
0.84 ⫾ 0.03
CW, column width; GLR, grey level index; CGLR, core gray level ratio; NS, neuropil space; LIN, linearity.
Fig. 5.
Graphic representations of data described in Table 1.
of differences between hemispheres across the species in this model, is the F-test for species by hemisphere interaction. The analysis including all species was followed by separate mixed-model ANOVAs
for each of the species. This approach allowed evaluation of species-specific differences between hemispheres, and thus investigation of the differences
underlying significant interaction terms in the overall analyses. Mixed-model ANOVA was also used to
evaluate differences among species, using measurements made on the left hemisphere only. The statistical test of interest in this analysis is the F-test for
the main effect of species. Post hoc t-tests were used
to evaluate pairwise differences between species in
analyses showing significant overall tests.
other nonhuman primate brains as reported in Table 1 and Figure 5.
Columns were significantly wider in human PT (P ⫽
0.0001) when compared across species. Differences between the individual species are given in Table 2
A,B,C. The mean width for human columns was 51
␮m, while that of the chimpanzee was 36.6 ␮m and
that of the rhesus monkey was 36.4 ␮m. Translation
into three dimensions would yield a volume twice as
large as a cell column in the nonhuman primates.
Even though the chimpanzee brain on average is
about four times larger than the rhesus, their minicolumns are the same size, at least in this area of cortex.
Neuropil space
The basic morphology of the minicolumn in humans is clearly distinguishable from that of the
The cell-poor space in the periphery of cell columns was also considerably enlarged in humans
Column width
TABLE 2. Individual comparisons between primates
2A: Human vs. chimp
2B: Human vs. rhesus
2C: Chimp vs. rhesus
% difference
P value
CW, column width; GLR, grey level index; CGLR, core grey level
ratio; NS, neuropil space; LIN, linearity.
compared to the chimpanzee and rhesus monkey
(P ⫽ 0.0008). No difference between the chimpanzee
and monkey NS was confirmed statistically (P ⫽
0.16). The increase in peripheral neuropil space in
human columns appeared closely related to the
larger width of the columns. A ratio of NS/CW revealed similar proportions for all three species. In
this regard, human columns are merely larger versions of nonhuman primate ones. Nonetheless, the
expansion of this area in human columns has provided more space for dendritic branching, spines,
and other features generally found in this part of the
Grey level ratio
An estimate of cell density demonstrated that in
human columns cell soma occupied much less space
than in either the chimpanzee or rhesus monkey
(P ⫽ 0.0001). The GLRs of the nonhuman primates
were indistinguishable from each other as demonstrated in Table 2C (P ⫽ 0.910). This means as a
whole, the human column contains fewer Nisslstained elements. While it is well known that larger
brains are less cell dense, we found that this occurs
within the context of the minicolumn (Buxhoeveden
et al., 1996).
Core grey level ratio
To examine potential differences in the core of the
column, we also measured the GLR (CGLR) for this
region only (Table 1). As with the overall GLR, the
human core was far less cell dense than that of the
other primates. However, by comparison to the overall GLR, the human core was over 2⫻ denser than
the GLR for the whole column, whereas it was 1.8⫻
more for the chimpanzee and 1.7⫻ more for the
monkey (P ⫽ 0.0001). Apparently, the core of the cell
column is more compact in humans. These findings
suggest that cells are more concentrated in the core
of the human column with fewer cell bodies in the
periphery (Fig. 2D). However, this does not mean
the minicolumns are more linear in humans, in fact,
the reverse is true (see below). The CGLR is a gross
measure of cell concentration versus cell-poor area.
Because human cell columns are much wider and
contain greater neuropil space in the periphery, the
concentration of cells in the core area is relatively
higher. This means the core area has not expanded
in proportion to the increase in width. Human Tpt
“emphasizes” neuropil space by limiting the relative
expansion of neurons in the periphery which results
in the core GLR being more relatively dense than in
the other primates.
While the human minicolumn is more compact
than in other primates, it is not as linear. This
agrees with earlier studies using a different measure of linearity (Buxhoeveden et al., 1996). Nonhuman primate columns are more vertically linear and
therefore closer to the ontogenetic unit.
