Clearance of Alzheimer’s amyloid-β_1-40 peptide from brain by LDL receptor–related protein-1 at the blood-brain barrier

Elimination of amyloid-β peptide (Aβ) from the brain is poorly understood. After intracerebral microinjections in young mice, ¹²⁵I-Aβ_1-40 was rapidly removed from the brain (t_1/2 ≤ 25 minutes), mainly by vascular transport across the blood-brain barrier (BBB). The efflux transport system for Aβ_1-40 at the BBB was half saturated at 15.3 nM, and the maximal transport capacity was reached between 70 nM and 100 nM. Aβ_1-40 clearance was substantially inhibited by the receptor-associated protein, and by antibodies against LDL receptor–related protein-1 (LRP-1) and α₂-macroglobulin (α₂M). As compared to adult wild-type mice, clearance was significantly reduced in young and old apolipoprotein E (apoE) knockout mice, and in old wild-type mice. There was no evidence that Aβ was metabolized in brain interstitial fluid and degraded to smaller peptide fragments and amino acids before its transport across the BBB into the circulation. LRP-1, although abundant in brain microvessels in young mice, was downregulated in older animals, and this downregulation correlated with regional Aβ accumulation in brains of Alzheimer’s disease (AD) patients. We conclude that the BBB removes Aβ from the brain largely via age-dependent, LRP-1–mediated transport that is influenced by α₂M and/or apoE, and may be impaired in AD.

Deposition of amyloid-β peptide (Aβ) in the brain occurs during normal aging and is accelerated in patients with Alzheimer’s disease (AD). Aβ is central to the pathology of AD, and is the main constituent of brain parenchymal and vascular amyloid (1–6). Aβ extracted from senile plaques contains mainly Aβ_1-40 and Aβ_1-42 (7), whereas vascular amyloid is predominantly Aβ_1-39 and Aβ_1-40 (8). Several sequences of Aβ were found in both lesions (9–11). A major soluble form of Aβ, which is present in the blood, cerebrospinal fluid (CSF) (12–14), and brain (15–16) is Aβ_1-40. In the circulation, CSF, and brain interstitial fluid (ISF), soluble Aβ may exist as a free peptide or be associated with different transport binding proteins such as apolipoproteins J (apoJ) (17–18) and E (apoE) (19), transthyretin (20), lipoproteins (21), albumin (22), and α₂-macroglobulin (α₂M) (23).

The neuronal theory argues that soluble brain-derived Aβ is a precursor of Aβ deposits. Neuronal cells secrete Aβ in culture (24), which supports this view. An increase in soluble Aβ in AD and Down syndrome brains precedes amyloid plaque formation (15, 25, 26) and correlates with the development of vascular pathology (27). Several cytosolic proteases that may degrade intracellular Aβ in vitro cannot degrade extracellular Aβ from brain ISF (28) or CSF (29) in vivo. An exception to this is enkephalinase, which may degrade Aβ_1-42 from brain ISF (28). However, the physiological importance of this degradation in vivo remains unclear, because the peptide was studied at extremely high pharmacological concentrations (30).

It has been suggested that decreased clearance of Aβ from brain and CSF is the main cause of Aβ accumulation in sporadic AD (31). Because Aβ is continuously produced in the brain, we hypothesized that an efficient clearance mechanism or mechanisms must exist at the blood-brain barrier (BBB) to prevent its accumulation and subsequent aggregation in the brain. Cell-surface receptors such as the receptor for advanced glycation end products (RAGE) (32–33), scavenger receptor type A (SR-A) (34), LDL receptor–related protein-1 (LRP-1) (35–38), and LRP-2 (39) bind Aβ at low (nanomolar) concentrations as free peptide (RAGE, SR-A) and/or in complex with α₂M, apoE, or apoJ (LRP-1, LRP-2). RAGE and SR-A regulate brain endothelial endocytosis and transcytosis of Aβ that is initiated at the luminal side of the BBB (33), whereas LRP-2 mediates BBB transport of plasma Aβ complexed to apoJ (39). The role of vascular receptors and BBB transport in the removal of brain-derived Aβ is unknown.

In this study, we developed a brain tissue clearance technique in mice based on a model used previously in the rabbit (40). This technique was used to determine in vivo the efflux rates of Aβ_1-40 from the CNS as a function of time and concentration of peptide, and to characterize vascular transport and/or any receptor-mediated efflux mechanisms involved in elimination of brain-derived Aβ across the BBB. The study focused on LRP-1 and its ligands, α₂M and apoE, first because they promote Aβ clearance in smooth muscle cells (35), neurons (36, 38), and fibroblasts (37); and second, because apoE4 is a definite risk factor, and α₂M is a possible risk factor for AD (41–42).

Figure 1a illustrates brain radioactivity-disappearance curves of [¹⁴C]inulin and ¹²⁵I-Aβ_1-40 (TCA-precipitable ¹²⁵I radioactivity) studied at a concentration of 60 nM. Clearance of inulin, a reference extracellular fluid (ECF) marker that is neither transported across the BBB nor retained by the brain (40, 43), approximated a single exponential decay, as expected from previous studies. The clearance curve reflecting total efflux of ¹²⁵I-Aβ_1-40 from brain was bi-exponential, and was much lower than that for inulin, indicating significant biological transport of Aβ_1-40 out of the brain. The two components of Aβ_1-40 efflux, rapid elimination by vascular transport across the BBB into the blood and slow elimination via the ISF flow, were computed from Figure 1a with equations 3 and 4 and are illustrated in Figure 1b. Figure 1b indicates significantly higher clearance of Aβ via BBB transport than via ISF bulk flow.

Figure 1

(a) Time-disappearance curves of [¹⁴C]inulin (open circles) and ¹²⁵I-Aβ_1-40 (60 nM; TCA-precipitable ¹²⁵I radioactivity, filled circles) from the CNS after simultaneous microinjections of tracers into the caudate nucleus in mice. Each point represents the mean ± SD of three to seven animals. (b) Two components of ¹²⁵I-Aβ_1-40 efflux, vascular transport across the BBB (filled triangles) and transport via ISF bulk flow (open triangles), were computed with equations 3 and 4 using data from a. (c) Relative contributions to Aβ_1-40 efflux by its transport across the BBB (open bar), diffusion via ISF bulk flow (filled bar), and retention (gray bar) in the brain were studied at 60 nM concentrations and calculated from the fractional coefficients given in Table 1.

