Increased proliferation reflects glial and vascular-associated changes, but not neurogenesis in the presenile Alzheimer hippocampus

Abstract

Adult proliferation and hippocampal neurogenesis are stimulated by injury. In agreement, aberrant cell-cycle-related protein expression has been reported in senile Alzheimer's disease (AD), where the hippocampus is particularly affected. Recently, increased expression of doublecortin (DCX), a neurogenesis marker, was reported in senile AD. Here, we addressed whether proliferative and neurogenic responses also occur in younger, i.e., presenile AD cases, using immunohistochemistry for Ki-67, GFAP and DCX. Increased numbers of Ki-67+ cells with a healthy, non-mature appearance were found in CA1–3. These were mainly due to glial and vasculature-associated changes, while DCX immunostaining appeared sensitive to postmortem breakdown. We found no indications for altered dentate gyrus neurogenesis. Our data obtained using validated methodology in a well-characterized, presenile cohort thus differ from previous data obtained in senile AD. They reflect clear differences in proliferative responsivity, particularly in the glia and vascular components, and suggest different underlying mechanisms in these groups.

Introduction

Alzheimer's disease (AD) is clinically characterized by progressive dementia. Neuropathologically, the hippocampus is one of the first and most severely damaged structures in AD. Extensive pathology, with large numbers of neurofibrillary tangles (NFTs) and β amyloid (Aβ) plaques, is characteristic of the AD hippocampus (Braak and Braak, 1991). Furthermore, prominent inflammatory responses take place in AD that involve the activation of microglia and astrocytes (Nagele et al., 2004, Meda et al., 2001, Akiyama et al., 2000). In addition, almost all AD patients (around 90%) show cerebral amyloid angiopathy (CAA) that correlates with cognitive impairment (Jellinger, 2002, Mandybur, 1975) and may play an important role in AD pathogenesis (Zlokovic, 2005, Castellani et al., 2004, Nicoll et al., 2004).

In addition to these well-known processes, various cell-cycle-related events have been identified in AD. For instance, an increased expression of various cyclins and cyclin-dependent kinases has been reported in the hippocampus in AD (Busser et al., 1998, McShea et al., 1997, Vincent et al., 1997, Arendt et al., 1996, Kondratick and Vandre, 1996, Smith and Lippa, 1995). Most authors suggest that this ectopic expression occurs in AD-affected degenerating cells. However, one recent publication shows increased expression of doublecortin (DCX), a marker for new neurons, in a senile cohort of AD patients (Jin et al., 2004a). These studies are especially interesting since the hippocampal dentate gyrus (DG) is one of the few brain areas where adult neurogenesis occurs, albeit at very low rates in aged subjects (Heine et al., 2004). In rodents, proliferation and neurogenesis are stimulated by hippocampal damage (Radley and Jacobs, 2003, Kuhn et al., 2001, Covolan et al., 2000, Parent et al., 1997). Interestingly, Aβ can also exert neurogenic effects on stem cells in vitro (Lopez-Toledano and Shelanski, 2004, Ohsawa et al., 1999). Changes in neurogenesis have furthermore been reported in some mouse models for AD (Chevallier et al., 2005, Jin et al., 2004b), although these results are still far from conclusive (Boekhoorn et al., 2006, Dong et al., 2004, Wang et al., 2004, Wen et al., 2004, Haughey et al., 2002).

Possible proliferative and compensatory responses are considered more likely to occur in younger presenile patients than in older AD patients, while the former condition generally also has a more aggressive nature and shorter disease duration. To address whether proliferative changes are indeed different in a younger versus older AD group, we studied a well defined cohort of presenile AD cases and performed a detailed quantitative immunohistochemical inventory on the hippocampus, using the proliferation marker Ki-67 (Gerdes et al., 1991). Subsequently, we investigated associations with gliogenesis and vasculature using GFAP and Von Willebrand Factor (VWF) immunohistochemistry, respectively.

Most reports have examined cell cycle changes predominantly in close association with AD-affected neurons (Busser et al., 1998, Arendt et al., 1996). This ectopic expression is presumed to be associated with cell death rather than cell birth. Since this was not the topic of our investigation, we here specifically focused on Ki-67 expression in apparently healthy, small, presumably precursor cells and excluded Ki-67 expression – if any – in mature profiles. DCX immunohistochemistry was further applied as a new marker for adult generated, migrating neurons in rodents. Unlike Bromodeoxyuridine (BrdU) immunocytochemistry, DCX does not require prior injections in live subjects. As such, DCX has an interesting potential to detect neurogenesis in human brain as well (Couillard-Despres et al., 2005, Jin et al., 2004a, Rao and Shetty, 2004, Brown et al., 2003, Francis et al., 1999).