ABC transporters function

       Using bioinformatic tools, we have identified 14 ABC transporters that are expressed in the brain. It is apparent that there are no brain-specific ABC transporters: all ABC transporters that have been identified as being expressed in brain are also expressed in at least one other tissue, and several are expressed quite widely. While Table 10.1 provides data about general tissue distribution, further analyses will be required to determine whether these transporters are localized to neurons, glia, or even to cerebral vasculature. The latter point is important, as ABC transporters such as MDR1 and MRP1 represent key elements of the blood-brain barrier and are prominently expressed in cerebrovascular endothelial cells (Cordon- Cardo et al., 1989). We have put forth the hypothesis that ABC transporters are involved in the process of detachment of AP from cellular membranes (Lam et al., 2001).

      A(3 is unlikely to aggregate while attached to the membrane, as the hydrophobic amino acids in the peptide COOH tail would be shielded by their association with the lipid bilayer. On the other hand, the likelihood of Ap aggregation increases substantially following membrane detachment. Thus, detachment of Ap} from the membrane may represent a critical change in the biophysical properties of the peptide, and may very well be a prerequisite to the aggregation events that are thought to be at the core of the pathology of Alzheimer's disease. In summary, we have identified the suite of ABC transporters that are expressed in the brain. Such bioinformatic analysis is a prerequisite to the functional expression of each of these transporters in model systems.

    With such information in hand, we will be able to determine which brainexpressed ABC transporters function as (3-amyloid efflux pumps. Such proteins represent novel targets for the development of drugs that can regulate p-amyloid levels in the brain.

ABC transporter genes

         Allikmets, Gerrard, Hutchinson, and Dean (1996) characterized a large number of human ABC transporter genes by searching the human EST database using sequences derived from ABC (Walker A, ABC signature, and Walker B) of MDR1, as well as the entire sequence of cystic fibrosis transmembrane conductane regulator (CFTR). Because many of the ESTs are derived from brain libraries, the data suggested that many of these ABC transporters were expressed in brain. Because the number of ESTs deposited in public databases has increased substantially in the past 4 years, we recently repeated this analysis in order to ensure that the full complement of brain-expressed ABC transporters was represented in the database.

      Moreover, to increase the power of our search, we developed consensus amino acid sequences for the Walker A motif, as well as the ABC signature and Walker B motifs, by comparing the sequences of the following human ABC transporters.

      The BLAST programs (Altschul, Gish, Miller, Myers, & Lipman, 1990) were used to search the human database EST using these consensus sequences, as well as the complete amino acid sequences of the human ABC-transporter proteins identified above. Human clones with the highest scores were retrieved and primers were designed using the program Mac Vector (Oxford Molecular, UK). Reverse transcriptase-polymerase chain reaction (RT-PCR) was performed using brain fetal and adult total RNA (Invitrogen) in order to confirm human brain expression, and PCR products were sequenced to authenticate the ESTs. Based on both the cDNA library source used to generate these ESTs, as well as the expression profile of RNA as seen in Northern blots (Allikmets et al. 1996, and unpublished data), we identified the 14 ABC transporters listed in Table 10.1 as being brain-expressed.

Neurofibrillary tangles, senile plaques, and neuronal death

       Alzheimer's disease is characterized by neurofibrillary tangles, senile plaques, and neuronal death. The neurofibrillary tangles contain paired helical filaments composed of hyperphosphorylated tau, while the senile plaques are comprised of an array of proteins deposited around a core of insoluble Ab peptide.

       The cause of neuronal death remains unknown, but considerable evidence suggests that it is secondary to an increase in the brain Ab load. The molecular events involved in the constitutive production of Ap are increasingly being understood.

     The first step appears to be cleavage of the amyloid precursor protein (APP) by (3-secretase (Vassar et al., 1999; Lin et al., 2000), yielding an extracellular fragment known as sAPP(3, which is shed into the extracellular space (Mills & Reiner, 1999). Cleavage of C99 within the membrane by an enzyme known as y-secretase, which appears to be identical to the presenilins (Wolfe et al., 1999; Lin et al., 2000), liberates intact A{3. Both the 40 and 42 amino acid versions of A(3 are amphipathic, consisting of 28 charged amino acids and either 12 or 14 hydrophobic amino acids (for A(3t_40 and A(3142, respectively).

