Intact blood-brain barrier transport of small molecular drugs in animal models of amyloid beta and alpha-synuclein pathology

(1)

Intact blood-brain barrier transport of small molecular drugs in animal models of amyloid beta and alpha-synuclein pathology

So ﬁa Gustafsson

^a^,^*

, Veronica Lindstr€om

^b

, Martin Ingelsson

^b

, Margareta Hammarlund-Udenaes

^a

, Stina Syv€anen

^b

aDepartment of Pharmaceutical Biosciences, Translational PKPD, Box 591, 751 24 Uppsala, Sweden

bDepartment of Public Health and Caring Sciences, Molecular Geriatrics, Daghammarsk€oldsv€ag 20, 751 85 Uppsala, Sweden

a r t i c l e i n f o

Article history:

Received 28 March 2017 Received in revised form 4 July 2017

Accepted 3 August 2017 Available online 5 August 2017

Chemical compounds studied in this article:

Digoxin (PubChem CID 2724385)

Paliperidone (PubChem CID 115237)

Levoﬂoxacin (PubChem CID 149096)

Diazepam (PubChem CID 3016)

Oxycodone hydrochloride (PubChem CID 5462350)

Keywords:

Blood-brain barrier Pharmacokinetics Drug transport Disease Amyloid beta Alpha-synuclein

a b s t r a c t

Pathophysiological impairment of the neurovascular unit, including the integrity and dynamics of the blood-brain barrier (BBB), has been denoted both a cause and consequence of neurodegenerative diseases. Pathological impact on BBB drug delivery has also been debated. The aim of the present study was to investigate BBB drug transport, by determining the unbound brain-to-plasma concentration ratio (Kp,uu,brain), in aged AbPP-transgenic mice,a-synuclein transgenic mice, and wild type mice.

Mice were dosed with a cassette ofﬁve compounds, including digoxin, levoﬂoxacin (1 mg/kg, s.c.), paliperidone, oxycodone, and diazepam (0.25 mg/kg, s.c.). Brain and blood were collected at 0.5, 1, or 3 h after dosage. Drug concentrations were measured using LC-MS/MS. The total brain-to-plasma concentration ratio was calculated and equilibrium dialysis was used to determine the fraction of unbound drug in brain and plasma for all compounds. Together, these three measures were used to determine the Kp,uu,brainvalue.

Despite Ab^ora-synuclein pathology in the current animal models, no difference was observed in the extent of drug transport across the BBB compared to wild type animals for any of the compounds investigated. Hence, the present study shows that the concept of a leaking barrier within neurodegenerative conditions has to be interpreted with caution when estimating drug transport into the brain. The capability of the highly dynamic BBB to regulate brain drug exposure still seems to be intact despite the presence of pathology.

1. Introduction

Since the early 1900's it has become increasingly recognized that a physical and highly dynamic barrier hinders the passage of constituents from the circulation into the brain (Saunders et al., 2014). Today this barrier is well known as the blood-brain barrier

(BBB), consisting of the thin but tightly joined layer of endothelial cells, constituting the capillary walls of the brain. Together with other brain parenchymal cells and structures, the endothelial cells form the neurovascular unit (NVU) (Neuwelt et al., 2011). The NVU plays an important role in maintaining the minute microenviron- ment required within the brain for normal neuronal functioning.

Active and passive transport mechanisms at the BBB aid in the uptake of essential molecules from the blood to the brain. However, much of the activity at the BBB is dominated by efficient efflux transport, acting in parallel with tight and adherens junctions, metabolism, low pinocytotic activity, bulkflow and the charge of the glycocalyx to restrict exogenous and endogenous molecules from entering the brain.

* Corresponding author. Translational PKPD, Department of Pharmaceutical Biosciences Uppsala University, Box 591, SE-751 24 Uppsala, Sweden.

E-mail addresses: soﬁa.gustafsson@farmbio.uu.se (S. Gustafsson), veronica.

lindstrom@pubcare.uu.se (V. Lindstr€om), martin.ingelsson@pubcare.uu.se (M. Ingelsson), mhu@farmbio.uu.se (M. Hammarlund-Udenaes), stina.syvanen@

pubcare.uu.se(S. Syv€anen).

Contents lists available atScienceDirect

Neuropharmacology

j o u rn a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / n e u ro p h a r m

https://doi.org/10.1016/j.neuropharm.2017.08.002

(2)

Pathophysiological impairment of the BBB and the NVU has been suggested in neurodegenerative diseases (Zhao et al., 2015).

Terms like “BBB breakdown” and “a leaking barrier” are also becoming increasingly acknowledged. Accordingly, NVU dysfunction, reduction of tight junctions, and impaired efﬂux transport in both in vitro and in vivo studies, speaks in favor of an affected BBB permeability in disorders like Alzheimer's disease and Parkinson's disease (Farkas et al., 2000; Kortekaas et al., 2005; Bartels et al., 2008b; Mehta et al., 2015; van de Haar et al., 2016). Conse- quently, an altered drug transport across the BBB under such disease conditions has been postulated, with enhanced or reduced effect or side effects as potential outcomes. However, conﬂicting evidence question how and to what extent the NVU is compromised and how this might affect drug delivery to the brain and ultimately the treatment of patients (Bourasset et al., 2009; Cheng et al., 2010; Mehta et al., 2013a; Bien-Ly et al., 2015).

Brain drug delivery can be investigated by the means of rate or extent of drug transport across the BBB (Hammarlund-Udenaes et al., 2008). While the rate could be important for a fast onset of effect, most dose regimens are aimed for continuous and daily treatment where the extent of drug transport into the brain might be of higher signiﬁcance (Hammarlund-Udenaes et al., 2008). For valid measurements of drug BBB transport it is of uttermost importance to investigate the unbound concentrations of drug that are present within brain and plasma, as it is only the unbound drug that can cross the barrier and later also drive the therapeutic effect (Gupta et al., 2006; Summerﬁeld et al., 2006; Kalvass et al., 2007a, 2007b; Liu et al., 2009).

While BBB transport of drug molecules is difﬁcult to study in humans, animal models are used in basic research and drug development to elucidate the pharmacokinetic properties of compounds under different disease conditions. One such disease model is the transgenic mouse tg-APP_ArcSwe, carrying both the Swedish (p.

KM670/671NL) and the Arctic (p. E693G) mutations of the amyloid- b precursor protein (AbPP) observed in familial forms of Alz- heimer's disease (Lord et al., 2006). The tg-APP_ArcSwemice express high brain levels of amyloid beta (Ab) and the proposed neurotoxic and soluble Aboligomers and protoﬁbrils (Nilsberth et al., 2001;

Lord et al., 2006). An advantage compared to many other models

of Abpathology, is that the plaques, that start to develop in the tg- APP_ArcSwebrain at an age of 6e7 months, are dense core plaques, resembling those observed in most forms of human Alzheimer's disease. In addition, the model also displays cerebral amyloid angiopathy (Philipson et al., 2009). Another disease model, tg- a^SYNA30P, used for the investigation ofa-synuclein related disorders, express a-synuclein with the missense mutation Ala30Pro that is found in one family with dominantly inherited Parkinson's disease (Kahle et al., 2001). This transgene exhibits overexpression of humana-synuclein, with the further formation of potentially toxic solublea-synuclein aggregates, such as oligomers and pro- toﬁbrils (Kahle et al., 2000; Neumann et al., 2002; Fagerqvist et al., 2013). Aggregateda-synuclein is found in the tg-a^SYNA30Pbrain from the age of 12 months.

By investigating the plasma and brain pharmacokinetic properties of a cassette of compounds in these animal models, the current study aimed to elucidate how pathological hallmarks of neurodegenerative diseases might influence the extent of transport of small molecular drugs across the BBB. The compounds investigated in the current study were chosen with regard to their net transport across the BBB, as indicated by their K_p,uu,brainvalues. In order to detect any change in passive permeability at the BBB, diazepam was chosen as a model compound, with a Kp,uu,brainclose to unity (Friden et al., 2009). Owing to extensivefindings of altered expression and functional changes of efflux transporters during CNS disorders, levofloxacin, paliperidone and digoxin were exam- ined as they all show a net active efflux at the BBB, where the two latter are confirmed substrates of P-glycoprotein (P-gp) (Schinkel et al., 1995; Wang et al., 2004; Friden et al., 2009). As little is known about the active uptake of small molecular drugs during disease, oxycodone was chosen as a marker of net active influx at the BBB (Bostrom et al., 2006).

2. Materials and methods 2.1. Chemicals

Oxycodone hydrochloride and diazepam were obtained from APL (Stockholm, Sweden). Digoxin was obtained from Apoteket AB (Solna, Sweden). Levoﬂoxacin and paliperidone, both of analytical grade, were purchased from Sigma Aldrich (St. Louis, MO, USA). The internal standards diazepam-D5 and oxycodone-D6, used in the sample analysis, were purchased from Cerilliant Corporation (Round Rock, Texas, USA). Phosphate buffered saline (PBS) tablets of 0.01 M, pH 7.4, were used for perfusions and purchased from Medicago (Uppsala, Sweden). Acetonitrile (ACN) (Merck, Darm- stadt, Germany) was of gradient grade and all other chemicals were of analytical grade. The water used was deionized in house and further puriﬁed with a Milli-Q Academic system (Millipore, Bed- ford, MA, USA).

2.2. Animals

In the current study, male and female (11:10) homozygous (Thy- 1)-h[A30P]a-SYN transgenic mice (tg-a^SYNA30P) (Kahle et al., 2000, 2001), overexpressing A30P mutated human a-synuclein, were used, as well as heterozygous male and female (8:9) (Thy-1)-h [E693G; KM670/671NL] AbPP transgenic mice (tg-APPArcSwe) (Lord et al., 2006), expressing AbPP with the human Arctic (E693G) (Nilsberth et al., 2001) and Swedish (KM670/671NL) (Mullan et al., 1992) mutations. Non-transgenic male and female (51:46), C57BL/

6J wild type (WT) mice were used for validation and as controls.

All animals were on a C57BL/6J background and all animals were bred in house. The animals were housed in a humidity and temperature controlled environment on a 12 h light/dark cycle Abbreviations

Ab Amyloid beta ACN Acetonitrile

Ab^PP ^Amyloid-bprecursor protein

AUC Area under the concentration-time curve BBB Blood-brain barrier

CNS Central nervous system

fu,brain Fraction of unbound drug in brain fu,hD Fraction of unbound drug in diluted brain

homogenate

fu,plasma Fraction of unbound drug in plasma Kp,brain Ratio of total brain to total plasma drug

concentrations

Kp,uu,brainRatio of unbound brain to unbound plasma drug concentrations

LC-MS/MS Liquid chromatography tandem mass spectrometry

NVU Neurovascular unit P-gp P-glycoprotein WT Wild type

(3)

with free access to food and water. The studies were approved by the local Animal Ethics Committee in Uppsala, Sweden (reference numbers C363/12, C17/14, and C92/14).

2.3. Experimental set up for cassette validation

Levoﬂoxacin, paliperidone, digoxin, diazepam and oxycodone were chosen as model compounds in the present study. Before used in disease models, the total brain-to-plasma concentration ratio (Kp,brain) was determined for all compounds after discrete or cassette dosing in order to validate the pharmacokinetic properties of the drugs when given alone or in combination with other drugs.

Theﬁve model compounds were dissolved in 2.5% ethanol, 0.21%

acetic acid and 97% PBS, both when prepared as a cassette and when compounds were prepared for discrete dosing. For the cassette validation, 2e5 months old WT mice were administered a subcutaneous dose of the compounds as discrete doses or as a cassette at a dose of 0.25 mg/kg for oxycodone and diazepam, and 1 mg/kg for digoxin and levoﬂoxacin. Paliperidone was administered at a dose of 1 mg/kg in the discrete dosing experiment but the dose was lowered to 0.25 mg/kg when included in the cassette in order to minimize CNS effects when three CNS active compounds were administered simultaneously. The volume of dose solution administered to all animals was 8mL/g body weight. The animals were euthanized at 0.5 h, 1 h, and 3 h after dose (n¼ 3e4 per time point). Blood was collected through cardiocentesis and brains were extracted following transcardial 0.01 M PBS perfusion. Blood samples were centrifuged at 7200 g for 5 min. Plasma was collected and directly frozen on dry ice. The brains were further dissected on ice.

The dissection was performed in order to extract areas of the brain with pathological relevance for the animal models later used in other experiments. Brieﬂy, the left and right hemispheres were separated and cerebellum and the olfactory bulb of the right hemisphere were removed. The right hemisphere was further divided in two pieces that generated a caudal part involving pathologically affected brain regions such as hippocampus and thalamus for the tg-APP_ArcSwemice, and the midbrain and brain- stem found to be affected in the tg-a^SYNA30Pmice (Freichel et al., 2007; Lillehaug et al., 2014). The samples were frozen on dry ice and all samples were stored at80C until analysis.

2.4. Experimental setup for estimation of Kp,brainin transgenes and WT animals

The Kp,brain was determined by terminal brain and plasma sampling in tg-a^SYNA30Pmice at an age of 16e19 months, as well as in 16 months old tg-APP_ArcSwe and WT mice. The animals were administered a subcutaneous dose of the cassette compounds at doses of 0.25 mg/kg for oxycodone, diazepam and paliperidone, and 1 mg/kg for digoxin and levoﬂoxacin. The animals were euthanized at 0.5 h (n tg-APPArcSwe ¼ 3, tg-a^SYNA30P ¼ 5, and WT¼ 3), 1 h (tg-APPArcSwe¼ 5, tg-a^SYNA30P¼ 5, and WT ¼ 5), and 3 h (tg-APP_ArcSwe¼ 3, tg-a^SYNA30P¼ 5, and WT ¼ 3) after dose.

Blood and brain were collected and processed as described in section2.3.

2.5. Brain and plasma sample preparation

Sample preparation was carried out just before analysis. Plasma samples were thawed in room temperature, vortexed and centrifuged for 1 min at 10 000 g. Brain samples were homogenized by diluting the tissue 5-fold in buffer (180 mM phosphate buffer, pH 7.4) and by the use of a high intensity ultrasonic processor (Vibra- Cell, Sonics & Materials Inc., Newtown, CT, USA). The brain homogenate was then brieﬂy vortexed. Fifty mL of sample was

precipitated with 150mL of ACN, containing 0.2% formic acid and the internal standards diazepam-D5 and oxycodone-D6 at concentrations of 20 ng/mL. The samples were mixed and centrifuged for 3 min at 10 000 g. One hundredfiftymL of supernatant was transferred to a new tube and evaporated under a stream of ni- trogen at 40C. For the plasma samples, the residual was dissolved in 30mL of ACN and 300mL of 5 mM ammonium formate buffer, pH 3.4. The residual of the brain homogenate samples were dissolved in 20mL of ACN and 200mL of 5 mM ammonium formate buffer. All samples were mixed and centrifuged at 10 000 g for 1 min and transferred into autosampler vials. For the analysis of levofloxacin and paliperidone in plasma, 15mL were injected on to a column, while 40 mL was injected for the analysis of diazepam and diazepam-D5. For the analysis of digoxin, oxycodone, and oxycodone-D6 in plasma, 20m^{L and 40}mL were injected onto the column, respectively. For concentration measurements in brain, 75mL was injected for the analysis of levofloxacin, paliperidone, diazepam, diazepam-D5, and digoxin. For the analysis of oxycodone and oxycodone-D6 in brain, 40mL was injected onto the column.