Chimpanzee versus rhesus monkey
Comparisons between the chimpanzee and rhesus
monkey revealed very little difference at the level we
were examining (Table 2C). Testing the basic morphology, our method revealed very similar columns
in the left hemisphere of the chimpanzee and rhesus
monkey, except for the smaller NS in the monkey.
As the only notable difference between the two, it
moves “away” from the human configuration.
In summary, the human cell columns are absolutely larger, the peripheral neuropil space is larger,
and the columns contain more non-cell space than
those of non-human primates. By contrast, in the
parameters we examined, only neuropil space distinguished the chimpanzee from the rhesus monkey
columns, and this was not statistically significant.
Overall size was the most obvious distinction between the morphology of columns in humans and
the nonhuman primates. Humans had much wider
cell columns than either the chimpanzee or rhesus
monkey, while the latter revealed essentially identical column sizes. The significance of size, whether
cortical volume or minicolumn, is very complex and
full of potential pitfalls (Deacon, 1997). Nonetheless,
the following discussion will briefly evaluate the
interaction of cortical volume and minicolumn size.
Since cortical thickness varies little between primate species, this discussion will focus upon the
width and not depth of columns.
First, column size has been generally noted as not
correlating with cortical volume when comparing
diverse species, e.g., columns in rabbits and mice are
as large or larger than humans in some areas of the
brain (Mountcastle, 1997). The size of cell columns
seemingly has less to do with brain size, and everything to do with function and organization. For ex-
ample, cell column size in the primary visual cortex
(Area 17) in monkeys and cats may be compared.
Functionally, the visual systems of cats and monkey
differ, as the monkey has better color and visual
acuity (Orban, 1984). The rhesus monkey has a
larger brain than the cat, but their minicolumns (in
this region) are much smaller, about 31 ␮m compared to 56 ␮m in the two-dimensional plane (Peters
and Sethares, 1991b; Peters and Yilmaz, 1993).
When extrapolated to three dimensions, the columns in the cat visual cortex are actually three
times larger than that of the monkey (Peters and
Yilmaz, 1993). Moreover, input from the monkey
thalamus differs from the cats’. Furthermore, its cell
columns contain relatively more cells. Overall, the
monkey’s cell columns are smaller and said to be
more ‘complex’ because they have relatively more
cells, providing more detailed specification of information than the cat; thus, the monkey’s orientation
and eye preference columns are better defined.
Nonetheless, the relationship of CW to brain size
may be significant if the same area is measured
between species that share a close evolutionary
bond. The generalized statement that cell columns
do not vary between species is based on a measure of
means. This assessment is taken from a sample of
very diverse organisms, often comparing totally different areas of the brain. We believe a stronger
correlation may exist if similar cortical areas among
closely related species are examined. Unfortunately,
we were not able to pursue this issue at the present
time since data were not available for brain size in
the chimpanzees and some of the monkeys. As our
data bank increases in regards to sample size and
availability of brain volumes, this question can be
Every minicolumn contains certain basic elements, including neurons, apical dendrites, myelinated efferents, double-bouquet bundles; thalamic,
callosal, and cortico-cortical input; dendritic arborizations and spines, synaptic contacts, as well as
glia. Brain volume may not correlate with column
size because neuronal elements such as cells and
axons require a minimum size to function. As a
result, their dimensions stay essentially the same
among species.
Although cell columns range from a minimum of
20 ␮m to a maximum of 80 ␮m (Mountcastle, 1997),
cortical volume varies far more. Unfortunately, very
little work has correlated cell column size with specific cortical areas. Ideally, it is as important to
compare column size to specific areas, as it is to
compare columns to overall brain size. Larger cell
columns potentially contain more neuronal elements at the cost of fitting fewer of them into a given
cortical space. In other words, by “downsizing” individual columns, a brain can pack in more of them.
Presumably, column size relates directly to the nature of their functions.