Discussion about neurobiologic and clinical events

        The neurobiologic and clinical events leading to the development of clear-cut AD are now being clarified. Emphasis in therapeutic trials is increasingly focused on strategies for preventing the development and progression of AD.

      Screening tools that can be used both in routine clinical practice and in research studies may play an important role in the development and effective use of these new therapies.

Materials and Methods

       The interview was designed to require no more than 5 minutes to administer and to be standardized so that a person with no clinical training could conduct the interview over the phone. The interview contains one item testing for the delayed recall of a list of three objects. The interview also contains some simple questions about cognitively demanding activities of daily living. Initially, the interview was given to a group of AD patients and agematched controls to determine its acceptability and whether there were obvious problems with telephonic administration and scoring.

      This pilot evaluation lead to rewording of some items and to clarifications in the administration procedure. We are now conducting a reliability and concurrent validity study. There are 90 participants in this study, all of whom have received a cognitive and diagnostic evaluation by a team of dementia experts in the past 6 months. Some of the participants have AD and others are elderly, nondemented controls.

      The telephone interviewers are blind to the participant's diagnosis and cognitive evaluation data. To assess reliability, each participant is being interviewed twice during a 1-month period. Concurrent validity will be assessed by determining the correspondence between scores on the screening interview and the participant's most recent clinical diagnosis.

      More detailed analyses will look at the ability of the screening instrument to identify AD patients with very mild disease. Although the sample size is too small to draw definitive conclusions about the possible influence of confounding variables, we will also examine the relationship of screening test scores to factors such as education, age, and medical comorbidities. These analyses will determine the reliability of the screening tool and provide a first look at its ability to identify cases of AD. Further validity studies will be conducted, subsequently, provided that the reliability and validity data in the current study are acceptable. One additional study will involve elderly patients enrolled in a managed care plan. None of the participants in this study will have received a full diagnostic evaluation for dementia. Participants will be screened over the telephone and determined to be cognitively normal or possibly impaired. Subsets from each group will then be examined clinically to determine the extent to which the scr

METHODS AND RESULTS

        Allikmets, Gerrard, Hutchinson, and Dean (1996) characterized a large number of human ABC transporter genes by searching the human EST database using sequences derived from ABC (Walker A, ABC signature, and Walker B) of MDR1, as well as the entire sequence of cystic fibrosis transmembrane conductane regulator (CFTR). Because many of the ESTs are derived from brain libraries, the data suggested that many of these ABC transporters were expressed in brain. Because the number of ESTs deposited in public databases has increased substantially in the past 4 years, we recently repeated this analysis in order to ensure that the full complement of brain-expressed ABC transporters was represented in the database. Moreover, to increase the power of our search, we developed consensus amino acid sequences for the Walker A motif, as well as the ABC signature and Walker B motifs, by comparing the sequences of the following human ABC transporters.  The consensus sequences that emerged from this analysis were (G(X)2G(X)GK(X)T(X)4L(X)2L(X)2PT(X)3G for the Walker A motif, and LSGG(X)4L(X)2A(X)AL(X)3PKV(X)2LDE(X)TS(X) for the ABC signature and Walker B motifs.

       The BLAST programs (Altschul, Gish, Miller, Myers, & Lipman, 1990) were used to search the human database EST using these consensus sequences, as well as the complete amino acid sequences of the human ABC-transporter proteins identified above. Human clones with the highest scores were retrieved and primers were designed using the program Mac Vector (Oxford Molecular, UK). Reverse transcriptase-polymerase chain reaction (RT-PCR) was performed using brain fetal and adult total RNA (Invitrogen) in order to confirm human brain expression, and PCR products were sequenced to authenticate the ESTs.

       Based on both the cDNA library source used to generate these ESTs, as well as the expression profile of RNA as seen in Northern blots (Allikmets et al. 1996, and unpublished data), we identified the 14 ABC transporters listed in Table 10.1 as being brain-expressed.

