The blood–brain barrier is intact after levodopa-induced dyskinesias in parkinsonian
primates—Evidence from in vivo neuroimaging studies
Arnar Astradssona,d, Bruce G. Jenkinsa,b, Ji-Kyung Choib, Penelope J. Halletta,d, Michele A. Levesquea,d,
Jack S. McDowella,d, Anna-Liisa Brownella,c, Roger D. Spealmana,d, Ole Isacsona,d,⁎
aHarvard University and McLean Hospital, NINDS Udall Parkinson's Disease Research Center of Excellence, Belmont, MA, USA
bMassachusetts General Hospital (MGH) Nuclear Magnetic Resonance Center, Athinoula A. Martinos Center for Biomedical Imaging, Boston, MA, USA
cMGH Positron Emission Tomography Center, Massachusetts General Hospital, Boston, MA, USA
dNew England Primate Research Center, Harvard Medical School, Southborough, MA, USA
a b s t r a c ta r t i c l ei n f o
Received 12 February 2009
Revised 6 May 2009
Accepted 28 May 2009
Available online 6 June 2009
It has been suggested, based on rodent studies, that levodopa (L-dopa) induced dyskinesia is associated with
a disrupted blood–brain barrier (BBB). We have investigated BBB integrity with in vivo neuroimaging
techniques in six 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) lesioned primates exhibiting L-
dopa-induced dyskinesia. Magnetic resonance imaging (MRI) performed before and after injection of
Gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA) revealed an intact BBB in the basal ganglia
showing that L-dopa-induced dyskinesia is not associated with a disrupted BBB in this model.
© 2009 Elsevier Inc. All rights reserved.
Levodopa (L-dopa) is currently the primary treatment of motor
symptoms in Parkinson's disease (PD). However, a major limitation of
chronic L-dopa treatment is the development of dyskinesias after
years of treatment (Fahn, 2003; Olanow et al., 2004). The pathophy-
siological mechanisms of L-dopa-induced dyskinesia are poorly
understood, though non-physiological release of synaptic dopamine
is likely to play a major role in its development (Obeso et al., 2000;
Olanow et al., 2004; Olanow and Obeso, 2000). Recently, it has been
suggested, based on studies in rodents, that L-dopa-induced dyskine-
sia may be associated with a disrupted blood–brain barrier (BBB)
(Westin et al., 2006) and that this may in turn contribute to its
pathophysiology, by further exacerbating dyskinesia (Westin et al.,
The purpose of the present study was to investigate the integrity of
the BBB using in vivo neuroimaging techniques in 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine (MPTP) lesioned parkinsonian primates
exhibiting L-dopa-induced dyskinesias.
Induction of parkinsonian and dyskinetic symptoms
Six adult male macaque monkeys (Macaca fascicularis), aged 6–
8 years and weighing 6–7 kg, were included in this study. Animals
were housed in individual home cages at the New England Primate
Research Center (NEPRC). All studies were approved by the Harvard
Medical School Institutional Animal Care and Use Committee (IACUC).
Parkinsonism was induced by weekly intravenous administration of
low doses of MPTP (Sigma-Aldrich®) diluted in normal saline. Doses
were given initially at 0.30 mg/kg to all animals but in some instances
subsequently reduced to 0.15 mg/kg, due to symptom severity and
individual sensitivity. Parkinsonian motor symptoms were rated
weekly during and after MPTP administration on a Parkinson's Rating
Scale (PRS) as developed for macaques (Imbert et al., 2000) and
modified from the motor subscale of the Unified Parkinson's Disease
Rating Scale (UPDRS) (Fahn, 2003), ranging from 0 to 24, with 24
being most severe. Stable PRS scores were obtained off L-dopa at least
3 months after the last MPTP dose and were considered stable if
standard deviation did not change more than ±2 over 6 weeks
(Table 1). All animals displayed a stable parkinsonian syndrome,
including tremor, rigidity, bradykinesia, hypokinesia and posture/
balance disturbances (Jenkins et al., 2004). Dopamine transporter loss
Neurobiology of Disease 35 (2009) 348–351
⁎ Corresponding author. Center for Neuroregeneration Research, McLean Hospital/
Harvard Medical School, MRC 130, 115 Mill St, Belmont, MA 02478, USA.
