Lewy bodies in grafted neurons in subjects with Parkinson's disease suggest host-to-graft disease propagation
Two subjects with Parkinson's disease who had long-term survival of transplanted fetal mesencephalic dopaminergic neurons (11-16 years) developed alpha-synuclein-positive Lewy bodies in grafted neurons. Our observation has key implications for understanding Parkinson's pathogenesis by providing the first evidence, to our knowledge, that the disease can propagate from host to graft cells. However, available data suggest that the majority of grafted cells are functionally unimpaired after a decade, and recipients can still experience long-term symptomatic relief.
Lewy bodies in grafted neurons
in subjects with Parkinson’s
disease suggest host-to-graft
, Elisabet Englund
, Janice L Holton
, Denis Soulet
, Andrew J Lees
, Tammaryn Lashley
Niall P Quinn
, Stig Rehncrona
, Anders Bjo
, Tamas Revesz
, Olle Lindvall
Two subjects with Parkinson’s disease who had long-term
survival of transplanted fetal mesencephalic dopaminergic
neurons (11–16 years) developed a-synuclein–positive Lewy
bodies in grafted neurons. Our observation has key implications
for understanding Parkinson’s pathogenesis by providing the
ﬁrst evidence, to our knowledge, that the disease can propagate
from host to graft cells. However, available data suggest that
the majority of grafted cells are functionally unimpaired after a
decade, and recipients can still experience long-term
Two sham surgery–controlled trials of neural transplantation in
Parkinson’s disease did not reach their primary endpoints
previous open-label trials with grafts of fetal ventral mesencephalic
tissue reported long-lasting functional beneﬁts in subjects with Par-
. Positron emission tomography (PET) studies
showed that grafted neurons were functionally integrated and released
dopamine for more than 10 years after surgery
. In postmortem
neuropathological studies performed 3–4 years after transplantation,
large numbers of dopaminergic neurons were found in the grafts
We now report that grafted cells can survive up to at least 16 years in
subjects with Parkinson’s disease. Large numbers of transplanted
dopaminergic neurons were found in two subjects who had undergone
bilateral implantation of fetal mesencephalic tissue into the putamen
(subject 3 in the Lund series; left graft 16 years before death, right graft
12 years before death) or both putamen and caudate nucleus (subject
8; left graft 13 years before death, right graft 11 years before death)
Graft survival was conﬁrmed by clinical improvement at 5 months up
to at least 3 years after surgery in subject 3 (ref. 6).
Both subjects died from causes unrelated to grafting (Supplemen-
tary Methods online). In their substantiae nigrae, the subjects had
histopathological changes characteristic of Parkinson’s disease: severe
© 2008 Nature Publishing Group http://www.nature.com/naturemedicine
Figure 1 Surviving dopaminergic neurons labeled with antibody to tyrosine
hydroxylase in a graft transplanted 16 years before death in subject 3.
(a–c) Surviving cells mainly exist in the periphery of the graft (a)with
classical morphology of dopaminergic neurons (b), extending long
processes into the host striatum (c). (d) Luxol fast blue staining shows good
morphology of grafted neurons and myelinated axons. (e) Girk2-positive
dopaminergic neurons are shown from a graft. Scale bars, 500 mmin
a; 100 mminb–e.
Received 3 December 2007; accepted 5 March 2008; published online 6 April 2008; doi:10.1038/nm1746
Neuronal Survival Unit, Wallenberg Neuroscience Center, Department of Experimental Medical Science, 221 84 Lund, Sweden.
Division of Neuropathology, Lund
University Hospital, 221 85 Lund, Sweden.
Queen Square Brain Bank for Neurological Disorders, Department of Molecular Neuroscience, UCL Institute of Neurology,
University College London, Queen Square, London, WC1N 3BG, UK.
Division of Neurology, Lund University Hospital, 221 85 Lund, Sweden.
Sobell Department of
Motor Neuroscience and Movement Disorders, UCL Institute of Neurology, University College London, Queen Square, London, WC1N 3BG, UK.
