Abstract
Amyloid plaques are formed in the extracellular space of Alzheimer's disease (AD) brain due to the accumulation of amyloid β (Aβ) proteins such as Aβ40. The relationship between Aβ40 pharmacokinetics and its accumulation within and clearance from the brain in both wild-type (WT) and AD transgenic mice (APP,PS1) was studied to understand the mechanism of amyloid plaque formation and the potential use of Aβ40 as a probe to target and detect amyloid plaques. In both WT and APP,PS1 mice, the 125I-Aβ40 tracer exhibited biexponential disposition in plasma with very short first and second phase half-lives. The 125I-Aβ40 was significantly metabolized in the liver kidney > spleen. Coadministration of exogenous Aβ40 inhibited the plasma clearance and the uptake of 125I-Aβ40 at the blood-brain barrier (BBB) in WT animals but did not affect its elimination from the brain. The 125I-Aβ40 was shown to be metabolized within and effluxed from the brain parenchyma. The rate of efflux from APP,PS1 brain slices was substantially lower compared with WT brain slices. Since the Aβ40 receptor at the BBB can be easily saturated, the blood-to-brain transport of Aβ40 is less likely to be a primary contributor to the amyloid plaque formation in APP,PS1 mice. The decreased elimination of Aβ40 from the brain is most likely responsible for the amyloid plaque formation in the brain of APP,PS1 mice. Furthermore, inadequate targeting of Aβ40 to amyloid plaques, despite its high BBB permeability, is due to the saturability of Aβ40 transporter at the BBB and its metabolism and efflux from the brain.
Development of amyloid plaques in the extracellular space of the brain parenchyma is considered a primary event in the pathogenesis of Alzheimer's disease (AD) (Selkoe, 2001). Amyloid plaques consist predominantly of the amyloid β (Aβ) proteins Aβ40 and Aβ42, which are produced continuously by cells in the nervous system and peripheral tissues. The higher concentration of soluble Aβ accumulates over time in the brain extracellular space, polymerizes into insoluble fibrils, and eventually forms amyloid plaques (Craft et al., 2002; Cirrito et al., 2003). Studies in AD patients indicated increased levels of peripherally circulating Aβ (Kuo et al., 1999; Matsubara et al., 1999). DeMattos et al. (2002) suggested that Aβ in plasma and CSF exist in equilibrium, which is controlled by a novel, yet unknown mechanism that shifts toward the brain during plaque development. Zlokovic (2004) proposed that the Aβ equilibrium between plasma and CSF is regulated at the blood-brain barrier (BBB) by an influx receptor (receptor for advanced glycation end products; RAGE) and an efflux receptor (low-density lipoprotein receptor-related protein).
In recent years, substantial effort has focused on the development of a pre-mortem diagnosis of AD, which involves detection of the plaques using various imaging techniques such as magnetic resonance imaging (MRI) and positron emission tomography. MRI used in conjunction with a contrast agent can resolve individual plaques and has the capability of differentiating plaques from other interfering structures such as blood vessels, myelinated fibers, and intracranial structures (Poduslo et al., 2002). Of the MRI contrast agents that are currently being developed for imaging amyloid plaques, the most notable is the amyloid protein itself, mostly Aβ40 (Wengenack et al., 2000a; Lee et al., 2002; Poduslo et al., 2002, 2004; Wadghiri et al., 2003). 125I-Aβ40 was reported to have high binding affinity to the amyloid plaques in human and double transgenic AD mouse brain slices in vitro and high in vivo permeability at the BBB; however, the plaque targeting ability of 125I-Aβ40 after i.v. injection in AD transgenic mice was low (Wengenack et al., 2000a). Hence, methods such as modifying the protein so that it can be actively transported across the BBB have been used to increase the targeting of Aβ40 after i.v. injection (Wengenack et al., 2000a; Poduslo et al., 2002, 2004). The inadequate targeting of Aβ40 to amyloid plaques despite its high permeability at the BBB could be due in part to 1) rapid elimination of Aβ40 from the systemic circulation, which leads to a reduction in its concentration at the BBB; 2) inadequate transcytosis across the capillary endothelium; 3) competing high level of efflux from the brain parenchyma; or 4) metabolism in the brain or uptake by various cells, which can deplete Aβ40 concentrations from the extracellular space.
Some of these questions are addressed in the present study using mechanism-based pharmacokinetic experiments using a tracer (125I-Aβ40) in both wild-type (WT) and AD transgenic mice (APP,PS1). Our investigation also addresses questions regarding the relationship between endogenous Aβ40 and the kinetics of Aβ accumulation and clearance from the brain. Such information is not only helpful in elucidating the disease pathology but also in optimizing the delivery of diagnostic probes derived from Aβ (Poduslo et al., 2004).
