Clinical and Research News
Cholesterol Metabolism May Provide Alzheimer’s Clue
Psychiatric News
Volume 38 Number 6 page 30-52

A new research report links an alteration in the gene that codes for an important brain enzyme—the enzyme controlling the rate-limiting step in the elimination of cholesterol from the brain—to a significantly increased risk for late-onset Alzheimer’s disease.

The report, along with recently published reports involving earlier-onset forms of Alzheimer’s disease, gives added support to the growing belief among researchers that the cellular mechanisms underlying the illness are centered around the brain’s metabolism of not only amyloid but also cholesterol.

If the new report can be substantiated, the specific change in the coding of the gene studied—known as a polymorphism—would be the second involved in cholesterol metabolism implicated as a major risk factor for late-onset Alzheimer’s disease (LOAD).

Alzheimer’s disease is now generally divided into three categories, defined by age of onset: "familial," or early-onset Alzheimer’s (EOAD), with onset occurring before age 65; LOAD, onset after age 65; and very late onset (VLOAD), after age 85.

While the resulting pathology—neuritic plaques and neurofibrillary tangles—are common to all forms of Alzheimer’s, researchers are learning that the mechanisms underlying the development of the disease at different ages are indeed unique.


A large amount of research in molecular genetics, neuropathology, and cell biology has led to the amyloid cascade hypothesis, which remains the basis of researchers’ current view of the pathology underlying all forms of Alzheimer’s disease.

Three genetic polymorphisms have been linked to an increased risk of the least-prevalent of the three forms of Alzheimer’s, EOAD: one involving the amyloid precursor protein (APP) and two recently discovered polymorphisms in genes coding for presenilin, which is one of several enzymes involved in the processing of APP.

Normally, most APP is broken down into two fragments, one that remains in the neuron and may play a part in gene transcription (although its function is not yet well understood) and a second, soluble form of amyloid that is secreted.

Each of the three polymorphisms linked to EAOD is thought to result in altered processing of APP, causing an increase in production of a much less soluble form of amyloid protein, known as beta amyloid, or Aβ. Aβ quickly accumulates and forms the plaques seen universally in people with Alzheimer’s, regardless of age at onset.

Amyloid plaques are directly toxic to nerve cells, causing cellular damage and eventually cell death. As part of that process, researchers believe, dying nerve cells release free radicals that activate enzymes in surrounding tissues. These enzymes lead to a chemical change in the tau protein, a critical intracellular protein that supports microtubules within the neurons. The altered tau dissociates from the microtubules—leading to neuronal collapse—and aggregates into the neurofibrillary tangles characteristic of Alzheimer’s. It is these pathologic cellular changes that are thought to be directly responsible for the cognitive decline and dementia that are clinically termed Alzheimer’s disease.

A new report from British researchers in the February issue of Neurology confirmed earlier reports of the link between EOAD and polymorphisms in the APP and presenilin genes. The team, led by J.C. Janssen, M.R.C.P., director of the Dementia Research Group at St. Mary’s Hospital in London, completed genetic analyses of 31 individuals with probable or definite Alzheimer’s whose age of onset was prior to 61. Of those patients with confirmed Alzheimer’s, 82 percent carried a suspect polymorphism in either the APP or presenilin gene. For those with probable Alzheimer’s, 77 percent had the polymorphism.

Jansen and his colleagues concluded that "because a molecular genetic diagnosis of an inherited disorder affects not only the patient, but also the entire family, genetic counseling must be an essential component of the diagnosis."

They wrote that counseling should be "confined to those with a clearly positive family history. . .and an early [age at onset] in all family members."


While early-onset Alzheimer’s disease is thought to be primarily an autosomal dominant disease directly involving amyloid processing—a patient needs only one copy of the polymorphism from either parent to potentially develop the disease—only recently have researchers been able to shed light on the genetic mechanisms potentially underlying LOAD.

Until now, the only gene clearly linked to LOAD was the gene coding for apolipoprotein E (APOE), which is the main cholesterol transport protein responsible for shuttling cholesterol back and forth as needed across the neuronal membrane. One particular APOE polymorphism (APOE-E4) has been shown to increase serum cholesterol levels and is a major risk factor for LOAD, including a reduction in the average age of onset by about 10 years.

