Host brain environmental influences on transplanted medial ganglionic eminence progenitors

Interneuron progenitor transplantation can ameliorate disease symptoms in a variety of neurological disorders. The strategy is based on transplantation of embryonic medial ganglionic eminence (MGE) progenitors. Elucidating how host brain environment influences the integration of interneuron progenitors is critical for optimizing this strategy across different disease states. Here, we systematically evaluated the influence of age and brain region on survival, migration, and differentiation of transplant-derived cells. We find that early postnatal MGE transplantation yields superior survival and more extensive migratory capabilities compared to transplantation during the juvenile or adult stages. MGE progenitors migrate more widely in the cortex compared to the hippocampus. Maturation to interneuron subtypes is regulated by age and brain region. MGE progenitors transplanted into the dentate gyrus sub-region of the early postnatal hippocampus can differentiate into astrocytes. Our results suggest that the host brain environment critically regulates survival, spatial distribution, and maturation of MGE-derived interneurons following transplantation. These findings inform and enable optimal conditions for interneuron transplant therapies.


Animals
All experiments were performed on mice maintained on a 12-hour light/12-hour dark cycle with no food or water restrictions.All procedures involving animals followed the guidelines of the National Institutes of Health and were approved by Institutional Animal Care and Use Committee at University of California, San Francisco (#AN181254-02B).Embryonic donor tissue was obtained by crossing wild-type CD1 mice (Charles River Laboratories, Cat Num #022) with homozygous beta-actin EGFP mice on a CD1 background.We used male and female CD1 and male C57BL/6J (Jackson laboratory; 000664) mice as recipients.The study was conducted in accordance with the ARRIVE (Animal Research: Reporting of in Vivo Experiments) guidelines.

Tissue dissection and transplantation
MGE progenitor cells were harvested from E13.5 GFP+ transgenic embryos, as previously described 10,13 .In brief, embryonic day 0.5 was defined when the vaginal plug was detected.To dissect MGE, embryonic brains were first removed, the telencephalon isolated and the two hemispheres separated by a sagittal cut.The ventral telencephalon was exposed, and the medial ganglionic eminence was isolated using the sulcus that clearly divide the medial from lateral ganglionic eminence.The dorsal MGE was collected after removing the preoptic area and mantle zone.MGE progenitor cells were collected in Leibovitz L-15 media, mechanically dissociated in a singlecell suspension in media containing DNase I (Roche, 100 ug/ml) using repeated pipetting, and concentrated by centrifugation (3 min at 800×g).Cells were kept at 4 °C until transplantation.
Concentrated GFP+ MGE cells (~ 600 cells per nl) were injected using 30° beveled glass micropipettes (80 μm diameter tip, Witerol 5 μl, Drummond Scientific) prefilled with mineral oil into the cortex or hippocampus of recipient pup (P2-4), juvenile (P30-40) and adult (P90-150) animals.Pipettes were attached to a microinjector (Narishige) mounted on a stereotaxic machine (Kopf).We quantified cell viability of MGE cells using Triptan Blue (Sigma), and a viability more than 80% was considered as cutoff to proceed with transplantation.Each cortical recipient animal received a single injection containing 5 × 10 4 cells.Hippocampal recipient animals received a single or a double injection per hemisphere containing 3 × 10 4 cells due to an increased risk of clumping when using higher volumes.
Pups were anesthetized by exposure to ice until pedal reflexes were absent.Around 50-100 nl of highly concentrated cells were injected at the following coordinates from bregma to target the cortex (1.2 mm anterior, 1.2 mm lateral, and 0.6 mm dorsal) or the hippocampus (1.2 mm anterior, 1.2 mm lateral, and 1.4 mm dorsal).After injection, pups were returned to their mothers.Juvenile and adult animals were anesthetized by a mixture of ketamine/xylazine.Cell injections were made into the cortex at the following coordinates from bregma: anterior-posterior (AP) -1.75 mm, medial lateral (ML) 1.75 mm, and dorsal-ventral (DV) 0.6 mm.For the hippocampus, injections to target the stratum radiatum of area CA3 or the stratum oriens of area CA1 were made at the following coordinates: AP -2 mm, ML 2.5 mm, DV 1.8 mm and AP -2 mm, ML 1.6 mm, DV 1.15 mm, respectively.