Identification of Brain-Expressed ABC Transporters That May Mediate Detachment of p-Amyloid From Biological Membranes

         Alzheimer's disease is characterized by neurofibrillary tangles, senile plaques, and neuronal death. The neurofibrillary tangles contain paired helical filaments composed of hyperphosphorylated tau, while the senile plaques are comprised of an array of proteins deposited around a core of insoluble A(3 peptide).

        The cause of neuronal death remains unknown, but considerable evidence suggests that it is secondary to an increase in the brain A(3 load. It has been known for nearly 10 years that the A|3 peptides are rapidly released from cells (Haass et al., 1992; Seubert et al., 1992; Shoji et al., 1992; Busciglio, Gabuzda, Matsudaira, & Yankner, 1993), but the hydrophobic amino acids at the COOH-terminus make it likely that the peptide will remain associated with the membrane following y-secretase cleavage. Thus, we hypothesized that an active process was required in order for A(3 to detach from the membrane. Selected members of the ATP-binding cassette (ABC) superfamily of transporters are responsible for the energydependent efflux of a variety of lipophilic and amphipathic molecules from cells, and the process bears a striking similarity to that which occurs with the AP peptide.
      ABC transporter known as MDR1 is an Ap efflux pump.

     We have identified a single ABC transporter, MDR1, as an AP efflux pump. Cells throughout the body constitutively produce and release Ap, yet the MDR1 protein is only expressed in a limited number of tissues, and is essentially undetectable in neurons

Discussion

       The neurobiologic and clinical events leading to the development of clear-cut AD are now being clarified.

     Emphasis in therapeutic trials is increasingly focused on strategies for preventing the development and progression of AD.
     Screening tools that can be used both in routine clinical practice and in research studies may play an important role in the development and effective use of these new therapies.

Materials and Methods

         The interview was designed to require no more than 5 minutes to administer and to be standardized so that a person with no clinical training could conduct the interview over the phone. The interview contains one item testing for the delayed recall of a list of three objects. 

    The interview also contains some simple questions about cognitively demanding activities of daily living.                                                                                                                                     

     Initially, the interview was given to a group of AD patients and agematched controls to determine its acceptability and whether there were obvious problems with telephonic administration and scoring. This pilot evaluation lead to rewording of some items and to clarifications in the administration procedure. We are now conducting a reliability and concurrent validity study. There are 90 participants in this study, all of whom have received a cognitive and diagnostic evaluation by a team of dementia experts in the past 6 months. 

    Some of the participants have AD and others are elderly, nondemented controls. The telephone interviewers are blind to the participant's diagnosis and cognitive evaluation data. To assess reliability, each participant is being interviewed twice during a 1-month period. 

     

      Concurrent validity will be assessed by determining the correspondence between scores on the screening interview and the participant's most recent clinical diagnosis. More detailed analyses will look at the ability of the screening instrument to identify AD patients with very mild disease. Although the sample size is too small to draw definitive conclusions about the possible influence of confounding variables, we will also examine the relationship of screening test scores to factors such as education, age, and medical comorbidities. 

     These analyses will determine the reliability of the screening tool and provide a first look at its ability to identify cases of AD. Further validity studies will be conducted, subsequently, provided that the reliability and validity data in the current study are acceptable. One additional study will involve elderly patients enrolled in a managed care plan. None of the participants in this study will have received a full diagnostic evaluation for dementia. Participants will be screened over the telephone and determined to be cognitively normal or possibly impaired. Subsets from each group will then be examined clinically to determine the extent to which the screening classification agrees with the results of a diagnostic evaluation.

TELEPHONE SCREENING FOR ALZHEIMER'S DISEASE

     A complete diagnostic evaluation for AD is time-consuming, expensive, and requires substantial expertise on the part of the examining physician.
    In clinical settings and in research studies, there is a need for brief screening tools that could help to identify those individuals who should receive a full diagnostic evaluation. Effective screening tools could help to ensure that scarce diagnostic services are focused on those individuals who are most likely to have dementia. Screening can be conducted in a variety of settings and, potentially, using a variety of different tools. As examples, screening could be done in clinics, doctors' offices, door-to-door surveys, by mail, or by telephone. Several mental-status interviews have been developed or adapted for use over the telephone (e.g., Brandt, Spencer, & Folstein, 1988; Roccaforte, Burke, Bayer, & Wengel, 1992). Most of these instruments are designed as telephonic versions of brief mental-status examinations, such as the MMSE and, therefore, have enough questions to measure the severity, as well as detect the presence of dementia.