E-mail address: email@example.com (O. Isacson).
Available online on ScienceDirect (www.sciencedirect.com).
0969-9961/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
Contents lists available at ScienceDirect
Neurobiology of Disease
journal homepage: www.elsevier.com/locate/ynbdi
in the posterior putamen was measured by positron emission
tomography (PET) studies and binding of the dopamine transporter
tracer11C-(2β-carbomethoxy-3β-(4-fluorophenyl) tropane) (CFT) at
the stable stage, at least 3 months after last MPTP administration, as
previously described (Brownell et al., 1998). Animals then received
daily intramuscular (i.m.) injections of L-dopa methylester (Sigma-
Aldrich®) in combination with the peripheral decarboxylase inhibitor
benserazide(Sigma-Aldrich®), dilutedin normalsalineandinjectedat
1 ml, for the induction of dyskinesia. L-dopa was administered
according to individual animal response and tolerance at 30, 60 or
120 mg/kg dailyfor15–36 weeks.Benserazidewasco-administered at
10–15 mg/kg per dose. Dyskinesia severity was rated weekly by two
independent observers at 30, 60 and 90 min after a single i.m.
administration of L-dopa (30 or 60 mg/kg) in combination with
benserazide (10–15 mg/kg). Abnormal movements were classified as
chorea (rapid, random flicking movements), athetosis (sinuous,
writhing distal limb movements) dystonia (sustained twisting
movements resulting in abnormal posturing), myoclonus (jerky) or
stereotypy (repetitive purposeless behavior). Severity was rated
according to the Dyskinesia Disability Severity scale as described
(Bezard et al., 2003; Pearce et al., 1995), ranging from 1 to 4, based
on frequency and interference with normal behavior by 0=absent;
1=mild, fleeting and dyskinetic movements and postures (b5 in
10 min); 2=moderate, more prominent and abnormal dyskinesia
but not interfering with normal behavior (∼5–20 in 10 min);
3=marked, frequent dyskinesia, intruding on normal behavior
(21–50 in 10 min); 4=severe, virtually continuous dyskinesia,
disabling the animal. Sum of dyskinesia scores (peak scores) at the
maximally effective dose and time point were obtained and severity
(disability) scores were calculated by dividing the total score by the
number of affected regions, as previously described (Sanchez-
Pernaute et al., 2007).
MRI studies with Gadolinium-diethylenetriamine pentaacetic acid
(Gd-DTPA) contrast enhancement
After developing reproducible dyskinesias, animals underwent
neuroimaging studies. L-dopa was administered until the morning of
the study. Animals were anesthetized with a Ketamine (10 mg/kg)/
Xylazine (1.5 mg/kg) combination i.m. Atropine was administered at
0.04 mg/kg i.m. Anesthesia was maintained with halothane (1–1.5%)
while the animal was intubated but free breathing. The animal was
heating blanket to maintain body temperature. Respiratory rate, heart
rate, SpO2and body temperature were constantly monitored through-
out the procedure. MRI studies were performed on a 3 TAllegra system
(Siemens®, Erlangen, Germany) using a transmit–receive 3 inch surface
coil. The animal's head was placed in the center of the surface coil such
that thecoil fitovertheskull, abovetheeyes.Aftercollectionofbaseline
images, Gd-DTPA was administered intravenously at 0.3 mmol/kg and
Animal characteristics, dosing and symptoms.