Cell counts and quantification
Images were acquired at 1024 pixels resolution using a Nikon confocal microscope.Quantification analysis was performed with NIS-Elements software (Nikon) on fluorescently labeled sections (50 µm) captured with 4×, 10×, and 40× objectives.All transplanted cells expressing GFP were counted in every sixth coronal section (spaced 300 µm apart) across all layers of the cortex or hippocampus, as previously described 21 .To estimate cell survival in each hemisphere, we counted the number of GFP+ cells in every 6-coronal section (spaced 300 µm apart) along the rostro-caudal axis.This count was multiplied by 6, and the survival rate was calculated as (total survived cells/numbers of transplanted cells) *100.To assess the distribution of GFP+ cells across cortical layers, GFP+ cells were assigned to one of 10 bins superimposed on the area between pial surface (bin 1) and white matter (bin 10).To determine the percentage of grafted GFP+ cells co-expressing NeuN, SST, PV, GFAP, Oligo or CC1 after transplantation (n = 3 mice per marker), we quantified the number of GFP+ cells co-expressing these antibody markers in every sixth coronal section (spaced 300 µm apart).

Statistical analysis
All analyses were performed using PRISM (GraphPad).Two-tailed t-tests, unpaired t-tests and one-way ANOVA's were employed as specified in each analysis to compare different groups.All plots with error bars are reported as mean ± SEM.No data were excluded from the analysis.

Data availability
The original data of this study are available from R.P. upon reasonable request.

Results
Murine embryonic MGE progenitor cells migrate widely following early postnatal transplantation into the recipient mouse brain 18 .Within the host brain, transplanted MGE-derived interneurons efficiently distribute across cortical layers [30][31][32] .However, residual clustering and limited migration away from the injection site have also been reported 29,33 .Since factors transiently expressed in the developing brain, e.g., clustered gamma-protocadherins 34,35 or CCCTC-binding factor 36 , can influence migration and integration of MGE progenitor cells, we investigated here the integration following transplantation into recipient mice at different developmental ages.Because factors expressed in local microcircuits, e.g., vesicular GABA transporters 32 or MTG8 37 , can also influence integration of these progenitors, we compared transplantations into the cortex and hippocampus (Fig. 1A).