The half-time (t_1/2) for brain efflux of Aβ_1-40 and inulin calculated from Figure 1a and equations 2 and 3 were 25.5 ± 2.0 minutes and 239.0 ± 12.5 minutes, respectively, a 9.4-fold difference (Table 1). The half-time of efflux of Aβ_1-40 across the BBB was 34.6 ± 3.6 minutes, 6.9-fold faster than by ISF bulk flow. In addition to efflux, there was also a slow, time-dependent retention of Aβ_1-40 in brain parenchyma with a t_1/2 of 164.5 minutes. As shown in Table 1, the rate k (min^-1) of clearance of Aβ_1-40 from the brain was 7.9-fold higher than that for inulin. The relative contributions of Aβ_1-40 efflux at 60 nM, by transport across the BBB and by ISF bulk flow based on 5-hour measurements, were 73.8% and 10.7% respectively, whereas 15.6% of the dose remained sequestered within the CNS (Figure 1c).

Table 1

Clearance rates (k) for ¹²⁵I-Aβ_1-40 and [¹⁴C]inulin

After CNS injection, both tracers reached the CSF; the CSF time-appearance curves are shown in Figure 2a. The amount of ¹²⁵I-Aβ_1-40 (TCA-precipitable ¹²⁵I radioactivity) measured in the CSF was lower than that for inulin at each studied timepoint, possibly reflecting active clearance of Aβ_1-40 from the CSF, as was suggested previously (28). It is noteworthy that at each studied timepoint, the ¹²⁵I-labeled material in the CSF was greater than 96% TCA-precipitable, indicating no degradation of the peptide. Both tracers also appeared in plasma (Figure 2b), and higher levels of TCA-precipitable ¹²⁵I-Aβ_1-40 radioactivity than of [¹⁴C]inulin radioactivity were consistent with active transport of Aβ_1-40 out of the CNS across the BBB. The absolute amounts of both tracers in the CSF and plasma were low, however, due to relatively rapid clearance from the CSF in comparison to slow ISF bulk flow (29), and significant systemic body clearance (48), respectively.

Figure 2

Time-appearance curves of [¹⁴C]inulin (open circles) and ¹²⁵I-Aβ_1-40 (60 nM; TCA-precipitable ¹²⁵I radioactivity, filled circles) in the CSF (a) and plasma (b) after simultaneous microinjections of tracers into the caudate nucleus in mice. Values are expressed as percentages of injected dose (%ID); each point is mean ± SD of three to seven animals.

Figure 3 (a and b) illustrates that ¹²⁵I-Aβ_1-40 was not significantly degraded in brain ISF before its transport across the BBB, as determined by TCA, HPLC, and SDS-PAGE analysis of ¹²⁵I radioactivity in brains. The TCA analysis suggests that only 4.2–9.9% of ¹²⁵I radioactivity in brains was not TCA-precipitable at different timepoints within 270 minutes of intracerebral microinjection of ¹²⁵I-Aβ_1-40 (Figure 3a). The HPLC analysis of brain radioactivity confirmed the TCA results by indicating that 93.7% of the peptide remains intact in brain ISF at 60 minutes (Figure 3b, right). It is noteworthy that ¹²⁵I-Aβ_1-40 was more than 97% intact at the time of injection, as determined both by the HPLC and TCA analyses. The results were corroborated by SDS-PAGE analysis of lyophilized aliquots of HPLC peaks of brain homogenates at different timepoints after ¹²⁵I-Aβ_1-40 injection, showing a single radioactive band at about 4 kDa (Figure 3b, left). The identity of the radioactive components on gels as Aβ_1-40 peptide was confirmed by Western blot analysis using anti-Aβ Ab and enhanced chemiluminescence as a detection system (not shown). More than 96% of ¹²⁵I radioactivity in the CSF was TCA-precipitable at studied timepoints between 15 minutes and 270 minutes (not shown). In contrast, degradation products of ¹²⁵I-Aβ_1-40 were found in plasma (Figure 3c); the amount of degraded ¹²⁵I-Aβ_1-40 corresponding to non–TCA-precipitable ¹²⁵I radioactivity increased from 37.6% to 58.3% during the period from 15 minutes to 120 minutes after intracerebral microinjection of intact ¹²⁵I-Aβ_1-40 (Figure 3c). It is noteworthy that the amount of radioactivity in plasma after 120 minutes was relatively small, and approached the limits of sensitivity of the TCA assay.

Figure 3

(a) Brain TCA-precipitable (open bars) and non–TCA-precipitable ¹²⁵I radioactivity (solid bars) after intracerebral microinjections of ¹²⁵I-Aβ_1-40 (60 nM) into the caudate nucleus in mice, expressed as a percentage of total ¹²⁵I radioactivity in the brain; mean ± SD of three to five animals. (b) Left panel shows HPLC elution profile of brain tissue 60 minutes after intracerebral microinjection of ¹²⁵I-Aβ_1-40 (60 nM). Separation was performed for 30 mg of brain tissue on a reverse-phase HPLC column, using a 30-minute linear gradient of 25–83% acetonitrile in 0.1% TFA, pH 2. ¹²⁵I-Aβ_1-40 eluted at 52%, corresponding to the elution time of Aβ_1-40 standard. Right panel shows SDS-PAGE analysis of brain tissue supernatant at 30 minutes (lane 1) and 60 minutes (lane 2) after intracerebral microinjection of ¹²⁵I-Aβ_1-40 (60 nM). The radioactivity in the brain eluted as a single peak on HPLC, with the same retention time as the Aβ_1-40 standard (data not shown). Aliquots of lyophilized sample were subjected to 10% Tris-tricine SDS-PAGE, transferred to a nitrocellulose membrane, and exposed to x-ray film. (c) Plasma TCA-precipitable (open bars) and non–TCA-precipitable ¹²⁵I radioactivity (filled bars) after intracerebral microinjections of ¹²⁵I-Aβ_1-40 (60 nM) into the caudate nucleus in mice, expressed as a percentage of total ¹²⁵I radioactivity in plasma; mean ± SD of three to five animals.