Column size be evaluated in the context of cortical
volume. If the volume of a particular region (like PT)
is much larger in a given brain than the relative
increase in column size, then the columns are actually smaller in relation to the space they occupy. A
50␮m column in a 20cc brain is very ‘large’ compared to the same size column in a 1400cc brain. The
human brain is about 13x larger than that of the
monkey and over 3x larger than the chimpanzee,
while its cell columns are only about a 2x larger in
human PT. From this perspective, columns in human PT are actually smaller than those of the chimpanzee and rhesus monkey, relative to brain volume. The chimpanzee brain is about 4x larger than
that of the rhesus monkey, yet their columns were
the same size. Thus, the chimpanzee columns appear relatively smaller than those in the rhesus
Increased brain size allows for more flexibility
regarding the size of particular cortical regions.
These regions may increase absolutely and relatively, or they may decrease relative to the large
brain while remaining considerably larger than
those found in a small brain.
The functional operation of the mammalian brain
is thought be distributed across many areas of the
brain rather than being strictly modular (Mountcastle, 1997). This means that individual columns
may serve multiple purposes depending on how they
are recruited. In this regard, small brains may reveal yet another limitation: delegating more columns to a particular function(s) reduces the number
available for something else. Thus because a small
brain contains fewer processing units, with fewer
interconnections, its total capabilities must be more
restricted than those of a larger brain. Thus, a
smaller brain will run out of options before a larger
one does.
It would appear that minicolumns and cortical
volume are co-dependents to some degree. The composition of an individual minicolumn depends on
many local factors. However, the functional capacity
of minicolumns in a given region is partly influenced
by the degree of potential complexity in the brain,
i.e., its interconnected pathways and regions. Cortical structures in the form of minicolumns and cortical volume exert a reciprocal effect on each other.
A bidirectional four-level hierarchy best describes
the relationships between the size of minicolumns
and brain volume: 1) The complexity of the individual column. The larger a cell column, the greater the
number and variety of its functional elements. Increases in column size may specifically accommodate a large increase in incoming myelinated pathways, or handle a generalized increase in a variety
of structures and functions. 2) The number of columns in a given region. A large brain can “afford” to
pack in not only more columns but also more complex ones if particular areas expand. 3) The number
and size of cortical regions connecting columns directly. A large brain may contain more, as well as
larger, cortical regions. As the size of cortical regions
increases, so does the number of columns they may
contain. These changes increase the amount of information regions can share. 4) Secondary connections. Areas lacking direct connections may exert
influence indirectly. For example, Tpt, a cortical region within the planum temporale, may not connect
directly to area 9. However, it does connect to area
46, which in turn connects to area 9 (Romanski et
al., 1999). Even if a column in one area lacks a direct
connection to another, it may influence it indirectly,
magnifying the effects of direct connections. Therefore, the greater number and intensity of direct connections within a brain, the greater may also be the
indirect influences.
In summary, our data suggest several conclusions.
First, because human cell columns contain considerably more peripheral neuropil space, they presumably utilize their increased size for more input/output pathways and more microcircuitry. Secondly,
relative to brain volume, human columns are actually smaller than those of the primates tested so that
more of them can be packed into an area like PT.
Third, they demonstrate that beneath the surface of
the PT, the organizational arrangement of neurons
and pathways has changed since the chimpanzee
and hominid split. This area of the brain has been
continuously implicated in language and is part of
Wernicke’s area. It is therefore a possibility that
these findings may represent an anatomical link to
this very human characteristic.
This material is based on work supported by the
Theodore and Veda Stanley Foundation and the Office of Research and Development, Department of
Veterans Affairs.
Armstrong E. 1990. Evolution of the brain. In: Paxinos G, editor.
The human nervous system. New York: Academic Press. p
Armstrong E, Buxhoeveden D. 1991. Development of morphological cortical cell columns in human cortex. Soc Neurosci Abstr
Armstrong E, Curtis M, Buxhoeveden D, Fregoe C, Zilles K,
Casanova MF, McCarthy WF. 1991. Cortical gyrification in the
rhesus monkey, a test of the mechanical folding hypothesis.
Cereb Cortex 1:426 – 432.
