For the brave patients who elected to have foreign fetal neurons implanted in their brain as an experimental treatment for Parkinson disease (PD), there must have been many agonizing thoughts. Apart from the obvious will-it-help-or-hurt questions, those volunteers must have wondered how long the grafts would hold up. The research community has wondered the same. Now it turns out that such transplants may be remarkably stable. Three separate studies, reported as brief communications in the April 6 Nature Medicine online, describe grafts surviving as long as 9-16 years in six different patients, even without immunosuppressive medication. But there is a caveat. Grafts in three patients showed signs of Lewy bodies, intracellular protein aggregates that are a hallmark of Parkinson’s, suggesting that transplanted neurons may be vulnerable to the same pathology that caused the disease in the first place. Given that the transplanted neurons were too young to develop this pathology on their own, that finding also suggests that Lewy body pathology is not cell-autonomous but induced by the microenvironment of the brain.

It is not yet clear how these findings will affect transplant research, or even whether these Lewy body-like entities have any impact on the overall efficacy of the grafts. Curt Freed, from the University of Colorado Health Science Center, and a pioneer of PD cell transplants, told Alzforum via e-mail that he finds this a positive rather than a cautionary story. Freed was not involved in these studies. “What would a kidney or liver transplant look like 14-16 years after transplant in a patient who did not receive immunosuppression? The kidney or liver would have been destroyed,” he wrote. “I find it remarkable that all three Nature Medicine reports and our experience in Colorado show that dopamine cell transplants survive and function almost indefinitely.”

The studies were led by Patrik Brundin at the Wallenberg Neuroscience Center, Lund, Sweden; Ole Isacson at McLean Hospital, Belmont, Massachusetts; and Jeffery Kordower at Rush University Medical Center, Chicago, Illinois. The European study (two patients) and the Rush study (one patient) both report signs of Lewy bodies in grafted neurons, whereas Isacson’s group found no such pattern in grafts from three different patients.

Jia-Yi Li and colleagues at the Wallenberg Neuroscience Center studied postmortem samples from two PD patients who had twice received grafts of fetal dopaminergic (DA) neurons. The first patient received a graft 16 years before death and then a second graft on the opposite side of the brain four years later. The second patient received the first graft 13 years before death and a second two years later. All grafts were densely positive for tyrosine hydroxylase (TH)-expressing neurons, suggesting good survival. The neurons had long processes and formed dense networks within the graft and the surrounding striatum (in all three studies the grafts were transplanted into the striatum to compensate for lost dopaminergic innervation resulting from damage to DA neurons in the substantia nigra). But Li and colleagues also noticed that in both cases, neurons in the graft contained α-synuclein- and ubiquitin-positive inclusions that have characteristics of Lewy bodies. α-synuclein, normally restricted to presynaptic terminals, also turned up in neuritic processes. In the patient with 12- and 16-year-old grafts, about 40 percent of TH-positive neurons contained detectable α-synuclein in the youngest graft, while about 80 percent of the cells were α-synuclein-positive in the older graft. The findings “support the notion that increased intracellular α-synuclein is time- or age-dependent, which is consistent with the fact that age could be a risk factor for Parkinson’s disease,” write the authors.

Kordower and colleagues report a similar finding from their study of a patient who received a graft 14 years before death. TH immunoreactivity showed strong survival of the DA neurons and innervation into the striatum. But again, some cells also tested positive for α-synuclein- and ubiquitin-positive bodies and also for α-synuclein in neuritic processes. For comparison, the researchers examined postmortem brain tissue from two patients who died four years after receiving grafts. Those tissue samples showed no signs of Lewy bodies. The researchers are now carrying out quantitative analysis to see just what fraction of grafted neurons is affected. “Preliminary counts suggest that more cells in the graft may display these markers than in the host,” Kordower told ARF.