Clearance of Aβ in young mice was concentration dependent (Figure 4a). The efflux transport system was half saturated (K_1/2) at 15.3 nM of Aβ_1-40. The plateau or maximal clearance capacity was reached between 70 nM and 100 nM, and further increases in Aβ concentration resulted in progressively greater retention of the peptide in the brain. In contrast, clearance of [¹⁴C]inulin did not change with increasing concentrations of Aβ, suggesting a physiologically intact BBB (Figure 4a).

Figure 4

(a) Concentration-dependent clearance of Aβ_1-40 from mouse brain. Clearance via BBB transport (filled circles) is shown separately from clearance via ISF bulk flow (open circles). Clearance was determined 30 minutes after simultaneous microinjection of ¹²⁵I-Aβ_1-40 at increasing concentrations (0.05–120 nM) along with [¹⁴C]inulin into the caudate nucleus. (b) Effects of anti–LRP-1 Ab R777 (60 μg/ml), RAP (0.2 and 5 μM), anti-α₂M Ab (20 μg/ml), and anti–LRP-2 Ab Rb6286 (60 μg/ml) on brain clearance of ¹²⁵I-Aβ_1-40 at 12 nM, determined 30 minutes after simultaneous microinjection of ¹²⁵I-Aβ_1-40 and [¹⁴C]inulin. (c) Effects of anti–LRP-1 Ab R777 (60 μg/ml), anti-RAGE Ab (60 μg/ml), fucoidin (100 μg/ml), and 2-amino-bicyclo[2.2.1]heptane-2-carboxylic acid (BCH; 10 mM) on brain clearance of ¹²⁵I-Aβ_1-40 at a higher load of 60 nM, determined 30 minutes after simultaneous microinjection of ¹²⁵I-Aβ_1-40 and [¹⁴C]inulin. Mean ± SD of three to four animals. ^AP < 0.05; NS, not significant.

The next set of experiments was designed to characterize the BBB transport system responsible for the transcytosis of Aβ. Brains were loaded with ¹²⁵I-Aβ_1-40 at either 12 nM (Figure 4b) or 60 nM (Figure 4c), and clearance was determined at 30 minutes in the absence and presence of several molecular reagents that may act as potential inhibitors of and/or competitors in export. Figure 4b indicates that both the LRP-1 Ab (60 μg/ml) and RAP (200 nM) produced significant (58% and 30%, respectively) reductions in Aβ clearance from the brain compared with vehicle-treated controls; increasing the concentration of RAP to 5 μM decreased Aβ clearance by 44%. A significant (25%) inhibition in Aβ clearance was also obtained in the presence of anti-α₂M Ab (20 μg/ml). In contrast, anti–LRP-2 Ab (Figure 4b) and anti-RAGE Ab (Figure 4c) did not affect Aβ clearance. Fucoidin, a specific ligand for SR-A, produced a modest increase in clearance, possibly by blocking the binding of Aβ to parenchymal SR-A receptors, thereby allowing more peptide to be available for clearance. At higher Aβ loads (Figure 4c), anti–LRP-1 Ab produced a 53% decrease in clearance, similar to that observed at a lower load (Figure 4b), but Aβ recovery approached that of [¹⁴C]inulin, suggesting drainage of the peptide almost exclusively via ISF bulk flow. Clearance of [¹⁴C]inulin was not affected by any of the studied molecular reagents. We also demonstrated that BCH, a substrate that specifically blocks the L system for amino acids, does not affect clearance of Aβ across the BBB. This excludes the possibility that ¹²⁵I-Aβ_1-40 is degraded to ¹²⁵I-tyrosine that is then transported out of the CNS instead of ¹²⁵I-Aβ_1-40.

Next, we studied the effect of apoE and aging by determining Aβ clearance in 2-month-old and 9-month-old apoE KO mice and wild-type mice, using two different loads of ¹²⁵I-Aβ_1-40, 12 nM (Figure 5a) and 60 nM (Figure 5b). Figure 5a shows that the clearance of Aβ was reduced by 30% in young apoE KO mice, and by about 55% and 40% in 9-month-old wild-type and apoE KO mice, respectively. These results were confirmed at a higher load of Aβ; the observed decrease in clearance was 46% in 9-month-old apoE KO mice (Figure 5b).

Figure 5

(a) Effect of apoE genotype and age on brain clearance of ¹²⁵I-Aβ_1-40. Brain clearance of ¹²⁵I-Aβ_1-40 in 2-month-old and 9-month-old wild-type mice and apoE KO mice studied at the lower load of 12 nM ¹²⁵I-Aβ_1-40 (a) and a higher load of 60 nM (b). In all studies, ¹²⁵I-Aβ_1-40 and [¹⁴C]inulin were injected simultaneously, and clearance was determined after 30 minutes. Mean ± SD of three to four animals. ^AP < 0.05; NS, not significant compared with 2-month-old wild-type mice.

Immunocytochemical studies confirmed abundant expression of LRP-1 in brain microvessels (including capillaries, small venules, and arterioles) in 2-month-old mice (Figure 6a, upper panel; and Figure 6b, upper panel), in addition to significant parenchymal cellular (including neuronal) staining (Figure 6a, upper panel). As shown in the lower panels of Figures 6a and 6b, there was a significant reduction in LRP-1–positive vessels in 9-month-old mice compared with 2-month-old mice; the number of LRP-1–positive vessels dropped from 94% in 2-month-old mice to 52% in 9-month-old mice (Figure 6d, upper panel). Quantitative analysis of LRP-1–positive parenchymal cells (excluding blood vessels) showed a trend toward reduced staining in older animals, though the difference was not statistically significant (Figure 6d, lower panel). Similarly, there was no difference between young and old mice in the number of α₂M-positive microvessels or parenchymal cells in the brain (Figure 6c, upper and lower panels, respectively; Figure 6d, lower panel). It is noteworthy that staining for α₂M was not able to distinguish between circulating α₂M and α₂M expressed on microvessels.

Figure 6

(a) LRP-1 immunoreactivity in brain microvessels of young (2-month-old; upper panel) and old (9-month-old; lower panel) wild-type mice. Many vessels in young mice stained positive for LRP-1, detected with anti–LRP-1 Ab R777 (5 μg/ml; arrows). There were relatively fewer positive vessels in old mice (arrows), and many weakly positive- or negative-staining vessels (arrowheads). There was no significant difference in the staining of parenchymal cellular elements (open arrows) between the young and old mice. Vessels in young mice stained strongly positive (b, upper panel) compared with the faint staining seen in old mice (b, lower panel). In contrast, there was no difference in staining for α₂M in brain cells (arrowheads) or microvessels (arrows) between young (c, upper panel) and old (c, lower panel) mice. (d) Comparison of LRP-1 and α₂M immunoreactivity in brain microvessels (upper panel) and parenchymal cellular elements (lower panel) in young and old wild-type mice. ^AP < 0.05; NS, not significant.