Brodmann K 1912. Neuere Ergebnisse über die vergleichende
histoligische Lokalisation der Grosshirnrinde mit besonderer
Berücksichtigung des Stirnhirns. Anat Anz 211 [cited in Holloway, 1968].
Brummelkamp R. 1939. Über dem Zusammenhang zwischen
Schädelkapazität und bestimmten Femurmassen, zugleich ein
Beitrag zur Cephalisationfrage von Pithecanthropus erectus.
Acta Neerl Morphol 2:379 – 400 [cited in Holloway, 1968].
Buldyrev SV, Cruz L, Gomez-Isla T, Gomez-Tortosa E, Havlin S,
Le R, Stanley HE, Urbanc B, Hyman BT. 2000. Description of
microcolumnar ensembles in association cortex and their disruption in Alzheimer and Lewy body dementias Proc Natl Acad
Sci USA 97:5039 –5043.
Buxhoeveden DP, Lefkowitz W, Loats P, Armstrong E. 1996. The
linear organization of cell columns in human and nonhuman
anthropoid Tpt cortex. Anat Embryol (Berl) 194:23–36.
Buxhoeveden DP, Switala AE, Roy E, Casanova MF. 2000. Quantitative analysis of cell columns in the cerebral cortex. J Neurosci Methods 97:7–17.
Calvin WH. 1998. April 30. Competing for consciousness: how
subconscious thoughts cook on the back burner [on-line]. Available at
Carpenter WJ, Bartka JJ, Carpenter CL, Strauss JL. 1976. Another view of schizophrenic subtypes. Arch Gen Psychiatry
33:508 –516.
Conroy GC, Weber GW, Seidler H, Tobias PV, Kane A, Brunsden
B. 1998. Endocranial capacity in an early hominid cranium
from Sterkfontein, South Africa. Science 280:1730 –1731.
Conroy GC, Falk D, Guyer J, Weber GW, Seidler H, Recheis W.
2000. Endocranial capacity in Sts 71 (Australopithecus africanus) by three-dimensional computer tomography. Anat Rec
Corballis PM, Funnell MG, Gazzaniga MS. 2000. An evolutionary
perspective on hemispheric asymmetries. Brain Cogn 43:112–
Corsellis JA, Alston RL, Miller AK. 1975. Cell counting in the
human brain: traditional and electronic methods. Postgrad
Med J 51:722–726.
Deacon T. 1997. The symbolic species. New York: W.W. Norton.
Falk D. 1978. Cerebral asymmetry in Old World monkeys. Acta
Anat 101:334 –339.
Falk D, Pyne L, Helmkamp RC, DeRousseau CJ.1988. Directional
asymmetry in the forelimb of Macaca mulatta. Am J Phys
Anthropol 77:1– 6.
Falk D, Hildebrot C, Vannier MW. 1989. Reassessment of the
Taung early hominid from a neurological perspective. J Hum
Evol 18:485– 492.
Falk D. 1993. Meningeal arterial patterns in great apes: implications for hominid vascular evolution. Am J Phys Anthropol
Falk D, Froese N, Sade DS, Dudek BC. 1999. Sex differences in
brain/body relationships of rhesus monkeys and humans. J
Hum Evol 36:233–238.
Favorov OV, Whitsel BI. 1988a. Spatial organization of the peripheral input to area 1 cell columns I: the detection of “segregates”. Brain Res Rev 13:25– 42.
Favorov OV, Whitsel BI. 1988b. Spatial organization of the peripheral input to area 1 cell columns II: the forelimb representation achieved by mosaic of segregates. Brain Res Review13:
Favorov OV, Diamond ME. 1990. Demonstration of discrete
place-defined columns-segregates-in the cat SI. J Comp Neurol
Favorov OV, Kelly G. 1994a. Minicolumnar organization within
somatosensory cortical segregates I: development of afferent
connections. Cereb Cortex 4:408 – 427.
Favorov OV, Kelly G. 1994b. Minicolumnar organization within
somatosensory cortical segregates II: emergent functional properties. Cereb Cortex 4:428 – 442.
