Positron emission tomography imaging of fibrillar parenchymal and vascular amyloid-β in TgCRND8 mice

Abstract

Few quantitative diagnostic and monitoring, tools are available to clinicians treating patients with Alzheimer's disease. Further, many of the promising quantitative imaging tools under development lack clear specificity toward different types of Amyloid-β (Aβ) pathology such as vascular or oligomeric species. Antibodies offer an opportunity to image specific types of Aβ pathology because of their excellent specificity. In this study, we developed a method to translate a panel of anti-Aβ antibodies, which show excellent histological performance, into live animal imaging contrast agents. In the TgCRND8 mouse model of Alzheimer's disease, we tested two antibodies, M64 and M116, that target parenchyma aggregated Aβ plaques and one antibody, M31, that targets vascular Aβ. All three antibodies were administered intravenously after labeling with both poly(ethylene glycol) to enhance circulation and (64)Cu to allow detection via positron emission tomography (PET) imaging. We were clearly able to differentiate TgCRND8 mice from wild type controls by PET imaging using either M116, the anti-Aβ antibody targeting parenchymal Aβ or M31, the antivascular Aβ antibody. To confirm the validity of the noninvasive imaging of specific Aβ pathology, brains were examined after imaging and showed clear evidence of binding to Aβ plaques.

Figure 1

Anti-amyloid-β antibodies bind to cerebral and vascular deposits of Aβ. Representative immunohistochemical staining of 5 month old TgCRND8 brain tissue. (A) M116 binds in the cortex. (B) Thioflavin S positive plaques are detected in the TgCRND8 tissue. (C) Merged images demonstrate that M116 plaques are Thioflavin S positive indicating the deposits to which M116 are binding are composed of Aβ. (D) M64 positive deposits are observed in TgCRND8 brain tissue. (E) Thioflavin S deposits are observed to (F) colocalize with the M64 deposits, confirming these are composed of Aβ. (G) M31 bound to circular shapes (white arrow), indicative of vascular Aβ. (H) M31 deposits are also Thioflavin S positive, and the (I) merged M31/Thioflavin image confirms that the deposits contain Aβ. (J) An M31 deposit stained positive for (K) CD31 and the (L) colocalized image of M31 and CD31 demonstrates that this Aβ deposit is in the vasculature. (M) Tissue stained with M31 and anti-rabbit IgG-Alexa564 shows round, hollow deposits. (N) Tissue stained with anti-rat IgG-Alexa488 does not colocalize with M31 deposits confirming the specificity of CD31 to vasculature. There is no evidence that CD31 positive structures result from nonspecific binding of the secondary antibody. Images (A)–(L) were taken at 20× magnification, and the scale bar is 60 μm. Images (M) and (N) were taken at 10× magnification, and the scale bar is 200 μm.

Figure 2

M116, M64, and M31 bind to Aβ deposits in human brain tissue. Human AD brain tissue was taken from Broadmann’s area 11, and serial sections were examined with each antibody using immunohistochemistry with DAB staining. (A) Aβ 6E10 reveals extensive plaque deposits and diffuse background staining. (B) M31 binds hollow, branched structures indicative of vascular Aβ which is similar to binding observed in TgCRND8 mouse brain tissue. (C) M64 binds to numerous Aβ plaques, but the background is high which is consistent with results in TgCRND8 mouse brain tissue. (D) M116 binds strongly to Aβ plaques in the parenchyma with very low background. All images are taken at 40× magnification, and the scale bar is 200 μm.

Figure 3

Intracerebral injections of M116 and M64 label Aβ plaques in mouse brain tissue. Labeling of (A) M116 at 4 h after intracortical injection with an anti-rabbit-Alexa548 secondary antibody reveals deposits. (B) Similar labeling of M116 in wild type tissue reveals no deposits. (C) Dense deposits of M64 are observed near the injection site in TgCRND8 animals, while no deposits are found in wild type tissue (D). (E) Intracortical injections of M31 revealed no binding to deposits which is consistent with the vascular plaque labeling. This suggests that intracortical injection of antibodies does not permit the antibodies to cross the vasculature to label the plaques. (F) No labeling of plaques or vasculature was observed in wild type tissue. N = 3 for each antibody and animal type. All images are taken at 10× magnification, the scale bar is 200 μm, Tg refers to transgenic TgCRND8 tissue, and WT refers to wild type tissue.