    Recently, we have initiated a study to test the reliability and validity of a very brief telephonic-screening interview that could be given by nonprofessionals in a few minutes. The interview is not designed to enable a
specific dementia diagnosis or to evaluate the severity of dementia, but simply to identify persons with a high probability of being demented.

NEUROBIOLOGY OF EARLY ALZHEIMER'S DISEASE

     Details of the studies on early AD conducted at our center have been published. The population for these studies are the residents of the Jewish Home and Hospital (JHH), a long-term care facility affiliated with the Mount Sinai School of Medicine in New York.

    The two main campuses of the JHH have approximately 1,600 residents, with an average age of over 85. Cognitive screening is a routine part of clinical care in the JHH, and Mini-Mental State Examination (MMSE)
scores are available for nearly all residents. For the early AD study, all consenting residents with MMSE scores of 15 or greater are given a thorough diagnostic evaluation and are assigned a score on the Clinical
Dementia Rating (CDR; Morris et al., 1993). Those with CDR scores of 0 (no dementia), 0.5 (questionable dementia), or 1.0 (mild dementia) are administered a battery of neuropsychological tests.

     The JHH routinely requests autopsy permission when a resident dies and, over the period of the early AD project, more than 50 autopsies have been obtained from residents who died with CDR scores of 0 to 1.0, and evidenced either no significant neuropathology or had neuropathologic lesions associated with AD only. As might be expected with cases from a long-term care facility, many more autopsies have been obtained from residents who died with more severe dementia or with comorbid neuropathologic lesions.

     Neuropathologic studies of cases dying with no dementia, questionable dementia, or mild dementia have shown that senile plaques are more abundant in most areas of the neocortex in patients with CDR 0.5 than in persons dying without any evidence of dementia (CDR 0) (Haroutunian et al., 1998). Examination of specific amyloid fragments has shown that the Afi peptides, especially the A^ 42 peptide, are markedly elevated even in the CDR 0.5 cases, compared with the CDR 0 cases (Naslund et al., 2000). By contrast, neurofibrillary tangles were evident in the entorhinal cortex and hippocampus of virtually all brains from
persons over age 80, even those from nondemented individuals; extensive neurofibrillary tangles in the neocortex were evident only in patients with CDR 2.0 (moderate dementia) or greater (Haroutunian et al., 1999). These data support the view that overproduction and accumulation of amyloid protein, particularly the ^ 42 fragment, is a very early manifestation of AD. Neurofibrillary tangles in entorhinal cortex and hippocampus are age-related phenomena that are further influenced by AD as dementia progresses
from severe to terminal, while the development of tangles in the neocortex is associated with progression to moderate and severe dementia.

     Other biologic manifestations of AD have also been examined in this series. Cholinergic markers, including choline acetyltransferase (ChAT) and acetylcholinesterase, were not diminished in early AD cases, but were reduced substantially in those patients dying with moderate-to-severe dementia (Davis et al., 1999a).

NEUROPSYCHOLOGY OF EARLY ALZHEIMER'S DISEASE

      Cross-sectional studies comparing patients with very mild AD to agematched,  nondemented control subjects have demonstrated that a deficit in memory is the earliest and most prominent neuropsychological deficit in patients with diagnosed AD (Welsh, Butters, Hughes, Mohs, & Heyman,1991). This deficit is most pronounced on tasks for which the patients are asked to recall previously learned information (such as a short list of words) after a brief delay during which the patient engages in other cognitive activity.

      Longitudinal studies of nondemented persons who are at risk for dementia have examined the question of whether there are neuropsychological deficits that are measurable before patients are impaired enough to warrant a diagnosis of AD. In these studies, baseline neuropsychological data are used to compare performance of patients who subsequently were diagnosed with AD with those who remain dementia free.