Neurosurgery, Lund University Hospital, 221 85 Lund, Sweden.
Neurobiology Unit and
Section of Restorative Neurology, Wallenberg Neuroscience Center,
221 84 Lund, Sweden.
These authors are senior authors. Correspondence should be addressed to P.B. (email@example.com).
NATURE ME DICI NE ADVANCE ONLINE PUBLICATION 1
loss of pigmented neurons and a-synuclein– and ubiquitin-positive
Lewy bodies and Lewy neurites in surviving neurons. Cortical regions
showed diffuse cortical Lewy body–type pathology in both subjects.
Numerous tyrosine hydroxylase–immunoreactive, presumed dopami-
nergic neurons were found primarily at the periphery of the trans-
plants in subject 3 (Fig. 1a–c). In both subjects, the majority of
dopaminergic neurons had long processes, forming dense networks
in the grafts and surrounding striatum (Fig. 1). Many neurons
contained neuromelanin and possessed myelinated axons (Fig. 1b and
Fig. 2a,b,e). The tyrosine hydroxylase–immunoreactive neurons,
grafted at different time points, were similar morphologically (Sup-
plementary Fig. 1 online). Some grafted neurons were immuno-
positive for Girk2, a marker of dopaminergic neurons in substantia
nigra pars compacta (Fig. 1e and Supplementary Fig. 2 online),
whereas others expressed calbindin, a marker of ventral tegmental area
and substantia nigra dorsal tier pars compacta neurons (Supplemen-
tary Fig. 2). Iba1-positive microglia accumulated around the grafts,
but CD68 immunohistochemistry did not indicate strong activation
(Supplementary Fig. 3 online). Quantitative analysis in subject 3
showed 12,100–29,500 tyrosine hydroxylase–immunoreactive neurons
per injection track on the ﬁrst grafted side and 14,400–27,600 tyrosine
hydroxylase–immunoreactive cells per injection track on the second
side, indicating that the numbers of long-surviving (12–16 years)
dopaminergic neurons were of the same magnitude as in subjects
surviving 18 months to 4 years after transplantation
In both cases, and similarly to surviving neurons in the substantia
nigra pars compacta, grafted neurons, some of which were clearly
pigmented, contained characteristic a-synuclein– and ubiquitin-
positive Lewy bodies and Lewy neurites (Fig. 2a,b and Supplementary
Fig. 4 online), which were also labeled by an antibody recognizing
a-synuclein phosphorylated at Ser129 (Fig. 2). A small number of
grafted neurons showed punctate cytoplasmic tau reactivity (pre-
tangle) with phospho-tau immunohistochemistry (Fig. 2e).
We found that 40% of tyrosine hydroxylase–positive cells contained
detectable amounts of a-synuclein in the 12-year-old implants, as did
80% of the tyrosine hydroxylase–positive cells in the 16-year-old grafts
in subject 3. Dopaminergic neurons in the host midbrain had similar
expression of a-synuclein (Fig. 2c,d). The difference in a-synuclein
expression between the grafts in subject 3 supports the notion
that increased intracellular a-synuclein is time- or age-dependent,
which is consistent with the fact that age could be a risk factor for
The most striking ﬁnding in both cases was the presence of Lewy
bodies and Lewy neurites in the long-surviving grafted dopaminergic
© 2008 Nature Publishing Group http://www.nature.com/naturemedicine
Figure 2 a-synuclein–positive Lewy bodies in
host substantia nigra and grafted dopaminergic
neurons in subject 3 (a–d) and subject 8 (e).
(a) Classical Lewy bodies and Lewy neurites in
neurons of the substantia nigra (left) and in the
grafts (middle and right) are immunoreactive for
a-synuclein in parafﬁn-embedded tissue sections.