MGE progenitors transplanted into cortex
To investigate the influence of host brain age on survival, migration, and maturation, we harvested MGE progenitor cells from E13.5 GFP-labeled donor embryos and transplanted into the neocortex of neonatal (P2-P4), juvenile (P30-40), and adult (P60-150) mice.At 30 days post-transplantation (DAT), we noted a significant difference in cell survival across the three age groups, with higher GFP+ cell survival rate following transplantation into neonatal host brains compared to juvenile or adult mice (P2-4, 17.2 ± 2.5%, n = 8 mice; P30-40, 0.8 ± 0.1%, n = 6 mice; P60-150, 0.7 ± 0.1%, n = 5 mice; p < 0.001, one-way ANOVA).We also found a significant difference in migration capacity away from the injection site.At 30 days after a single cortical injection of MGE-GFP progenitors in neonates (n = 8), GFP + cells showed extensive distribution along the antero-posterior, medial-lateral, and dorsal-ventral axes (Fig. 1B-E).In the anterior-posterior direction, cells migrated up to 3000 µm from the injection site in both directions, covering the entire antero-posterior cortical structures (up to anterior cingulate and visual cortex in the anterior and posterior axes, respectively).Along the dorsal-ventral axis, in addition to neocortex, MGE-GFP cells reached into striatum, dorsal subiculum and in some cases, CA1 regions of the hippocampus (Fig. 1D).Across the medial-lateral extent of neocortex, MGE-GFP cells reached lateral cortical areas, such as the auditory cortex.In juvenile and adult host brains, although transplanted MGE-GFP progenitor cells migrated away from the injection site, the distance reached in antero-posterior axis was more limited (up to 600 µm) (Fig. 2).The total anterior-posterior cortical distances covered were 3825 ± 239 µm (neonate), 850 ± 92 µm (juvenile), and 900 ± 95 µm (adult) (p < 0.001, one-way ANOVA).
Next, to assess the laminar distribution of MGE-derived GFP+ cells within specific cortical sub-layers of the host brain, we divided the radial axis of the cortex into ten equal bins and assigned each GFP+ cell to a single bin (see Methods).In the neonatal (P2-P4) transplantation group, GFP+ cells were distributed across the entire neocortex, from Layer II to VI (Fig. 3), and no cells were identified in Layer I (corresponding to bin 1).This distribution matched the endogenous distribution of MGE-derived interneurons in the cortex [30][31][32] .In the juvenile (P30-40) and adult (P60-150) transplantation groups, we observed a similar distribution of GFP+ cells across layers.However, an increased number of transplant-derived cells were observed in the deep cortical layers (bin 10) near injection sites in adult compared to pup and juvenile mice (Fig. 3A,B).We also found a significant increase in GFP+ cells along the medial-lateral axis that migrate laterally from the injection site after transplantation in neonatal host brains compared to juveniles and adults (P2-4, 5229 ± 80.1 mm, n = 8 mice; P30-40, 970 ± 66.6 mm, n = 6 mice; P60-150, 994 ± 62 mm, n = 5 mice; p < 0.001, one-way ANOVA) (Fig. 3C).Interestingly, in juvenile and adult brains, transplanted-derived cells appeared to have a columnar distribution, while in pup brains, they were broadly distributed.
To evaluate the temporal migration profile after transplantation, we analyzed the cortical grafted GFP+ cells in the adult (P60-150) transplantation groups at 7 and 14 DAT.As previously shown in pup transplantation 18 , the antero-posterior migration of the grafted MGE-derived cells is already completed by 7 DAT, with GFP+ cells located up to 600 µm from the injection site.Additionally, we noted that the GFP+ cell survival rate following transplantation was similar to the rate observed at 30 DAT (7 DAT, 2.1 ± 0.7%, n = 5 mice; 14 DAT, 1 ± 0.4%, n = 5 mice; 30 DAT, 0.7 ± 0.1%, n = 5 mice; p > 0.05 one-way ANOVA).These findings indicate that the majority of the GFP+ cells in the cortex have reached their final destination at 7 DAT and are already initiating the maturation process.

MGE progenitors transplanted into hippocampus
To investigate whether local host brain microcircuits influence the survival, migration, or maturation of transplanted MGE progenitor cells, we targeted the hippocampus.This complex brain region has a unique anatomical structure and is widely implicated in the pathophysiology of neurological diseases 39 .MGE progenitors from E13.5 GFP-labeled donor embryos were transplanted into the hippocampus of neonatal (P2-P4), juvenile (P30-40), and adult (P60-150) mice.At 30 DAT, we noted a significant difference in cell survival across the three age groups.
A higher hippocampal GFP+ cell survival rate was noted in pups compared to juvenile and adult mice (P2-4, 8.7 ± 0.6%, n = 6 mice; P30-40, 0.98 ± 0.2%, n = 6 mice; P60-150, 0.7 ± 0.08%, n = 7 mice; p < 0.001, one-way ANOVA).Migration away from the injection site was also different across the three age groups.In particular, after a single injection of MGE progenitor cells in the pup hippocampus, transplanted cells distributed widely across the entire antero-posterior axis of the hippocampus (up to 2400 μm from the injection site), with GFP+ cells spreading into the dorsal and ventral hippocampus, as shown in Fig. 6A-C.In some cases (4 out of 6 mice), a few GFP+ transplanted cells migrated and spread into cortical areas and transition subiculum (Fig. 6B); note that the average GFP+ survival rate outside the hippocampus was 5.9 ± 1.5% (n = 6 mice).Migration in juvenile and adult mice (Fig. 7A) was more limited, reaching up to 900 µ m from the injection site (Fig. 7B,C).In all age groups (pup, juvenile, and adult), the transplant-derived cell distribution within the hippocampus was not uniform, and an increased number of GFP+ cells were observed in specific subregions.For example, at 30 DAT in the hippocampal dentate gyrus (DG) of pups, GFP+ cells were primarily found in DG, with only few cells in CA1 and CA3 subregions (74%, 21.4%, and 4.5%, respectively; p < 0.001, one-way ANOVA) (Fig. 8A).In adult mice, cells transplanted in area CA3 migrated mostly within the CA3 subregions (70%), with only few cells reaching CA1 or DG (Fig. 8B).In one case, we observed extensive migration of GFP+ cells in DG after injection in CA3 stratum radiatum.These results highlight that, in addition to a host brain age effect, within the hippocampus, targeted subregions also influence the migration of transplanted MGE progenitor cells.

