Elsevier

Neurobiology of Aging

Volume 21, Issue 6, November–December 2000, Pages 797-806
Neurobiology of Aging

Beta-amyloid activated microglia induce cell cycling and cell death in cultured cortical neurons

https://doi.org/10.1016/S0197-4580(00)00219-0Get rights and content

Abstract

Immunocytochemical studies of postmortem human tissue have shown that the neurons at risk for degeneration in Alzheimer’s are marked by the ectopic expression of several cell cycle components. The current work investigates the roles that β-amyloid activated microglia might play in leading neurons to re-express cell cycle components. Stable cultures of E16.5 mouse cortical neurons were exposed to β-amyloid alone, microglial cells alone, or microglial cells activated by β-amyloid. Increased cell death was found in response to each of these treatments, however, only the amyloid activated microglial treatment increased the number of neurons that were positive for cell cycle markers such as PCNA or cyclin D and incorporation of BrdU. Double labeling with BrdU and TUNEL techniques verified that the ‘dividing’ neurons were dying, most likely through an apoptotic mechanism. The identity of the soluble factor(s) elaborated by the microglia remains unknown, but FGF2, a suspected neuronal mitogen, was ruled out. These results further support a model in which microglial activation by β-amyloid is a key event in the progression in Alzheimer’s disease.

Introduction

Alzheimer’s disease is a late-onset dementia that affects over 40% of all individuals over the age of 85 [12]. Viewed at autopsy, the brain presents characteristic pathological features including amyloid plaques and neurofibrillary tangles. Consistent neuronal cell loss is also reported in cortex (particularly entorhinal, hippocampal and frontal areas) as well as in several subcortical nuclei, specifically, the basal nucleus of Meynart, the dorsal raphe and the locus coeruleus [49], [54]. It is this neuronal loss and its antecedent reduction in synapses that are believed to be the most direct cause of the clinical expression of Alzheimer’s disease.

The cell biology and biochemistry of the cell death process in general and neuronal death in particular is increasingly well understood [7], [11], [16], [31], [32], [50], [53]. The involvement of cell cycle proteins in the cell death process, though seemingly paradoxical, is increasingly recognized as an important mechanism by which programmed cell death is regulated. In vitro, the death of postmitotic neuronal cells caused by either trophic factor withdrawal [36], [40], [43] DNA damage [37], [38], [39], [42], or KCl withdrawal from cerebellar granule cell cultures [35] is accompanied by activation of several cell cycle regulators.

In addition to these in vitro studies, a variety of animal and human studies have leant strong support to the role of ectopic cell cycle activation in the death of nerve cells. In the early embryo, young postmitotic neurons appear to require a member of the retinoblastoma family to avoid an apoptotic death [6], [19], [21], [22], [23]. This apoptosis is accompanied by evidence of abnormal BrdU incorporation in regions normally occupied exclusively by post-mitotic neurons [21], [23]. Later in development, the cell death during numerical matching also appears to involve cell cycling. Dying neuronal cells, in both normal and pathological situations, can be labeled with BrdU as well as with a variety of cell cycle protein antibodies [17]. Further, if post-mitotic neurons are experimentally forced to re-enter the cell cycle by expression of an oncogene, the result is neuronal cell death accompanied by an incorporation of BrdU [1], [13].

Experiments such as these have led several laboratories to investigate the involvement of cell cycle proteins in the neurodegeneration that occurs in Alzheimer’s disease [2], [4], [9], [28], [29], [30], [33], [34], [41], [46], [47]. These investigations have all shown a strong positive correlation between ectopic activation of cell cycle proteins and the death of susceptible neurons in autopsy material. Experiments in our laboratory have shown that the cell cycle markers, cyclin B, cyclin D, as well as PCNA mark neurons at risk not only in hippocampus but also in the locus coeruleus and the dorsal raphe [4]. Indeed, the combined evidence suggests that cell cycle-related proteins are the most reliable markers yet identified for the neurons at risk for cell death in Alzheimer’s disease.

These findings raise many important questions, but clearly one central question is the nature of the signal that triggers an abortive mitosis in nerve cells that have been mitotically inactive for decades. An important clue to the answer comes both from the pathology and epidemiology of the disease. If the regions of the Alzheimer brain that are most invested by plaques are examined for non-neuronal cells, the presence of reactive microglial cells is a prominent feature [27]. Activated microglia secrete a diverse array of bioactive molecules including cytokines and mitogens. This observation is particularly significant since epidemiologic studies show that individuals on large doses of non-steroidal anti-inflammatory drugs have a nearly 40% lower incidence of Alzheimer’s disease [26], [44]. The signal transduction pathways underlying the microglial response to fibrillar forms of β-amyloid have now been identified [8], [24], [25] and they appear to act through a unique cell surface receptor complex [3]. We wished to explore the hypothesis that the reactive microglia represent a potential source of the mitotic pressure that drives the AD neurons to their demise. We report here a tissue culture model of this interaction and present evidence that β-amyloid activated microglial cells secrete a substance that triggers ectopic cell cycling and cell death in cultured mouse cortical neurons.