Contrasting these two studies is the data from McLean. First author Ivar Mendez and colleagues carried out postmortem analysis on tissue from two patients who had received grafts nine years before death, and a third patient who had a graft for 14 years. Similar to the other studies, Mendez found that the grafts were well integrated, with TH-positive cells extending into the putamen. But unlike the other two studies, Mendez and colleagues found no signs of Lewy bodies or morphological signs of neurodegeneration. They found no α-synuclein, ubiquitin, or lipofuscin inclusions, and even though immunosuppressant medication was withdrawn six months after the transplant operations, the autopsy showed no major immune reaction to the foreign tissue.

What explains the different findings? “We tried very hard to find Lewy bodies. In fact as a byline to the other stories, Dr. Kordower sent tissues from his case so that we could apply the same techniques, and I did, indeed, find a few cells that had some kind of protein aggregate,” said Isacson in an interview with ARF. He suggested neuroinflammation as one possible explanation for the differences. Whereas Isacson’s group found no signs of inflammation, Kordower and colleagues found intense activation of microglia in the grafts, which vastly exceeded that seen in the patient’s own striatum. Brundin’s groups found that microglia surrounded the grafts, but were not strongly activated. Isacson suggested that differences in transplant technology might partly explain the different data, since solid, or tissue chunk, transplants, the kind used by Kordower’s group, are known to elicit a more extensive immune response (see Freed et al., 2001).

Kordower is not convinced that inflammation due to differences in technique explains the Lewy body pattern, even though Isacson used dispersed cell suspensions rather than the solid tissue method. “Dr. Isacson’s technique and Dr. Brundin’s are virtually identical, so I don’t think that is what’s different,” said Kordower. He also added that lots of things cause inflammation but don’t cause Lewy bodies. “This is a very specific Parkinsonian pathology, and there’s no doubt that it occurs in grafted neurons,” he said.

A separate question that needs to be answered is how representative these few cases are. Freed weighed in on the no-pathology side. “In Colorado, I believe that we have done the largest series of dopamine cell transplants in the world, a total of 61 patients since 1988, and we have seen no protein deposits in dopamine neurons up to 14 years after transplant,” he wrote to ARF. Isacson suggested that individual responses of patients may be part of the explanation for the differences. “It could simply be that we haven’t had a patient who had that kind of reaction,” he said. Kordower added that “this is one of the problems with case studies. They don’t tell you what happens; they tell you what can happen, and this data tells you that Parkinson pathology can happen in grafted neurons.”

What does all this mean for transplant treatments? “It may not be relevant, in a sense. Our patient did very well for 10 years, and that has tremendous value,” said Kordower. There is also some question as to whether the aggregates seen in the grafts are true Lewy bodies. John Trojanowski, University of Pennsylvania, Philadelphia, told ARF that ultrastructural studies would have to be carried out to be sure. Trojanowski, who collaborated with Isacson and is a coauthor on his paper, also said that there are other circumstances when synuclein turns up in the cytoplasm. During development, for example, α-synuclein is normally in the cell body. “Toward the end of term there’s a shift of expression of α-synuclein from the cell body to processes, so why couldn’t that normal developmental occurrence fail?” he suggested (see Galvin et al., 2001). Whether removing fetal tissue and transplanting it into a new environment causes such a failure remains to be tested. Freed also echoed Trojanowski’s caution about the bodies seen in the grafts. “Since the precipitation of α-synuclein and ubiquitin protein is not unique to Parkinson’s, there is no indication that the cells have developed a Parkinson condition,” he wrote.

The Future of Transplants
Though no double-blind trial has shown dopaminergic transplants to be effective in PD patients, there is evidence that patients do well in open-label follow-up, and all the researchers seemed optimistic that transplants offer good therapeutic potential. What form those transplants will take is another question. Therapeutic cloning may yield new approaches, though technical hurdles and ethical objections are holding back that technology. An alternative possibility, which avoids generating embryos, is to reprogram adult cells to assume neural cell fates. In this week’s PNAS, Isacson, collaborator Rudolf Jaenisch at the Whitehead Institute for Biomedical Research, Cambridge, and colleagues demonstrate how this strategy can work to rescue PD-like symptoms in rats.

The researchers capitalized on the recent discovery that expression of a handful of key proteins is all that is needed to turn differentiated adult cells into pluripotent stem cells. Previously, researchers at the University of Kyoto in Japan, had shown that four transcription factors, Oct4, Sox2, Kif4, and c-Myc, can reset the genome to resemble that of pluripotent stem cells (see Takahashi and Yamanaka, 2006). In this study, first author Marius Wernig and colleagues took this strategy a step further, coaxing induced pluripotent stem cells (iPS) derived from rat fibroblasts to form neural precursor cells. After persuading these precursors to form dopaminergic neurons, the researchers grafted the cells into rats missing DA neurons on one side of the brain. These animals have movement problems and tend to rotate on one side when given a stimulant such as amphetamine. However, animals that received grafts showed significantly fewer rotations per hour after eight weeks than sham-operated animals.—Tom Fagan


  1. I think the impression created by the Li and Kordower papers that there is “PD pathology in the transplants ” is unfortunate. The papers give the impression of an “either/or” proposition of "pathology" in the transplant cases, but neither paper provides a clear description of proportion (such as percent) of neurons with “Lewy bodies.”

    The facts are that in the material of Kordower et al., the vast majority of dopamine neurons in each transplant case (containing many thousands of new dopamine neurons) do not have any α-synuclein aggregates, which I have personally studied with primary data and the same staining used in Mendez et al., as stated in this news article by Tom Fagan. We estimate that less than 1 percent of dopamine neurons in the one Kordower et al. case have any protein inclusions. Moreover, as Dr. Trojanowski points out, we do not know if such α-synuclein inclusions are necessarily permanent and direct evidence of PD, or simply a low-level dynamic shift in protein distribution and aggregation that may also occur in the normal brain.

    Even more misleading are the percentages of α-synuclein stained neurons in the Li et al. paper. Fagan correctly cites this on Alzforum: “In the patient with 12- and 16-year-old grafts, about 40 percent of TH-positive neurons contained detectable α-synuclein in the youngest graft, while about 80 percent of the cells were α-synuclein-positive in the older graft.” Here the misleading notion is born (as evidently misunderstood by every commentary on these papers so far) that such staining represents Lewy body pathology. Instead, the vast majority of neurons in the Li et al. case (as in the Kordower et al. case) are not staining for α-synclein inclusions. In the Li et al. case, too, this is not fully quantified. The authors simply show α-synclein protein staining, which is expected and normal, as Dr. Trojanowski points out. α-synuclein is a normal protein with both cellular and synaptic distribution and function.

    In summary, so far all of our cases and Dr. Freed’s seven cases up to 14 years after surgery have no signs of α-synuclein pathology. In transplants surviving for up to 16 years, the papers by Li et al. and Kordower et al. demonstrate that a small fraction of neurons contain α-synuclein stained inclusions. From these data, one cannot conclude that there is a PD process transferred from the patient to the transplant, or that PD is simply caused by a non-cell-autonomous process.

    From a potentially important future treatment perspective, all the evidence actually shows that it is reasonable to assume that similar dopamine neurons obtained from other cell sources in the future, for example, stem cells, could survive without pathology and function for at least 15 years without significant problems generated by the PD of the host.

  2. These studies elegantly demonstrate that transplanted fetal grafts can survive and thrive in human brains for prolonged periods, clearly moving the field in new, exciting directions. However, the view that the presence of α-synuclein in these grafted neurons represents a “host to graft” disease progression needs further scrutiny.

    Though it may seem parsimonious to assume that the accumulation of α-synuclein in these grafts is due to the environment in which the grafts are, a closer inspection of the facts argue that the “accumulation” of α-synuclein seen in these grafted neurons may be related to the biology of the protein.