Materials and Methods
Animals. The double transgenic mice were bred in our mouse colony at Mayo. Hemizygous transgenic mice (mouse strain C57B6/SJL; i.d. no. Tg2576) expressing mutant human amyloid precursor protein (APP695) (Hsiao et al., 1996) were mated with a second strain of hemizygous transgenic mice (mouse strain Swiss-Webster/B6D2; i.d. no. M146L6.2) expressing mutant human presenilin 1 (PS1) (Holcomb et al., 1998). These double transgenic mice have been shown to exhibit an accelerated phenotype with amyloid deposits and behavioral deficits by 12 weeks of age (Holcomb et al., 1998; Wengenack et al., 2000b). WT mice (B6/SJL) were obtained from The Jackson Laboratory (Bar Harbor, ME) at 6 to 8 weeks of age and were the same background strain as the transgenic mice. The animals were housed in a virus-free, indoor, light- and temperature-controlled barrier environment and were provided ad libitum access to food and water. All procedures with animals were in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Mayo Institutional Animal Care and Use Committee.
Synthesis of Aβ40. Human Aβ40 was synthesized by the Mayo Protein Core Facility (Rochester, MN) on an ABI 433A peptide synthesizer (Applied Biosystems, Foster City, CA), using standard solid phase methods and procedures. Each peptide was purified by reverse phase high-performance liquid chromatography on a Jupiter C18 column (250 × 21.2 mm, 15μ; Phenomenex, Torrance, CA) in 0.1% trifluoroacetic acid/water with a 50-min gradient from 10 to 70% acetonitrile/0.1% trifluoroacetic acid. The integrity of the protein was verified by electrospray ionization mass analysis on a PerkinElmer Sciex API 165 mass spectrometer (Applied Biosystems, Foster City, CA). Protein concentration was determined using a bicinchoninic acid protein assay reagent kit (Pierce Chemical, Rockford, IL) and bovine serum albumin (BSA) standard.
Radioiodination of Proteins. Carrier-free Na125I and Na131I were obtained from PerkinElmer Life and Analytical Sciences (Boston, MA). Human Aβ40 (500 μg) and BSA (500 μg) were labeled with 125I and 131I, respectively, using the chloramine-T procedure as described previously (Poduslo et al., 1994). Free radioactive iodine was separated from the radiolabeled protein by dialysis against 0.01 M phosphate-buffered saline at pH 7.4 (Sigma-Aldrich, St. Louis, MO). Purity of the radiolabeled proteins was determined by trichloroacetic acid (TCA) precipitation. The radiolabeled protein was determined to be acceptable if the TCA precipitable counts were greater than 95% of the total counts. The final radioactivity associated with 125I labeled Aβ40 was determined to be 4 mCi/mg protein.
Aβ40 Pharmacokinetic Studies. Before the beginning of experiment each mouse was weighed (WT = 18–21 g; APP,PS1 = 20–23 g), and the femoral vein and artery were catheterized under general anesthesia (isoflurane = 1.5% and oxygen = 4 l/min). The 125I-Aβ40 (100 μCi; 100 μl) was administered intravenously in the femoral vein. Blood was sampled (20 μl) from the femoral artery at various intervals. At the end of the experiment, an aliquot of 131I-BSA (100 μCi; 100 μl) was injected to serve as a measure of residual plasma volume (Vp). One minute after the 131I-BSA injection, the final blood sample was collected, and the animal was sacrificed. The blood samples, diluted to a volume of 100 μl using normal saline, were centrifuged, and the supernatant was obtained. After TCA precipitation, the samples were assayed for 125I and 131I radioactivity in a two-channel gamma counter (Cobra II; Amersham Biosciences Inc., Piscataway, NJ). The measured activity was corrected for the background and crossover of 131I activity into the 125I channel.
The plasma pharmacokinetics of 125I-Aβ40 was determined by collecting serial blood samples (20 μl) from the femoral artery over a period of 15 min at time points of 0.25, 1, 3, 5, 10, and 15 min. The accumulation of 125I-Aβ40 in the peripheral organs such as liver, kidney, and spleen was determined by perfusing the animals with PBS at the end of the experiment. The linearity of 125I-Aβ40 disposition was determined by repeating the experiment by coadministering 1 or 2 mg of cold Aβ40 with 100 μCi of 125I-Aβ40.
