Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Nov 3;30(44):14595-609.
doi: 10.1523/JNEUROSCI.2257-10.2010.

Changes in Prefrontal Axons May Disrupt the Network in Autism

Affiliations
Free PMC article

Changes in Prefrontal Axons May Disrupt the Network in Autism

Basilis Zikopoulos et al. J Neurosci. .
Free PMC article

Abstract

Neural communication is disrupted in autism by unknown mechanisms. Here, we examined whether in autism there are changes in axons, which are the conduit for neural communication. We investigated single axons and their ultrastructure in the white matter of postmortem human brain tissue below the anterior cingulate cortex (ACC), orbitofrontal cortex (OFC), and lateral prefrontal cortex (LPFC), which are associated with attention, social interactions, and emotions, and have been consistently implicated in the pathology of autism. Area-specific changes below ACC (area 32) included a decrease in the largest axons that communicate over long distances. In addition, below ACC there was overexpression of the growth-associated protein 43 kDa accompanied by excessive number of thin axons that link neighboring areas. In OFC (area 11), axons had decreased myelin thickness. Axon features below LPFC (area 46) appeared to be unaffected, but the altered white matter composition below ACC and OFC changed the relationships among all prefrontal areas examined, and could indirectly affect LPFC function. These findings provide a mechanism for disconnection of long-distance pathways, excessive connections between neighboring areas, and inefficiency in pathways for emotions, and may help explain why individuals with autism do not adequately shift attention, engage in repetitive behavior, and avoid social interactions. These changes below specific prefrontal areas appear to be linked through a cascade of developmental events affecting axon growth and guidance, and suggest targeting the associated signaling pathways for therapeutic interventions in autism.