Favorov OV, Kelly G. 1994c. Minicolumnar organization within
somatosensory cortical malformations. Prog Brain Res 73:15–
Fleischauer K, Petsche H, Wittkowski W. 1972. Vertical bundles
of dendrites in the neocortex. Z Anat Entwickl Gesch 136:213–
Galuske RA, Schlote W, Bratzke H, Singer W. 2000. Interhemispheric asymmetries of the modular structure in human temporal cortex. Science 289:1946 –1949.
Gannon P, Holloway RL, Broadfield DC, Braun AR. 1998. Asymmetry of chimpanzee temporale: humanlike pattern of Wernicke’s brain language area homolog. Science 279:220 –222.
Gazzaniga MS. 2000a. Cerebral specialization and interhemispheric communication: does the corpus callosum enable the
human condition? Brain 123:1293–1326.
Gazzaniga MS. 2000b. Neuroscience. Regional differences in cortical organization. Science 289:1887–1888.
Hackett TA, Stepniewska I, Kaas JH. 1999. Prefrontal connections of the parabelt auditory cortex in macaque monkeys.
Brain Res 817:45–58.
Haug H. 1986. History of neuromorphometry. J Neurosci Methods 18:1–17.
Heilbroner P, Holloway RL. 1989. Anatomical brain asymmetries
in monkeys: frontal, temporoparietal, and limbic cortex in Macaca. Am J Phys Anthropol 80:203–211.
Holloway RL Jr. 1968. The evolution of the primate brain: some
aspects of quantitative relations. Brain Res 7:121–172.
Holloway RL. 1973. New endocranial values for the East African
early hominids. Nature 243:97–99.
Holloway RL. 1974. The casts of fossil hominid brains. Sci Am
231:106 –115.
Holloway RL, De La Costelareymondie MC. 1982. Brain endocast
asymmetry in pongids and hominids: some preliminary findings on the paleontology of cerebral dominance. Am J Phys
Anthropol 58:101—110.
Hopkins WD, Marino L, Rilling JK, MacGregor LA. 1998. Planum
temporale asymmetries in great apes as revealed by magnetic
resonance imaging MRI. Neuroreport 9:2913–2918.
Jerison HJ. 1977. The theory of encephalization. Ann NY Acad
Sci 299:146 –160.
Jerison HJ. 1985. Animal intelligence as encephalization. Philos
Trans R Soc London (Biol) 308:21–35.
Jones EG, Burton H. 1974. Cytoarchitecture and somatic sensory
connectivity of thalamic nuclei other than the ventrobasal complex in the cat. J Comp Neurol 1544:395– 432.
Jones EG. 2000. Microcolumns in the cerebral cortex. Proc Natl
Acad Sci USA 97:5019 –5021.
Juliano S, Whitsel BL. 1987. A combined 2-deoxyglucose and
neurophysiological study of primate somatosensory cortex.
J Comp Neurol 263:514 –525.
Kaas JH, Nelson RJ, Sur M, Merzenich MM. 1981. Organization
of somatosensory cortex in primates. In: Schmitt FO, Worden
FG, Adelman G., Dennis SG, editors. The Organization of the
Cerebral Cortex. Cambridge, MA: MIT Press, p 237-261.
Kohn A, Pinheiro A, Tommerdahl MA, Whitsel BL. 1997. Optical
imaging in vitro provides evidence for the minicolumnar nature
of cortical response. Neuroreport 8:3513–3518.
Krmpotic-Nemanic J, Kostovic I, Nemanic D. 1984. Prenatal and
perinatal development of radial cell columns in the human
auditory cortex. Acta Otoloarygol. Stockh 97:489 – 495.
Krmpotic-Nemanic J, Kostovic I, Bogdanovic A, Fucic A, Judas M.
1988. Cytoarchitectonic parameters of developmental capacity
of the human associative auditory cortex during postnatal life.
Acta Otolaryngol Stockh 105:463– 466.
LeBoucq G. 1928. Le rapport poids-surface dans le cerveau des
singes. CR. Ass Anat (Prague) 2:3–56 [cited in Holloway, 1968].