Now, according to a new report in the January issue of Archives of Neurology, a common, single base substitution within the gene coding for CYP46 significantly increases the risk of developing late-onset Alzheimer’s. CYP46, found only in the brain, is a member of the cytochrome P450 family of proteins. It is responsible for the addition of a hydroxyl group to cholesterol, which results in cholesterol becoming much more soluble and allowing it to cross the blood-brain barrier to exit the brain.

The CYP46 enzyme regulates levels of brain cholesterol, researchers believe, through a feedback mechanism that allows the amount of soluble cholesterol leaving the brain to match closely the amount of cholesterol that is normally synthesized or routinely recycled (through APOE) within the brain. The end result, under normal conditions, is a fairly constant level of cholesterol in brain tissues.

A team led by Andreas Papassotiropoulos, M.D., director of the division of psychiatry research at the University of Zurich, believes the polymorphism he and his team studied—a change of just one base in the CYP46 gene from cytosine (C) to thymine (T)—leads to a decrease in functioning of CYP46, causing cholesterol levels in the brain and cerebrospinal fluid to reach higher-than-normal levels.

Because other research has shown that depletion of brain cholesterol leads to a reduction in Aβ, and some cholesterol-lowering medications have been linked to lower prevalence of LOAD (see box below), the team hypothesized that the increase in cholesterol levels due to the single base polymorphism would translate into an increase in risk for LOAD.

Papassotiropoulos and colleagues report that the CYP-TT polymorphism (substitution of T for C in both copies of the subject’s CYP46 gene) is fairly common, occurring in 44 percent of the overall population they studied. They first studied postmortem brain tissue samples from 55 nondemented elderly patients, measuring the amount of Aβ deposition (termed Aβ load.)

Aβ load in the brain tissue samples was significantly linked to the CYP46-TT polymorphism. Interestingly, the load was highest in tissue samples that were positive for both CYP46-TT and APOE-E4. Aβ load was lowest in subject tissues without either the CYP46-TT or APOE-E4 polymorphisms while it was intermediate—and roughly equal—in those which had one or the other, but not both.

The researchers next looked at levels of Aβ and tau in the cerebrospinal fluid (CSF) of 38 living patients with known Alzheimer’s and 25 control subjects. CSF levels of Aβ were highest in Alzheimer’s patients who were positive for the CYP46-TT polymorphism, even significantly elevated from levels of Aβ in patients known to have Alzheimer’s, but were negative for CYP46-TT.

CSF levels of tau were markedly higher in patients with Alzheimer’s than in the control group. Those subjects with CYP46-TT and APOE-E4 with clinically diagnosed Alzheimer’s had the highest levels of CSF tau, followed by those that had both polymorphisms but no diagnosis. Again, roughly equal and intermediate levels were found in patients who had one or the other polymorphism, but not both, suggesting a relatively equal effect from each polymorphism to elevate both Aβ and tau. The lowest levels were seen in patients that were negative for both polymorphisms.

Finally, Papassotiropoulos performed genetic-association studies on two separate populations involving 201 patients with Alzheimer’s and 248 control subjects. Subjects who carried the CYP46-TT polymorphism were 2.2 times more likely to have LOAD than those who were CYP46-TT negative, while those who were positive for APOE-E4 were 4.4 times more likely to be diagnosed with LOAD. Subjects carrying both polymorphisms were 9.6 times more likely to have LOAD.

Papassotiropoulos and his coauthors concluded that not only is CYP46-TT a novel genetic risk factor for LOAD that warrants further study in larger populations, but that the polymorphism works in a synergistic way with APOE-E4 to increase risk for the disease drastically.

In an editorial accompanying the report by Papassotiropoulos, Benjamin Wolozin, M.D., Ph.D., associate professor of pharmacology at Loyola University Medical Center, noted that the CYP46-TT linkage to Alzheimer’s "integrates comfortably with the model of LOAD based on regulation of Aβ production by cholesterol." Wolozin also says that the CYP46-TT study may work to focus attention on the "potential importance of cholesterol metabolism in LOAD." The results, he wrote, "suggest the possibility that LOAD, the most common degenerative disease of the brain, is a general end point for abnormalities that increase the amount of cholesterol in the central nervous system. If so, inhibiting cholesterol metabolism in the brain might represent a viable treatment for LOAD."

An abstract of "Increased Brain β-Amyloid Load, Phosphorylated Tau, and Risk of Alzheimer Disease Associated With an Intronic CYP46 Polymorphism" is posted on the Web at http://archneur.ama-assn.org/issues/v60n1/abs/noc20166.html.

Arch Neurol20036029

Interactive Graphics


Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).