Dodart JC, Marr RA, Koistinaho M, Gregersen BM, Malkani S, Verma IM, Paul SM.
Gene delivery of human apolipoprotein E alters brain Abeta burden in a mouse model of Alzheimer's disease.
Proc Natl Acad Sci U S A. 2005 Jan 25;102(4):1211-6.
Please login to recommend the paper.
To make a comment you must login or register.
In their recent PNAS paper, Dodart and colleagues have shown that modulation of brain ApoE through a lentiviral gene delivery approach alters the level of hippocampal Aβ and its deposition in the PDAPP mouse model of AD. Their results confirm and extend the now extensive literature demonstrating effects of ApoE level and isoform on Aβ metabolism in vivo, with expression of ApoE4 leading to increases in Aβ and plaque burden, and the less well-characterized effect of ApoE2 leading to reductions in these pathologies.
A few findings in the present study are particularly interesting. First, the effects are acute and relatively quick-acting, i.e., effective within a 5-12 week time period, or perhaps even quicker—the investigators didn’t look earlier. Such a rapid effect makes acute modulation of ApoE in the adult brain a potentially viable therapeutic strategy, at least in principle. Second, the dominant-negative effect of human ApoE2 over mouse ApoE was striking, somewhat surprising, and therefore interesting. Barring alternative interpretations based on technical issues (i.e., the inability to quantify expression of the human ApoE isoforms in PDAPP mice expressing mouse ApoE), this result begs the question as to whether acute increases in ApoE2 can also overcome effects of ApoE3 or ApoE4, the situation that would be encountered in humans. Finally, it’s also interesting that the effects were observed with neuronal expression of ApoE. One wonders whether modulating expression by the normal, glial source of ApoE would have the same effects on Aβ metabolism. Since neurons are the primary source of APP (and hence Aβ) in PDAPP mice, it is also conceivable that neuronal expression of ApoE could be actually modulating the production of APP and/or Aβ in these experiments, rather than influencing Aβ clearance mechanisms. This study did not investigate this possibility. Regardless of the mechanism(s) of the effect, the present findings are promising in that they put ApoE gene delivery on the map as a possible therapeutic strategy for AD amyloidosis.
Dodart and coworkers use hippocampal injection of lentiviral vectors expressing the human isoforms of ApoE to demonstrate that ApoE2 reduces amyloid burden in PDAPP mice. These results suggest that ApoE2 may be a viable candidate for gene therapy in the treatment of AD. While increased amyloid burden occurs with the expression of ApoE4 both in the presence and absence of mouse ApoE, the reduced amyloid burden elicited with ApoE2 is seen only in the presence of mouse ApoE. Indeed, on an ApoE knockout background, amyloid burden under expression of ApoE2 is not significantly different from that when ApoE4 is expressed, while with expression of ApoE3 the burden is significantly reduced. While intriguing, the requirement that mouse ApoE must be present for the ApoE2 effect is puzzling.
This paper raises an additional issue that is the subject of continued debate: whether ApoE is expressed by neurons, glia, or both, depending on the conditions (i.e., injury may induce neuronal ApoE expression). More than likely, lentiviral expression occurred in neurons as previously demonstrated by a number of in vivo, ex vivo, and in vitro studies. And indeed, this paper suggests that lentiviral expression of ApoE is occurring in neurons because ApoE colocalizes by immunohistochemical analysis with a neuronal and not a glial marker. However, ApoE is a secretory protein and a number of careful studies in human tissue have demonstrated that while neurons in normal brain are almost always ApoE immunopositive, in situ hybridization shows expression only in glia and primarily astrocytes. The function of this apparent uptake by ApoE receptors and sequestration of ApoE by neurons is unclear. Thus, the neuronal expression of ApoE in this model system provides more fodder for the neuron vs. glia debate.