Lee CJ, Whitsel BL. 1992. Mechanisms underlying somatosensory cortical dynamics, 1. In vivo studies. Cerebral Cortex 2:81–
LeMay M, Geschwind N. 1975. Hemispheric differences in the
brains of great apes. Brain Behav Evol 11:48 –52.
LeMay M. 1976. Morphological asymmetry in modern man, fossil
man, and non-human primates. Ann NY Acad Sci 280:349 –366.
LeMay M. 1985. Asymmetries of the brains and skulls of nonhuman primates. In: Glick SD, editor. Cerebral lateralization in
nonhuman species, behavioral biology series. New York: Academic Press. p 223–245.
Lorente De Nó, R. 1938. Cerebral cortex, architecture, intracortical connections, motor programs. In: Fulton JF, editor. Physiology of the nervous system. New York: Oxford University
Press. p 274 –301.
Mountcastle VB. 1957. Modality and topographic properties of
single neurons of cat’s somatic sensory cortex. J. Neurophysiol
20:408 – 434.
Mountcastle VB. 1978. An organizing principle for cerebral function, the unit module and the distributed system. In: Edelman
GM, Mountcastle VB, editors. The mindful brain: cortical organization and the group-selective theory of higher brain function. Cambridge, MA: MIT Press. p 7–51.
Mountcastle VB. 1997. The columnar organization of the neocortex. Brain 120:701–722.
Ong WY, Carey LJ. 1990. Neuronal architecture of the human
temporal cortex. Anat Embryol (Berl) 51–369.
Orban GA. 1984. Studies in brain function. VolumeII. Neuronal
operation in the visual cortex. Berlin: Springer.
Peters A, Walsh TM. 1972. A study of the organization of apical
dendrites in the somatic sensory cortex of the rat. J Comp
Neurol 1442:253–268.
Peters A, Sethares C. 1991a. Organization of pyramidal neurons
in area 17 of monkey visual cortex. J Comp Neurol 306:1–23.
Peters A, Sethares C. 1991b. Layer IVA of rhesus monkey primary visual cortex. Cereb Cortex 1:445– 462.
Peters A, Yilmaz E. 1993. Neuronal organization in area 17 of cat
visual cortex. Cereb Cortex 3:49 – 68.
Peters A, Payne BR. 1993. Numerical relationships between
geniculocortical afferents and pyramidal cell modules in cat
primary visual cortex. Cereb Cortex 3:69 –78.
Peters A, Sethares C. 1996. Myelinated axons and the pyramidal
cell modules in monkey primary visual cortex. J Comp Neurol
Powell TPS, Mountcastle VB. 1959. Some aspects of the functional organization of the cortex of the postcentral gyrus of the
monkey: a correlation of findings in a single unit analysis with
cytoarchitecture. Bull Johns Hopkins Hosp 105:133–162.
Preuss TM, Huixin Q, Kaas JH. (1999). Proc Natl Acad Sci USA
Quester R, Schroder R. 1997.The shrinkage of the human brain
stem during formalin fixation and embedding in paraffin.
J Neurosci Methods 75:81– 89.
Radinsky L. 1972. Endocasts and studies of primate brain evolution. In: Tuttle R, editor. The functional and evolutionary biology of primates. Chicago: Aldine-Atherton.
Radinsky L. 1973. Aegyptopithecus endocasts: oldest record of a
pongid grain. Am J Phys Anthropol 39:239 –247.
Radinsky L. 1975. Primate brain evolution. M Scient 63:656 – 663.
Radinsky LB. 1979. The fossil record of primate brain evolution
(49th James Arthur Lecture, New York). New York: American
Museum of Natural History.
Rakic P. 1972. Mode of Cell Migration to the superficial layers of
fetal monkey cortex. J. Comp Neurol 145:61– 84.
Rakic P. 1988a. The specification of cerebral cortical areas: the
radial unit hypothesis. Science 241:928 –931.
Rakic P. 1988b. Defects of neuronal migration and the pathogenesis of cortical malformations. Prog Brain Res 73:15–37.
