Outcome of cell suspension allografts in a patient with Huntington's disease

For patients with incurable neurodegenerative disorders such as Huntington's (HD) and Parkinson's disease, cell transplantation has been explored as a potential treatment option. Here, we present the first clinicopathological study of a patient with HD in receipt of cell‐suspension striatal allografts who took part in the NEST‐UK multicenter clinical transplantation trial. Using various immunohistochemical techniques, we found a discrepancy in the survival of grafted projection neurons with respect to grafted interneurons as well as major ongoing inflammatory and immune responses to the grafted tissue with evidence of mutant huntingtin aggregates within the transplant area. Our results indicate that grafts can survive more than a decade post‐transplantation, but show compromised survival with inflammation and mutant protein being observed within the transplant site. Ann Neurol 2018;84:950–956

For patients with incurable neurodegenerative disorders such as Huntington's (HD) and Parkinson's disease, cell transplantation has been explored as a potential treatment option. Here, we present the first clinicopathological study of a patient with HD in receipt of cell-suspension striatal allografts who took part in the NEST-UK multicenter clinical transplantation trial. Using various immunohistochemical techniques, we found a discrepancy in the survival of grafted projection neurons with respect to grafted interneurons as well as major ongoing inflammatory and immune responses to the grafted tissue with evidence of mutant huntingtin aggregates within the transplant area. Our results indicate that grafts can survive more than a decade posttransplantation, but show compromised survival with inflammation and mutant protein being observed within the transplant site. ANN NEUROL 2018;84:950-956 H untington's disease (HD) is an autosomal-dominant neurodegenerative disorder that presents with a combination of motor, cognitive, and psychiatric problems that progress over a 20-year period to death. It is currently incurable, and although many therapies have been the subject of clinical trials, none have been shown to alter the natural history of this condition. 1 More than 20 years ago, work commenced on a novel strategy that involved allografting fetal tissue into the striatum of patients with midstage disease to attempt to repair the damaged circuitry, as had been shown preclinically using excitotoxic lesions of the rat and monkey striatum. 2 To date, seven small open-label studies of neural transplants have been conducted worldwide assessing the feasibility, safety, and tolerability of this procedure in patients with HD. [3][4][5][6][7][8][9] This approach has yielded mixed results including postmortem analysis of transplanted patients. Here, we present long-term histological data on 1 of the 5 patients treated with a fetal striatal cell suspension allograft as part of the UK study. 4

Trial Information
This patient was 1 of 5 who were part of the NEST-UK multicenter study, which was initiated in 1998 to evaluate the safety and efficacy of bilateral fetal striatal transplantation in patients with mild HD (ISRCTN no 36485475). 4 The operations were undertaken in Cambridge between 2000 and 2003, and this report deals with patient number 5, the only one to receive bilateral transplants of fetal striatal tissue in a single operation. The trial was approved by the Cambridge Regional Ethics Committee (REC ref: 95/086), as was the postmortem analysis (REC ref:01/117).

Image Acquisition and Quantification
Brightfield photomicrographs were taken using the Picture Frame software (MicroBrightField Bioscience, Williston, VT) attached to an E800 Nikon microscope (Nikon Instruments, Tokyo, Japan) and prepared for illustration using Adobe Photoshop CS5 and final figure panels assembled using Adobe Illustrator CS5 (Adobe Systems, San Jose, CA).
All cell quantifications were performed on two sections of the cortex and striatum containing the graft. The perimeters of the grafted areas were delineated using the tracing contour option in Stereo Investigator (NeuroExplorer, version 10.0; MicroBrightField Bioscience) as described in a previous publication. 16 To measure cholinergic activity, images of AChE stainings were processed with ImageJ software (NIH, Bethesda, MD) and P-zones, non-P-zones (NP-zones), and the host striatum were individually traced. These areas were measured for their mean gray values of color intensity.

Statistical Analysis
For all cell quantifications, an unpaired Student t test was performed using Prism (6.0; GraphPad Software Inc., La Jolla, CA).

Clinical Course
The patient first noticed problems in 1995, at the age of 37, with a slight change in his mood and his family became aware of his problems in 1997 when he developed nightmares and depression. He had a family history for HD and went on to have a positive genetic test with an expansion of 47 CAG repeats in exon 1 of the huntingtin gene. In 2003, he was selected for neural grafting and underwent a bilateral transplant procedure without complications and was followed up according to the CAPIT-HD protocol until his death in 2015. His clinical history showed no obvious change in his disease course after grafting, either clinically or on positron emission tomography imaging, as detailed in a previous work. 4 Postmortem Graft Evaluation Graft Location and Cytoarchitecture. Macroscopically grafts were easily identified. In total, six grafts were located in the left hemisphere with two in the caudate and four in the putamen (Fig 1A,B 0 ) whereas in the right hemisphere only one and two grafts were found in these structures respectively (Fig 1C-E).
General Graft Health. All the typical phenotypes of striatal interneurons were observed within the grafted tissue including cells positive for calretinin (CR), parvalbumin (PV), NADPH, and ChAT (Fig 1F-I). They showed healthy morphologies and extensive dendritic arborizations. Stereological quantifications for each of these cell types revealed similar counts within the transplanted area in comparison to the host striatum for CR, PV, and NADPH interneurons (Fig 1F 0 -I 0 ). Despite rare examples of healthy MAP2 staining within the graft (Fig 1J), the presence of DARPP-32-positive cells was almost exclusively observed within the host striatum (Fig 1K). Grafted projections neurons, as also identified with calbindin (CB), were largely necrotic (Fig 1L,L 0 ) and there were fewer cells within the transplant area. Because of the tested antibody being discontinued, we used AChE staining to measure cholinergic enzyme levels within the P-zones and compared them to the host striatum, where they looked similar (Fig 1I 0 ).