The brain uptake studies of 125I-Aβ40 were conducted by collecting serial blood samples (20 μl) from the femoral artery over a period of 15 min at time points 0.25, 1, 3, 5, 10, and 15 min. At the end of the experiment, the brain of the animal was removed from the cranial cavity; dissected into the anatomical regions, cortex, caudate putamen, hippocampus, thalamus, brain stem, and cerebellum; and assayed for 125I and 131I radioactivity. The brain regions were lyophilized, and dry weights were determined with a microbalance and converted to wet weights using wet weight/dry weight ratios determined previously. The saturability of 125I-Aβ40 transport at the BBB was determined by coadministering 0.5, 1, or 2 mg of cold Aβ40 with 100 μCi of 125I-Aβ40.
To determine the influence of high circulating levels of Aβ40 on the elimination of 125I-Aβ40 from the brain, 100 μCi of 125I-Aβ40 was administered to the animal intravenously followed by four i.v. bolus injections of 0.5 mg of cold Aβ40 at 15-(Tmax of 125I-Aβ40 in the brain), 30-, 45-, and 60-min intervals. Blood samples were collected over a period of 60 min at time points 0.25, 1, 3, 5, 10, 15, 30, 45, and 60 min. At the end of 90 min, 131I-BSA was administered; the animal was sacrificed a minute later to obtain the brain regions, which were assayed for 125I and 131I radioactivity as described above.
Metabolism of 125I-Aβ40. To determine the metabolism of 125I-Aβ40 in the plasma of WT or AD transgenic animals, 0.1 μCi of 125I-Aβ40 was added to 350 μl of plasma and incubated at 37°C. Aliquots from the mixture (20 μl) were taken at regular intervals up to 60 min, and the amount of intact 125I-Aβ40 was determined by TCA precipitation. The metabolism of 125I-Aβ40 in the presence of brain, liver, kidney, and spleen slices was determined by obtaining the organs from WT and AD transgenic mice after perfusion with PBS. The organs were weighed, cut into 1-mm-thick slices using a tissue slicer (Stoelting Co., Wood Dale, IL), and placed in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) prewarmed to 37°C. 125I-Aβ40 (0.1 μCi) was added to the medium containing tissue slices and maintained at 37°C under 5% CO2 for the entire length of the experiment. Aliquots (20 μl) of the medium were obtained at regular intervals and assayed for the intact protein using TCA precipitation method.
Efflux of 125I-Aβ40 from Brain Slices. WT and APP,PS1 transgenic mice were killed by decapitation under general anesthesia. The brains were rapidly removed, washed with PBS, and cut into 1-mm-thick cortical slices, containing hippocampus, using the tissue slicer. After equilibrating in oxygenated (95% O2/5% CO2) Krebs-Ringer bicarbonate buffer (KRB) for 15 min at 37°C, each slice was incubated in 1 ml of donor solution (0.6 μCi of 125I-Aβ40 in 1 ml of KRB) at 37°C for 30 min. The loaded brain slices were washed with KRB, and the efflux rate of 125I-Aβ40 from each brain slice was determined by incubating it in 5 ml of receiver medium (KRB or KRB + 1 mM 2,4-dinitrophenol) at 37°C. The receiver medium was replaced every 30 min to maintain sink conditions. One brain slice was sampled at each time interval and assayed for 125I radioactivity.
Data Analysis. The Aβ40 plasma concentration profile after a single i.v. bolus dose of 125I-Aβ40 was best described by a biexponential disposition function: where C(t) = 125I-Aβ40 microcuries per milliliter of plasma, A and B are the intercepts, and α and β are the slopes of the biexponential curve. Pharmacokinetic parameters were estimated by nonlinear curve fitting using Gauss-Newton (Levenberg and Hartley) algorithm and iterative reweighting (WinNonlin Professional, version 4.1; Pharsight, Mountain view, CA). Secondary parameters such as the maximum plasma concentration (Cmax), the first and second phase half-lives [t1/2(α) and t1/2(β), respectively], the plasma clearance (Cl), the steady-state volume of distribution (Vss), and area under the plasma concentration curve (AUC) were also calculated using WinNonlin. The mean values of controls and treatments were compared by Student's t test using GraphPad Prism version 3.03 (GraphPad Software Inc., San Diego, CA).
