Persistent Cortical Angiogenesis and Neuronal Migration in the Young Autism Brain

Background:   We propose a prolonged period of angiogenesis and neurodevelopment in our collection of postmortem brains from autistic donors (2.8-20.8 years of age).  In autism, the brain shows accelerated postnatal growth (Redcay and Courchesne, 2005), and neurons are smaller (Jacot-Descombes et al, 2012; van Kooten et al, 2008; Fatemi et al, 2002) and more numerous (Casanova et al, 2004). Serotonin stimulates the release of a neurite extension factor from astrocytes (Whitaker-Azmitia and Azmitia, 1994; Eriksen et al, 2002; Deng et al, 2007) and increases postnatal brain growth (Akbari et al, 1994). Serotonin fibers reach the developing cortex while neurons and glial cells migrate to form cortical layers (Foote and Morrison, 1984; Wallace and Lauder, 1983).  Serotonin promotes angiogenesis in rats (Warner-Schmidt and Duman, 2008) and humans (Boldrini et al, 2012). We reported increased serotonin staining of axons in the cortex of autism donors (Azmitia et al, 2011).

Objectives:   We aim to detect alterations in proliferation and migration of blood vessels, astrocytes and neurons using nestin and vimentin antibodies in cortices of autistic subjects. Other antibodies examined include a-SMA (mature pericytes, immature neurons), CD146 (immature pericytes), UEA1 (endothelial cells), GFAP (mature astrocytes) and iba-1 (microglia).

Methods: Brain tissue samples were obtained from the NYS Brain Bank for Developmental Disabilities and Aging (Staten Island, New York), the Autism Tissue Program (Princeton, New Jersey), and the NICHD Brain and Tissue Bank for Developmental Disorders (University of Maryland, Baltimore). The brains were hemisected and one hemisphere was fixed in 10% formalin. The tissue was embedded in polyethylene glycol and cut into 50 mm-thick serial sections. The sections were stored in 70% ethanol. Details of our immunocytochemical method are published (Azmitia et al, 2011).

Results:   In autism brains, nestin heavily labeled precapillaries and capillary vessels in all layers of superior temporal cortex at all ages analyzed (2.8-20.8 years, n=9).  In control brains, nestin labeled vessels only at the youngest ages (1.8-2.1 years, n=2), while older donors (8.6-25.6 years, n=9) showed no nestin staining. Other vascular antibodies tested showed no significant differences between autism and control individuals.  UEA1 stained endothelial cells in all cortical vessels; CD146 stained small pericytes in precapillary arterioles; and a-SMA stained immature neurons and pericytes in arterioles and capillaries. GFAP and iba-1 positive cells were frequently in close proximity to blood vessels and more reactive in autism than in control brains.

Vimentin antibodies react with filaments in immature astrocytes, neurons and pericytes.  The immunostained pericytes in arterioles and capillaries were abundant in control and autism cortices. Immature astrocytes, especially in layers I-III, were more numerous in autism compared to control.  Vimentin positive migrating neurons were seen in autism cortices at all ages examined, heaviest in layers IV and VI, but only seen in young control cortices.

Conclusions:  Our results indicate sustained angiogenesis and neuronal migration in autism that may impact cortical size and function, integrity of the blood brain barrier and microglial infiltration. The potential involvement of extended angiogenesis may open up a new avenue for understanding and treating autism.