     Results of these studies indicate that there are measurable deficits in memory and, to a lesser extent, in language and cognitive-processing speed at least 1 year before patients meet diagnostic criteria for dementia  Our own results from the JHH study confirm that patients with a CDR score of 0.0 who convert to CDR 0.5 1 year later have baseline memory scores that are poorer than the baseline memory scores of patients who remain CDR 0.0 on follow-up. Thus, data from a variety of sources indicates that poor scores on a test of memory, particularly delayed-recall memory, are a sensitive indicator of early dementia.

Early Clinical and Biological Manifestations of Alzheimer's Disease: Implications for Screening and Treatment

        Alzheimer's disease (AD) is a slowly progressive disease with an average of more than 10 years from first manifestations to death. Both the underlying biology of AD and its clinical manifestations change substantially over the course of the illness. This chapter reviews briefly some recent studies investigating the neurobiologic changes in the brains of AD patients over the course of the illness, with an emphasis on the earliest changes.

     Cognitive deficits also change over the course of illness and an understanding of the earliest cognitive changes may enable the development of screening instruments for AD. An ongoing set of studies designed to determine the reliability and validity of telephonic screening instruments is described. An instrument of this type could be useful in clinical trials of agents for the primary prevention of AD, as well as in health service-delivery organizations where there is a need to identify cases of AD that have not yet been diagnosed.

Positron Emission Tomography Imaging of Amyloid Senile Plaques and Neurofibrillary Tangles

      Several groups have been developing new, small-molecule probes to image the amyloid NPs and NFTs.        
     Current methods for measuring brain amyloid, such as histochemical stains, require tissue fixation on postmortem or biopsy material. Available in vivo methods for measuring NPs or NFTs are indirect. Studies that may lead to direct in vivo, human AP imaging include various radiolabeled probes using small organic and organometallic molecules capable of detecting differences in amyloid-fibril structure or amyloid-protein sequences. Investigators also have used chrysamine-G, a carboxylic acid analogue of Congo red, an amyloid-staining histologic dye, serum amyloid-P component, a normal plasma  glycoprotein that binds to amyloid-deposit fibrils (Lovat, O'Brien, Armstrong, et al., 1998), or monoclonal antibodies (Majocha et al, 1992).
      Methodological difficulties that hinder progress with these techniques include poor blood-brain barrier crossing and limited specificity and sensitivity. In addition, most approaches do not measure both NPs and NFTs. Recently, Barrio et al. (1999) reported using a hydrophobic, radiofluorinated derivative of l,l-dicyano-2-[6-(dimethylamino)naphthalen-2-yllpropene (FDDNP) (Jacobson, Petric, Hogenkamp, Sinur, & Barrio,1996) with PET to measure the cerebral localization and load of NFTs and SPs in AD patients. The probe showed visualization of NFTs, NPs, and diffuse amyloid in AD brain specimens using in vitro fluorescence microscropy, which matched results using conventional stains (e.g., thioflavin S) in the same tissue specimens. Such approaches may ultimately aid in the early detection of AD and brain-function monitoring during antidementia treatment trials, particularly those designed to interrupt accumulation of NPs and NFTs.

Preclinical Detection: Benefits and Strategies

      Although no cure exists for AD, preclinical disease detection has several benefits. When early detection assessments are negative, people with mild memory complaints can be reassured that their forgetfulness reflects a normal age-related change that probably will not progress.

     In addition, many people would like to know about a poor prognosis while still in a mildly impaired state in order to plan their futures while mental faculties remain. Perhaps the most compelling argument for preclinical detection strategies is to identify candidates for novel antidementia treatments before the dementing process causes extensive neuronal death, since new antidementia treatments are more likely to delay the dementing process than to reverse neuronal death.
      Although current cholinergic treatments have been shown to result in symptomatic, rather than disease-altering or structural effects, it would certainly be of interest to initiate treatments very early when searching for a disease-modifying effect. Moreover, both the expense and potential risks of treatment make it reasonable to reserve treatment only for those people who are at the greatest risk for developing the disease.