    First, as pointed out previously, it has been shown that α-synuclein is present in the perikarya of fetal neurons (Galvin et al., 2001; Raghavan et al., 2004), and that over time, the protein is predominantly localized to presynaptic boutons/terminals. This is especially apparent in a reduced system like cultured hippocampal neurons, where the protein is initially abundant in the perikarya, and over time, as synapses develop, most of the protein is concentrated at the boutons (Withers et al., 1997;Roy et al., 2007). Indeed, this behavior is not unique for α-synuclein, as most presynaptic proteins tend to behave similarly in human brains (Galvin et al., 2001) as well as in cultured neurons (Fletcher et al., 1991). Thus, the perikaryal α-synuclein accumulation may simply be a manifestation of the immature state of these neurons.

    One prediction of this model would be that other presynaptic proteins would also show a similar perikaryal distribution. Though it can be argued that the time-course of the grafted neurons should allow the fetal neurons to mature enough so that their presynaptic proteins can be redistributed from the perikarya to synapses, the onus is upon the authors to show that this indeed is the case.

    A second point worth considering is that α-synuclein does not spontaneously appear at the presynapses, but is continuously synthesized at the perikarya and then transported along the axon up to the synapses (Jensen et al., 1999; Li et al., 2004; Roy et al., 2007) throughout the life of the neuron, much like any other synaptic or non-synaptic protein including ubiquitin. As the grafted neurons in these studies seem to grow with no apparent organization, it is conceivable that this atypical organization of grafted neurons is not conducive to physiologic axonal transport in these neurons, causing proteins (including α-synuclein) to accumulate in cell bodies and axons. Both these scenarios seem reasonable, and can perhaps more easily explain the phenotype, as opposed to the alternative but more exotic proposal of disease-induction in the grafted neurons.

    In summary, I would definitely recommend these papers, but would argue that we must apply critical judgment in our interpretation of the α-synuclein story.


    . The distribution of synapsin I and synaptophysin in hippocampal neurons developing in culture. J Neurosci. 1991 Jun;11(6):1617-26. PubMed.

    . Differential expression and distribution of alpha-, beta-, and gamma-synuclein in the developing human substantia nigra. Exp Neurol. 2001 Apr;168(2):347-55. PubMed.

    . Axonal transport of synucleins is mediated by all rate components. Eur J Neurosci. 1999 Oct;11(10):3369-76. PubMed.

    . Axonal transport of human alpha-synuclein slows with aging but is not affected by familial Parkinson's disease-linked mutations. J Neurochem. 2004 Jan;88(2):401-10. PubMed.

    . Alpha-synuclein expression in the developing human brain. Pediatr Dev Pathol. 2004 Sep-Oct;7(5):506-16. PubMed.

    . Rapid and intermittent cotransport of slow component-b proteins. J Neurosci. 2007 Mar 21;27(12):3131-8. PubMed.

    . Delayed localization of synelfin (synuclein, NACP) to presynaptic terminals in cultured rat hippocampal neurons. Brain Res Dev Brain Res. 1997 Mar 17;99(1):87-94. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Transplantation of embryonic dopamine neurons for severe Parkinson's disease. N Engl J Med. 2001 Mar 8;344(10):710-9. PubMed.
  2. . Differential expression and distribution of alpha-, beta-, and gamma-synuclein in the developing human substantia nigra. Exp Neurol. 2001 Apr;168(2):347-55. PubMed.
  3. . Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006 Aug 25;126(4):663-76. PubMed.

Further Reading

Primary Papers

  1. . Dopamine neurons implanted into people with Parkinson's disease survive without pathology for 14 years. Nat Med. 2008 May;14(5):507-9. PubMed.
  2. . Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson's disease. Nat Med. 2008 May;14(5):504-6. PubMed.
  3. . Lewy bodies in grafted neurons in subjects with Parkinson's disease suggest host-to-graft disease propagation. Nat Med. 2008 May;14(5):501-3. PubMed.
  4. . Neurons derived from reprogrammed fibroblasts functionally integrate into the fetal brain and improve symptoms of rats with Parkinson's disease. Proc Natl Acad Sci U S A. 2008 Apr 15;105(15):5856-61. PubMed.