This chapter proposes to develop novel probes, which may be useful as in vivo imaging agents, for studying beta-amyloid aggregates in living human brains. These novel probes are based on highly conjugated styrene derivatives labeled with I-125/I-123. An initial binding study showed that these agents displayed high-binding affinity and distinctively labeled beta-amyloid aggregates in the postmortem brain sections of patients with Alzheimer's disease (AD). The proposed agents may provide simple and effective tools to diagnose and monitor patients with Alzheimer's disease (AD).
It is well known that AD is a neurodegenerative disease associated with regional neuronal loss (Selkoe,1998; Selkoe,1999; Sinha & Lieberburg, 1999). Clinical symptoms of AD include but are not limited to cognitive decline, irreversible memory loss, disorientation, and language impairment (Selkoe, 1997). Postmortem brain tissue examinations show neuropathology observations-the presence of senile plaques (Sps), neurofibrillary tangles, and neurophil threads containing beta-amyloid aggregates and highly phosphorylated tau proteins (Lee, 1996; Trojanowski & Lee, 1994). Several genomic factors have been linked to AD. Familial Alzheimer's disease has been reported to have mutations encoding beta-amyloid precursor protein (APP), apolipoprotein E4 (apo E-e4), presenilin 1 (PS1), and presenilin 2 (PS2) genes (Selkoe, 1997). The exact mechanisms of these four mutations, which lead to the development of AD, are not fully understood; however, the hallmark of AD is the formation of beta-amyloid aggregates. It is likely that P-amyloid precursor proteins are degraded by several proteases, among which the catabolism reactions of beta- and gamma secretases on APP lead to the production of excess Abeta (beta-amyloid). The excessive burden of A(3, produced by various normal or abnormal mechanisms, may represent the starting point of neurodegenerative events. Formation. of P-amyloid aggregates in the brain may be the pivotal event, which produces various toxic effects in neuronal cells, leading to the formation of neuritic plaques. The plaques consist of extracellular masses of A(3 filaments intimately associated with dystrophic dendrites and. axons, activated microglia, and reactive astrocytes (Selkoe, 1997).
It is, therefore, of great scientific interest to develop ligands that can specifically bind to the P-amyloid aggregates (Selkoe, 2000; Styren, Hamilton, Styren, Klunk, 2000; Wengenack, Curran, & Poduslo, 2000). The new ligands will not only be useful as in vivo diagnostic tools, but may also provide in vitro labeling agents for studying the beta-amyloid aggregates. They can be further modified as potential agents for inhibiting the formation of the P-amyloid aggregates. The proposed new ligands in this chapter may be useful as in vivo diagnostic tools for monitoring the formation of P-amyloid aggregates. Advances in early diagnosis of P-amyloid formation and aggregation may also lead to development of inhibitors for treatment of AD.
In the past few years, several interesting reports on developing P-amyloid aggregate-specific imaging agents have appeared in the literature (Ashburn, Han, McGuinness, Lansbury, 1996; Han, Cho, Lansbury, 1996; Klunk, Debnath, & Pettegrew, 1995; Klunk et al., 1997; Mathis, Mahmood, Debnath, Klunk,1997; Styren et al.; 2000; Wengenack et aL., 2000). By far, the most attractive approach is based on highly conjugated chrysamine-G (CG) and congo red (CR), normally used for fluorescent staining of the plaques and tangles in postmortem brain sections of AD patients. It was demonstrated that the binding of CG and CR (also 3'-bromo and 3'-iodo derivatives; see Table 8.1) appeared to be selective toward the beta-amyloid aggregates in vitro (A(beta1-40) peptide aggregated) or in AD brain tissues with confirmed P-amyloid aggregates.