Frontal cortex of all AD patients revealed moderate to marked neuritic plaques and Aβ deposits; two of the three showed parenchymal and vascular amyloid. Controls revealed no neuritic plaques or Aβ in the parenchyma, and only meningeal vascular Aβ in one of the three patients. As seen in Figure 7a, staining for LRP-1 in the frontal cortex of control patients revealed moderate vascular staining in capillaries and arterioles, as well as neuronal staining. There was reduced LRP-1 staining in AD tissues, including regions with Aβ_1-40-positive or Aβ_1-42-positive plaques and vessels (Figure 7b). However, the immediate subcortical white matter showed more robust vascular staining for LRP-1, and absence of staining for Aβ, in both AD patients and controls (not shown). Anti-CD105, which identifies vascular endothelium, revealed ample staining of capillaries and arterioles of frontal cortex in controls, and moderately reduced numbers of stained vessels in AD tissues. Cerebellum revealed equivalent vascular staining with anti–LRP-1 and anti-CD105 in AD and control sections (not shown). No anti-Aβ_1-40–positive or Aβ_1-42–positive staining was seen in either AD or control tissues.

Figure 7

LRP-1 expression in human frontal cortex. Brain sections (Brodmann’s area 10) of controls (a and c) reveal well-defined staining of capillaries (arrowheads) and arterioles (arrows) by LRP-1, detected with anti–LRP-1 mAb 8G1 (5 μg/ml) (a) and CD105 (c). No Aβ staining was present in double-labeled or serially labeled sections (not shown). In contrast, double-labeled sections from AD patients show vessels and plaque cores stained positive with anti-Aβ_1-40 (brown stain), reduced numbers and intensity of LRP-1 staining of vessels (b), and reduced numbers of CD105-labeled vessels (d).

This study demonstrates the importance of vascular transport across the BBB in clearing Aβ from the brain into the circulation. Moreover, we provide evidence that this transport mechanism is mediated mainly via LRP-1 in brain microvascular endothelium, and suggest that transport of brain-derived Aβ out of the CNS may be influenced by the LRP-1 ligands α₂M and apoE. This vascular clearance mechanism for Aβ is age dependent, and lower clearance rates in older animals correlate with decreased vascular abundance of LRP-1.

The capability of BBB to remove Aβ was significant in younger animals. The elimination time, t_1/2, for Aβ_1-40 at 60 nM was 25 minutes, or 9.4-fold faster than for inulin, an ECF marker used to determine the ISF bulk flow rate (40). The major component of the CNS efflux of Aβ was transport across the BBB into the vascular system. The clearance of Aβ across the BBB was dependent on both time and concentration. At very low concentrations, i.e., less than 2 nM as found normally in mouse brain (54–55), Aβ_1-40 was eliminated from brain at a rate that was on average 3.5-fold faster than when it was present at a load of 60 nM. This may be comparable to concentrations of Aβ_1-40 found in the brains of transgenic APP animals at 3–4 months of age (54–55). The efflux transport system was half saturated at 15.3 nM of Aβ_1-40, and appears to be fully saturated at concentrations between 70 nM and 100 nM. Thus, this efflux transporter may be completely saturated by higher levels of Aβ, as are found in the brains of older transgenic APP animals (54–55), which in turn may lead to vascular accumulation of Aβ and development of prominent deposits of cerebrovascular amyloid, as recently described (56–57).

In this study, we did not observe significant metabolism or degradation of Aβ_1-40 within 5 hours, in contrast to a recent report suggesting that Aβ_1-42 is degraded by enkephalinase (neprilysin) in the brain within minutes (28). It may be that Aβ_1-40 and Aβ_1-42 are processed differently in the brain. However, the physiological relevance of the proposed degradation mechanism for Aβ_1-42 (28) remains unclear, because the peptide was studied at extremely high concentrations (∼240 μM) that are not found even in brains with severe β-amyloidosis (30). As shown by pharmacological studies, these high concentrations of Aβ may impair local BBB integrity (58–59), which in turn may contaminate brain ISF with blood and/or plasma that possesses Aβ-degrading activity (48), as confirmed in this study.

Consistent with the hypothesis that cytosolic peptidases have little access to Aβ peptides secreted or injected into brain ISF (28) or CSF (29), it has been reported recently that insulin-degrading enzyme (IDE) cannot not degrade Aβ in the brain in vivo after intracerebral injection of radiolabeled peptide (28). This is in contrast to in vitro degradation of ¹²⁵I-Aβ_1-40 by IDE from brain and liver cytosol fractions (60). Because IDE is an intracellular protease, it is not surprising that IDE may not be able to process Aβ from brain ISF, particularly if peptide clearance is faster than its cellular uptake, as suggested by our data and a previous study (28). It is noteworthy that brain endothelial cells in vitro (33) and astrocytes (61) do not catabolize Aβ, in contrast to activated microglial cells that secrete a specific metalloproteinase that degrades Aβ in vitro (61). Neuronal cells in vitro metabolize Aβ via an LRP-1–dependent mechanism that may require apoE or α₂M (38). The rate of this degradation, however, is about 50- to 100-fold slower than that occurring via transport across the BBB in vivo.

Transport of Aβ out of the CSF was not associated with significant degradation of peptide in the CSF (29). Lower levels of Aβ_1-40 than inulin in the CSF may suggest active transport of Aβ from the CSF to the blood, possibly across the choroid plexus or leptomeningeal vessels, as shown previously (29). Higher levels of radiolabeled Aβ in the plasma relative to inulin confirm vascular transport of the peptide out of the CNS. Although our results indicate that brain-derived Aβ could contribute to the pool of circulating peptide, its degradation in plasma, systemic metabolism, and body clearance tend to reduce the levels of circulating peptide, as shown previously (48). Under our experimental conditions, the levels of radiolabeled Aβ in the circulation were two to three orders of magnitude lower than the brain levels. This makes re-entry of radiolabeled Aβ into the brain very unlikely, because the blood-to-brain transport of Aβ normally operates down the concentration gradient (39, 62–65). In addition, the apoJ system that transports blood-borne Aβ into the brain is saturated under the physiological conditions (64) that may facilitate the efflux of Aβ from the brain. Previous studies have shown that circulating free Aβ is also metabolized during its transport across the BBB (48, 49, 65, 66), possibly by pericytes, which represent a major enzymatic barrier for the transport of several peptides and proteins across the BBB (67).