The residual brain region plasma volume (Vp, microliters per gram) and the cerebrovascular permeability-surface area product (PS) values were calculated as described previously by Poduslo (1993). where qp is the 131I-BSA content (cpm) of tissue, Cv is the 131I-BSA concentration (cpm per milliliter) in plasma, W is the dry weight (grams) of the brain region, and R is the wet weight/dry weight ratio for mice of a defined age group. From the total 125I-Aβ40 content (qT) (cpm) of the brain region, the amount of 125I-Aβ40 that enters the brain region extravascular space (q) (cpm per gram) is calculated as follows: where Ca is the final 125I-Aβ40 concentration (cpm per milliliter) in plasma. The PS (milliliters per gram per second) at the BBB is calculated as follows: where t is the circulation time, q(t) is the extravascular 125I activity in the brain region at time t, and ∫t0Cpdt is the plasma concentration time integral of 125I-Aβ40.
The rate of 125I-Aβ40 efflux (k) from the brain slices in vitro was determined by curve-fitting (one-phase exponential decay model) the amount of radioactivity retained in the brain slices versus time using GraphPad Prism version 3.03 as follows: where Yo is the amount of radioactivity (cpm) in the brain slice at 0 min, k is the decay rate constant, and t is the time in minutes. The adequacy of fit was determined by F-test.
The PS values of 125I-Aβ40 were plotted versus the log concentration (log C) of Aβ40, and the inhibitor concentration 50% (IC50) of Aβ40 in various brain regions were calculated by fitting one-site competitive binding equation to the data using GraphPad Prism version 3.03: where PSmax is the maximum PS value, which was obtained at the lowest Aβ40 concentration. PSmin is the minimum PS value obtained at the highest Aβ40 concentration.
Results
Recent reports have suggested that Aβ40 exhibits bidirectional transport between the peripheral circulation and brain via the BBB (Shibata et al., 2000; Deane et al., 2003). According to Zlokovic (2004), such a bidirectional transport regulates the Aβ40 equilibrium between central nervous system and peripheral circulation and contributes to the formation of amyloid plaques in the brain parenchyma. If peripherally circulating Aβ40 can so directly impact AD pathogenesis, then it becomes important to study the plasma pharmacokinetics of 125I-Aβ40 in WT and AD mice. Differences in peripheral distribution and elimination between the two strains could have a profound influence on plasma and brain steady-state Aβ concentrations, as well as on the time course of brain 125I-Aβ40 concentration after i.v. administration. The information obtained from such a comparative study could help optimize Aβ40 delivery (or its derivatives) as a diagnostic probe and elucidate the physiological parameters responsible for plaque formation in AD mice.
125I-Aβ40 Plasma Pharmacokinetics and Metabolism. After i.v. administration, the 125I-Aβ40 concentration in the plasma of WT as well as APP,PS1 mice declined rapidly exhibiting a biexponential disposition with short first (t1/2,α) and second phase (t1/2,β) half-lives (Fig. 1A; Table 1). The plasma pharmacokinetic profile of 125I-Aβ40 in 8-week-old APP,PS1 mice (no amyloid plaque formation), was significantly different from that of the 8-week-old WT mice. However, this difference was not statistically significant when the animals were 24 weeks old (Fig. 1B; Table 1), an age with substantial amyloid burden (Wengenack et al., 2000b). Although no significant differences in the plasma pharmacokinetic parameters were observed between 8- and 24-week-old WT mice, significantly lower clearance and higher AUC was observed in 24-week-old APP,PS1 mice compared with that of 8-week-old mice (Table 1). These differences are most likely due to the increased Aβ40 levels with age in APP,PS1 mice that could saturate peripheral elimination.
A substantial amount of 125I-Aβ40 was found in the liver, kidney, and spleen of both WT and APP,PS1 animals perfused with PBS at the termination of the experiment. The accumulation of 125I-Aβ40 was higher in the kidney than in the liver or spleen. Moreover, no significant differences in the accumulation of 125I-Aβ40 in these organs was observed between WT and APP,PS1 animals. To investigate the role of peripheral metabolism on the rapid elimination of 125I-Aβ40 from the systemic circulation, in vitro metabolism studies in the presence of plasma, liver, kidney, and spleen obtained from WT and APP,PS1 mice were conducted. Although the 125I-Aβ40 metabolism in the plasma was slightly higher than the degradation in Dulbecco's modified Eagle's medium at 37°C, less than 10% of the initial amount of 125I-Aβ40 was degraded in 60 min (Fig. 2). The metabolism of 125I-Aβ40 in the liver slices, however, was so rapid that Aβ40 was degraded substantially before the initial sample (t = 0 min) could even be obtained, and proceeded to near completion by 60 min (Fig. 2). The metabolism in the APP,PS1 mouse peripheral tissues was not significantly different from that of WT mouse tissues. However, the 125I-Aβ40 metabolism in the brain slices of APP,PS1 mice is significantly lower compared with that in wild-type brain slices (Fig. 2).