MF 1MF 2 MF 3MF 4 MF 5 MF 6
MPTP (total mg)
Daily L-dopa dose (mg/kg)
Duration of L-dopa
Maximal effective L-dopa
Maximal effective time
point after L-dopa (min)
Dyskinesia peak score
Dyskinesia severity score
60 60 60 6030 60
90 90 90 9090 90
Fig.1. The BBB of the basal ganglia is intact as shown by Gadolinium-DTPA (Gd-DTPA) MRI studies in dyskinetic monkeys. T1 weighted axial brain MRI images before (upper panel)
and after (lower panel) peripheral injection of Gadolinium-DTPA, 0.3 mmol/kg to a dyskinetic macaque. Post Gd-DTPA there is marked signal enhancement of the hypothalamic/
pituitary region, structures lacking a BBB, and the sagittal sinus, but no signal enhancement of the basal ganglia, including the putamen, caudate, globus pallidus and substantia nigra
is observed. Put=putamen; Cd=caudate nucleus; GPe=globus pallidus externa, GPi=globus pallidus interna SN=substantia nigra, Pit/Hyp=pituitary/hypothalamic region.
L=left; R=right. Bright spots near SN are contrast filled blood vessels.
A. Astradsson et al. / Neurobiology of Disease 35 (2009) 348–351
and short TE (TR/TE=235/4.5 ms) with 30 s temporal resolution, and
high resolution (0.65 mm isotropic) T1-weighted sequence (TR/TI/
TE=1910/1100/3.1 ms) were collected. MRI data acquisition occurred
were extubated and placed in a warmed cage until fully recovered.
Regionsof interest (ROIs) were handdrawn of the SN,the putamen, the
caudate, the pituitary–hypothalamic region, the sagittal sinus and jaw
muscle on the various MRI images, and the average image intensity was
used for the quantitative analysis using the serial gradient echo images
as a function of time after injection of Gd-DTPA and delayed
enhancement was analyzed using the high resolution T1 weighted
images. Statistical analyses were performed with the GraphPad Prism®
program, version 5.01.
Animals received a weekly low dose of the neurotoxin MPTP for a
total of 4–34 weeks, with a total cumulative MPTP dose of 7.2–
39.5 mg. This resulted in moderate to severe parkinsonian symptoms
in all six animals, with an average PRS score of 18±2.7 (range 14–22),
that remained stable at least 3 months after the last MPTP dose
(Table 1). All animals displayed a significant loss of dopamine
transporter binding in the putamen with an average reduction of 59
±7.4 % (t-test; pb0.001), as measured by PET and the dopamine
treatment, all six animals developed dyskinesias as defined by the
presence of abnormal involuntary movements, mainly choreiform,
dystonic and stereotypic movements affecting limbs, axial body, tail,
and orolingual muscles (Table 1).
Animals then underwent an MRI brain scan with Gd-DTPA. Visual
inspection of high resolution T1 weighted images, revealed no
increase in signal intensity post Gd-DTPA in the basal ganglia,
11C-CFT. After 15–36 weeks of daily L-dopa
including the substantia nigra, in any animal. Signal enhancement
was observed in structures lacking a BBB, namely the pituitary/
hypothalamus region (pit/hyp), in addition to the sagittal sinus and
jaw muscles, thus serving as an internal control of Gd-DTPA delivery
(Fig.1). A region of interest (ROI) quantitative analysis (see Methods),
confirmedan intact BBBin thebasalgangliain all sixanimals.Oneway
ANOVA across brain regions showed that there were no significant
differences between images of the caudate nucleus (Cd), putamen
(Put), SN or occipital cortex (OccCx) with either the serial gradient
echo sequences (F23,3=1.27; pN0.3) or the high resolution (0.65 mm
isotropic) T1-weighted sequence (F23,3=1.40; pN0.25) whereas
there were highly significant differences between the Cd, Put, SN or
OccCx and either jaw muscle, pit/hyp or sagittal sinus, as expected
(Figs. 2A and B).
This is the first in vivo demonstration of the integrity of the BBB
in parkinsonian primates exhibiting L-dopa-induced dyskinesia.