The affinity of neprilysin for its physiological substrates (e.g., enkephalins, tachykinins, atrial natriuretic peptide) and/or different synthetic peptides is in the low millimolar range (68). In contrast, the levels of Aβ in the brain are normally in the low nanomolar range, and in transgenic mouse models of brain amyloidosis they vary from 40 nM to 250 nM/kg from 3 to 12 months of age (54). Thus, under physiological and/or pathological conditions, Aβ will likely bind to its high-affinity cell-surface receptors (such as RAGE and/or SR-A), and/or high-affinity transport binding proteins (α₂M, apoE, and apoJ), which all react with low nanomolar levels of peptide corresponding to their K_D values.

In this study, anti–LRP-1 antibodies inhibited Aβ_1-40 clearance by about 55%, both at lower (12 nM) and higher loads (60 nM) of the peptide, suggesting the involvement of LRP-1 in vascular elimination of Aβ from the brain. RAP, a chaperone protein that facilitates proper folding and subsequent trafficking of LRP-1 and LRP-2 (69), also inhibited Aβ clearance. RAP binds to multiple sites on LRP and antagonizes binding of all known LRP ligands to both LRP-1 and LRP-2 in vitro (69), as well as to LRP-2 in vivo at the blood side of the BBB (39). In this study, RAP at higher concentrations produced inhibition of Aβ clearance comparable to that produced by an anti–LRP-1 Ab. It is interesting that anti–LRP-1 Ab inhibited vascular transport of Aβ almost completely at higher concentrations of peptide, which may indicate that LRP-1 could be of primary importance in eliminating the peptide from the brain. At a lower load of the peptide (i.e., 12 nM), neither of the molecular reagents was able to abolish clearance of Aβ, which suggests that in addition to LRP-1, there may be an alternative, highly sensitive BBB transport mechanism (or mechanisms) that eliminates the peptide from the brain at very low concentrations. The molecular nature of this putative second transport system is not presently known, although data from this study suggest that RAGE and LRP-2 are unlikely to be involved in rapid elimination of Aβ from brain. The fact that fucoidin, an SR-A ligand, moderately increased clearance of Aβ suggests that inhibition of SR-A receptors in the brain may decrease CNS sequestration of the peptide, thus allowing more peptide to be available for enhanced clearance across the BBB.

A role for LRP-1 in promoting Aβ clearance in vitro in smooth muscle cells, neurons, and fibroblasts via α₂M and apoE has been suggested (35–38), although at significantly slower rates than the Aβ transport across the BBB demonstrated in this study. High-affinity in vitro binding of Aβ to α₂M and to lipidated apoE3 and apoE4, and a lower-affinity binding to delipidated apoE isoforms has been well documented (23, 70). Binding/uptake studies in mouse embryonic fibroblasts (wild type and deficient in LRP-1), confirmed that free Aβ is not a ligand for LRP-1 (37, 45). The possible role of the two LRP-1 ligands in elimination of Aβ via vascular transport is suggested by inhibition of Aβ clearance with anti-α₂M antibodies and by significantly reduced clearance in apoE KO animals (by 30% and 46% at 2 months and 9 months of age, respectively, compared with young wild-type controls). In relation to these findings, it is interesting to note that recent studies have indicated that lack of endogenous mouse apoE in both the APP^V717F and APPsw mouse models of AD results in less Aβ deposition and no fibrillar Aβ deposits in the brain (57, 71). This suggests that mouse apoE strongly facilitates Aβ fibrillogenesis. It is possible that mouse apoE also plays a role in clearance of soluble Aβ across the BBB, as suggested by the current study, but that its ability to influence Aβ aggregation in APP transgenic mice is dominant. In contrast to the effects of mouse apoE, a recent study demonstrates that human apoE isoforms suppress early Aβ deposition in APP^V717F mice (72). Further studies in this model will be useful to determine whether this suppressive effect of human apoE isoforms on early Aβ deposition is secondary to effects on facilitating Aβ transport across the BBB. Although our study does not rule out the possibility that Aβ clearance by neurons, vascular smooth muscle cells, and fibroblasts shown in vitro (35–38) may also occur in vivo, vascular transport across the BBB seems to be of primary importance for rapid elimination of Aβ from brain in vivo.

Because normal aging is associated with Aβ accumulation in the brain (1), and there is a significant, time-dependent, and progressive accumulation of the peptide with age in transgenic APP animals (54, 55), we postulated that Aβ clearance mechanisms might be impaired in older animals, and are also possibly impaired in elderly humans. Our findings of about 55–65% inhibition of Aβ clearance in 9-month-old wild-type animals compared with young (2-month-old) animals confirmed this hypothesis. Immunocytochemical studies indicated a significant reduction in the number of LRP-1–positive cerebral blood vessels, from 94% in 2-month-old mice to 52% in 9-month-old mice, which correlated well with the observed reductions in the clearance capacities in the two age groups. Interestingly, downregulation of vascular LRP-1 correlated well with regional parenchymal and vascular accumulation of Aβ in brains of Alzheimer’s patients compared with age-matched controls. In brain areas where LRP-1 vascular expression remains prominent, such as in the white matter, no accumulation of Aβ was found in Alzheimer’s brains.

In conclusion, our data support the concept that the vascular system plays an important role in regulating the levels of Aβ in the brain. Our findings further suggest that if the levels of Aβ in brain extracellular space exceed the transport capacity of the clearance mechanism across the BBB, or if the vascular transport of the peptide were impaired, for example by downregulation of LRP-1, this would result in accumulation of Aβ in the brain, and possibly formation of amyloid plaques. Previous studies from our laboratory and others have demonstrated a major role of the BBB in determining the concentrations of Aβ in the CNS by regulating transport of circulating Aβ (33, 39, 49–51, 62–66). This study extends this hypothesis by showing that vascular transport out of the brain across the BBB may represent a major physiological mechanism that prevents accumulation of Aβ and amyloid deposition in the brain.