2.6. Brain tissue and plasma protein binding

The fraction of unbound drug was determined in plasma and in brain tissue from the caudal right hemisphere of 16e19 months old tg-a^SYNA30P, and of 16 months old tg-APPArcSweand WT mice, and in the rostral right hemisphere of the same tg-APPArcSweand WT mice (number of animals for plasma¼ 6e8, and brain ¼ 6 per group). Dialysis membrane strips with a molecular weight cut off of 12e14 kDa (HTDialysis LLC, Gales Ferry, CT, USA) were conditioned in 180 mM phosphate buffer, pH 7.4, overnight and rinsed again with fresh phosphate buffer before use. Brain tissue homogenate was prepared from the rostral and caudal part of the right hemisphere by diluting the tissue samples 5-fold with buffer (180 mM phosphate buffer, pH 7.4). The tissue samples were homogenized using a high intensity ultrasonic processor (Vibra-Cell, Sonics &

Materials Inc., Newtown, CT, USA). Due to the limited amount of tissue and plasma from each animal, samples were pooled within each group, in order to avoid extensive dilution of the samples and to retrieve sufficient technical replicates. The plasma was adjusted to 37C in a water bath and the pH was adjusted to 7.2 by the use of 1 M phosphoric acid. After handling of the samples and hence after exposure to air, the pH reached 7.4. The brain homogenate and plasma from all groups were spiked with thefive cassette compounds to afinal concentration of 1m^{M, and 125}mL aliquots were loaded infive replicates into a 96-well equilibrium dialysis appa- ratus (HTDialysis LLC, Gales Ferry, CT, USA). Dialysis against the same volume of 180 mM phosphate buffer, pH 7.4 was carried out for 6 h at 37C, at 200 rpm in a MaxQ 4450 benchtop shaker (Thermo Scientific, Waltham, MA, USA). Adhesive sealing film (HTDialysis LLC, Gales Ferry, CT, USA) was used to cover the samples to prevent evaporation, pH changes, and contamination during dialysis. At the end of the incubation period, samples were collected from all chambers and the matrix in brain homogenate, plasma or buffer was balanced with the corresponding oppositefluid before stored at20C until analysis.

2.7. Equilibrium dialysis sample preparation

Sample preparation was carried out just before analysis. Plasma samples were vortexed and centrifuged for 1 min at 10 000 g.

Homogenate samples were vortexed only. FiftymL of sample was precipitated with 150mL of ACN, containing 0.2% formic acid, and the internal standards diazepam-D5 and oxycodone-D6 at concentrations of 20 ng/mL. The samples were mixed and centrifuged for 3 min at 10 000 g. One hundredmL of supernatant was mixed

(4)

with 600mL 5 mM ammonium formate buffer, pH 3.4. For the later analysis of compounds, 40mL was injected onto the column.

2.8. Liquid chromatography-tandem mass spectrometry analysis

The quantitative analysis was carried out on a liquid chromatography-tandem mass spectrometry (LC-MS/MS) system.

The analytical methods were developed by optimizing the parameters to obtain a high resolution in a reasonable analysis time. The LC system included a SIL-HTc autosampler (Schimadzu, Kyoto, Japan) and one or two LC-10ADvp pumps connected to the column.

A Zorbax Eclipse XDB-CN column (4.6 150 mm; Agilent Tech- nologies, Santa Clara, CA, USA), maintained at 50C, was used when analyzing oxycodone and oxycodone-D6. The mobile phase consisted of 45% ACN in 5 mM ammonium acetate and a constantﬂow rate of 1.0 mL/min was used. Theﬂow was split, allowing 0.2 mL/

min to enter the detector. The detector was a Quattro Ultima Pt Triple Quadrupole Mass Spectrometer (Waters, Milford, MA, USA).

When analyzing levofloxacin, paliperidone, diazepam, diazepam- D5, and digoxin a HyPURITY C18 column (50 4.6 mm; Thermo Hypersil-Keystone, Bellefonte, PA, USA) was used with a guard column in the same material placed before the analytical column. A gradient run was applied when analyzing levofloxacin, paliperidone, diazepam and diazepam-D5. Mobile phases A (MPA) consisted of 5 mM ammonium formate buffer, pH 3.4, and mobile phase B (MPB) consisted of ACN:5 mM ammonium formate buffer (90:10). The gradient started with 90% MPA and 10% MPB, switching to 80% MPA and 20% MPB, and to 10% MPA and 90% MPB after 1 min and 4 min, respectively, returning back to 90% MPA and 10% MPB at 4.5 min. Digoxin was analyzed under isocratic conditions using the same two aforementioned mobile phases, MPA and MPB (66:34). The flow rate was set to 0.8 mL/min and split to 0.3 mL/min before entering the mass spectrometer. The analysis was carried out in a positive ionization mode using multiple reaction monitoring for all of the compounds. The transition modes for oxycodone and oxycodone-D6 were set to m/z 316.0 to m/z 297.9 and 322.1 to m/z 303.9, respectively. The transition modes for diazepam and diazepam-D5 were set to m/z 284.9 to m/z 154 and 290 to m/z 154, respectively. The transition modes for levofloxacin, paliperidone and digoxin were set to m/z 362.05 to m/z 318.05, m/z 427 to m/z 207, and m/z 798.2 to m/z 651.3, respectively. Mass spectrometry control and spectral processing for determination of compound concentrations were carried out using MassLynx soft- ware version 4.1 in combination with the Application Manager QuanLynx (Waters Corporation, Milford, MA, USA). The standard curves for all compounds showed good linearity (coefficient of determination > 0.990) and all samples were within the linear range.

2.9. Ab^anda-synuclein ELISA

Brain concentrations of total Ab were measured in the tg- APP_ArcSweand the WT animals, and totala-synuclein was measured in the tg-a^SYNA30Pand WT animals as described previously, with modifications (Fagerqvist et al., 2013; Sehlin et al., 2016). Briefly, for ELISA measurement of Ab^{1-40 and A}b1-42, 96-well plates were coated with 100 ng per well of polyclonal rabbit anti-Ab40 or anti- Ab42 (Agrisera, Umeå, Sweden). For ELISA measurement ofa^-synuclein, 96-well plates were coated with 50 ng per well of Syn-1 (BD Biosciences, Franklin lake, NJ, USA). All plates were coated overnight at 4C. Unspecific binding was blocked with 1% BSA the next day. The left hemisphere of each animal involved in the study was homogenized at a 1:5 weight:volume ratio in TBS with complete protease inhibitors (Roche, Basel, Switzerland). The tissues were

homogenized in 2 mL homogenization tubes containing 1.4 mm ceramic beads using a Precellys Evolution homogenizer (Bertin Technologies, Montigny-le-Bretonneux, France). 62mL of the TBS extract was mixed with 167mL of concentrated formic acid, to a ﬁnal formic acid concentration of 70%, followed by vortexing and centrifugation for 1 h at 16 000 g, at 4C. Formic acid extracts were neutralized with 2 M Tris and diluted 2000 or 20 000 times, for Ab^{1-42 and A}b1-40, respectively, and 5000 times fora-synuclein.

Samples were loaded on the plates and incubated overnight at 4C, followed by detection with biotinylated 82E1, (0.25mg/mL; IBL In- ternational/Tecan Trading AG, Switzerland) and streptavidin-HRP (1:5000; Mabtech AB, Nacka Strand, Sweden) for Ab^{1-42 and} Ab1-40. For the detection ofa-synuclein a rabbit IgG anti-a^-synuclein polyclonal antibody, FL140 (1: 1000; Santa Cruz Biotech- nology, Santa Cruz, CA, USA) and a polyclonal HRP-coupled goat- anti-rabbit IgG antibody (1: 1000; Pierce, Rockford, IL, USA) were used. Signals were developed with K blue aqueous TMB substrate (Neogen Corp., Lexington, KY, USA), the reaction was stopped using 1 M H2SO4, and the signal was read with a spectrophotometer at 450 nm. All samples and secondary antibodies were diluted in ELISA incubation buffer (PBS with 0.1% BSA, 0.05% Tween and 0.15%

proclin). The wells were washed three times between each step with ELISA washing buffer (phosphate-buffered NaCl with 0.1%

Tween 20 and 0.15% proclin).

2.10. Data analysis

The total brain-to-plasma concentration ratio, K_p,brain, was calculated (Eq.(1)) based on the area under the concentration-time curves (AUC) from 0 to 3 h of the brain and plasma proﬁles, estimated by the linear-trapezoidal method. As the AUC estimates are based on mean concentrations of terminal brain and plasma samples at each time point, the Kp,brainis presented as a single value without standard deviation (SD). However, the coefﬁcient of variation (CV) in the brain-to-plasma concentration ratio at each time point was approximately 20% (Supplementary Table 1).

K_p;brain¼ AUCbrain; 03h

AUCplasma; 03h (1)

The fraction of unbound drug in plasma, fu,plasma, was calculated, from the equilibrium dialysis experiments, as the buffer-to-plasma concentration ratio (Eq.(2)).

f_u;plasma¼ Cbuffer

.

C_plasma (2)

The fraction of unbound drug in diluted brain homogenate, fu,hD, was calculated as the buffer-to-homogenate concentration ratio (Eq.(3)). fu,hDwas further used to calculate the fraction of unbound drug in brain, fu,brain(Eq.(4)), while also accounting for the dilution factor, D, of the brain homogenate.

f_u;hD¼ C_buffer.

C_homogenate (3)

f_u;brain¼1 D

, 1 f_u;hD

!

1

! þ1

D (4)

The unbound brain-to-plasma concentration ratio, Kp,uu,brain, used to estimate net BBB transport, was calculated by combining the K_p,brainestimates with the fractions of unbound drug in brain and plasma as given by Equation(5).

K_p;uu;brain¼ f_u;brain.

f_u;plasma Kp (5)

(5)

3. Results

3.1. Cassette validation

The five drugs, levofloxacin, paliperidone, digoxin, diazepam, and oxycodone were investigated in WT animals, at an age of 2e5 months, for the presence or absence of drug-drug interactions, as they were to be administered as a cassette to reduce the number of animals required in the following experiments. The Kp,brainvalues, representing total brain-to-plasma concentration ratios, estimated from the cassette dosing, were within 2-fold of the values of the discrete dosing for all of the five compounds investigated.

Concluding that the compounds could be administered as a cassette (Fig. 1).

3.2. BBB transport in tg-APPArcSweand tg-a^SYNA30Pversus WT mice

The BBB transport in transgenic mice, expressing Ab^{(16 months} of age) or a-synuclein (16e19 months of age) pathology, was compared to age matched (16 months) healthy WT animals, by a comparison of Kp,uu,brainvalues (Table 1,Fig. 2). Digoxin and paliperidone are targets of P-gp and together with levofloxacin showed a net efflux at the BBB, indicated by Kp,uu,brainvalues below unity, as expected. No difference in the BBB transport was observed for these three compounds when comparing tg-APPArcSwemice and tg-a^SYNA30Pmice with age matched controls. Statistical comparison of the values was not applicable due to terminal sampling and pooling of data at each time point for the estimation of brain and plasma AUC, used for the Kp,braincalculations. However, the differences in K_p,uu,brainvalues between the groups were so small that the likeliness of an affected efflux transport for these compounds can be regarded as insignificant. The plasma concentration-time pro- files did not differ between healthy animals and transgenes for any of the compounds.

Diazepam is a lipophilic drug with an estimated Kp,uu,brainvalue close to unity in the current study, supporting passive diffusion as the dominating transport of diazepam. The extent of transport of diazepam across the BBB did not differ for any of the two transgenes compared to WT animals (Table 1,Fig. 2) and there were no pronounced differences in the plasma concentration proﬁles when comparing WT animals to transgenes.

Oxycodone, displaying net active uptake, showed K_p,uu,brain values around 2 for all groups (Table 1,Fig. 2), with no differences in the plasma concentration-time proﬁles. Hence, independent of transgene no differences in BBB transport was observed, indicating a preserved active uptake transport of oxycodone at the BBB.

3.3. Brain tissue binding in rostral and caudal brain regions

In order to investigate any differences in drug binding properties due to the brain distribution of Abpathology or due to het- erogeneity in regional brain composition, drug tissue binding was investigated in rostral and caudal brain areas from 16 months old tg-APPArcSweand WT mice (Fig. 3). There was no difference in the tissue binding when comparing rostral and caudal brain for paliperidone, digoxin, diazepam and oxycodone in neither WT nor transgenic animals. Nor was there any difference in tissue binding when comparing WT with transgenes. Levoﬂoxacin was excluded in this part of the study due to technical difﬁculties caused by very low binding.

3.4. Ab^anda-synuclein pathology in transgenic mice

The presence of pathology was conﬁrmed in the tg-APPArcSwe

mice with total Ab^{40 and A}b42 levels of 285± 85 and 8.9 ± 2.7 ng/

mg brain (mean± SD), respectively (Fig. 4a), and with undetectable amounts in WT animals.a-synuclein pathology was also conﬁrmed with totala-synuclein levels of 1656± 591 ng/mg brain in the tg- a^SYN^A30Pmice and 129± 21 in WT mice (Fig. 4b). Thea^-synuclein detected in WT mice is most probably endogenous mousea^-synuclein detected by the ELISA.

4. Discussion

Accumulating evidence suggests an impairment of the NVU, and in particular the BBB, during CNS related disorders (Daneman, 2012; Zhao et al., 2015), which would indicate an altered drug transport at the BBB. By investigating drugs with different routes of BBB transport in animal models expressing Abôra-synuclein pathology, our current study shows that the extent of BBB transport of small molecular drugs was unaffected in disease compared to healthy conditions. Hence, the present study suggests that the concept of a leaking barrier should be considered with caution for small molecular drugs under neurodegenerative conditions. It also indicates that the capacity of the highly dynamic BBB is sufficient for normal regulation of drug transport despite pathological influence.

In order to reduce the number of animals used in this study, the compounds were administered as a cassette. Cassette dosing also adds to the advantage of investigating all BBB transport mechanisms in the same animals, thereby minimizing group-related differences. The drugs in the cassette can also act as internal controls in the analysis. While some of the drugs constitute targets of the same transporter at the BBB, simultaneous administration of the drugs did not indicate any interactions affecting the neuropharmacokinetic properties of the compounds compared to discrete dosing. Furthermore, terminal blood and brain sampling was used in the current study, resulting in pooling of data where the inter-individual variability is lost when using non- compartmental analysis. However, despite the lack of inter- individual variation, we believe that any profound difference in drug neuropharmacokinetic properties in transgenes compared to WT would be noted under the current experimental conditions, given the similarity in pharmacokinetic proﬁles and the high expression of toxic Ab^ora-synuclein species.

Pathological impact on efﬂux transport was investigated by the Fig. 1. Relationship between the total brain-to-plasma concentration ratio, Kp,brain,

determined from discrete and cassette dosing of theﬁve model compounds in WT animals, 2e5 months old. The solid line represents the line of identity. The dashed lines represent a 2-fold deviation from unity. Dig, digoxin; Lev, levoﬂoxacin; Pal, paliperidone; Diaz, diazepam; Oxy, oxycodone.

(6)

use of digoxin, paliperidone, and levofloxacin, which all display a net active efflux at the BBB and where the two former constitutes substrates of P-gp. Despite the confirmed presence of toxic Ab species and plaques in the brain of the tg-APPArcSwemice, no difference was observed in the extent of transport of these compounds across the BBB compared to healthy, age matched controls.

Numerous studies report Abⁱⁿﬂuence on BBB integrity and P-gp expression. In vitro ﬁndings, using brain endothelial cells have shown a reduced expression of tight junction proteins and increased permeability of high molecular weight compounds upon treatment with different Ab species (Marco and Skaper, 2006;

Gonzalez-Velasquez et al., 2008; Tai et al., 2010; Carrano et al., 2011). An altered expression of P-gp on both mRNA and protein levels in endothelial cells were also shown but seem to be dependent on Abconcentrations (Kania et al., 2011; Qosa et al., 2014).

Furthermore, in vivo studies in transgenic models of Alzheimer's disease report disruption of tight junctions and an approximately 60% reduction of P-gp expression in isolated brain microvessels of Tg2576 mice, as well as a 42% reduction of P-gp expression in the

3xTg AD mouse model (Hartz et al., 2010; Biron et al., 2011; Mehta et al., 2013a). However, despite the reduction of P-gp expression in the 3xTg mice, there was no change in the rate of transport of digoxin across the BBB, which is in line with our currentfindings (Mehta et al., 2013a). Together with our present study, the results in the 3xTg mice suggest that the remaining transporter capacity is sufficient to maintain an appropriate drug efflux at the BBB. In contrary to the aforementionedfindings, Do et al. observed an age dependent upregulation of P-gp in brain capillaries of the 3xTg mice and a tendency towards reduced Oatp1a4 was also reported (Do et al., 2016). Since the net transport of digoxin, a substrate of both transporters, was not changed in the current study, it was not possible to distinguish individual transporter impact in the present study. In line with Abpathology, aggregated a-synuclein did not influence net transport of digoxin, levofloxacin, and paliperidone.

Hence, these results indicate that the expression of Ab^ora^-synuclein pathology alone do not affect the investigated efﬂux properties to such an extent that drug transport is altered at the BBB at in vivo relevant doses.

Table 1

Estimated values of Kp,brain, fu,brain, and fu,plasma, and calculated Kp,uu,brain. The data represents measurements in the right caudal hemisphere and plasma from old, aged matched WT (16 months), tg-APPArcSwe(16 months), and tg-aSYNA30P(16e19 months) mice.

Parameters Levoﬂoxacin Paliperidone Digoxin Diazepam Oxycodone

Kp,braina

WT 0.144 0.371 0.0269 3.44 3.71

tg-App ArcSwe 0.151 0.413 0.0195 3.24 3.86

tg-aSYN A30P 0.160 0.358 0.0256 3.42 3.70

fu,brain(mean± SD)^b

WT 0.611± 0.067 0.0989± 0.0068 0.0893± 0.0072 0.0380± 0.0011 0.481± 0.022

tg-App ArcSwe 0.634± 0.152 0.100± 0.009 0.0908± 0.0073 0.0396± 0.0044 0.486± 0.039

tg-aSYN A30P 0.563± 0.159 0.0921± 0.0086 0.0891± 0.0058 0.0384± 0.0012 0.468± 0.030

fu,plasma(mean± SD)^b

WT 0.877± 0.038 0.339± 0.023 0.381± 0.029 0.130± 0.005 0.864± 0.020

tg-App ArcSwe 0.866± 0.092 0.343± 0.039 0.366± 0.013 0.120± 0.004 0.845± 0.020

tg-aSYN A30P 0.848± 0.070 0.296± 0.032 0.331± 0.047 0.116± 0.004 0.823± 0.022

Kp,uu,braina

WT 0.101 0.108 0.00629 1.01 2.06

tg-App ArcSwe 0.111 0.120 0.00484 1.07 2.22

tg-aSYN A30P 0.106 0.111 0.00689 1.13 2.10

Kp,brain, total brain-to-plasma concentration ratio; fu,brain, unbound fraction in brain; fu,plasma, unbound fraction in plasma; Kp,uu,brain, unbound brain-to-plasma concentration ratio.

aNo standard deviation (SD) as the values are based on AUC0-3hfrom terminal sampling.

b The mean and SD represent technical replicates.

Fig. 2. Relationship between the unbound brain-to-plasma concentration ratio, Kp,uu,brain, estimated in (a) tg-APPArcSweand WT (all 16 months of age), or (b) tg-aSYNA30P(16e19 months) and WT (16 months) for all of theﬁve model compounds. The solid line represents the line of identity. The dashed lines represent a 2-fold deviation from unity. Dig, digoxin; Lev, levoﬂoxacin; Pal, paliperidone; Diaz, diazepam; Oxy, oxycodone.

(7)

Fig. 3. Fraction of unbound drug in brain, fu,brain, of (a) paliperidone, (b) digoxin, (c) oxycodone, and (d) diazepam in caudal and rostral regions from 16 months old WT and tg- APPArcSwemice. Data are presented as technical replicates with lines representing mean values.

Fig. 4. Measurements of (a) total Ab40 and Ab42 in 16 months old tg-APPArcSwemice (n¼ 17), and of (b) totala-synuclein in 16e19 month old tg-aSYNA30Pmice (n¼ 21) and 16 months old WT mice (n¼ 17). Ablevels were undetectable in WT animals. The data is presented as mean± SD.

(8)

Diazepam was used in the present study as a marker of transcellular transport by the means of passive diffusion, supported by a Kp,uu,brainclose to unity. No difference was observed in the transport of diazepam across the BBB in any of the transgenic models compared to controls. Hence, the present Ab^ora-synuclein pathology does not have a major impact on the extent of transcellular passive drug transport across the BBB. In support of ourﬁndings, no difference was observed in passive diffusional BBB transport in a previous study using animal models exhibiting Abplaque disposi- tion or stimulated tau pathology (Cheng et al., 2010). A reduction in the rate of diazepam transport across the BBB has been observed in the 3xTg mouse model of Alzheimer's disease, said to be caused by a thickening of the basement membrane (Mehta et al., 2013a).

However, in a different study, using the 3xTg model, no difference was found in diazepam transport rate compared to controls (Bourasset et al., 2009).

To examine possible changes in net active uptake at the BBB, oxycodone was investigated in the present study. However, independent of transgene, the transport of oxycodone was the same as in WT mice. Oxycodone is transported by an energy-dependent, proton-coupled antiporter at the BBB (Bostrom et al., 2006; Okura et al., 2008). Memantine is a drug that also shows active uptake at the BBB and is transported by the same system as oxycodone (Hesselink et al., 1999; Mehta et al., 2013b; Higuchi et al., 2015). The rate of BBB transport of memantine has been studied and was decreased in an animal model of Alzheimer's disease, possibly due to a thickening of the basement membrane (Mehta et al., 2013c).

While we did not observe any change in oxycodone brain uptake when considering the extent of transport, this points to the importance of investigating both rate and extent of drug transport at the BBB. Hence, the uptake mechanisms appear to be intact under current disease conditions and at the administered doses.

Overall, our study shows that increased Ab ôr a^-synuclein expression alone does not affect the BBB to such an extent that the transport of the drugs studied is altered. This is partly in line with recent findings in humans with cognitive impairment, which indicate that amyloid pathology alone is not sufficient for BBB permeability changes (Janelidze et al., 2016). In further support of our data, studies show no difference in the distribution of large molecules to the brain in a variety of Alzheimer’s disease mouse models compared to controls (St-Amour et al., 2013; Alata et al., 2015; Bien-Ly et al., 2015). Still, it is important to stress that the animal models used in the current study, display only parts of the respective diseases. It cannot be fully ruled out that changes in brain drug delivery of compounds are altered in patients with Alzheimer's disease, where tau pathology is also extensive and where the disease is dependent on a multitude of factors, or in patients with different forms ofa-synucleinopathies. In a study by Janelidze et al., permeability changes during Alzheimer's disease were not shown to be associated with APOE genotype in humans (Janelidze et al., 2016), while this has previously been suggested by Halliday et al. (2016). However, as studies of APOE4 animal models also show cerebrovascular changes and altered NVU properties, the impact of APOE on small molecular drug transport needs further evaluation in both animals and humans (Bell et al., 2012; Alata et al., 2015). In addition, the influence of tau pathology on BBB functioning and the extent of brain drug delivery needs further investigation as abnormalities in brain vascular volume and BBB transporter expression has been presented in the 3xTg mouse model, also exhibiting tau pathology (Do et al., 2014, 2016).

Furthermore, in a study by Vogelgesang et al., an inverse correlation between cerebrovascular P-gp expression and the deposition of Ab was observed in aged human tissue (Vogelgesang et al., 2002). In later studies, P-gp has also been suggested to be reduced in human brain tissue from Alzheimer patients (Wijesuriya et al., 2010; Chiu

et al., 2015). Clinical positron emission tomography studies further suggest a reduced P-gp function at the BBB in patients with Alzheimer's disease, and while P-gp function does not seem to be impaired in early stages of Parkinson's disease, a reduction seems to be apparent in more advanced Parkinson patients (Kortekaas et al., 2005; Bartels et al., 2008a, 2008b; van Assema et al., 2012; Deo et al., 2014). A thickening of the basement membrane was also observed in patients with Alzheimer's or Parkinson's disease (Farkas et al., 2000), which would primarily inﬂuence the rate of drug delivery to the brain as observed by Mehta and coworkers (Mehta et al., 2013c, a). As differences in BBB transport properties on a regional level have been suggested under both healthy and disease conditions, studies relating brain regional pathology to local drug concentrations is also essential in both animals and humans (Jeynes and Provias, 2013; Loryan et al., 2016). In the present study, areas of the brain subjected to pathological hallmarks were analyzed for differences in drug brain binding. How- ever, no profound differences were observed for any of the compounds investigated.

In conclusion, no difference was observed in the extent of transport of small molecular drugs, transported across the BBB by several active or passive transport mechanisms, in animals displaying Ab^ora-synuclein pathology compared to healthy WT animals. The present study highlights the importance of further neuropharmacokinetic assessment of drugs in patients. Hence, altered NVU function and its impact on BBB drug transport at clinically relevant drug concentrations needs to be evaluated in humans. The present study also indicates that the human BBB may be capable of balancing drug transport in spite of pathological in- sults. In addition, changes in pharmacodynamic properties as well as pharmacokinetics must be investigated if altered effect or side effect proﬁles are observed under pathological conditions.

Acknowledgments

Philipp Kahle is acknowledged for having provided the tg- a^SYNA30Pmice and Lars Nilsson who developed the tg-APPArcSwe

model. The authors would also like to acknowledge Nebojsa Mihajlica for the cooperative work in attaining baseline data and for the technical assistance of Britt Jansson, Jessica Dunhall, and Dag Sehlin.

Conﬂict of interest

The authors declare that they have no conﬂict of interests.

Appendix A. Supplementary data

Supplementary data related to this article can be found at https://doi.org/10.1016/j.neuropharm.2017.08.002.

References

Alata, W., Ye, Y., St-Amour, I., Vandal, M., Calon, F., 2015. Human apolipoprotein E varepsilon4 expression impairs cerebral vascularization and blood-brain barrier function in mice. J. Cereb. Blood Flow. Metab. 35, 86e94.

Bartels, A.L., van Berckel, B.N., Lubberink, M., Luurtsema, G., Lammertsma, A.A., Leenders, K.L., 2008a. Blood-brain barrier P-glycoprotein function is not impaired in early Parkinson's disease. Park. Relat. Disord. 14, 505e508.

Bartels, A.L., Willemsen, A.T., Kortekaas, R., de Jong, B.M., de Vries, R., de Klerk, O., van Oostrom, J.C., Portman, A., Leenders, K.L., 2008b. Decreased blood-brain barrier P-glycoprotein function in the progression of Parkinson's disease, PSP and MSA. J. Neural Transm. (Vienna) 115, 1001e1009.

Bell, R.D., Winkler, E.A., Singh, I., Sagare, A.P., Deane, R., Wu, Z., Holtzman, D.M., Betsholtz, C., Armulik, A., Sallstrom, J., Berk, B.C., Zlokovic, B.V., 2012. Apoli- poprotein E controls cerebrovascular integrity via cyclophilin A. Nature 485, 512e516.

Bien-Ly, N., Boswell, C.A., Jeet, S., Beach, T.G., Hoyte, K., Luk, W., Shihadeh, V.,

(9)

Ulufatu, S., Foreman, O., Lu, Y., DeVoss, J., van der Brug, M., Watts, R.J., 2015. Lack of widespread BBB disruption in Alzheimer's disease models: focus on therapeutic antibodies. Neuron 88, 289e297.

Biron, K.E., Dickstein, D.L., Gopaul, R., Jefferies, W.A., 2011. Amyloid triggers extensive cerebral angiogenesis causing blood brain barrier permeability and hypervascularity in Alzheimer's disease. PLoS One 6, e23789.

Bostrom, E., Simonsson, U.S., Hammarlund-Udenaes, M., 2006. In vivo blood-brain barrier transport of oxycodone in the rat: indications for active inﬂux and implications for pharmacokinetics/pharmacodynamics. Drug Metab. Dispos. 34, 1624e1631.

Bourasset, F., Ouellet, M., Tremblay, C., Julien, C., Do, T.M., Oddo, S., LaFerla, F., Calon, F., 2009. Reduction of the cerebrovascular volume in a transgenic mouse model of Alzheimer's disease. Neuropharmacology 56, 808e813.

Carrano, A., Hoozemans, J.J., van der Vies, S.M., Rozemuller, A.J., van Horssen, J., de Vries, H.E., 2011. Amyloid Beta induces oxidative stress-mediated blood-brain barrier changes in capillary amyloid angiopathy. Antioxid. Redox Signal 15, 1167e1178.

Cheng, Z., Zhang, J., Liu, H., Li, Y., Zhao, Y., Yang, E., 2010. Central nervous system penetration for small molecule therapeutic agents does not increase in multiple sclerosis- and Alzheimer's disease-related animal models despite reported blood-brain barrier disruption. Drug Metab. Dispos. 38, 1355e1361.

Chiu, C., Miller, M.C., Monahan, R., Osgood, D.P., Stopa, E.G., Silverberg, G.D., 2015. P- glycoprotein expression and amyloid accumulation in human aging and Alz- heimer's disease: preliminary observations. Neurobiol. Aging 36, 2475e2482.

Daneman, R., 2012. The blood-brain barrier in health and disease. Ann. Neurol. 72, 648e672.

Deo, A.K., Borson, S., Link, J.M., Domino, K., Eary, J.F., Ke, B., Richards, T.L., Mankoff, D.A., Minoshima, S., O'Sullivan, F., Eyal, S., Hsiao, P., Maravilla, K., Unadkat, J.D., 2014. Activity of P-Glycoprotein, a beta-amyloid transporter at the blood-brain barrier, is compromised in patients with mild Alzheimer disease.

J. Nucl. Med. 55, 1106e1111.

Do, T.M., Alata, W., Dodacki, A., Traversy, M.T., Chacun, H., Pradier, L., Scherrmann, J.M., Farinotti, R., Calon, F., Bourasset, F., 2014. Altered cerebral vascular volumes and solute transport at the blood-brain barriers of two transgenic mouse models of Alzheimer's disease. Neuropharmacology 81, 311e317.