The induction of dyskinesia by the administration of daily high
dose L-dopa over several months to MPTP lesioned, parkinsonian
primates, did not lead to a leaking BBB. It is conceivable that in the
case of a disrupted BBB, this could lead to high and uncontrolled
levels of L-dopa entering the brain following systemic L-dopa
therapy, further exacerbating non-physiological synaptic release of
dopamine (Olanow et al., 2004; Westin et al., 2006). Also, the BBB is
usually impermeable to carbidopa, a peripheral L-dopa decarboxylase
inhibitor, and if disrupted and rendered permeable, this could
compromise physiological L-dopa decarboxylation in the brain (Carvey
et al., 2005). Finally, in gene therapy, a dysfunctional BBB could
possibly result in a different distribution of secreted gene products
(Isacson and Kordower, 2008) or in the case of cell transplantation,
exposure to immune factors and rejection (Isacson and Kordower,
Fig. 2. Quantitative results following injection of Gadolinium-DTPA in dyskinetic monkeys show an intact BBB of the basal ganglia. (A) Averaged plot across all animals showing the
effects of 0.3 mmol/kg Gd-DTPA as a function of time using serial gradient echo imaging with a flip angle alpha of 25° and short TE (TR/TE=235/4.5 ms) with 30 s temporal
resolution. Injections were made during serial imaging for comparison effects and are shown in the sagittal sinus vein (Sag sinus), the pituitary/hypothalamic region (Pit/Hyp) and
the substantia nigra(SN). Thereis no increase in the SNaside froma small contribution attributable tothe intrinsic blood volume.(B) Barplotshowingthe averagesacrossall animals
for signal enhancement at an average of 16 min after GD-DTPA injection using a high resolution (0.65 mm isotropic) T1-weighted sequence (TR/TI/TE=1910/1100/3.1 ms). The
regions shown are the same as in (A) but also include putamen (Put), caudate (Cd), jaw muscle (muscle), and occipital cortex (OccCx) as a control gray matter region. One way
ANOVA across brain regions showed that there were no significant differences between images of the Cd, Put, SN and OccCx with either the gradient echo data in (A) (F23,3=1.27;
pN0.3) or in (B) (F23,3=1.40; pN0.25). As expected, there were highly significant differences between the latter four regions and either muscle, pit/hyp or sagittal sinus.
A. Astradsson et al. / Neurobiology of Disease 35 (2009) 348–351
2008). The findings of an intact BBB in the present study may Download full-text
therefore have implications for existing and new therapies for PD.
BBB integrity has also been studied in clinical and experimental
models of Parkinson's disease. For example, a PET study of
verapamil uptake in the brain demonstrated a decreased function of
the P-glycoprotein (P-gp) transporter in the BBB of PD patients
(Kortekaas et al., 2005). Findings from a rodent study have suggested
that L-dopa-induced dyskinesia may be associated with a compro-
mised BBB (Westin et al., 2006). Postmortem analysis of 6-OHDA
lesioned rats rendered dyskinetic after a 2 week course of L-dopa,
revealed a BBB with long-term structural changes in the basal ganglia,
particularly in its output regions; the entopeduncular nucleus and the
substantia nigra pars reticulata, as demonstrated by increased
immunostaining for albumin and a reduction in endothelial barrier
antigen (EBA) expression (Westin et al., 2006). However, no external
tracer such as horseradish-peroxide (HRP) was administered (Westin
et al., 2006). HRP is a glycoprotein with a small molecular weight that
produces a fluorimetric or luminescent derivative of the labeled
molecule, and can be administered intravenously, subsequently
allowing it to be histologically detected and quantified and has been
widely used as a histological marker of BBB integrity (Harris et al.,
2002). EBA is rodent specific and may not be applicable to the clinical
setting (Sternberger and Sternberger, 1987). Finally, Westin et al.
found a high rate of cell proliferation in the basal ganglia and newly
born microvessels (Westin et al., 2006). These observations were
specifically associated with the development of dyskinesia and not L-
dopa treatment alone (Westin et al., 2006).