This work was supported in part by NIH grants AG-16223 and NS-33466 (to B.V. Zlokovic), AG-05891 (to B. Frangione), AG-05142 (to C. Finch and C.A. Miller), AG-13956 (to D.M. Holtzman), NS-38777 (to J. Ghiso), and HL-50784 (to D.K. Strickland). We acknowledge the excellent technical assistance of Celia Williams.

1. Wisniewski, T, Ghiso, J, Frangione, B. Biology of Aβ amyloid in Alzheimer’s disease. Neurobiol Dis 1997. 4:311-328.
   View this article via: CrossRef Google Scholar
2. Selkoe, DJ. Alzheimer’s disease: genotype, phenotype, and treatments. Science 1997. 275:630-631.
   View this article via: PubMed CrossRef Google Scholar
3. Selkoe, DJ. The cell biology of β-amyloid precursor protein and presenilin in Alzheimer’s disease. Trends Cell Biol 1998. 8:447-453.
   View this article via: PubMed CrossRef Google Scholar
4. Younkin, SG. The role of Aβ42 in Alzheimer’s disease. J Physiol (Paris) 1998. 92:289-292.
   View this article via: PubMed CrossRef Google Scholar
5. Roses, AD. Alzheimer disease: a model of gene mutations and susceptibility polymorphisms for complex psychiatric diseases. Am J Med Genet 1998. 81:49-57.
   View this article via: PubMed CrossRef Google Scholar
6. Hardy, J, Duff, K, Hardy, KG, Perez-Tur, J, Hutton, M. Genetic dissection of Alzheimer’s disease and related dementias: amyloid and its relationship to tau. Nat Neurosci 1998. 1:355-358.
   View this article via: PubMed CrossRef Google Scholar
7. Masters, CL, et al. Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci USA 1985. 82:4245-4249.
   View this article via: PubMed CrossRef Google Scholar
8. Prelli, F, Castano, EM, Glenner, GG, Frangione, B. Differences between vascular and plaque core amyloid in Alzheimer’s disease. J Neurochem 1988. 51:648-651.
   View this article via: PubMed CrossRef Google Scholar
9. Roher, AE, et al. β-amyloid (1–42) is a major component of cerebrovascular amyloid deposits: implications for the pathology of Alzheimer disease. Proc Natl Acad Sci USA 1993. 90:10836-10840.
   View this article via: PubMed CrossRef Google Scholar
10. Shinkai, Y, et al. Amyloid β-proteins 1-40 and 1-42(43) in the soluble fraction of extra- and intracranial blood vessels. Ann Neurol 1995. 38:421-428.
    View this article via: PubMed CrossRef Google Scholar
11. Castaño, EM, et al. The length of amyloid-β in hereditary cerebral hemorrhage with amyloidosis, Dutch type. Implications for the role of amyloid-β 1-42 in Alzheimer’s disease. J Biol Chem 1996. 271:32185-32191.
    View this article via: PubMed CrossRef Google Scholar
12. Seubert, P, et al. Isolation and quantification of soluble Alzheimer’s amyloid-β peptide from biological fluids. Nature 1992. 359:325-327.
    View this article via: PubMed CrossRef Google Scholar
13. Shoji, M, et al. Production of the Alzheimer amyloid-β protein by normal proteolytic processing. Science 1996. 258:126-129.
    View this article via: PubMed CrossRef Google Scholar
14. Vigo-Pelfrey, C, Lee, D, Keim, P, Lieberburg, I, Schenk, DB. Characterization of β-amyloid peptide from human cerebrospinal fluid. J Neurochem 1993. 61:965-968.
15. Tabaton, M, et al. Soluble amyloid β-protein is a marker of Alzheimer amyloid in brain but not in cerebrospinal fluid. Biochem Biophys Res Commun 1994. 200:1598-1603.
    View this article via: PubMed CrossRef Google Scholar
16. Kuo, YM, et al. Water-soluble Aβ (N-40, N-42) oligomers in normal and Alzheimer disease brains. J Biol Chem 1996. 271:4077-4081.
    View this article via: PubMed CrossRef Google Scholar
17. Ghiso, J, et al. The cerebrospinal-fluid soluble form of Alzheimer’s amyloid-β is complexed to SP-40, 40 (apolipoprotein J), and inhibitor of the complement membrane-attack complex. Biochem J 1993. 293:27-30.
    View this article via: PubMed Google Scholar
18. Matsubara, E, Frangione, B, Ghiso, J. Characterization of apolipoprotein J-Alzheimer’s Aβ interaction. J Biol Chem 1995. 270:7563-7567.
    View this article via: PubMed CrossRef Google Scholar
19. Yang, DS, Smith, JD, Zhou, Z, Gandy, SE, Martins, RN. Characterization of the binding of amyloid-β peptide to cell culture-derived native apolipoprotein E2, E3, and E4 isoforms and to isoforms from human plasma. J Neurochem 1997. 68:721-725.
    View this article via: PubMed Google Scholar
20. Schwarzman, AL, et al. Transthyretin sequesters amyloid β-protein and prevents amyloid formation. Proc Natl Acad Sci USA 1994. 91:8368-8372.
    View this article via: PubMed CrossRef Google Scholar
21. Matsubara, E, et al. Lipoprotein-free amyloidogenic peptides in plasma are elevated in patients with sporadic Alzheimer’s disease and Down’s syndrome. Ann Neurol 1999. 45:537-541.
    View this article via: PubMed CrossRef Google Scholar
22. Biere, AL, et al. Amyloid β-peptide is transported on lipoproteins and albumin in human plasma. J Biol Chem 1996. 271:32916-32922.
    View this article via: PubMed CrossRef Google Scholar
23. Du, Y, et al. β2-Macroglobulin as a β-amyloid peptide-binding plasma protein. J Neurochem 1997. 69:299-305.
    View this article via: PubMed Google Scholar
24. Busciglio, J, Gabuzda, DH, Matsudaira, P, Yanker, BA. Generation of β-amyloid in the secretory pathway in neuronal and non-neuronal cells. Proc Natl Acad Sci USA 1993. 90:2092-2096.
    View this article via: PubMed CrossRef Google Scholar
25. Naslund, J, et al. Relative abundance of Alzheimer Aβ amyloid peptide variants in Alzheimer disease and normal aging. Proc Natl Acad Sci USA 1994. 91:8378-8382.
    View this article via: PubMed CrossRef Google Scholar
