22 October 2007. The Pittsburgh Compound-B (PIB) ligand is one of the most promising diagnostic tools for Alzheimer disease (AD). But even as PIB-based positron emission tomography (PET) in humans has blossomed around the globe, it has also been plagued by a downside. PIB-PET imaging has been unable to distinguish normal mouse brains from those laden with amyloid plaques—something researchers and drug developers would clearly like. In the October 10 Journal of Neuroscience, researchers led by Makoto Higuchi at the National Institute of Radiological Sciences, Chiba, Japan, report solving that dilemma. The researchers have generated a carbon 11 form of PIB with much higher radioactivity than has been achieved before. With this super hot PIB, they not only managed to image amyloid plaques in transgenic mice, but track amyloid reduction in response to anti-Aβ treatment as well. The study has drawn praise from Bill Klunk and Chet Mathis at the University of Pittsburgh, who developed PIB (see comment below).
Klunk and colleagues had suggested that the inability to detect mouse plaques with PIB is due to a dearth of high-affinity PIB binding sites in mouse amyloid (Klunk et al., 2005; see ARF related news story). One potential solution to that, Klunk told ARF via e-mail, is to use PIB with higher specific radioactivity. The hotter the molecule, the fewer of them are needed to get the same radioactivity, and fewer molecules means less background and better sensitivity. This is just the approach taken by the Japanese researchers. First author Jun Maeda and colleagues synthesized C-11 PIB that is around an order of magnitude hotter than that used in previous studies of transgenic mice.
Using the hotter tracer, Maeda and colleagues were able to detect amyloid plaques in APP23 mice by PET. In keeping with the known pathology in this model, they found that the PIB signal was detectable by 17 months, and grew more intense as the animals aged (up to 29 months). They also were able to use PIB to monitor a passive immunization protocol. One and 2 weeks after injecting anti-Aβ antibodies into one side of the animals’ hippocampi, the researchers detected a decrease in PIB retention in the same side. They also detected a concomitant increase in signal from a PET ligand, [18F]-FE-DAA1106, that binds to the peripheral benzodiazepine receptor, a marker of glial cell activation. Wild-type mice retained little [18F]-FE-DAA1106 after passive immunization, suggesting the glial response was due to the presence of amyloid.
The difficulty in using PIB in mice suggests that there may be something qualitatively different about mouse and human amyloid deposits, but what? That’s a key question, according to Klunk. “Per unit volume, mice have as many or more plaques than human AD, and per mg brain, Tg mice have 10-20 times more insoluble Aβ than in human AD brain. But, each mole of that mouse Aβ only contains 1/500th the number of PiB binding sites,” Klunk wrote to ARF. To find out what that qualitative difference between human and mouse amyloid might be, Maeda and colleagues measured PIB binding to a variety of Aβ subtypes. They found that, in both human and mouse brain, PIB retention correlated with levels of an N-terminal, pyroglutamate derivative of Aβ (see ARF related news story) and that this AβN3-pyroglutamate had much higher affinity for PIB in vitro. Because the formation of AβN3-pyroglutamate is a slow enzymatic process, the researchers suggest that accelerated production of Aβ in transgenic mice does not promote development of “AD-like” plaques, which are enriched in the pyroglutamate derivative.
The Pacific Rim has produced other recent advances on PIB. Researchers led by Christopher Rowe at the Centre for PET, Austin Health, Heidelberg, Australia, report in the October 10 Brain online that the ligand labels amyloid deposits in elderly people who, to all intents and purposes, appeared cognitively normal. On closer inspection, it turned out that those people have poor episodic memory. The finding supports the idea that amyloid deposition occurs long before any clinical signs of AD, but more importantly, perhaps, suggests that PIB binding portends full-blown AD.
First author Kerryn Pike and colleagues (including Klunk), measured PIB signals in 31 AD patients, 33 patients with mild cognitive impairment (MCI), and 32 healthy older adults (mean age 71.6 years, MMSE score 29.2 +/- 0.9). Not surprisingly, they found increased cortical PIB binding in 97 and 61 percent of AD and MCI patients, respectively. But they also found that 22 percent of the controls had increased PIB binding. Pike and colleagues found that those PIB-positive controls performed worse in a test of episodic memory than PIB-negative controls. They also found a correlation between episodic memory and PIB binding. This correlation was much stronger in the MCI and control groups.
Because memory loss is one of the earliest and most predictive changes associated with AD, the authors suggest that Aβ gets deposited early in the pathological process. This is compatible with postmortem data showing that people who reported memory complaints but were clinically non-AD had some AD pathology on autopsy (see ARF related news story).
Last but not least, another Australian team, this one by Victor Villemagne at the University of Melbourne, report in the September 26 Journal of Neuroscience that though PIB does bind α-synuclein fibrils, PIB-PET signals in dementia with Lewy bodies (DLB) are due almost entirely to amyloid plaques. The finding could have diagnostic ramifications.
First author Michelle Fodero-Tavoletti and colleagues compared binding of tritiated PIB with synthetic α-synuclein fibrils and synthetic Aβ fibrils. They found that both fibrils had high- and low-affinity binding sites for PIB, but the high-affinity Aβ site bound more than 10 times tighter to the ligand than α-synuclein’s high-affinity site. Next, the researchers measured PIB binding in amyloid-positive and amyloid-negative tissue taken postmortem from DLB patients. They found that while PIB bound tightly to the amyloid-positive tissue, it failed to bind to amyloid-negative samples.
The finding indicates that PIB can be used to detect purely Aβ amyloid, and that signals will not be complicated by contributions from α-synuclein or tau aggregates (previous work showed that PIB does not significantly bind to neurofibrillary tangles, see Klunk et al., 2003). Of course, the downside, as the authors point out, is that because of the overlapping pathology between DLB and AD, PIB will be incapable of differentiating between the two without additional clinical diagnosis.—Tom Fagan.
Maeda J, Ji B, Irie T, Tomiyama T, Maruyama M, Okauchi T, Staufenbiel M, Iwata N, Ono M, Saido TC, Suzuki K, Mori H, Higuchi M, Suhara T. Longitudinal, quantitative assessment of amyloid, neuroinflammation, and anti-amyloid treatment in a living mouse model of Alzheimer’s disease enabled by positron emission tomography. J. Neurosci. 2007 Oct 10; 27:10957-10968. Abstract
Pike KE, Savage G, Villemagne VL, Ng S, Moss SA, Maruff P, Mathis CA, Klunk WE, Masters CL, Rowe CC. Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer’s disease. Brain 2007 Oct 10; advanced access. Abstract
Fodero-Tavoletti MT, Smith DP, McLean CA, Adlard PA, Barnham KJ, Foster LE, Leone L, Perez K, Cortes M, Culvenor JG, Li Q-X, Laughton KM, Rowe CC, Masters CL, Cappai R, Villemagne VL. In vitro characterization of Pittsburgh compound-B binding to Lewy bodies. J Neurosci. 2007, Sep 26;27:10365-10371. Abstract