To determine whether the elimination of 125I-Aβ40 from the peripheral circulation is saturable, the plasma kinetics of 125I-Aβ40 was studied by coadministering various amounts of unlabeled Aβ40 (0.125–4 mg). When 0.125 mg of Aβ40 was coadministered with 100 μCi of 125I-Aβ40, no significant changes in the plasma pharmacokinetic parameters were observed (Table 2). Upon coadministration of 1, 2, or 4 mg of Aβ40, the pharmacokinetic parameters of clearance and Vss of 125I-Aβ40 decreased significantly, whereas the AUC increased significantly (Table 2).
125I-Aβ40 Brain Uptake. Upon i.v. administration in 24-week-old WT mice, 125I-Aβ40 showed rapid brain uptake (Tmax of ∼15 min) (Fig. 3). 125I-Aβ40 pharmacokinetic profile in the brain was substantially different from that of the plasma. Elimination of 125I-Aβ40 from the brain was not as rapid as it was from the plasma.
To verify whether the uptake of 125I-Aβ40 at the BBB is receptor-mediated, various amounts of unlabeled Aβ40 was coadministered along with 125I-Aβ40 in 24-week-old WT mice. When 28.8 nM (0.125 mg) of Aβ40 was coadministered, the decrease in the PS values of 125I-Aβ40 in all brain regions was statistically significant (p < 0.05) (Table 3). Further increments of unlabeled Aβ40 to 115.4 nM (0.5 mg), 231 (1 mg), 462 (2 mg), or 924 nM (4 mg) resulted in a nonlinear decrease in the PS value (Fig. 4A). The IC50 values of 125I-Aβ40 PS values in cortex, caudate putamen, hippocampus, thalamus, brain stem, and cerebellum were 73.0, 58.2, 108.9, 49.9, 81.2, and 78.7 nM, respectively (Fig. 4B). The Vp values, however, did not change significantly with the coadministration of unlabeled Aβ40 (Table 3).
The PS values of 125I-Aβ40 at the BBB in various brain regions of 8- and 24-week-old APP,PS1 transgenic mice were significantly lower than that in age matched WT mice (Table 4). There was no significant difference in the Vp values of 125I-Aβ40 in various brain regions except in hippocampus and brain stem (Table 4). These findings differ from that of our previous report where no differences in the PS values of 125I-Aβ40 between AD transgenic and normal mice were observed (Poduslo et al., 2001). In the previous study, we used 131I-Aβ40 as vascular space marker, but in this study 131I-BSA was used to estimate the Vp values of 125I-Aβ40 in both WT and AD transgenic mice. Using the same protein labeled with a different isotope helps correct for potential artifacts such as nonspecific adherence to vessel walls and allows for an accurate estimation of residual brain plasma volume (Vp). However, further studies conducted in our laboratory demonstrated that rapid elimination of proteins such as Aβ40 (unpublished data), whose plasma concentrations are reduced by ∼50% in 45 s, can lead to an underestimation of PS values. Moreover, differences in the circulating levels of endogenous Aβ40 in WT and AD transgenic mice further confounds the determination of these parameters (Poduslo et al., 2001). Another reason for the differences between the two studies is the manner in which the blood samples were processed. In the former study, the TCA precipitation to quantify the intact protein was conducted on whole blood, whereas in the current study, the blood was diluted and centrifuged to remove cells, and the TCA precipitation was conducted on the diluted plasma.
The APP,PS1 mice at the age of 6 months develop distinct plaques, mostly in the cortex and hippocampus (Wengenack et al., 2000). They also carry significantly higher levels of Aβ40 (13.9 pmol/ml) in the peripheral circulation compared with wild-type mice (1.07 pmol/ml) (Poduslo et al., 2001). If one of the primary elimination pathways of Aβ40 from the brain involves efflux into peripheral circulation across the BBB via low-density lipoprotein receptor-related protein 1 located on the abluminal surface (Shibata et al., 2000), the higher levels of Aβ40 in the peripheral circulation of the transgenic mice could reduce Aβ40 efflux by saturating the efflux transporter. To investigate this hypothesis, a pharmacokinetic study based on the absorption and elimination profile of 125I-Aβ40 in the extravascular brain tissue (Fig. 3) was designed. A bolus dose of 125I-Aβ40 was administered to WT mice intravenously starting at the Tmax of 125I-Aβ40 in the brain (∼15 min); four i.v. bolus doses of unlabeled Aβ40, each 0.5 mg, were injected at 15-min intervals up to 60 min; and the brain levels of 125I-Aβ40 were then measured at the end of 90 min. Control experiments were performed on a similar set of three WT animals by injecting saline instead of unlabeled Aβ40. The assumptions behind this experiment were as follows: 1) most of the uptake of 125I-Aβ40 into endothelial cells or brain parenchyma takes place during the absorption phase (0–15 min); and 2) if the efflux of the absorbed 125I-Aβ40 from brain is inhibited by higher concentrations of Aβ40 (resulting from multiple bolus doses) on the luminal side, higher amounts of 125I-Aβ40 remain in the brain compared with the control.
