Heterocyclic compounds as imaging probes of tau pathology

Our pharmacokinetics department offers the clients a broad spectrum of high quality of services in the areas of in vitro ADME, in vivo pharmacokinetics and bioanalysis services, ranging from small molecules to large molecules, such as protein and antibody. The animal species involved in our services are non-human primate, canine, mice, rat, rabbit and hamster. Meanwhile, non-human primate experimental platform and isotope platform for protein/antibody are certified by the Shanghai Government. Email:marketing@medicilon.com.cn Web:www.medicilon.com

A number of novel compounds were screened for their ability to bind tau-i- neurofibrillary tangles in Alzheimer disease brain tissue, in vitro ADME properties and brain uptake in vivo. Taken together, the results demonstrate that a selection of compounds of the invention bind preferentially to tangles, are metabolically stable in vitro, can be radiolabeled, and have a high brain uptake in rodent models. Thus, such compounds display the desired characteristics for a tau imaging agent.

1 Material and methods

1.1 Human tissue

Human brain tissue samples from the entorhinal cortex of patients with Alzheimer's disease (AD) and healthy controls were obtained from Tissue Solutions . Tissue samples were collected after informed written consent, and specimens were dissected at the tissue bank and snap-frozen for cryopreservation at a time interval of between 3 and 18 h after death. The frozen tissue was embedded in TissueTek® (VWR) and sectioned using a Microm HM560 cryostat. Serial 12 μιη sections were mounted onto SuperFrost®+ glass slides (VWR) and stored a -70°C.

1.2 Immunohistochemistry

To confirm presence and location of tau-i- neurofibrillary tangles (NFTs) and β-amyloid (Αβ)+ plaques in the human tissue sections, every 20th tissue section throughout the specimens was processed for immunohistochemical labelling with antibodies raised against aggregated and hyperphosphorylated tau, and aggregated Αβ. Briefly, tissue sections were defrosted and fixed in ice-cold 70% ethanol. All tissue sections were rinsed with PBS after fixation and between all subsequent incubation steps.

Following fixation, tissue sections were incubated first with H202 (En Vision™ kit, Dako). Tissue sections to be processed for Αβ immunohistochemistry were further treated for antigen retrieval by incubation in 70% formic acid (Sigma- Aldrich) for 10 min. All tissue sections were then incubated with 10% normal goat serum (Vector Labs) to block nonspecific labelling. After the blocking steps, the tissue sections were incubated with primary antibodies raised against tau (AT8, mouse monoclonal antibody, 1 :20 dilution, Invitrogen) or Αβ (4G8, mouse monoclonal antibody, 1:100 dilution, Covance) for 1 h at room temperature (RT).

Following incubation with primary antibodies, the tissue sections were incubated with secondary antibodies conjugated to horseradish peroxidase (HRP) directed against mouse IgG for 30 min at RT. This was followed by incubation with the chromogen 3,3'- diaminobenzidine (DAB) for 2-3 min. En Vision™ HRP kits were used for secondary labelling (Dako). Finally the sections were counterstained with haematoxylin (Merck), dehydrated and mounted in DPX mounting media (VWR). Images of tissue sections labelled with tau and Αβ were captured using a Nikon digital camera connected to a Leica microscope and using the NIS Elements D software (Nikon). Images were further processed with the Photoshop® software (Adobe).

1.3 Gallyas silver stain

Conventional immunohistochemistry rely on the presence and detection of specific antigen by primary antibodies. For example, the tau antibody (AT8) used for the immunohistochemical detection of NFTs in 1.2 detects a specific conformation of hyperphosphorylated tau aggregates, but it will not detect less mature tau aggregates (Augustinack et al., 2002). Likewise, further phosphorylation results in conformational changes and loss of the AT8 specific tau antigen (Augustinack et al., 2002, Jeganathan et al., 2008). It has therefore been suggested that using a different method, such as Gallyas silver stain, that doesn't rely on one antigen is a more sensitive and accurate method to detect and label NFTs (Rosenwald et al., 1993, Cullen et al., 1996, Uchihara et al., 2001, Uchihara, 2007). Therefore, in addition to tau-i- and Αβ+ immunohistochemistry, tissue sections adj acent to the slides used for immunohistochemistry where processed for Gallyas silver stain. Briefly, tissue sections were defrosted and fixed for 10 min in neutral buffered formalin (VWR) and washed first in PBS and then dH20. Unless stated otherwise, tissue sections were rinsed in dH20 between each of the subsequent incubation steps. First, the tissue sections were incubated in 5% periodic acid for 5 min, and then for 1 min in an alkaline silver iodide solution. This was followed by a 10 min wash in 0.5% acetic acid, and then the tissue sections were incubated in developer solution for 5-30 min. The tissue sections were then washed in 0.5% acetic acid and rinsed in dH20. This was followed by incubation for 5 min in a 0.1 % gold chloride solution, and then 5 min in 1 % sodium thiosulphate solution. The tissue sections were then rinsed in tap water and counterstained with 0.1% nuclear fast red for 2 min.