Section snippets

Animals

Female and male C57BL/6J mice were obtained from Harlan (Indianapolis, IN). Mating pairs were established and checked daily for breeding by inspection for a vaginal plug. The morning that the plug was found was assumed to be embryonic day 0.5 (E0.5).

Embryonic neocortical cultures

After 16.5 days of gestation, pregnant C57BL/6J dams were sacrificed and their embryos delivered by C-section. Cerebral cortices were dissected into chilled buffer, then treated with 0.25% trypsin–EDTA (Sigma) for 15 min at 37°C. Washing the tissue

Results

Cells from E16.5 embryonic cortex were dissociated and plated in N2 medium with B27 supplement. After 5 days the cultures were stable in terms of both cell number and cell type. Of the original 40,000 cells, 80% were typically found after 5-days in untreated cultures. Of these 32,000, most were neurons. Astrocytes and NG2-positive oligodendrocyte progenitors each made up less than 1% of the total. The state of the culture and the morphology of the constituent cells are illustrated in Fig. 1.

Discussion

The results presented here illustrate four main points. First, the activation of either THP-1 monocytes or mouse brain microglia by 1 μM concentrations of fibrillar forms of β-amyloid leads to an enhanced neuronal cell loss in E16.5 mouse cortical neuronal cultures. Second, this effect is mediated through soluble factors secreted by the microglia/monocytes since the same result is observed when conditioned medium is used instead of direct cell addition. Third, the increased loss of nerve cells

Acknowledgements

The authors wish to express their gratitude to the Blanchette H. Rockefeller foundation for their generous support of this work. Thanks also to the NIH for funding received through grants NS20591 to KH and the Alzheimer’s Disease Research Center of University Hospitals and Case Western Reserve University (AG08012). Special thanks to Dr. Akiko Nishiyama for providing the NG2 antibody.

References (54)

  • T Arendt et al.

    Expression of the cyclin-dependent kinase inhibitor p16 in Alzheimer’s disease

    Neuroreport

    (1996)
  • M Bamberger et al.

    The B-class scavenger receptor CD36 and the integrin associated protein CD47 mediate interactions with beta-amyloid fibrils in THP-1 monocytes

    Neurosci Abst

    (1999)
  • Busser J, Geldmacher D, Herrup K. Ectopic cell cycle proteins predict the sites of neuronal cell death in Alzheimer’s...
  • C.C Chao et al.

    Tumor necrosis factor (TNF)-alpha potentiates glutamate neurotoxicity in human fetal brain cell cultures

    Dev Neurosci

    (1994)
  • A Clarke et al.

    Requirement for a functional Rb-1 gene in murine development

    Nature

    (1992)
  • G Cohen

    Caspasesthe executioners of apoptosis

    Biochem J

    (1997)
  • C.K Combs et al.

    Identification of microglial signal transduction pathways mediating a neurotoxic response to amyloidogenic fragments of beta-amyloid and prion proteins

    J Neurosci

    (1999)
  • A Copani et al.

    Mitotic signaling by beta-amyloid causes neuronal death

    FASEB J

    (1999)
  • F de Bock et al.

    Release of TNF alpha in the rat hippocampus following epileptic seizures and excitotoxic neuronal damage

    Neuroreport

    (1996)
  • M Deshmukh et al.

    Programmed cell death in neuronsfocus on the pathway of nerve growth factor deprivation-induced death of sympathetic neurons

    Mol Pharmacol

    (1997)
  • D Evans et al.

    Prevalence of Alzheimer’s disease in a community population of older personshigher than previously reported

    JAMA

    (1989)
  • F Gonzalez-Scarano et al.

    Microglia as mediators of inflammatory and degenerative diseases

    Ann Rev Neurosci

    (1999)
  • A Gross et al.

    BCL-2 family members and the mitochondria in apoptosis

    Genes Devel

    (1999)
  • K Herrup et al.

    The induction of multiple cell cycle events precedes target-related neuronal death

    Development

    (1995)
  • L Iversen et al.

    The toxicity in vitro of β-amyloid protein

    Biochem J

    (1995)
  • T Jacks et al.

    Effects of an Rb mutation in the mouse

    Nature

    (1992)
  • C Lagenaur et al.

    An L1-like molecule, the 8D9 antigen, is a potent substrate for neurite extension

    Proc Natl Acad Sci USA

    (1987)
  • Cited by (0)

    View full text