The amount of 125I-Aβ40 in various brain regions of mice injected with unlabeled Aβ40 (treatment) was not significantly different from the amount present in the brain regions of mice injected with saline (control) (Table 5). No significant differences were found in the Vp values as well. These results suggest that the higher concentration of Aβ40 in the peripheral circulation does not significantly influence the brain efflux of 125I-Aβ40.
Efflux of 125I-Aβ40 from Brain Slices. In the past, investigators have demonstrated that intracerebrally injected 125I-Aβ40 was effluxed into CSF and blood (Ghersi-Egea et al., 1996; Shibata et al., 2000). Although the mechanism of efflux is still not clear, it is reasonable to assume that the transport across the brain parenchyma plays a significant role in the efflux of 125I-Aβ40 from brain. Hence, the efflux rate of 125I-Aβ40 from the brain slices of WT and APP,PS1 animals was determined in vitro. The efflux of 125I-Aβ40 from the brain slices of WT (in the presence and absence of 2,4-DNP) and APP,PS1 transgenic animals (24 weeks) was significantly different from each other (Fig. 5) (F-test, p < 0.0001). The efflux rate of 125I-Aβ40 from the WT mouse brain slices was 0.06 ± 0.01 min-1 and was significantly higher than that of brain slices obtained from APP,PS1 transgenic animals (0.02 ± 0.009 min-1) (mean ± S.E.M.). In the presence of a metabolic inhibitor like 2,4-DNP, which interferes with the energy metabolism, the efflux of 125I-Aβ40 from the WT mouse brain slices was reduced to 0.02 ± 0.003 min-1 (mean ± S.E.M.).
Discussion
The objective of this work was to elucidate the role of absorption, distribution, metabolism, and elimination characteristics of Aβ40 in the plasma and brain of APP,PS1 mice. This objective was accomplished with the resulting conclusions: 1) accumulation of Aβ40 in the systemic circulation of APP,PS1 mice is not due to impaired plasma clearance; 2) neither Aβ40 influx nor efflux across the BBB plays a pivotal role in the accumulation of Aβ40 in the brain; and 3) accumulation of Aβ40 in the brains of APP,PS1 animals is due to ineffective efflux in the brain parenchyma and/or reduced metabolism.
Our experiments demonstrated no major differences in the plasma pharmacokinetics or peripheral metabolism of 125I-Aβ40 between WT and APP,PS1 animals. The high level of circulating Aβ40 in APP,PS1 animals could saturate the transporter at the BBB and hence contribute insignificantly to the amyloid plaque formation in the brain. Also, these kinetic experiments provided no direct evidence for the existence of Aβ40 efflux transport at the BBB. On the other hand, 125I-Aβ40 efflux studies conducted in the brain slices, used as an vitro model for brain parenchymal transport demonstrated substantially higher efflux rate from WT brain slices compared with that of APP,PS1 brain slices. Moreover, the rate of 125I-Aβ40 degradation in the presence of APP,PS1 brain slices is substantially lower compared with that of WT brain slices. The conclusions from our experimental studies that lead to these key findings are further elaborated below.
After i.v. administration, our studies have demonstrated that the tracer 125I-Aβ40 was eliminated rapidly from the plasma. Such rapid clearance of Aβ40 from the systemic circulation could adversely affect its permeability at the BBB and limit its utility as a diagnostic probe. The distribution and elimination half-lives of 125I-Aβ40 in WT mice are lower than those reported in nonhuman primate model of cerebral β-amyloidosis (Mackic et al., 2002). Similar results are expected with AD transgenic mice used in this study, because they have higher plasma Aβ40 levels than the WT animals (Poduslo et al., 2001). The higher plasma levels of Aβ40 could saturate the receptors and/or enzymes involved with the distribution and metabolism of Aβ40, resulting in higher distribution and elimination half-lives. Surprisingly, 125I-Aβ40 clearance and AUC were significantly higher in APP,PS1 mice (8 weeks) compared with that in 8-week-old WT mice. However, at 24 weeks of age, when the peripherally circulating Aβ40 in APP,PS1 mice is 12 times higher than in WT mice (Poduslo et al., 2001), the 125I-Aβ40 plasma profiles were similar.