Inflammatory-Immune Response to Cell Suspension
Grafts. A ring of densely packed, maximally hyperactive microglia surrounded each of the transplanted sites (Fig  2A,B) and was also present within the graft itself. Throughout this ring, the spectrum of microglial activation was represented from dystrophic ( Fig 2C,D), clustered ( Fig 2E) and rod (Fig 2H) cells. Grafted regions lacking striatal markers (NP-zones) had qualitatively much lower microglial density, though microglial morphology and Iba1 intensity remained consistent with a proinflammatory phenotype (Fig 2F,G). As senescence continues, these processes become fragmented and severe dystrophy was confirmed by spherical microglial cell bodies with no processes, surrounded by small dots of cytoplasm apparently disconnected from the soma (Fig 2I). The outer borders of the graft contained more dystrophic microglia, with gnarled, beaded processes (Fig 2J,K). The host tissue, on the other hand, was almost entirely populated by dystrophic and nonfunctional microglia. The P-zones contained the highest ratio of ramified microglia and the fewest dystrophic cells (Fig 2L), although there were still many more hyperactive cells than normally present in healthy tissue.

Graft Innervation and Presence of HD-Related Pathology
The grafts received modest innervation, as evidenced by the sparse dopaminergic fibers observed near the transplants (Fig 3A,B). Several mHtt + aggregates were observed within the graft site and at the graft/host interface ( Fig  3C,D). The mHtt + aggregates were not only observed within the extracellular matrix of the grafted area, as previously reported in solid tissue transplants, 16 but they were also found within cell bodies in the grafts (Fig 3F-H,L). A particularly interesting observation was that some mHtt + aggregates were found along, or even within, the December 2018 dendritic tree of cells within the transplant (Fig 3J,K). Notably, striatal host cells expressing mHtt + extended their dendritic trees into the grafts (Fig 3I,I 0 ).

Discussion
This is the first histological report on striatal cell suspension allografts originating from the UK NEST striatal transplantation HD project (NEST-HD). 5,17 Postmortem evaluation revealed that (1) it was easy to find striatal transplants within which grafted interneurons were largely spared in contrast to the grafted projection neurons that were rarely observed; (2) there was a striking microglial response around the graft; (3) there were mHtt aggregates within the grafted tissue as well as within cells of the transplanted area and finally; and (4) there was infiltration of immune cells within the grafted tissue that also contained mHtt, both of which may have further contributed to poor graft survival. These observations are similar to those that we have reported previously in HD patients who receiving solid tissue pieces, but, importantly, there were notable differences in that we saw a much more marked infiltration of immune cells in the cell suspension grafts along with more florid mHtt cellular pathology within the transplant sites. This latter observation relating to mHtt expression within the grafts requires further comment. One could argue that the cells expressing mHtt within the boundaries of the grafted areas are host cells that have migrated into the transplant. Alternatively, one could speculate that innervation of the grafts by host striatal neurons containing mHtt + results in the transport of the pathological protein from the host to the graft, given that mature striatal cells have little if any migratory capacity. 18 Unfortunately, the underlying explanation for this finding cannot be

954
Volume 84, No. 6 explored in this context, because the technical limitations inherent to human postmortem analysis do not allow one to make mechanistic conclusions. Additionally, other routes of transport should be considered in light of the fact that mHtt + was observed, although rarely, within CD8 + cells. The significant infiltration of CD4 + and CD8 + cells, notable within the grafted areas, may have indeed been facilitated by the more complex and efficient vascularization that we observed within the cell suspension grafts (data not shown).
In conclusion, this study again highlights that fetal striatal allografts can survive long term in the human HD brain. However, although interneurons within them survive, projection neurons degenerate, and this is all associated with inflammation around and in the transplant as well as the expression of mHtt pathology at the graft site. The relevance and mechanistic consequences of these observations awaits clarification, but raises questions as to whether cell-based approaches for repairing the HD brain can ultimately repair the dysfunctional networks observed in this condition. 16,[19][20][21][22][23][24][25][26] Acknowledgment