The distribution and density of Huntingtin inclusions across the Huntington disease neocortex: regional correlations with Huntingtin repeat expansion independent of pathologic grade

Abstract

Huntington disease is characterized by progressive neurodegeneration, especially of the striatum, and the presence of polyglutamine huntingtin (HTT) inclusions. Although HTT inclusions are most abundant in the neocortex, their neocortical distribution and density in relation to the extent of CAG repeat expansion in the HTT gene and striatal pathologic grade have yet to be formally established. We immunohistochemically studied 65 brains with a pathologic diagnosis of Huntington disease to investigate the cortical distributions and densities of HTT inclusions within the calcarine (BA17), precuneus (BA7), motor (BA4) and prefrontal (BA9) cortices; in 39 of these brains, a p62 immunostain was used for comparison. HTT inclusions predominate in the infragranular cortical layers (layers V-VI) and layer III, however, the densities of HTT inclusions across the human cerebral cortex are not uniform but are instead regionally contingent. The density of HTT and p62 inclusions (intranuclear and extranuclear) in layers V-VI increases caudally to rostrally (BA17 < BA7 < BA4 < BA9) with the median burden of HTT inclusions being 38-fold greater in the prefrontal cortex (BA9) than in the calcarine cortex (BA17). Conversely, intranuclear HTT inclusions prevail in the calcarine cortex irrespective of HTT CAG length. Neocortical HTT inclusion density correlates with CAG repeat expansion, but not with the neuropathologic grade of striatal degeneration (Vonsattel grade) or with the duration of clinical disease since motor onset. Extrapolation of these findings suggest that HTT inclusions are at a regionally-contingent, CAG-dependent, density during the advanced stages of HD. The distribution and density of HTT inclusions in HD therefore does not provide a measure of pathologic disease stage but rather infers the degree of pathogenic HTT expansion.

Keywords: Aggregation; Cortex; HD; Huntingtin; Huntington disease; Inclusion; Neuropathologic staging; Striatum; Trinucleotide repeat.

Fig. 1

Intracellular distribution of HTT and p62 inclusions within HD. a Numerous HTT or p62 cytoplasmic inclusions can be found within some neurons, such as in this Betz cell of a 48-year-old woman with 45 CAG repeats. b Intranuclear HTT inclusions are occasionally found in nuclear clusters, particularly in individuals with > 50 CAG repeats. This example was from the calcarine cortex (BA17) of a 30-year-old man with 61 CAG repeats. c Dual staining for HTT (brown)/p62 (red) highlights p62 colocalization with HTT and wrapping of p62 around HTT. Inset (c′) shows a fortuitous dystrophic neurite with a long extension of p62 that emanates from and enwraps an extranuclear HTT inclusion. d The proportion of intranuclear HTT inclusions relative to total HTT inclusions varies in the neocortex and prevails in caudal neocortices, especially the calcarine cortex (BA17). Vertical lines indicate the median value for the respective neocortical region. Note that in the instance of BA17 where the absolute number of inclusions is least of all four cortical regions, only 54 individuals could have calculated ratios because 11 brains had no inclusions in BA17. Brown arrows indicate HTT labeling; red arrows indicate p62 labeling. Brown arrows with a red edge indicate p62 colocalizing with HTT. Scale bars: a–c: 50 µm, c′: 10 µm

Fig. 2

HTT and p62 inclusion density follows an increasing caudal to rostral gradient in parasagittal neocortical regions of HD. a Macroscopic photographs of the formalin fixed right half brain of a 56-year-old man with adult-onset HD (CAG: 47/21) demonstrating the lateral and medial aspects. The brain (fresh weight: 981.6 g) appeared diffusely small. Vertical lines indicate the plane of section to obtain the coronal slices of images (b)–(d). b–d Select coronal slices of this brain from rostral to caudal with rectangles showing the Brodmann areas of interest, BA9 (superior frontal cortex, b), BA4 (motor cortex, c) and BA7/ BA17 (precuneal cortex and calcarine/visual cortex, respectively, d). e–l Immunohistochemical stains against HTT (e–h) and against p62 (i–l) demonstrating gradually increasing inclusion densities from BA17, BA7, BA4 to BA9. Brown arrows indicate a subset of the inclusions seen in these micrographs. Original magnifications: e–l: 400×, scale bars of e–l: 50 µm

Fig. 3

Violin plots of total HTT inclusion densities (a) and p62 inclusion densities (b) across neocortical regions showing successive increases from caudal to rostral cortices (BA17 → BA9). Statistical tables for these respective proteins are provided beneath each graph. a, b Dots indicate individual cortical regions; horizontal dashed lines indicate median values while horizontal dotted lines indicate the first and third quartile values

Fig. 4

Bivariate regression plots showing correlations between CAG repeat expansion and neocortical HTT/ p62 inclusion density. a HTT inclusion density correlates with CAG repeat expansion and increasing gradient of HTT burden along a caudal to rostral direction. b The x-intercepts of the extrapolated linear regression lines between HTT and CAG imply that inclusions may be formed in individuals with intermediate CAG expansion. c p62 inclusion density significantly correlates with CAG repeat expansion in BA17, BA7 and BA9

Fig. 5

a–d: Selected examples of brains that ranged from grade 1 to grade 4. a–d High power image of the dorsal head of the caudate nucleus. Arrows indicate neurons and demonstrate increasing neuronal loss with higher grades and worsening gliosis in higher grades. e–h HTT/p62 immunostains demonstrate similar inclusion densities between the prefrontal neocortices (BA9) despite increasing grade. i–l Violin plots showing that HTT inclusion density is stable across neocortical regions, except between grades 2 and 4 in BA7. Horizontal dashed lines indicate the median. m–n Bivariate regression plots demonstrate correlations between CAP score and HTT inclusion density in BA17, BA7 and BA9 (m, n = 65), however, no correlations were found between the clinical disease duration and HTT inclusion densities (n) across BA17, BA7, BA4 and BA9 (n = 22). Scale bar: 50 µm