(b) Lewy bodies in the grafts are immunoreactive
for ubiquitin in parafﬁn-embedded tissue
sections. Arrows (left) point to dopaminergic
neurons containing a large number of pigmented
granules. (c) Double immunolabeling shows
colocalization of tyrosine hydroxylase (TH, green)
and a-synuclein (red) in some dopaminergic
neurons of the substantia nigra (arrowheads). One
dopaminergic neuron with good morphology
(arrows) does not contain detectable a-synuclein.
(d) Double immunolabeling shows colocalization
of tyrosine hydroxylase (green) and a-synuclein
(red) in a graft (arrowheads). One dopaminergic
neuron (arrows) does not contain detectable a-
synuclein. (e) Histological sections of grafts from
subject 8 stained by immunohistochemistry for
neuroﬁlament (left), human tau (middle) and
phosphorylated S129 on a-synuclein (right).
Neuroﬁlament staining shows numerous axons
and neurons containing pigmented granules in a
graft (left). Neurons with punctate tau positivity
are suggestive of pretangles (middle). A typical
Lewy body stained with an antibody to phospho–
a-synuclein is shown in a grafted neuron (right).
Scale bars, 40 mm, except in e (middle and
right), 15 mm. Grafting procedures in both
subjects were performed with informed consent.
Informed consent was also obtained for
postmortem investigation, including for the use of
brain tissue for research in both cases. Grafting
and all procedures related to postmortem tissue
handling were approved by the Regional Ethical
Review Board in Lund, Sweden and by the Joint
Research Ethics Committee of the National
Hospital for Neurology and Neurosurgery and UCL Institute of Neurology, UK. Brain donations to the Queen Square Brain Bank for Neurological Disorders are
approved by a London Multi-Centre Research Ethics Committee.
Ubiquitin Ubiquitin TH
2 ADVANCE ONLINE PUBLICATION NATURE ME DICI NE
neurons that were morphologically indistinguishable from those seen
in the substantia nigra pars compacta neurons in Parkinson’s disease.
The Lewy bodies in grafted cells were a-synuclein and ubiquitin
immunoreactive and were also stained with an antibody recognizing
a-synuclein phosphorylated at Ser129, strongly suggesting the pre-
sence of disease-related, post-translationally modiﬁed and aggregated
The mechanisms leading to the initiation and spread of a-synuclein
pathology in Parkinson’s disease are not well understood, and those
underlying similar pathological changes in grafts are also obscure.
Because the dopaminergic neurons were grafted into an ectopic site,
they may have been exposed to an unfavorable microenvironment.
Studies in rodents have shown that grafted dopaminergic neurons
can receive corticostriatal afferents
, and glutamatergic excitotoxic
damage might have occurred in the grafts. Exposure to amphetamine
or the toxins MPTP and rotenone can lead to an unfavorable
microenvironment and upregulation of a-synuclein in nigral neurons,
promoting its aggregation
. Moreover, it has been suggested that
Parkinson’s disease is coupled to impaired neurotrophic support
and grafts may lack appropriate trophic signaling in the striatum
in Parkinson’s disease. Finally, microglial activation occurs in the
basal ganglia in Parkinson’s disease and, combined with the
low-level inﬂammation indicated by graft-related microglia (Supple-
mentary Fig. 3), could contribute to a-synuclein aggregation in
Stereotypic topographical progression of Lewy body pathology has
been suggested to be a characteristic feature in Parkinson’s disease
On the basis of experimental data, including those showing the
acceleration of amyloid deposition after exogenous seeding
hypothesis of ‘permissive templating’ has been proposed to explain
disease propagation along neuronal pathways in neurodegenerative
diseases, including Parkinson’s disease
. Our observations may pro-
vide support for such a hypothesis, implicating a ‘prion-like’ mechan-
ism. One could, therefore, speculate that the a-synuclein aggregation
and deposition observed in the transplanted dopaminergic neurons
was triggered by misfolded a-synuclein in the host, which was
transmitted into grafted cells.