Do, T.M., Dodacki, A., Alata, W., Calon, F., Nicolic, S., Scherrmann, J.M., Farinotti, R., Bourasset, F., 2016. Age-dependent regulation of the blood-brain barrier inﬂux/

efﬂux equilibrium of amyloid-beta peptide in a mouse model of Alzheimer's disease (3xTg-AD). J. Alzheimers Dis. 49, 287e300.

Fagerqvist, T., Lindstrom, V., Nordstrom, E., Lord, A., Tucker, S.M., Su, X., Sahlin, C., Kasrayan, A., Andersson, J., Welander, H., Nasstrom, T., Holmquist, M., Schell, H., Kahle, P.J., Kalimo, H., Moller, C., Gellerfors, P., Lannfelt, L., Bergstrom, J., Ingelsson, M., 2013. Monoclonal antibodies selective for alpha-synuclein oligomers/protoﬁbrils recognize brain pathology in Lewy body disorders and alpha-synuclein transgenic mice with the disease-causing A30P mutation.

J. Neurochem. 126, 131e144.

Farkas, E., De Jong, G.I., de Vos, R.A., Jansen Steur, E.N., Luiten, P.G., 2000. Patho- logical features of cerebral cortical capillaries are doubled in Alzheimer's disease and Parkinson's disease. Acta Neuropathol. 100, 395e402.

Freichel, C., Neumann, M., Ballard, T., Muller, V., Woolley, M., Ozmen, L., Borroni, E., Kretzschmar, H.A., Haass, C., Spooren, W., Kahle, P.J., 2007. Age-dependent cognitive decline and amygdala pathology in alpha-synuclein transgenic mice.

Neurobiol. Aging 28, 1421e1435.

Friden, M., Winiwarter, S., Jerndal, G., Bengtsson, O., Wan, H., Bredberg, U., Ham- marlund-Udenaes, M., Antonsson, M., 2009. Structure-brain exposure re- lationships in rat and human using a novel data set of unbound drug concentrations in brain interstitial and cerebrospinalﬂuids. J. Med. Chem. 52, 6233e6243.

Gonzalez-Velasquez, F.J., Kotarek, J.A., Moss, M.A., 2008. Soluble aggregates of the amyloid-beta protein selectively stimulate permeability in human brain microvascular endothelial monolayers. J. Neurochem. 107, 466e477.

Gupta, A., Chatelain, P., Massingham, R., Jonsson, E.N., Hammarlund-Udenaes, M., 2006. Brain distribution of cetirizine enantiomers: comparison of three different tissue-to-plasma partition coefﬁcients: K(p), K(p,u), and K(p,uu). Drug Metab. Dispos. 34, 318e323.

Halliday, M.R., Rege, S.V., Ma, Q., Zhao, Z., Miller, C.A., Winkler, E.A., Zlokovic, B.V., 2016. Accelerated pericyte degeneration and blood-brain barrier breakdown in apolipoprotein E4 carriers with Alzheimer's disease. J. Cereb. Blood Flow.

Metab. 36, 216e227.

Hammarlund-Udenaes, M., Friden, M., Syvanen, S., Gupta, A., 2008. On the rate and extent of drug delivery to the brain. Pharm. Res. 25, 1737e1750.

Hartz, A.M., Miller, D.S., Bauer, B., 2010. Restoring blood-brain barrier P-glycoprotein reduces brain amyloid-beta in a mouse model of Alzheimer's disease. Mol.

Pharmacol. 77, 715e723.

Hesselink, M.B., De Boer, B.G., Breimer, D.D., Danysz, W., 1999. Brain penetration and in vivo recovery of NMDA receptor antagonists amantadine and memantine: a quantitative microdialysis study. Pharm. Res. 16, 637e642.

Higuchi, K., Kitamura, A., Okura, T., Deguchi, Y., 2015. Memantine transport by a proton-coupled organic cation antiporter in hCMEC/D3 cells, an in vitro human blood-brain barrier model. Drug Metab. Pharmacokinet. 30, 182e187.

Janelidze, S., Hertze, J., Nagga, K., Nilsson, K., Nilsson, C., Swedish Bio, F.S.G., Wennstrom, M., van Westen, D., Blennow, K., Zetterberg, H., Hansson, O., 2016.

Increased blood-brain barrier permeability is associated with dementia and diabetes but not amyloid pathology or APOE genotype. Neurobiol. Aging 51,

104e112.

Jeynes, B., Provias, J., 2013. P-glycoprotein altered expression in Alzheimer's disease:

regional Anatomic variability. J. Neurodegener. Dis. 2013, 257953.

Kahle, P.J., Neumann, M., Ozmen, L., Muller, V., Jacobsen, H., Schindzielorz, A., Okochi, M., Leimer, U., van Der Putten, H., Probst, A., Kremmer, E., Kretzschmar, H.A., Haass, C., 2000. Subcellular localization of wild-type and Parkinson's disease-associated mutant alpha -synuclein in human and transgenic mouse brain. J. Neurosci. 20, 6365e6373.

Kahle, P.J., Neumann, M., Ozmen, L., Muller, V., Odoy, S., Okamoto, N., Jacobsen, H., Iwatsubo, T., Trojanowski, J.Q., Takahashi, H., Wakabayashi, K., Bogdanovic, N., Riederer, P., Kretzschmar, H.A., Haass, C., 2001. Selective insolubility of alpha- synuclein in human Lewy body diseases is recapitulated in a transgenic mouse model. Am. J. Pathol. 159, 2215e2225.

Kalvass, J.C., Maurer, T.S., Pollack, G.M., 2007a. Use of plasma and brain unbound fractions to assess the extent of brain distribution of 34 drugs: comparison of unbound concentration ratios to in vivo p-glycoprotein efﬂux ratios. Drug Metabolism Dispos. Biol. Fate Chem. 35, 660e666.

Kalvass, J.C., Olson, E.R., Cassidy, M.P., Selley, D.E., Pollack, G.M., 2007b. Pharmaco- kinetics and pharmacodynamics of seven opioids in P-glycoprotein-competent mice: assessment of unbound brain EC50,u and correlation of in vitro, preclinical, and clinical data. J. Pharmacol. Exp. Ther. 323, 346e355.

Kania, K.D., Wijesuriya, H.C., Hladky, S.B., Barrand, M.A., 2011. Beta amyloid effects on expression of multidrug efﬂux transporters in brain endothelial cells. Brain Res. 1418, 1e11.

Kortekaas, R., Leenders, K.L., van Oostrom, J.C., Vaalburg, W., Bart, J., Willemsen, A.T., Hendrikse, N.H., 2005. Blood-brain barrier dysfunction in parkinsonian midbrain in vivo. Ann. Neurol. 57, 176e179.

Lillehaug, S., Syverstad, G.H., Nilsson, L.N., Bjaalie, J.G., Leergaard, T.B., Torp, R., 2014.

Brainwide distribution and variance of amyloid-beta deposits in tg-ArcSwe mice. Neurobiol. Aging 35, 556e564.

Liu, X., Vilenski, O., Kwan, J., Apparsundaram, S., Weikert, R., 2009. Unbound brain concentration determines receptor occupancy: a correlation of drug concentration and brain serotonin and dopamine reuptake transporter occupancy for eighteen compounds in rats. Drug Metabolism Dispos. Biol. Fate Chem. 37, 1548e1556.

Lord, A., Kalimo, H., Eckman, C., Zhang, X.Q., Lannfelt, L., Nilsson, L.N., 2006. The Arctic Alzheimer mutation facilitates early intraneuronal Abeta aggregation and senile plaque formation in transgenic mice. Neurobiol. Aging 27, 67e77.

Loryan, I., Melander, E., Svensson, M., Payan, M., Konig, F., Jansson, B., Hammarlund- Udenaes, M., 2016. In-depth neuropharmacokinetic analysis of antipsychotics based on a novel approach to estimate unbound target-site concentration in CNS regions: link to spatial receptor occupancy. Mol. Psychiatry 21, 1527e1536.