We have developed a slow, progressive model of L-dopa-induced
dyskinesia, by the administration of L-dopa over several months, to
chronically MPTP lesioned non-human primates (Jenkins et al., 2004;
Sanchez-Pernaute et al., 2007). Whereas Parkinson's disease patients
usually develop dyskinesias only after several years of L-dopa
treatment, we have used substantially higher doses of L-dopa than
clinicallyapplied, for the induction of dyskinesias in primates, in order
to shorten the length of the induction phase (Sanchez-Pernaute et al.,
2007). Nevertheless, this model may more realistically simulate the
progressive pathogenesis of dyskinesia in clinical PD, than current
rodent models of L-dopa-induced dyskinesias do.
MRI studies with Gd-DTPA enhancement are widely used to detect
BBB changes in a variety of neurological conditions, such as multiple
sclerosis (Kermode et al.,1990; Soonet al.,2007), includingsubtle BBB
changes associated with non-enhancing lesions (Soon et al., 2007), as
well as stroke (Wardlaw et al., 2008), intracerebral neoplasm
(Ludemann et al., 2002) and head injury (Beaumont et al., 2000).
We have chosen to use Gd-DTPA MRI to detect BBB integrity in our in
vivo model of L-dopa-induced dyskinesia of primates, as it is a well
established, clinically useful marker to evaluate BBB integrity. It has
the advantage over HRP and albumin, that it can be readily used in
vivo, whereas the analysis of HRP and albumin leakage is suitable for
postmortem studies. Furthermore, Gd-DTPA is a much smaller
molecule than both albumin and HRP and therefore should be more
sensitive to subtle BBB permeability changes (Harris et al., 2002;
Schmiedl et al., 1991). Notably, if a molecule as large as albumin can
leak across the BBB it must indicate a very high permeability surface
area product (Westin et al., 2006). Giventhatwe could not see leakage
of a small molecule like Gd-DTPA in the present study, it must mean
that there was minimal opening of the BBB in our model.
While we found no evidence of BBB damage after chronic L-dopa
administration in our study, it cannot be excluded that other
microvascular effects of L-dopa treatment might have occurred in
this model. For example, the possibility of L-dopa-induced micro-
vascular proliferation and increased cerebral blood volume cannot be
excluded (Westin et al., 2006). Furthermore, it cannot be excluded, as
was recently demonstrated, that L-dopa treatment is associated with
increased cerebral blood flow and dissociation of cerebral blood flow
and metabolism in the striatum (Hirano et al., 2008).
In conclusion, in primates rendered parkinsonian with MPTP,
repeated L-dopa treatmentordyskinesia did not disrupt the BBB in the
basal ganglia, as detected with MRI neuroimaging using Gd-DTPA.
These findings contrast with studies of the BBB in rodent models of L-
This work was supported by the US National Institutes of Health
NINDS Udall Parkinson's Disease Research Center of Excellence (P50
NS39793), The Michael Stern Foundation, the Consolidated Anti-
Aging Foundation, the Orchard Foundation, and an NIH base grant to
NEPRC (RR00168). The authors declare no financial conflict of interest.
We thank Angela Carville and Shannon Luboyeski for veterinary
Beaumont, A., et al., 2000. The permissive nature of blood brain barrier (BBB) opening
in edema formation following traumatic brain injury. Acta. Neurochir. Suppl. 76,
Bezard, E., et al., 2003. Attenuation of levodopa-induced dyskinesia by normalizing
dopamine D3 receptor function. Nat. Med. 9, 762–767.
Brownell, A.L., et al., 1998. Combined PET/MRS brain studies show dynamic and long-
term physiological changes in a primate model of Parkinson disease. Nat. Med. 4,
Carvey, P.M., et al., 2005. 6-Hydroxydopamine-induced alterations in blood–brain
barrier permeability. Eur. J. Neurosci. 22, 1158–1168.