Increased a-synuclein abundance and the formation of Lewy bodies
could be detrimental to grafted dopaminergic neurons and potentially
limit the duration of efﬁcacy of cell replacement therapy. Indeed, an
inverse correlation has been shown between levels of a-synuclein and
tyrosine hydroxylase in substantia nigra dopaminergic neurons
F-dopa and raclopride PET imaging, however, shows that mesen-
cephalic grafts can still synthesize and release normal amounts of
dopamine 10 years after transplantation
. Therefore, even though
some grafted dopaminergic neurons undergo pathological changes
similar to those found in Parkinson’s disease, the majority do not seem
functionally impaired, and recipients may still experience long-term
symptomatic relief after a decade.
Note: Supplementary information is available on the Nature Medicine website.
This work was supported by grants from the Swedish Research Council, Swedish
Parkinson Foundation, the Nordic Center of Excellence on Neurodegeneration
and The Strong Research Environment of the Swedish Research Council
(NeuroFortis). The Queen Square Brain Bank is supported by the Reta Lila
Weston Institute of Neurological Studies and the Progressive Supranuclear Palsy
(Europe) Association. T.R. and J.L.H. are supported by grants from the
Parkinson’s Disease Society, UK, the Alzheimer’s Research Trust and the Sarah
Matheson Trust. The authors wish to thank B.-M. Lindberg and A. Persson
for their excellent technical support. E.E. and J.L.H. contributed equally as the
J.-Y.L. designed and performed the detailed morphological analysis of most of the
material from subject 3. E.E. and J.L.H. did autopsy and routine neuropathology
of subjects 3 and 8, respectively. J.L.H. and T.R. designed and T.L. performed the
detailed morphological analysis of subject 8. A.B. provided expertise in neural
transplantation. D.S. provided expertise regarding graft reconstruction, imaging
and microglia staining and also generated ﬁgures. P.H. assisted in the care of
subject 3 and provided data related to clinical follow-up. H.W. and N.P.Q. took
care of subjects 3 and 8, respectively. A.J.L. provided clinical evaluation for
subject 8. S.R. operated on both subjects. P.B. dissected and prepared tissue for
both surgeries and participated in the morphological assessment of subject 3.
O.L. took care of subject 3 and headed the clinical transplantation program.
J.-Y.L., J.L.H., T.R., O.L. and P.B. wrote the manuscript. All authors gave input
to the manuscript.
Published online at http://www.nature.com/naturemedicine
Reprints and permissions information is available online at http://npg.nature.com/
1. Olanow, C.W. et al. Ann. Neurol. 54, 403–414 (2003).
2. Freed, C.R. et al. N. Engl. J. Med. 344, 710–719 (2001).
rklund, A. et al. Lancet Neurol. 2, 437–445 (2003).
4. Piccini, P. et al. Nat. Neurosci. 2, 1137–1140 (1999).
5. Mendez, I. et al. Brain 128, 1498–1510 (2005).
6. Hagell, P. et al. Brain 122, 1121–1132 (1999).
7. Kordower, J.H. & Sortwell, C.E. Prog. Brain Res. 127, 333–344 (2000).
8. Chu, Y. & Kordower, J.H. Neurobiol. Dis. 25, 134–149 (2007).
9. Anderson, J.P. et al. J. Biol. Chem. 281, 29739–29752 (2006).
10. Doucet, G. et al. Exp. Neurol. 106,1–19(1989).
11. Vila, M. et al. J. Neurochem. 74, 721–729 (2000).
12. Mogi, M. et al. Neurosci. Lett. 270, 45–48 (1999).
13. Sigurdsson, E.M., Wisniewski, T. & Frangione, B. Trends Mol. Med. 8, 411–413
14. Braak, H. et al. Neurobiol. Aging 24, 197–211 (2003).
15. Meyer-Luehmann, M. et al. Science 313, 1781–1784 (2006).
16.Hardy,J.Biochem. Soc. Trans. 33, 578–581 (2005).
© 2008 Nature Publishing Group http://www.nature.com/naturemedicine
NATURE ME DICI NE ADVANCE ONLINE PUBLICATION 3