Discussion
Our results suggest that the host brain environment plays a crucial role in regulating survival, migration, and maturation of cells derived from the transplantation of embryonic MGE progenitors.Although previous work suggested that interneuron survival following transplantation is guided by an intrinsic maturational program rather than the developmental state of the host brain itself 41 , our data effectively demonstrate robust survival and extensive migration when MGE progenitors are transplanted early in development (but not later) or into cortex (compared to hippocampus).The maturation profile of MGE-derived interneurons is also regulated by the developmental state and brain region.We observed a two-fold greater percentage of SST-positive interneurons following hippocampal transplantation into juvenile or adult brains (compared to neonates).In contrast, the percentage of PV-positive interneurons decreased nearly two-fold with age following cortical transplantation into the adult brain compared to neonates.Unexpectedly, we also found sub-regions of pup hippocampus-namely the dentate gyrus-where transplant-derived MGE progenitors locally differentiate into astrocytes.Considering the emerging therapeutic potential of transplanted embryonic MGE progenitors in various neurological disorders 10,11,13,21,25,42 , these results hold important clinical implications.migrating into the cortex, they transiently participate in cortical neurodevelopment before being replaced by other glia cell populations 57 .Furthermore, earlier studies showed the absence of proliferative markers following transplantation of embryonic MGE progenitors 22,58 .While unlikely, the presence of proliferating neural cells cannot be fully excluded.Future studies focused on additional markers and lineage analysis will be necessary to identify factor(s) driving astrocyte survival following transplantation into the DG microenvironment.
In our hands, MGE transplantation at postnatal day 2 consistently resulted in widespread migration, in all directions, within the host brain parenchyma 18,26,30 .The transplanted MGE progenitor cells also show clear morphologies consistent with migrating cells 18 .Our results confirm and extend these reports.While the possibility that some MGE-GFP cells reached distant sites via the meninges or leakage into the cerebral spinal fluid cannot be entirely excluded due to the small size of the mouse brain 59 , the migratory profiles and properties of MGE-derived cells both in vivo and in vitro do not support this conclusion.However, it is interesting that in wild-type juvenile or adult mice, the migratory ability is more limited.The differences in molecular, cellular, and extracellular profiles in the adult versus pup environments 60,61 may explain this reduced migration in the mature brain compared to early developmental stages.Such difference, rather than being a limitation of the technique, highlights the translational applicability of this approach.The ability to modify host circuits with regional and neuronal specificity in the mature brain contrasts with the broad, non-specific circuit changes observed after pups' transplantation.Indeed, adult MGE transplantation has already proven to be an effective disease-modifying therapy in adult mouse models representing acquired epilepsy or traumatic brain injury 10,12,23 .
It is worth noting that a significant proportion of individuals with epilepsy, autism spectrum disorders, or schizophrenia are considered to have "inter-neuropathies", define by frank reductions or dysfunction in PV+ and/or SST+ interneurons 3,[62][63][64] .Given that these conditions manifest at different stages of development and that interneuron deficits can be localized to specific cortical or hippocampal brain regions, a better understanding of the influence of local environment on transplanted MGE progenitors is necessary.Indeed, several studies using embryonic MGE progenitors 65 or other embryonic cell sources 66,67 for transplantation also support our findings, emphasizing the significant impact of local environmental factors on the integration of these cells in the host brain.Understanding advantages and limitations of where (and when) MGE progenitors integrate following transplantation is critical for establishing effective and targeted cell therapies for patients suffering from various neurological disorders.

Figure 1 .
Figure 1.Timeline and distribution of transplanted MGE progenitors in pup cortex.(A) schematic of the overall experimental design.(B) Coronal section of a pup recipient mouse (30 DAT) labeled for transplanted GFP+ neurons (green).Transplanted cells widely dispersed across the cortex.(C) Schematic at 30 DAT showing the distribution of GFP+ cells (red) across brain slices in the antero-posterior axis (300 μm apart) after transplantation in pup cortex.Region divisions were adapted from the Paxinos atlas.(D) Higher magnification of the hippocampal region outlined in B. (E) Plot showing the distribution of GFP+ cells along the anteroposterior axis 30 DAT cortex injection in pup (P2-4) mice; n = 8 brains.DAT, days after transplantation.Data are represented as mean ± SEM.