To investigate the effect of higher endogenous plasma Aβ40 concentrations on the elimination of Aβ40 from the plasma, 125I-Aβ40 was coadministered with various amounts of unlabeled Aβ40 into the systemic circulation of WT mice. In the presence of high plasma concentrations of Aβ40, the clearance and Vss of 125I-Aβ40 decreased significantly, which suggest that 125I-Aβ40 exhibits nonlinear disposition in the systemic circulation. In vitro metabolism studies conducted using WT and APP,PS1 mouse tissue slices indicated that 125I-Aβ40 is substantially metabolized in liver and kidney. However, no significant difference in the extent of metabolism between WT and APP,PS1 animals was observed. These results demonstrate that the lower clearance of Aβ40 in 24-week-old APP,PS1 mice compared with that in 8-week-old animals is due to the higher circulating Aβ40 levels saturating the peripheral elimination. Despite the absence of any differences in the metabolism of Aβ40 in the peripheral tissues such as plasma, liver, kidney, and spleen of APP,PS1 and WT mice, the increase in the plasma levels of Aβ40 with age in APP,PS1 mice indicates that Aβ40 is replenished at a rate faster than it is removed from the peripheral circulation. Excess Aβ40 in the peripheral circulation could be due to overproduction by the peripheral tissues and/or probable continual efflux from the brain.
For Aβ40 in the peripheral circulation to directly contribute to the plaque formation in the brain parenchyma, it has to be transported across the BBB. Several investigators have reported previously that Aβ40 is actively transported at the BBB (Zlokovic et al., 1993; Poduslo et al., 1997, 1999). Brain pharmacokinetic profile obtained in the present study demonstrated that 125I-Aβ40 is absorbed at the cerebral vasculature very rapidly without any lag time, which is expected if the transport is receptor mediated. A decrease in the PS value of 125I-Aβ40 in the presence of various amounts of unlabeled Aβ40 provides further evidence that 125I-Aβ40 exhibits receptor-mediated transport at the BBB.
Deane et al. (2003) claimed that RAGE mediates Aβ transport across the BBB and accumulation in brain. They reported that RAGE is up-regulated in cerebral vasculature of patients with Alzheimer's disease and in AD mouse model and hypothesized that inhibition of RAGE at the BBB may limit accumulation of Aβ in the brain. If the Aβ40 transport across the BBB is mediated primarily by RAGE, we would expect to see higher 125I-Aβ40 PS values in APP,PS1 mice compared with WT mice consistent with the RAGE up-regulation. On the contrary, the current study demonstrates that the PS value of 125I-Aβ40 is significantly lower in APP,PS1 mice compared with WT mice most likely due to the saturation of uptake receptors at the BBB with increasing levels of Aβ40 in the plasma.
Although researchers in the past have reported that 125I-Aβ40 have reasonable permeability at the BBB in APP,PS1 mice and nonhuman primates (Poduslo et al., 1999; Mackic et al., 2002), our laboratory and others have realized that the transport of Aβ40 into brain parenchyma is poor (Wengenack et al., 2000a; Lee et al., 2002; Poduslo et al., 2002; Wadghiri et al., 2003). The PS value represents the rate at which a protein is transferred from the blood to the endothelial cell, but it offers no information on the amount of protein delivered to the brain parenchyma. Hence, care must be taken not to overinterpret this parameter. Wengenack et al. (2000a) reported that no detectable signal due to 125I-Aβ40 was observed on the plaques when 250 μg of 125I-Aβ40 was administered intravenously. Based on the pharmacokinetic parameters determined in this study, such a dose can produce plasma concentrations, at least 50 times greater than the physiological concentrations usually observed in APP,PS1 mice. On a similar note, Selkoe and colleagues (Craft et al., 2002), based on mathematical simulations, have reported that a 100-fold reduction in plasma to brain transport of Aβ can only reduce the brain Aβ burden by less than 0.2%. These studies indicate, therefore, that plasma-to-brain transport of Aβ is not an important factor for plaque generation in the brain parenchyma of APP,PS1 transgenic mice. One may hypothesize, however, that very small amounts of Aβ40 transported across the BBB over a long period can accumulate in high enough quantities to form plaques. This is kinetically feasible if the clearance mechanism of Aβ40 from the brains of AD mice is impaired.