At equilibrium, the binding of [C-14]CG toward beta-amyloid aggregates is a saturable and reversible process similar to that observed for receptor-ligand binding. The unique observation of the interaction between beta-amyloid aggregates (normally existing as an antiparallel beta-sheet. structure) and small, negatively charged and highly conjugated ligands may represent a novel opportunity to design specific ligands for single photon emission computed tomography (SPECT) imaging (Klunk, Debnath, & Pettegrew, 1994; Munk et al., 1995). Recently, a significant advance has been demonstrated by Klunk and Mathis, who reported the used of a compound called "X-34" (Styren et al., 2000). Replacing the diazo group with a simple vinyl group (both CG and CR are diazo dyes) appears to preserve the binding affinity (Mathis et al., 1997). This substitution has several advantages over the diazo derivatives: (a) the molecular weight decreases, smaller molecules may show preferable ability to penetrate intact blood-brain barrier (BBB); (b) the vinyl group improve the in vitro and in vivo stabilities; (c) X-34 showed a similar separation of two negative charges by highly conjugated aromatic rings; and (d) derivatives can be prepared via a versatile reaction scheme, which is amenable to additional substitution.
Results of a preliminary study suggest that the initial synthesis of a 3'bromo-derivative of X-34 (BSB) showed excellent binding affinity to beta-amyloid aggregates comparable to CG (Ki = 300 to 500 nM).
It is postulated that the basic requirements for designing a beta-amyloid aggregate-specific ligand are: (a) a highly conjugated back bone; (b) two negative charges that are highly conjugated; (c) the negative charges are associated with salicylate moieties; (d) it is likely that there is bulk tolerance 3-,4- and 3'-positions on the CG or X-34; and (e) the molecular weight of the tracer will be smaller than 750 for penetrating the intact bloodbrain barrier.
Based on the requirements listed above, a novel fluorescent probe, (E,E)-1-bromo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styryl-benzene (BSB), and two iodinated ligands, (E,E)-1-iodo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styryl-benzene (ISB) and (E,E)-1-iodo-2, 5-bis-(3-hydroxycarbonyl-4-methoxy)styryl-benzene (1MSB), were synthesized (see Figure 8.1).
Preliminary results on the new fluorescent compound, BSB, were reported recently (Skovronsky et al., 2000). Basic conclusions of initial studies in cell lines and in transgenic mice suggest that: (a) BSB sensitively labels SPs in AD brain sections; (b) BSB permeates living cells in culture and binds specifically to intracellular A(3 aggregates; (c) following intracerebral injection in living transgenic mouse models of AD amyloidosis, BSB labels SPs composed of human A(3 with high sensitivity and specificity; and (d) BSB crosses the BBB and labels numerous AD-like SPs throughout the brain of the transgenic mice following intravenous injection.
Under the sponsorship of this project, we have developed two radioiodinated ligands ([I-123/125]ISB and ([I-123/125]IMSB). These "prototype" compounds were tested for their in vitro and in vivo binding characteristics to beta-amyloid. The proposed initial studies are designed to test and assess the parameters required for a radioiodinated ligand as a suitable biological marker for detecting senile plaques in living human brain. It is hypothesized that when labeled with I-123 (T-1/2 = 13 hr, 159 KeV), the beta amyloidselective ligands with the proper sensitivity and in vivo pharmacokinetics will be useful as imaging agents in conjunction with SPECT. These ligands, if successfully developed, will be useful for the detection of Abeta aggregates prior to the onset of the disease and they ultimately may be applicable for quantitation of senile plaques in AD patients. We believe that early detection of the formation of plaques may be beneficial for the older population in which AD is relatively prevalent.
Aggregated Abeta((1-40)) or Abeta((1.42)) in solution was used as the model system for initial screening and characterization of ligand binding to amyloid fibrils (Figure 8.2). High-binding affinities were observed with both [I-125]ISB and [I-125]IMSB for Abeta((1-40)) aggregates. Similarly, [I-125]IMSB displayed specific binding to Abeta((1-42)) aggregates, but with a lower affinity as compared with its binding to Abeta((1-40)) aggregates (0.70 nM vs. 0.13 nM; see Figure 8.2).