Figures

Figure 1.
Figure 1.
Map of prefrontal areas studied and segmentation of the white matter. A, Medial (top) and lateral (bottom) views of the human brain show the three prefrontal areas studied; ACC (A32, red; anterior A24, yellow); OFC (A11, green); LPFC (A46, blue). Dotted lines indicate the levels (L1, L2) used for analysis. B, 1-cm-thick slabs of frontal cortex show the areas sampled (color-coded dotted-line squares: A32, red; A24, yellow; A11, green; A46, blue). C, Matched levels from the brain atlas from the Autism Tissue Portal. D, Coronal view of a representative ACC (A32) tissue slab. Dotted lines indicate gross (macroscopic) distinction of superficial (SWM) and deep (DWM) white matter, based on subsequent microscopic analysis. E, F, Fluorescent photomicrographs of coronal sections from A32 white matter after labeling of axons with NFP-200 (green). Light microscopic segmentation of superficial (E) and deep (F) white matter is based on the distinct orientation of axons at different depths from the gray matter. Axons in the superficial white matter travel mainly perpendicular to the surface of the cortex (E, axons appear mainly as thin lines), whereas in the deep white matter most axons travel parallel to the cortical surface (F, axons appear mainly as green dots). G, H, EM photomicrographs show the prevalence of elongated axon profiles in the superficial white matter (G) and the prevalence of circular axon profiles in the deep white matter (H).
Figure 2.
Figure 2.
Altered axons below ACC in autism. A–H, Deep white matter: EM photomicrographs and respective plots (color coded) deeply below ACC (A32) in control and autistic cases show the distribution of small, medium, large and extra-large axons. I, The relative density of extra-large axons (±SEM) is significantly lower (*p = 0.03) in autistic individuals. J, Superficial white matter: the relative density of small (thin) axons (±SEM) just below ACC is significantly higher (*p = 0.01) in the autistic cases. K, Same information as in I plotted as a fingerprint of axons. L–S, Superficial white matter: EM photomicrographs and respective plots (color coded) in control and autistic cases show distribution of the four size groups of axons.
Figure 3.
Figure 3.
Neuronal density and cortical thickness in ACC (A32) were not affected in the autistic cases. A, B, Photomontages of adjoining high-magnification images of Nissl-stained coronal sections, from the pial surface to the white matter of ACC, in a control (A) and an autistic (B) case. Dotted lines indicate borders (from top to bottom) between layers I, II/III, V/VI, and the white matter. A32 does not have a well delineated layer IV, comparable to the same area in the rhesus monkey (Barbas and Pandya, 1989). C, Estimated overall neuronal density ±SEM in ACC (A32) based on stereologic analysis. D, Plot of the laminar neuronal density ± SEM in A32. E, Mean cortical gray matter thickness ±SEM in sulcal, straight, and gyral parts of A32 and overall average thickness.
Figure 4.
Figure 4.
Increased branching of axons in the superficial white matter below ACC in autism. A, Average (±SEM) of all axons with branches is significantly higher (*p = 0.03) in the autistic cases. B, Average relative number of axons with branches (±SEM) grouped by size. Medium-sized axons have significantly more branches (*p = 0.03) in the autistic group. C, Collapsed confocal image of myelinated axons (green) in an autistic case labeled with NFP-200. Yellow arrows show some branching points. D, Images from a three-dimensional confocal stack that was used for the branching analysis. The left column shows a 3D projection of the confocal stack and rotation in the y-axis. A branching axon is pseudo-colored with orange/yellow hue for visualization. The right column (images z1–z7) shows the same axon (red arrow) in serial images (0.4 μm apart) from the z-axis confocal stack that was used to create the three-dimensional projections. E, Three-dimensional reconstruction of axons (gray) and their branches (color coded by size of the parent axon) from uninterrupted series of ultrathin sections (150 nm) at the EM in a control (left) and an autistic case (right), which has more axons with branches.
Figure 5.
Figure 5.
GAP-43 is elevated in the superficial white matter below ACC in autism. A, Myelinated axons (percentage ±SEM) labeled with the axon marker NFP-200 that also express GAP-43 is over twofold higher (*p = 0.02) in autistic than in control cases. B, C, Control case: low (B) and high (C) magnification of confocal images from double immunofluorescence show GAP-43 (red) in axons labeled with NFP-200 (green). Some myelinated axons contain GAP-43 in their axolemma, which is transported to axon terminals and branching points. D, E, Autistic case: low (D) and high (E) magnification of confocal images show increased number of axons with GAP-43.
Figure 6.
Figure 6.
Decreased thickness of myelin in the superficial white matter below OFC (A11) in autism. A, G-ratio plot (inner/outer axon diameter ±SEM) in all areas and cases examined. The decreased myelin thickness, found only in the superficial white matter below OFC, increased the g-ratio (red squares and line) above normal levels (blue diamonds and line). B, Fingerprint plot shows that in OFC there was significant (*p = 0.01) overall decrease in the thickness of myelin in axons of all sizes. C, D, EM photomicrographs show differences in myelin thickness between control (C) and autistic (D) cases, apparent in all axon size groups.
Figure 7.
Figure 7.
The structural features of axons and their density identify distinct prefrontal areas in control and autistic cases. A–F, Fingerprint plots of the superficial (s) and deep (d) white matter below areas 32, 11, and 46. *Significant differences between control and autistic cases (p < 0.05).
Figure 8.
Figure 8.
Profile of prefrontal areas based on their axon features. NMDS based on all measures of axon features shows a clear separation of three prefrontal areas in controls (blue), and an altered relationship in the autistic cases (red). Alienation coefficient = 0.039.
Figure 9.
Figure 9.
Changes in structural features of axons and density alter the relationship of ACC, OFC, and LPFC in autism. Interareal differences were assessed by subtraction of corresponding normalized values for each pair of areas. A, Differences between A32 (ACC) and A11 (OFC); positive numbers indicate higher values for A32, and negative numbers indicate higher values for A11. B, Differences between A46 (LPFC) and A11; positive numbers indicate higher values for A46, and negative numbers indicate higher values for A11. C, Differences between A32 and A46; positive numbers indicate higher values for A32, and negative numbers indicate higher values for A46. Blue dotted line and magenta-shaded area indicate mean and range for the control cases. Orange dotted line and yellow-shaded area show mean and range for autistic cases. Asterisks indicate significant differences between control and autistic cases (p < 0.05).
Figure 10.
Figure 10.
Relationship of axonal features to developmental events. Model relates three developmental events: neurogenesis/migration (gray), expression of GAP-43 (red), and myelination (green), based on data from nonhuman and human primates (Flechsig, 1901; Von Bonin, 1950; Yakovlev and Lecours, 1967; Milosevic et al., 1995; Kanazir et al., 1996; Oishi et al., 1998; Rakic, 2002). Truncated rows show the proposed vulnerability interval in autism. Top rows: ACC (A32) develops first (Rakic, 2002) but myelinates late (Flechsig, 1901; Yakovlev and Lecours, 1967). High levels of GAP-43 for a prolonged period in autism help explain the increased branching, but myelination is unaffected because it starts much later (Flechsig, 1901), when GAP-43 levels drop. Middle rows: OFC completes neurogenesis/migration after ACC but myelinates before ACC, shortening the interval between the two developmental events. A small increase in GAP-43 during development may be sufficient to inhibit myelin growth but not affect branching, as seen in OFC in the autistic cases. Bottom rows: LPFC neurogenesis/migration are completed later, when levels of GAP-43 are comparatively low, so neither axon branching nor myelination is affected in autism.

Similar articles

See all similar articles

Cited by 117 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback