1. health and fitness
2. disease

Cell Stem Cell
In Translation
Graft-Induced Dyskinesias
in Parkinson’s Disease: What Is It All About?
Roger A. Barker1,* and Wei-Li Kuan1
1Brain Repair Centre and Department of Neurology, University of Cambridge, Cambridge CB2 0PY, UK
*Correspondence: [email protected]
DOI 10.1016/j.stem.2010.07.003
The treatment of Parkinson’s disease with grafts of fetal ventral mesencephalic tissue has shown some
success, but can result in graft-induced dyskinesias (GIDs). Recently in Science Translational Medicine,
Politis et al. (2010) demonstrate that GIDs may originate from serotoninergic neurons that are cografted in
these transplants.
Parkinson’s disease (PD) is a common termed graft-induced dyskinesias (GIDs)
neurodegenerative disorder of the central (reviewed in Goya et al., 2007). GIDs are
nervous system (CNS). PD is character- phenomenologically different from LIDs
ized by the loss of neurons at a number and are not obviously related to L-dopa
of CNS and non-CNS sites, which is medication and persist in the absence of
accompanied by the formation of a-synu- anti-PD medication. The etiology of GIDs
clein-positive Lewy bodies (Braak
et al., 2006). The etiology of this
condition remains obscure in the
Normal
A
majority of cases, although nearly 50
DA
5HT
years ago it was recognized that
patients experience degeneration of
the nigrostriatal dopaminergic pathway, and that using dopaminergic
DAT
5-HTT
drugs to substitute for this loss can
ameliorate many of the symptoms in
the short term. However, these
agents are not curative and with time
complications arise with their use,
Postsynaptic
c dopamine
dopam
receptors
including the development of unpreGID
B
dictable motor responses (on-off phenomena) and the genesis of
DA
5HT
abnormal, involuntary dyskinetic
movements (L-dopa-induced dyskinesias, LIDs). As a result, more definDAT
T
5-HTT
5
itive reparative strategies have been
sought, one of which involves using
grafts containing fetal dopaminergic
neuroblasts derived from the developing human ventral mesencephalon
Postsynaptic
c dopamine
dopam
receptors
(VM). These transplants, when placed
into the striatum of patients with PD,
Buspirone Treatment
C
can produce clear clinical benefits,
albeit inconsistently (Goya et al.,
DA
5HT
2007). The heterogeneity of patient
responses has always been an area
of concern, but an additional serious
DAT
AT
5HT1A
problem with this approach was iden5-HTT
5-H
tified in the early part of this century
during two double-blind, placebobuspirone
controlled trials. These studies
revealed that the grafts themselves
Postsynaptic dopamine receptors
could induce dyskinetic movements,
148 Cell Stem Cell 7, August 6, 2010 ª2010 Elsevier Inc.
has been the subject of intense debate,
and a predominant hypothesis under
examination names dopamine (DA) as
the causative agent and suggests that
the grafts either release too much dopamine or release it unevenly across the
striatal complex (Figure 1; Ma et al.,
2002). Recently, however, attention
has turned toward the population of
serotoninergic or 5-hydroxytryptamine (5HT) neurons within the graft,
which arise developmentally just
caudal to the VM and are therefore
also transplanted.
This suggestion that the 5HT cells
within the VM transplant may be important has its origins in recent work
on LIDs, where it has been shown
that exogenously supplied L-dopa
Figure 1. Proposed Mechanism
of Buspirone Attenuation of GID
(A) In the graft of all patients there is a mixture
of dopaminergic and serotonergic neurons,
the release of dopamine (orange dots) can
be regulated by presynaptic dopamine transporters (DAT), and the release of 5HT (blue
dots) by serotonin transporters (5-HTT). In
patients without GIDs, these neurons take
up their own neurotransmitter and there is no
real dialog between them.
(B) In patients with GIDs, the serotonergic
terminals hyperinnervate the striatum in the
grafted patient, and large amounts of 5HT
are released, which may be able to reverse
or alter the capacity of the DAT, thereby
causing a greater release of dopamine from
the grafted dopaminergic nerve terminals. In
addition, the 5HT transporter can also take
up dopamine and then release it as a false
transmitter from the 5HT nerve terminal.
(C) Buspirone, a selective 5HT1A agonist, was
used to activate the presynaptic 5HT1A
receptor to dampen the release of serotonin
from serotonergic terminals. Reduced levels
of 5HT leads to a change in DAT and 5-HTT
activity and restores more regulated dopamine release from the grafted dopaminergic
neurons.
Cell Stem Cell
In Translation
can be converted, stored, and released
by the serotonergic terminals (Tanaka
et al., 1999) but not inactivated because
such cells lack dopamine transporters.
Nevertheless, targeting the presynaptic
inhibitory serotonergic receptors can
attenuate LIDs in a rat model of PD (Mun˜oz et al., 2008), and selective lesioning
of serotonergic afferent projections to
the striatum can almost completely
abolish their development (Carta et al.,
2007). It is therefore logical to assume
that some similar process may be taking
place in the VM transplants (Mendez
et al., 2008), and as such these cells could
be at least partially responsible for the
development of GID. In a new study published recently in Science Translational
Medicine (Politis et al., 2010), the authors
use PET imaging and the specific serotonin transporter ligand, 11C-DASB, to
explore this possibility in two VM-grafted
PD patients that exhibit GIDs. Their
results indicate that these patients have
a 5HT hyperinnervation of the grafted
striatum, which in itself is not surprising
given what we already know about the
pathology of such grafts. However, the
authors extend their findings to show
that a specific 5HT1A agonist, buspirone,
lessens the GIDs without any worsening
of the patients’ underlying Parkinsonism.