The clearance of Aβ40 from the brain is known to be mediated by enzymes such as insulin degrading enzyme that metabolize the protein (Qiu et al., 1998) and the receptors (Shibata et al., 2000; Lam et al., 2001) that are involved in the efflux of the protein from central nervous system to plasma. Our in vitro efflux studies conducted in brain slices of 24-week-old normal and APP,PS1 transgenic mice demonstrated a significantly higher rate of 125I-Aβ40 efflux in the normal mice brain slices compared with the AD mice. The efflux rate decreased in the presence of 2,4-DNP (metabolic inhibitor), suggesting that 125I-Aβ40 efflux across the brain slices of normal mice is carrier-mediated. The lower rate of 125I-Aβ40 efflux across the brain slices of APP,PS1 mice could be due to 1) binding of 125I-Aβ40 to amyloid plaques, 2) impaired metabolism of 125I-Aβ40, or 3) inefficient efflux transport. More experiments are needed to elucidate the extent each of these factors contributes to the elimination kinetics of Aβ40 from the brain.
It has been reported that Aβ40 concentration in peripheral circulation and brain exist in dynamic equilibrium, with the Aβ present in the peripheral circulation being transported to brain parenchyma to compensate for the reduction in CSF Aβ levels due to plaque formation (DeMattos et al., 2002). Similarly, Aβ from brain is transported to peripheral circulation if plasma Aβ levels are depleted due to the presence of antibodies that could sequester Aβ (DeMattos et al., 2002). Based on these studies, higher plasma concentrations of Aβ40 in APP,PS1 animals could reduce the clearance of Aβ40 from the brain. However, our experiments demonstrated that the clearance of 125I-Aβ40 from the brain is not affected by the high levels of peripheral Aβ40, suggesting once again that the influx of Aβ40 at the luminal surface of the cerebrovascular endothelium is not significant enough to impact the capacity of the efflux transporter supposedly located on the abluminal surface.
Our studies demonstrate that the higher level of Aβ40 in the peripheral circulation in APP,PS1 mice compared with WT mice is not due to differences in the plasma, hepatic, or renal metabolism of the protein. The higher plasma concentration of exogenously added Aβ40 in WT animals inhibited the uptake of 125I-Aβ40 at the BBB, but it did not affect its elimination from the brain. The decreased permeability of 125I-Aβ40 at the BBB in APP,PS1 animals compared with WT animals therefore is a direct result of higher Aβ40 levels in the plasma. Since the Aβ40 receptor at the BBB can be easily saturated, the blood-to-brain transport of Aβ40 is less likely to be a primary contributor to the amyloid plaque formation in older (6-month) APP,PS1 animals, as claimed by other investigators. The 125I-Aβ40 was shown to be metabolized and effluxed in the brain parenchyma. The rate of 125I-Aβ40 efflux in APP,PS1 brain slices was substantially lower compared with WT brain slices. Also, the metabolism of 125I-Aβ40 was significantly lower in APP,PS1 brain slices compared with WT brain slices. Therefore, the decreased efflux of Aβ40 from the brain and its possible decreased metabolism are reasonable explanations for the Aβ accumulation and subsequent amyloid plaque formation in the brain of APP,PS1 transgenic mice. This study also demonstrates that inadequate targeting of Aβ40 to amyloid plaques despite its high BBB permeability is due to the saturability of Aβ40 transporter at the BBB. This saturability coupled with metabolism and efflux of Aβ40 in the brain parenchyma significantly affects the plaque targeting of Aβ40. The knowledge gained from these studies is being used in the development of new Aβ derivatives with improved BBB permeability and plaque targeting.
Acknowledgments
We thank Dr. Karen Duff for the PS1 transgenic mouse line and Dawn Gregor for excellent technical assistance.
Footnotes
-
This work is supported by National Institute on Aging Grant AG22034 (to J.F.P.) and the Mayo Foundation.
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.104.081901.
-
ABBREVIATIONS: AD, Alzheimer's disease; Aβ, amyloid β; CSF, cerebrospinal fluid; BBB, blood-brain barrier; RAGE, receptor for advanced glycation end products; MRI, magnetic resonance imaging; APP, amyloid precursor protein; PS1, presenilin 1; WT, wild-type; BSA, bovine serum albumin; TCA, trichloroacetic acid; PBS, phosphate-buffered saline; KRB, Krebs-Ringer bicarbonate; AUC, area under the plasma concentration curve; PS, cerebrovascular permeability coefficient-surface area product; 2,4-DNP, 2,4-dinitrophenol.
- Received December 8, 2004.
- Accepted February 25, 2005.
- The American Society for Pharmacology and Experimental Therapeutics