26. Teller, JK, et al. Presence of soluble amyloid β-peptide precedes amyloid plaque formation in Down’s syndrome. Nat Med 1996. 2:93-95.
    View this article via: PubMed CrossRef Google Scholar
27. Suzuki, N, et al. High tissue content of soluble amyloid-β 1-40 is linked to cerebral amyloid angiopathy. Am J Pathol 1994. 145:452-460.
    View this article via: PubMed Google Scholar
28. Iwata, N, et al. Identification of the major Aβ_1-42-degrading catabolic pathway in brain parenchyma: suppression leads to biochemical and pathological deposition. Nat Med 2000. 6:143-150.
    View this article via: PubMed CrossRef Google Scholar
29. Ghersi-Egea, JF, et al. Fate of cerebrospinal fluid-borne amyloid β-peptide: rapid clearance into blood and appreciable accumulation by cerebral arteries. J Neurochem 1996. 67:880-883.
    View this article via: PubMed Google Scholar
30. Zlokovic, BV, Yamada, S, Holtzman, D, Ghiso, J, Frangione, B. Clearance of amyloid β-peptide from brain: transport or metabolism? Nat Med 2000. 6:718-719.
    View this article via: PubMed Google Scholar
31. Rosenberg, RN. The molecular and genetic basis of AD: the end of the beginning. Neurology 2000. 54:2045-2054.
    View this article via: PubMed Google Scholar
32. Yan, SD, et al. RAGE and amyloid-β peptide neurotoxicity in Alzheimer’s disease. Nature 1996. 382:685-691.
    View this article via: PubMed CrossRef Google Scholar
33. Mackic, JB, et al. Human blood-brain barrier receptors for Alzheimer’s amyloid-β_1-40: asymmetrical binding, endocytosis and transcytosis at the apical side of brain microvascular endothelial cell monolayer. J Clin Invest 1998. 102:734-743.
    View this article via: JCI PubMed CrossRef Google Scholar
34. Paresce, DM, Ghosh, RN, Maxfield, FR. Microglial cells internalize aggregates of the Alzheimer’s disease amyloid β-protein via a scavenger receptor. Neuron 1996. 17:553-565.
    View this article via: PubMed CrossRef Google Scholar
35. Urmoneit, B, et al. Cerebrovascular smooth muscle cells internalize Alzheimer amyloid β-protein via a lipoprotein pathway: implications for cerebral amyloid angiopathy. Lab Invest 1997. 77:157-166.
    View this article via: PubMed Google Scholar
36. Jordan, J, et al. Isoform-specific effect of apolipoprotein E on cell survival and β-amyloid-induced toxicity in rat hippocampal pyramidal neuronal cultures. J Neurosci 1998. 18:195-204.
    View this article via: PubMed Google Scholar
37. Narita, M, Holtzman, DM, Schwartz, AL, Bu, G. α₂-macroglobulin complexes with and mediates the endocytosis of β-amyloid peptide via cell surface low-density lipoprotein receptor-related protein. J Neurochem 1997. 69:1904-1911.
    View this article via: PubMed Google Scholar
38. Qiu, Z, Strickland, DK, Hyman, BT, Rebeck, GW. α₂-macroglobulin enhances the clearance of endogenous soluble β-amyloid peptide via low-density lipoprotein receptor-related protein in cortical neurons. J Neurochem 1999. 73:1393-1398.
    View this article via: PubMed CrossRef Google Scholar
39. Zlokovic, BV, et al. Glycoprotein 330/megalin: probable role in receptor-mediated transport of apolipoprotein J alone and in a complex with Alzheimer’s disease amyloid β at the blood-brain and blood-cerebrospinal fluid barriers. Proc Natl Acad Sci USA 1996. 93:4229-4236.
    View this article via: PubMed CrossRef Google Scholar
40. Yamada, S, dePasquale, M, Patlak, CS, Cserr, HF. Albumin outflow into deep cervical lymph from different regions of rabbit brain. Am J Physiol 1991. 261:H1197-H1204.
    View this article via: PubMed Google Scholar
41. Strittmater, W, Roses, A. Apolipoprotein E and Alzheimer’s disease. Annu Rev Neurosci 1996. 19:53-77.
    View this article via: PubMed CrossRef Google Scholar
42. Blacker, D, et al. α₂-macroglobulin is genetically associated with Alzheimer’s disease. Nat Genet 1998. 19:357-360.
    View this article via: PubMed CrossRef Google Scholar
43. Zlokovic, BV, Davson, H, Preston, JE, Segal, MB. Effects of aluminum on cerebrospinal fluid secretion. Exp Neurol 1987. 98:436-452.
    View this article via: PubMed CrossRef Google Scholar
44. Kounnas, MZ, et al. The α₂-macroglobulin receptor/low density lipoprotein receptor-related protein binds and internalizes Pseudomonas exotoxin A. J Biol Chem 1992. 267:12420-12423.
    View this article via: PubMed Google Scholar
45. Kounnas, MZ, et al. LDL receptor-related protein, a multifunctional ApoE receptor, binds secreted β-amyloid precursor protein an mediates its degradation. Cell 1995. 82:331-340.
    View this article via: PubMed CrossRef Google Scholar
46. Mikhailenko, I, Kounnas, MZ, Strickland, DK. Low density lipoprotein receptor-related protein/α 2-macroglobulin receptor mediates the cellular internalization and degradation of thrombospondin. A process facilitated by cell-surface proteoglycans. J Biol Chem 1995. 270:9543-9549.
    View this article via: PubMed CrossRef Google Scholar
47. Kounnas, MZ, Haudenschild, CC, Strickland, DK, Argraves, WS. Immunological localization of glycoprotein 330, low density lipoprotein receptor related protein and 39 kDa receptor associated protein in embryonic mouse tissues. In Vivo 1994. 8:343-351.
    View this article via: PubMed Google Scholar
48. Mackic, JB, et al. Cerebrovascular accumulation and increased blood-brain barrier permeability to circulating Alzheimer’s amyloid β-peptide in aged squirrel monkey with cerebral amyloid angiopathy. J Neurochem 1998. 70:210-215.
    View this article via: PubMed Google Scholar
49. Maness, LM, Banks, WA, Podlisny, MB, Selkoe, DJ, Kastin, AJ. Passage of human amyloid-β protein 1-40 across the murine blood-brain barrier. Life Sci 1994. 55:1643-1650.
