Our study demonstrates the use of an anti-Aβ antibody to distinguish Aβ bearing TgCRND8 mice from wild type littermates. TgCRND8 mice can clearly be differentiated from wild type mice using systemically delivered 6E10, with the detection enhanced with 6E10-PEG. PET imaging of brain tissue after intravenous delivery of 6E10 resulted in significantly elevated brain concentrations of 6E10 in the TgCRND8 mice when compared to wild type controls. Brain concentrations of 6E10 remained constant over the 12 h period examined with PET imaging. After 6E10 was conjugated to PEG, we saw increased transport into brain tissue and evidence of Aβ binding.
6E10-PEG concentrations in brain tissue were elevated at 10 min and 2 h, but returned to concentrations observed in wild type controls at 4 h and 12 h, indicating antigen-mediated binding and clearance. In contrast, the concentration of 6E10-PEG in a reference tissue, where no antibody-antigen interactions are expected, remained constant over the entire time frame examined. Further, when the blood-brain barrier was disrupted by infusion with a hyperosmotic sugar solution, and thereby reduced (or eliminated) any differences in vasculature due to Aβ pathology, there is clear accumulation of 6E10 at early time points. Notwithstanding that hyperosmotic disruption of the blood-brain barrier is not a clinical strategy due to the invasive nature of this technique, there is higher accumulation of 6E10 in the brain. Moreover, the clearance profile of 6E10 in the TgCRND8 animal is consistent with the clearance profile observed after intravenous delivery of 6E10-PEG. It is possible that higher uptake of the tracer would be observed in TgCRND8 brain tissue, but the maximum concentration appears to happen before measurements are taken and limits the conclusions that can be drawn. These experiments demonstrate that intravenous delivery of 6E10-PEG can provide a non-invasive assessment of Aβ pathology in living animals.
These results advance the development of antibody-based contrast agents in Alzheimer's disease by demonstrating differentiation of mice bearing Aβ pathology from wild type littermate controls using a systemically delivered contrast agent and non-invasive live-animal medical imaging techniques. These results are consistent with the impressive binding to Aβ plaques using gadolinium labelled antibodies reported by Ramakrishnan et al 
. More recently, Koffie et al used an antibody-coated nanocarrier to label plaques, but also used histological techniques to quantify the signal 
. Koffie et al made an important advance by showing non-invasive detection of nanoparticle accumulation in wild type brain tissue. Herein, we build on that discovery by non-invasively detecting pathology without having to use histological techniques that were instrumental in previous studies.
Importantly, it is well-established that if 6E10 is able to gain access to the central nervous system it will bind to plaques. This was substantiated in our previous study where 6E10 labelled plaques were demonstrated in TgCRND8 mice after cortical injection 
. Increased accumulation of 6E10 in the TgCRND8 mice was observed with two delivery routes (intravenous and intracarotid) and two different formulations of 6E10 (native and PEG-modified). For these reasons, the increased accumulation of 6E10 in the brain of TgCRND8 mice is a reasonable proxy for Aβ pathology detection.
Conjugation of PEG to macromolecules is known to prolong circulation through reduced protein adsorption 
and reduced binding to receptors that may lead to clearance 
. Prolonged circulation results in greater serum concentrations that drive greater accumulation of an antibody in low permeability organs such as the brain. The higher concentrations of 6E10-PEG in brain tissue observed with PET in this study are likely partly driven through this process. Since it is unlikely that PEG would provide an enhanced protective effect in the TgCRND8 animals over wild type controls, enhanced accumulation of 6E10-PEG in the TgCRND8 brain is attributed to retention after binding to Aβ plaques.
In addition to the passive transport of 6E10-PEG driven through increased serum levels of antibody, PEG can also bind to LDL-receptors on the BBB and promote active transport of macromolecules. PEG is proposed to work through two pathways: selective adsorption of lipoproteins and direct binding to LDL-receptors 
. While LDL receptors are not upregulated in Alzheimer's disease 
, the enhanced accumulation of 6E10-PEG in the TgCRND8 mouse brain over wild type controls at early time points is likely due to greater binding to Aβ plaques.
A potential source of signal in the brain tissue of wild type and transgenic animals is antibody circulating in the brain vasculature that has not penetrated into the brain tissue. In order to quantify the contribution of the brain vasculature, we estimated the amount of blood in the brain tissue and compared this to the signal measured with PET imaging. The typical blood volume of a mouse is estimated at 1.5 mL while the typical mass of the mouse brain is approximately 400 mg with an estimated 0.01 mL of blood circulating in 1 gram of brain tissue 
. If 100% of the dose were in the blood, this would correspond to a measurement of 0.67% of the injected dose per gram of brain tissue, which is considerably lower than PET measurements made in all animals. Further, it is unlikely that 100% of the dose remains in circulation at any of the time points measured because antibodies are known to rapidly permeate into a variety of organs 
. Therefore, we are able to conclude that the signal measured is not due to the plasma pool.
Anti-Aβ antibodies have been widely studied as therapeutics to remove Aβ accumulation in the brain through passive or active immunization 
. Thus 6E10-PEG has the potential to be utilized as both a therapeutic and diagnostic agent. Antibodies can remove Aβ plaques from the brain through a variety of chemical and biological processes. While the effect of 6E10-PEG on the plaque load in the TgCRND8 mice was not quantified in this study, this is an interesting question for further examination as removing Aβ from the brain can alter brain vasculature and impact biocompatibility 
Copper-64 was chosen as a model radionuclide for this study primarily because its relatively long half-life of 12 h allowed the PET measurements to be made at this time point. In this study earlier time points (10 min and 2 h) were more useful in differentiating between wild type and transgenic animals and therefore demonstrate that an isotope with a shorter half-life could also be utilized. While the dosages of Copper-64 used in this study are already low, the selection of a different isotope could further improve biocompatibility.
This study provides a platform for further research with antibodies that have specificity to Aβ isoforms and are pathologically relevant, such as Aβ oligomers, or earlier stage Aβ plaques. In previous studies, 6E10 was shown to bind to plaques in 8 month old TgCRND8 mice 
, which have late-stage plaques yet unclear pathological relevance. The imaging methods described herein will allow relevant antibodies to be studied and thereby provide both a greater understanding of Aβ pathology and a diagnostic and monitoring tool.