Figure 4

Thioflavin S staining of TgCRND8 tissue that received intracortical injections confirms deposits are composed of Aβ. (A) M116 positive deposits are observed around the injection site. (B) Thioflavin S positive deposits are also observed near the injection site and the merged image (C) shows colocalization of signal confirming M116 deposits are composed of Aβ. (D) M64 positive deposits and (E) Thioflavin S positive deposits are observed around the injection site after intracortical injections. (F) The M64 and Thioflavin S signals show colocalization confirming deposits are composed of Ab. All images are taken at 10× magnification, the scale bar is 200 μm, and Tg refers to transgenic TgCRND8 tissue.

Figure 5

PEG- and ⁶⁴Cu-modified anti-amyloid-β antibodies M116 and M31 are able to distinguish between TgCRND8 and wild type mice using noninvasive PET imaging. Quantitation of PET images (% of injected dose per gram tissue) in the brain region 5 min, 2 h, and 4 h after intravenous injection of ⁶⁴Cu-labeled and PEG-modified: (A) M116 shows progressive accumulation of M116 in the TgCRND8 brain vs a lower, constant amount in the WT brain. (B) M64 shows no difference in accumulation between TgCRND8 and wild type mice at any time point. (C) M31 shows greater accumulated amount in TgCRND8 mice than WT, but at a constant amount. (D) Comparing the amounts of M116, M64, and M31 in the TgCRND8 brain relative to the wild type control shows linear increases of M116 while M64 and M31 remain constant of the time period examined. (E) Post-mortem quantitation of brain tissue at 4 h with gamma counting confirms PET quantitation data shown in (A), (B,C) Greater accumulation of M116 and M31 in TgCRND8 vs WT mice, yet no difference in M64. N = 3 for each antibody and animal type.

Figure 6

No differences in biodistribution (outside of the brain) were observed between transgenic and wild-type mice: M116 and M31 share similar biodistribution, yet M64 is different. (A) Blood concentrations of M64 are much lower than M116 and M31 in both TgCRND8 and wild type mice at 4 h, demonstrating faster clearance of M64. (B) Large intestine concentrations of M116 and M31 are lower than M64 at 4 h in both TgCRND8 and wild type animals, consistent with the blood concentrations observed. M64 is not significantly elevated in the liver (B) or kidney (C). Higher clearance of M64 appears to be responsible for the reduced blood concentrations. Mean ± standard deviation, n = 3.

Figure 7

Representative immunohistochemical staining after intravenous delivery of PEG-modified, ⁶⁴Cu-labeled M116, M64, and M31 confirms M116 and M31 labels Aβ deposits in brain tissue. (A) Staining of PEG-modified M116 with an anti-rabbit-Alexa546 secondary antibody found plaques in the brains of TgCRND8 animals after intravenous injection. (B) Thioflavin S positive plaques were found in the TgCRND8 tissue, and (C) some of the plaques colocalized with M116 deposits. (D) No deposits were found in the TgCRND8 tissue that received intravenous injections of PEG-modified and ⁶⁴Cu-labeled M64. (E) While this tissue did stain positive for Thioflavin S, (F) no colocalization was evident. (G) Staining of PEG-modified, ⁶⁴Cu-labeled M31with the anti-rabbit secondary antibody found deposits shaped like blood vessels. (H) Some Thioflavin S positive deposits were found and (I) these deposits localized with the M31 deposit. Tissue sections from wild type tissue were stained with an anti-rabbit IgG-Alexa-564 to detect (J) M116, (K) M64, or (L) M31 after intravenous delivery and PET imaging. No deposits were found in the wild type tissue with any of the three antibodies delivered. Images (A) through (I) were taken at 20× magnification, and the scale bars are 60 μm. Images (J) through (L) were taken at 10× magnification, and the scale bar is 200 μm. N = 3 for each antibody and animal type.