Finally, the tissue sections were rinsed in tap water, dehydrated and mounted in DPX mounting media (VWR). All reagents for the Gallyas silver stain were procured from Sigma-Aldrich unless otherwise stated. Images of tissue sections labelled with tau and Αβ were captured using a Nikon digital camera connected to a Leica microscope and using the NIS Elements D software. Images were further processed with the Photoshop® software (Adobe). 1.4 Tissue binding assay

The binding of compounds to tau+ NFTs and Αβ+ plaques in human AD tissue were evaluated based on fluorescence. All test compounds have an innate fluorescence, and binding of the compounds to NFTs/plaques in AD tissue can therefore be detected using a fluorescence microscope. Two reference compounds were included in the screen; PiB (Pittsburgh compound B and FDDNP. PiB has been reported to bind with a preference to Αβ+ plaques, whereas FDDNP binds to both NFTs and plaques . In addition, an aminothienopyrazidine compound (ATPZ), a tau aggregation inhibitor first reported by Ballatore et al (Ballatore et al., 2010), was also screened on tissue. Briefly, tissue sections were defrosted and fixed in ice-cold 70% ethanol. All tissue sections were rinsed with PBS after fixation and between all subsequent incubation steps. To quench autofluorescence, tissue sections were incubated first with 0.25% KMn04 (Sigma- Aldrich) in PBS for 12 min, and then with 0.1 % K2S2O5/0.1 % oxalic acid (both reagents from Sigma- Aldrich) in PBS for 1 min. The tissue sections were blocked with 2% BSA in PBS for 10 min, and then incubated with the test compounds at 100 μΜ concentration for 1 h at RT. Compounds with positive binding at 100 μΜ was further tested in subsequent assays using lower test concentrations, 10 μΜ and 1 μΜ. Finally the tissue sections were rinsed in PBS, and mounted in SlowFade® mounting media (Invitrogen).

1.5 In vitro ADME screening

Test compounds were screened using a panel of in vitro ADME assays for prediction of in vivo properties. The following assays were used; parallel artificial membrane permeability assay (PAMPA) to determine cell membrane passage, compound stability in the presence of human or rat plasma, compound stability in the presence of human or rat liver microsomes, and determination of binding to proteins in human plasma and rat brain homogenates. To enable comparison with two compounds reported to have high brain uptake in vivo, PiB and ATPZ were included in the screen.

1.5.1 PAMPA

The PAMPA assay is used to determine how well a compound crosses a cell membrane by measuring its passage through a phosphotidyl choline barrier. A permeability coefficient >- 6 indicates high permeability across lipid membranes and is indicative of a compounds ability to cross the blood brain barrier. A 10 μΜ solution was incubated on a PDVF membrane coated with a 2% phoshotidyl choline solution for 5h at RT. Membrane penetration was measured using LC-MS. 1.5.2 Protein binding assays

The protein binding assays provide an estimate of free fraction of the compound in the blood or brain in vivo. High protein binding of a compound within the blood indicates that it is potentially unavailable for passage across the blood brain barrier and could compromise its metabolism or excretion, whereas high binding to proteins in the brain is indicative of non-specific binding and slow excretion. The desirable criterion for this assay is < 99% of test compound bound. Test compounds were first dissolved in DMSO to a concentration of 50 μΜ. This was followed by incubation in samples of human plasma and rat brain homogenates (final test concentration 1 μΜ). Binding of compounds to proteins was determined in the samples by rapid equilibrium dialysis after 5 and 30 min of incubation. 1.5.3 Liver microsome stability assay.

The liver microsome stability assay provides an estimate of compound stability and rate of metabolism in vivo. The desirable criterion for this assay is > 50% parent compound after 30 min. A 1 μΜ compound solution was incubated with rat or human liver microsomes (20mg/ml) at 37C and the amount of parent compound remaining following the incubation was determined after 5 and 30 min of incubation using LC-MS.

1.6 In vivo cold bio-distribution

All animal studies were in compliance with local rules and regulations. Test compounds were administered by intravenous injection through the tail vein of naive male Wistar rats. The animals were sacrificed by dislocation of the neck at 2, 10, 30, 60 min post-injection. The brain and plasma were collected from each animal. The concentration of test compound was measured in the plasma and brain homogenates using LS-MS, and calculated as % compound/g (%ID/g). 1.7 In vivo bio-distribution with radio belled compounds

All animal studies were in compliance with local rules and regulations. [F]-radiolabelled compounds were injected i.v through the tail vein of naive male C57B1/6 mice (2MBq/mouse). The animals were sacrificed by dislocation of the neck at 2, 10, 30 and 60 min p.i. Next, the animals were dissected and the radioactivity of organs, tissue and blood was measured using a Wallac γ counter (Perkin-Elmer). The compound concentration in the specimens was calculated as ID/g.