Alzheimer's disease (AD) is a largely sporadic neurodegenerative disorder that rarely strikes before the 7^th decade with primary neuronal losses in hippocampus, frontal cortex and certain subcortical nuclei. Ataxia telangiectasia (A-T), by contrast, is a multisystemic disease caused by mutations in the ATM (A-T mutated) gene. It strikes before age 5 and is characterized by dysfunctions in many tissues including the CNS where it leads to neurodegeneration, primarily in cerebellum. Despite these differences, AD and A-T share several characteristics including neurodegeneration associated with ectopic neuronal cell cycle event (CCE). This led me to explore the hypothesis that ATM reduction plays a role in AD pathogenesis.

Partial ATM deficiency is able to drive epigenetic phenotypes and...[ Read more ]

Alzheimer's disease (AD) is a largely sporadic neurodegenerative disorder that rarely strikes before the 7^th decade with primary neuronal losses in hippocampus, frontal cortex and certain subcortical nuclei. Ataxia telangiectasia (A-T), by contrast, is a multisystemic disease caused by mutations in the ATM (A-T mutated) gene. It strikes before age 5 and is characterized by dysfunctions in many tissues including the CNS where it leads to neurodegeneration, primarily in cerebellum. Despite these differences, AD and A-T share several characteristics including neurodegeneration associated with ectopic neuronal cell cycle event (CCE). This led me to explore the hypothesis that ATM reduction plays a role in AD pathogenesis.

Partial ATM deficiency is able to drive epigenetic phenotypes and neuronal CCE, which can serve as markers for ATM reduction. I found that in AD mouse models, neurons under stress show evidence for a loss of ATM. In human AD, reduced ATM immunostaining is found in the same groups of neurons that are positive for the ATM loss-of-function markers in multiple brain regions where degeneration is prevalent. Though these ATM-deficient neurons represent only a fraction of the total cells in each affected region, their numbers significantly correlate with disease stage. This suggests that failure of ATM function is involved in AD pathology and may be an important contributor to the death of neurons.

To understand its mechanism, we generated mice that were double heterozygotes for ATM mutations and AD-causing human transgenes. I found that the addition of the AD transgene to the heterozygous, Atm^+/-, mice led to an exaggerated reduction in ATM protein level of frontal cortex. Compared to AD mice, the double heterozygotes have a shorter lifespan, reduced motor ability but no obvious aggravation of cognitive impairment.

At the molecular level, ATM reduction suppresses the PI3/Akt pathway, leading to GSK3β activation and tau hyperphosphorylation. Another classic AD pathology, amyloid deposition, is increased even more dramatically by adding ATM deficiency to the AD mouse model. At a cellular level, double heterozygosity exacerbates cell death related processes including CCE, DNA damage response, and changes in the epigenetic landscape. By contrast, double heterozygotes have no significant changes in their inflammatory response. Significantly, only by combining aging with APP (PS/APP) genes does ATM reduction reveal its ability to accelerate AD progression. This is consistent with the case in human AD, which requires multiple factors rather than a single one to begin. I found that while many markers of degeneration are enhanced, the cells also try to protect themselves. Thus I document an increase of the NFkB subunit p50, which is neuroprotective. Hence neuronal fates are determined by the balance between protection and exacerbation. Taken together, ATM reduction is involved in AD pathogenesis via multiple pathologies including tau, amyloid deposits and neuronal death.

Given my clinical background and knowing the complexity of AD – including the absence of any disease-modifying drug treatment – I extended my studies to living human subjects with advanced AD and explored a non-pharmacological approach. The intervention involved environment enhancement, specifically, the viewing of a Japanese garden. My hypothesis was that this simple procedure would improve the behavioral symptoms and life quality of individuals with advanced dementia. We validated previous work and showed that the design elements of the garden relieve stress and improve mood. Anecdotal evidence for tapping into old memories was common. To our surprise, we discovered that most of these beneficial effects were lost when the glass door to the garden area was closed. This was important as the view of the garden was largely unimpeded. This provided us a new insight into dementia therapy and suggests a highly cost-effect environmental approach to non-pharmacological therapy for individuals who are severely compromised by dementia.

In conclusion, my studies show that ATM reduction contributes to AD pathogenesis via multiple cellular processes including amyloid, tau pathology, neuronal loss, neuroinflammation etc. This suggests that partial ATM reduction may be a risk factor and potential therapeutic target of AD. However, the most impressive lesson I learned throughout my AD study is that it is a prevalent disease full of complexity and a devastating result of multiple risk factors. Turning to practical therapy for AD, adding one piece to the puzzle does not provide enough to solve the rapidly increasing prevalence of dementia and AD in the population, especially when medical resources are limited. Improvement in their life quality is possible due to relieved stress, improved behavior, and enhanced memory through observation of a natural Japanese garden. Environment enrichment hence is potentially efficient in benefiting millions of AD patients.