Fahn, S., 2003. Description of Parkinson's disease as a clinical syndrome. Ann. N. Y. Acad.
Sci. 991, 1–14.
Harris, N.G., et al., 2002. MRI measurement of blood–brain barrier permeability
following spontaneous reperfusion in the starch microsphere model of ischemia.
Magn. Reson. Imaging 20, 221–230.
Hirano, S., et al., 2008. Dissociation of metabolic and neurovascular responses to
levodopa in the treatment of Parkinson's disease. J. Neurosci. 28, 4201–4209.
Imbert, C., et al.,2000. Comparison of eight clinical rating scales used for the assessment
of MPTP-induced parkinsonism in the Macaque monkey. J. Neurosci. Methods 96,
Isacson, O., Kordower, J.H., 2008. Future of cell and gene therapies for Parkinson's
disease. Ann. Neurol. 64 (Suppl 2), S122–S138.
Jenkins, B.G., et al., 2004. Mapping dopamine function in primates using pharmacologic
magnetic resonance imaging. J. Neurosci. 24, 9553–9560.
Kermode, A.G., et al., 1990. Heterogeneity of blood–brain barrier changes in multiple
sclerosis: an MRI study with gadolinium-DTPA enhancement. Neurology 40,
Kortekaas, R., et al., 2005. Blood–brain barrier dysfunction in parkinsonian midbrain in
vivo. Ann. Neurol. 57, 176–179.
Ludemann, L., et al., 2002. Pharmacokinetic modeling of Gd-DTPA extravasation in brain
tumors. Invest. Radiol. 37, 562–570.
Obeso, J.A., et al., 2000. Pathophysiology of levodopa-induced dyskinesias in
Parkinson's disease: problems with the current model. Ann. Neurol. 47,
S22–S32. Discussion S32–S34.
Olanow, C.W., et al., 2004. Levodopa in the treatment of Parkinson's disease: current
controversies. Mov. Disord. 19, 997–1005.
Olanow, C.W., Obeso, J.A., 2000. Preventing levodopa-induced dyskinesias. Ann. Neurol.
47, S167–S176. Discussion S176–S178.
Pearce, R.K., et al., 1995. Chronic L-DOPA administration induces dyskinesias in the 1-
methyl-4- phenyl-1,2,3,6-tetrahydropyridine-treated common marmoset
(Callithrix Jacchus). Mov. Disord. 10, 731–740.
Sanchez-Pernaute, R., et al., 2007. In vivo evidence of D3 dopamine receptor
sensitization in parkinsonian primates and rodents with L-DOPA-induced dyskine-
sias. Neurobiol. Dis. 27, 220–227.
Schmiedl, U.P., et al., 1991. MRI of blood–brain barrier permeability in astrocytic
gliomas: application of small and large molecular weight contrast media. Magn.
Reson. Med. 22, 288–292.
Soon, D., et al., 2007. Quantification of subtle blood–brain barrier disruption in non-
enhancing lesions in multiple sclerosis: a study of disease and lesion subtypes.
Mult. Scler. 13, 884–894.
Sternberger, N.H., Sternberger, L.A., 1987. Blood–brain barrier protein recognized by
monoclonal antibody. Proc. Natl. Acad. Sci. U. S. A. 84, 8169–8173.
Wardlaw, J.M., et al., 2008. Changes in background blood–brain barrier integrity
between lacunar and cortical ischemic stroke subtypes. Stroke 39, 1327–1332.
Westin, J.E., et al., 2006. Endothelial proliferation and increased blood–brain barrier
permeability in the basal ganglia in a rat model of 3,4-dihydroxyphenyl-L-alanine-
induced dyskinesia. J. Neurosci. 26, 9448–9461.
A. Astradsson et al. / Neurobiology of Disease 35 (2009) 348–351