Marco, S., Skaper, S.D., 2006. Amyloid beta-peptide1-42 alters tight junction protein distribution and expression in brain microvessel endothelial cells. Neurosci.

Lett. 401, 219e224.

Mehta, D.C., Short, J.L., Nicolazzo, J.A., 2013a. Altered brain uptake of therapeutics in a triple transgenic mouse model of Alzheimer's disease. Pharm. Res. 30, 2868e2879.

Mehta, D.C., Short, J.L., Nicolazzo, J.A., 2013b. Memantine transport across the mouse blood-brain barrier is mediated by a cationic inﬂux Hþ antiporter. Mol.

Pharm. 10, 4491e4498.

Mehta, D.C., Short, J.L., Nicolazzo, J.A., 2013c. Reduced CNS exposure of memantine in a triple transgenic mouse model of Alzheimer's disease assessed using a novel LC-MS technique. J. Pharm. Biomed. Anal. 85, 198e206.

Mehta, D.C., Short, J.L., Hilmer, S.N., Nicolazzo, J.A., 2015. Drug access to the central nervous system in Alzheimer's disease: preclinical and clinical insights. Pharm.

Res. 32, 819e839.

Mullan, M., Crawford, F., Axelman, K., Houlden, H., Lilius, L., Winblad, B., Lannfelt, L., 1992. A pathogenic mutation for probable Alzheimer's disease in the APP gene at the N-terminus of beta-amyloid. Nat. Genet. 1, 345e347.

Neumann, M., Kahle, P.J., Giasson, B.I., Ozmen, L., Borroni, E., Spooren, W., Muller, V., Odoy, S., Fujiwara, H., Hasegawa, M., Iwatsubo, T., Trojanowski, J.Q., Kretzschmar, H.A., Haass, C., 2002. Misfolded proteinase K-resistant hyper- phosphorylated alpha-synuclein in aged transgenic mice with locomotor deterioration and in human alpha-synucleinopathies. J. Clin. Invest 110, 1429e1439.

Neuwelt, E.A., Bauer, B., Fahlke, C., Fricker, G., Iadecola, C., Janigro, D., Leybaert, L., Molnar, Z., O'Donnell, M.E., Povlishock, J.T., Saunders, N.R., Sharp, F., Stanimirovic, D., Watts, R.J., Drewes, L.R., 2011. Engaging neuroscience to advance translational research in brain barrier biology. Nat. Rev. Neurosci. 12, 169e182.

Nilsberth, C., Westlind-Danielsson, A., Eckman, C.B., Condron, M.M., Axelman, K., Forsell, C., Stenh, C., Luthman, J., Teplow, D.B., Younkin, S.G., Naslund, J., Lannfelt, L., 2001. The 'Arctic' APP mutation (E693G) causes Alzheimer's disease by enhanced Abeta protoﬁbril formation. Nat. Neurosci. 4, 887e893.

Okura, T., Hattori, A., Takano, Y., Sato, T., Hammarlund-Udenaes, M., Terasaki, T., Deguchi, Y., 2008. Involvement of the pyrilamine transporter, a putative organic cation transporter, in blood-brain barrier transport of oxycodone. Drug Metab.

Dispos. 36, 2005e2013.

Philipson, O., Hammarstrom, P., Nilsson, K.P., Portelius, E., Olofsson, T., Ingelsson, M., Hyman, B.T., Blennow, K., Lannfelt, L., Kalimo, H., Nilsson, L.N., 2009. A highly insoluble state of Abeta similar to that of Alzheimer's disease brain is found in Arctic APP transgenic mice. Neurobiol. Aging 30, 1393e1405.

Qosa, H., LeVine 3rd, H., Keller, J.N., Kaddoumi, A., 2014. Mixed oligomers and monomeric amyloid-beta disrupts endothelial cells integrity and reduces

(10)

monomeric amyloid-beta transport across hCMEC/D3 cell line as an in vitro blood-brain barrier model. Biochim. Biophys. Acta 1842, 1806e1815.

Saunders, N.R., Dreifuss, J.J., Dziegielewska, K.M., Johansson, P.A., Habgood, M.D., Mollgard, K., Bauer, H.C., 2014. The rights and wrongs of blood-brain barrier permeability studies: a walk through 100 years of history. Front. Neurosci. 8, 404.

Schinkel, A.H., Wagenaar, E., van Deemter, L., Mol, C.A., Borst, P., 1995. Absence of the mdr1a P-Glycoprotein in mice affects tissue distribution and pharmacokinetics of dexamethasone, digoxin, and cyclosporin A. J. Clin. Invest 96, 1698e1705.

Sehlin, D., Fang, X.T., Cato, L., Antoni, G., Lannfelt, L., Syvanen, S., 2016. Antibody- based PET imaging of amyloid beta in mouse models of Alzheimer's disease.

Nat. Commun. 7, 10759.

St-Amour, I., Pare, I., Alata, W., Coulombe, K., Ringuette-Goulet, C., Drouin-Ouellet, J., Vandal, M., Soulet, D., Bazin, R., Calon, F., 2013. Brain bioavailability of human intravenous immunoglobulin and its transport through the murine blood-brain barrier. J. Cereb. Blood Flow. Metab. 33, 1983e1992.

Summerﬁeld, S.G., Stevens, A.J., Cutler, L., del Carmen Osuna, M., Hammond, B., Tang, S.P., Hersey, A., Spalding, D.J., Jeffrey, P., 2006. Improving the in vitro prediction of in vivo central nervous system penetration: integrating permeability, P-glycoprotein efﬂux, and free fractions in blood and brain. J. Pharmacol.

Exp. Ther. 316, 1282e1290.

Tai, L.M., Holloway, K.A., Male, D.K., Loughlin, A.J., Romero, I.A., 2010. Amyloid-beta-

induced occludin down-regulation and increased permeability in human brain endothelial cells is mediated by MAPK activation. J. Cell Mol. Med. 14, 1101e1112.

van Assema, D.M., Lubberink, M., Bauer, M., van der Flier, W.M., Schuit, R.C., Windhorst, A.D., Comans, E.F., Hoetjes, N.J., Tolboom, N., Langer, O., Muller, M., Scheltens, P., Lammertsma, A.A., van Berckel, B.N., 2012. Blood-brain barrier P- glycoprotein function in Alzheimer's disease. Brain 135, 181e189.

van de Haar, H.J., Burgmans, S., Jansen, J.F., van Osch, M.J., van Buchem, M.A., Muller, M., Hofman, P.A., Verhey, F.R., Backes, W.H., 2016. Blood-brain barrier leakage in patients with early Alzheimer disease. Radiology 281, 527e535.

Vogelgesang, S., Cascorbi, I., Schroeder, E., Pahnke, J., Kroemer, H.K., Siegmund, W., Kunert-Keil, C., Walker, L.C., Warzok, R.W., 2002. Deposition of Alzheimer's beta-amyloid is inversely correlated with P-glycoprotein expression in the brains of elderly non-demented humans. Pharmacogenetics 12, 535e541.

Wang, J.S., Ruan, Y., Taylor, R.M., Donovan, J.L., Markowitz, J.S., DeVane, C.L., 2004.

The brain entry of risperidone and 9-hydroxyrisperidone is greatly limited by P- glycoprotein. Int. J. Neuropsychopharmacol. 7, 415e419.

Wijesuriya, H.C., Bullock, J.Y., Faull, R.L., Hladky, S.B., Barrand, M.A., 2010. ABC efﬂux transporters in brain vasculature of Alzheimer's subjects. Brain Res. 1358, 228e238.

Zhao, Z., Nelson, A.R., Betsholtz, C., Zlokovic, B.V., 2015. Establishment and dysfunction of the blood-brain barrier. Cell 163, 1064e1078.