This result suggests that the drug works
relatively selectively by activating the
presynaptic 5HT autoreceptor (Figure 1),
which in turn leads to a reduction in dopamine release from the transplanted
neurons.
This hypothesis is intriguing, yet it
remains unclear how exactly this dialog
between the serotonin and dopaminergic
neurons leads to GIDs. For example, one
would anticipate that the grafted dopaminergic neurons should take up any
excessive dopamine released within the
transplant. However, the authors propose
that the defect stems from changes in how
the dopamine transporter behaves during
serotoninergic hyperinnervation. There is
evidence from other studies to suggest
that the dopamine transporter can change
its activity and direction of dopamine flux
as a function of 5HT, but activity modulation of this sort has not been shown in
this study in these patients. Furthermore,
it is not clear why only some transplanted
patients develop GIDs, given that all VM
grafts contain a proportion of developing
5HT neurons. Indeed, even in the two
cases reported by Politis and colleagues,
the extent of GIDs is similar despite the
presence of different levels of 5HT reinnervation within the striatum. Finally, according to the authors’ model, it would be
expected that GIDs should worsen with
L-dopa therapy, and yet this pattern has
not been observed thus far.
Overall, the findings reported in this
paper highlight the importance of nondopaminergic networks in the genesis of
dyskinesias, and therefore attention
must be paid to this relationship in new
trials of fetal VM transplants for PD. In
contrast, these findings may present
a less serious issue for the world of stem
cell transplantation in PD, given that the
goal of the field has always been to produce a relatively uniform population of
true nigral dopaminergic neurons.
Achieving this goal seems increasingly
likely, although no consensus yet exists
as to the definition of a true nigral dopaminergic neuron (Li et al., 2008). Nevertheless, multipotent stem cell sources can
be engineered to selectively generate
specific cell fates and the current data
from Lindvall and colleagues suggests
that it will be important to avoid the
genesis of 5HT neurons in this setting.
However, it is likely that Politis et al.
have only partly explained the source of
GIDs and that the location of the transplanted dopaminergic neurons within the
striatal complex and their local innerva-
tion capabilities also contribute to the
genesis of these abnormal movements
(Carlsson et al., 2006).
Finally, it is worth noting that this study
shows that dopaminergic cell transplants
produce benefits on motor control that
may take years to mature, as is evident
by the reduction in UPDRS more than
13–16 years after grafting. Thus, judging
the success or otherwise of (stem) cellular
repair approaches to PD will, and cannot,
be ultimately achieved via studies with
short follow-up periods.
REFERENCES
Braak, H., Bohl, J.R., Mu¨ller, C.M., Ru¨b, U., de Vos,
R.A., and Del Tredici. Stanley Fahn, K. (2006).
Lecture 2005. Mov. Disord. 21, 2042–2051.
Carlsson, T., Winkler, C., Lundblad, M., Cenci,
M.A., Bjo¨rklund, A., and Kirik, D. (2006).
Neurobiol. Dis. 21, 657–668.
Carta, M., Carlsson, T., Kirik, D., and Bjo¨rklund, A.
(2007). Brain 130, 1819–1833.
Goya, R.L., Kuan, W.L., and Barker, R.A. (2007).
Expert Opin. Biol. Ther. 7, 1487–1498.
Li, J.Y., Christophersen, N.S., Hall, V., Soulet, D.,
and Brundin, P. (2008). Trends Neurosci. 31,
146–153.
Ma, Y., Feigin, A., Dhawan, V., Fukuda, M., Shi, Q.,
Greene, P., Breeze, R., Fahn, S., Freed, C., and
Eidelberg, D. (2002). Ann. Neurol. 52, 628–634.
Mendez, I., Vin˜uela, A., Astradsson, A., Mukhida,
K., Hallett, P., Robertson, H., Tierney, T., Holness,
R., Dagher, A., Trojanowski, J.Q., and Isacson, O.
(2008). Nat. Med. 14, 507–509.
Mun˜oz, A., Li, Q., Gardoni, F., Marcello, E., Qin, C.,
Carlsson, T., Kirik, D., Luca, M.D., Bjo¨rklund, A.,
Bezard, E., and Carta, M. (2008). Brain 131,
3380–3394.
Politis, M., Wu, K., Loane, C., Quinn, N.P., Brooks,
D.J., Rehncrona, S., Bjorklund, A., Lindvall, O., and
Piccini, P. (2010). Science Translational Medicine.
Published online June 23, 2010. 10.1126/scitranslmed.3000976.
Tanaka, H., Kannari, K., Maeda, T., Tomiyama, M.,
Suda, T., and Matsunaga, M. (1999). Neuroreport
10, 631–634.
Cell Stem Cell 7, August 6, 2010 ª2010 Elsevier Inc. 149