    View this article via: PubMed CrossRef Google Scholar
50. Poduslo, JF, Curran, GL, Haggard, JJ, Biere, AL, Selkoe, DJ. Permeability and residual plasma volume of human, Dutch variant, and rat amyloid β-protein 1-40 at the blood-brain barrier. Neurobiol Dis 1997. 4:27-34.
    View this article via: PubMed CrossRef Google Scholar
51. Martel, CL, et al. Isoform-specific effects of apolipoproteins E2, E3, E4 on cerebral capillary sequestration and blood brain barrier transport of circulating Alzheimer’s amyloid β. J Neurochem 1997. 69:1995-2004.
    View this article via: PubMed Google Scholar
52. Hyman, BT, Trojanowski, JQ. Consensus recommendations for the postmortem diagnosis of Alzheimer disease from the National Institute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. J Neuropathol Exp Neurol 1997. 56:1095-1097.
    View this article via: PubMed CrossRef Google Scholar
53. Strickland, DK, et al. Sequence identity between the α₂-macroglobulin receptor and low density lipoprotein receptor-related protein suggests that this molecule is a multifunctional receptor. J Biol Chem 1990. 265:17401-17404.
    View this article via: PubMed Google Scholar
54. Hsiao, K, et al. Correlative memory deficits, Aβ elevation, and amyloid plaques in transgenic mice. Science 1996. 274:99-102.
    View this article via: PubMed CrossRef Google Scholar
55. Holcomb, L, et al. Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor protein and presenilin 1 transgenes. Nat Med 1998. 4:97-100.
    View this article via: PubMed CrossRef Google Scholar
56. Calhoun, ME, et al. Neuronal overexpression of mutant amyloid precursor protein results in prominent deposition of cerebrovascular amyloid. Proc Natl Acad Sci USA 1999. 96:14088-14093.
    View this article via: PubMed CrossRef Google Scholar
57. Holtzman, DM, et al. ApoE facilitates neuritic and cerebrovascular plaque formation in the APPsw mouse model of Alzheimer’s disease. Ann Neurol 2000. 47:739-747.
    View this article via: PubMed CrossRef Google Scholar
58. Blanc, EM, Toborek, M, Mark, RJ, Hennig, B, Mattson, MP. Amyloid β-peptide induces cell monolayer albumin permeability, impairs glucose transport, and induces apoptosis in vascular endothelial cells. J Neurochem 1997. 68:1870-1881.
    View this article via: PubMed Google Scholar
59. Thomas, T, Thomas, G, McLendon, C, Sutton, T, Mullan, M. β-amyloid mediated vasoactivity and vascular endothelial damage. Nature 1996. 380:168-171.
    View this article via: PubMed CrossRef Google Scholar
60. Kurochkin, IV, Goto, S. Alzheimer’s β-amyloid peptide specifically interacts with and is degraded by insulin degrading enzyme. FEBS Lett 1994. 345:33-37.
    View this article via: PubMed CrossRef Google Scholar
61. Mentlein, R, Ludwig, R, Martensen, I. Proteolytic degradation of Alzheimer’s disease amyloid β-peptide by a metalloproteinase from microglia cells. J Neurochem 1998. 70:721-726.
    View this article via: PubMed Google Scholar
62. Zlokovic, BV, et al. Blood-brain barrier transport of circulating Alzheimer’s amyloid-β. Biochem Biophys Res Commun 1993. 197:1034-1040.
    View this article via: PubMed CrossRef Google Scholar
63. Ghilardi, JR, et al. Intra-arterial infusion of [¹²⁵I]Aβ_1-40 labels amyloid deposits in the aged primate brain in vivo. Neuroreport 1996. 7:2607-2611.
    View this article via: PubMed Google Scholar
64. Shayo, M, McLay, RN, Kastin, AJ, Banks, WA. The putative blood-brain barrier transporter for the β-amyloid binding protein apolipoprotein J is saturated at physiological concentrations. Life Sci 1997. 60:L115-L118.
65. Martel, CL, Mackic, JB, McComb, JG, Ghiso, J, Zlokovic, BV. Blood-brain barrier uptake of the 40 and 42 amino acid sequences of circulating Alzheimer’s amyloid-β in guinea pigs. Neurosci Lett 1996. 206:157-160.
    View this article via: PubMed CrossRef Google Scholar
66. Saito, Y, Buciak, J, Yang, J, Pardridge, WM. Vector-mediated delivery of ¹²⁵I-labeled β-amyloid peptide Aβ 1-40 through the blood-brain barrier and binding to Alzheimer’s disease amyloid to the Aβ_1-40/ vector complex. Proc Natl Acad Sci USA 1995. 92:10227-10231.
    View this article via: PubMed CrossRef Google Scholar
67. Krause, D, Kunz, J, Dermietzel, R. Cerebral pericytes: a second line of defense in controlling blood-brain barrier peptide metabolism. Adv Exp Med Biol 1993. 331:149-152.
    View this article via: PubMed Google Scholar
68. Hersh, LB, Morihara, K. Comparison of subsite specificity of the mammalian neutral endopeptidase 24.11 (enkephalinase) to the bacterial neutral endopeptidase thermolysin. J Biol Chem 1986. 261:6433-6437.
    View this article via: PubMed Google Scholar
69. Bu, G, Rennke, S. Receptor-associated protein is a folding chaperone for low-density lipoprotein receptor-related protein. J Biol Chem 1996. 271:22218-22224.
    View this article via: PubMed CrossRef Google Scholar
70. Tokuda, T, et al. Lipidation of apolipoprotein E influences its isoform-specific interaction with Alzheimer’s amyloid β-peptides. Biochem J 2000. 348:359-365.
    View this article via: PubMed CrossRef Google Scholar
71. Bales, KR, et al. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer’s disease. Proc Natl Acad Sci USA 1999. 96:15233-15238.
    View this article via: PubMed CrossRef Google Scholar
72. Holtzman, DM, et al. In vivo expression of apolipoprotein E reduces amyloid β deposition in a mouse model of Alzheimer’s disease. J Clin Invest 1999. 103:R15-R21.
    View this article via: JCI PubMed CrossRef Google Scholar