Rakic P. 1990. Principles of neural cell migration. Experientia
469:882– 891.
Rilling JK, Insel TR. 1999. The primate neocortex in comparative
perspective using magnetic resonance imaging. J Hum Evol
Romanski, LM, Bates JF, Goldman-Rakic PS. 1999. Auditory belt
and parabelt projections to the prefrontal cortex in the rhesus
monkey. J Comp Neurol 403:141–157.
Schlaug G, Schleicher A, Zilles K. 1995. Quantitative analysis of
the columnar arrangement of neurons in the human cingulate
cortex. J Comp Neurol 351:441– 452.
Schleicher A, Zilles K, Wree A. 1986. A quantitative approach to
cytoarchitectonics: software and hardware aspects of a system
for evaluation and analysis of structural inhomogeneities in
nervous tissue. J Neurosci Methods 18:221–235.
Schmolke C. 1989. The ontogeny of dendritic bundles in rabbit
visual cortex. Anat Embryol (Berl) 180:371–381.
Seldon HL. 1981a. Structure of human auditory cortex I: cytoarchitectonics and dendritic distributions. Brain Res 229:277–
Seldon HL. 1981b. Structure of human auditory cortex II: axon
distributions and morphological correlates of speech perception. Brain Res 229:295–310.
Seldon HL. 1985. The anatomy of speech perception. In: Jones
EG, Peters A, editors. Cerebral cortex, volume 4. New York:
Plenum Press. p 273—327.
Semendeferi K, Damasio H, Frank R, Van Hoesen GW. 1997. The
evolution of the frontal lobes: a volumetric analysis based on
three-dimensional reconstructions of magnetic resonance scans
of human and ape brains. J Hum Evol 324:375– 88.
Semendeferi I, Armstrong E, Schleicher A, Zilles K, Van Hoesen
GW. 1998. Limbic frontal cortex in hominoids: a comparative
study of area 13. Am J Phys Anthropol 106:129 –155.
Semendeferi K, Damasio H. 2000. The brain and its main anatomical subdivisions in living hominoids using magnetic resonance imaging. J Hum Evol 38:317–332.
Swindale NV 2000. How many maps are there in visual cortex?
Cereb Cortex 10:633– 643.
Szentagothai J. 1983. The modular architectonic principle of neural centers. Rev Physiol BiochemPharmacol 98:11– 61.
Preuss TM, Huixin Q, Kaas JH. 1999. Distinctive compartmental
organization of human primary visual cortex. Proc Natl Acad
Sci USA 96:11601–11606.
Tobias PV, Falk D. 1988. Evidence for a dual pattern of cranial
venous sinuses on the endocranial cast of Taung (Australopithecus africanus). Am J Phys Anthropol 76:309 –312.
Tommerdahl M, Favorov O, Whitsel BL, Nakhle B, Gonchar YA.
1993. Minicolumnar activation patterns in cat and monkey S1
cortex. Cereb Cortex 35:399 – 411.
Yeni-Komshian GH, Benson DA. 1976. Anatomical study of cerebral asymmetry in the temporal lobe of humans, chimpanzees,
and rhesus monkeys. Science 192:387–389.
Yuste R, Peinado A, Katz LC. 1992. Neuronal domains in developing neocortex. Science 257:665– 668.
Zilles K, Schleicher A, Stephan H. 1982. Quantitative cytoarchitectonics of the cerebral cortices of several prosimian species.
In: Armstrong E, Falk D, editors. Primate brain evolution:
methods and concepts. New York: Plenum.
Zilles K, Armstrong E, Schleicher A, Kretschmann, HJ. 1988. The
human pattern of gyrification in the cerebral cortex. Anat Embryol (Berl) 179:173–179.
Zilles K, Armstrong E, Moser KH, Schleicher A, Stephan H. 1989.
Gyrification in the cerebral cortex of primates. Brain Behav
Evol 34:143–150.
Без категории
Размер файла
345 Кб
primate, differences, nonhuman, minicolumns, corte, morphological, human
Пожаловаться на содержимое документа