Figure 2 .
Figure 2. Transplanted MGE progenitors in juvenile and adult cortex.(A) Representative images from 30 DAT brains receiving MGE progenitor cells in the cortical area at P30 (juvenile, top) and P90 (adult, bottom).(B) Plot showing the distribution of GFP+ cells in the antero-posterior axis from the injection site in juvenile (P30-40) mice; n = 6 brains and adult (P60-150) mice, n = 5 brains.(C) Schematic at 30 DAT showing the distribution of GFP+ cells (red) across brain slices in the antero-posterior axis (300 μm apart) after transplantation in juvenile (left panel) and adult (right panel) cortex.Region divisions were adapted from the Paxinos atlas.DAT, days after transplantation.Data are represented as mean ± SEM.

Figure 3 .
Figure 3. Distribution of MGE-derived cells following transplantation in cortex.(A) Representative coronal sections at 30 DAT after pup (top), juvenile (middle), and adult (bottom) transplantation.(B) Plot showing the distribution of MGE-derived cells along the rostro-caudal axis divided into 10 bins from pia to white matter (WM).Note comparable distribution across different age groups, except for an increased number of GFP+ cells close to the injection site in adults (bin 10).(C) Plot of the mediolateral distribution of cells across age groups.Note a highly mediolateral migration after pup transplantation.DAT, days after transplantation.Data are represented as mean ± SEM; one-way ANOVA *p < 0.05.

Figure 4 .
Figure 4. Maturation profile for MGE-derived interneurons transplanted in cortex.(A) Representative magnified cell images showing co-localization between GFP and PV (left side) and SST (right side) at 30 DAT after pup and adult transplantation.(B) Plot showing the quantification of marker expression in GFP-labeled cells (n = 3 mice per marker) across age groups.Note a decreased number of GFP co-expressing PV cells in juvenile and adult groups.Data are represented as mean ± SEM; one-way ANOVA *p < 0.05.DAT, days after transplantation.NeuN, neuronal nuclei; SST, somatostatin; PV, parvalbumin; GFAP, glial fibrillary acid protein; Oligo, oligodendrocyte.

Figure 5 .
Figure 5. Endogenous astrocyte expression at the injection site.(A) Representative images collected at the injection site of an animal at 30 DAT.Note the presence of endogenous astrocytes (GFAP) around the cortical injection site and a few GFP+ astrocytes (white arrow).(B) Representative images collected 300 μm away from the injection site from the same animal.GFP+ cells show mature neuronal morphology.DAT, days after transplantation.GFAP, glial fibrillary acid protein.

Figure 6 .Figure 7 .
Figure 6.Distribution of MGE-derived cells following transplantation in the hippocampus.(A) Representative images of dorsal and ventral hippocampal sections with GFP+ cells.(B) Schematic of the distribution of GFP+ cells (marked in red) sectioned 300 μm apart.Region divisions were adapted from the Paxinos atlas.(B) Percentage of GFP+ cells across the antero-posterior axis (6 mice, from 3 or more experiments).(D) Plot showing the quantification of the total number of GFP+ cells in the hippocampus at 30 DAT of a single injection.DAT, days after transplantation.Data are represented as mean ± SEM.

Figure 8 .
Figure 8. Distribution of transplant-derived MGE-GFP cells within hippocampal subfields.(A) Representative hippocampal sections of the subfields CA1, CA3, and DG at 30 DAT after pup injections.Plot showing the quantification of GFP+ cells across the subfields of the hippocampus (6 mice from 3 or more experiments).Note the increased number of cells in the dentate gyrus compare to CA1 and CA3.(B) Representative hippocampal sections of the subfields CA1, CA3, and DG at 30 DAT showing the distribution of GFP + cells after adult injections.Quantification of GFP+ cells across the subfields of the hippocampus (6 mice from 3 or more experiments).Note the increased number of cells in CA3 compare to CA1 and DG.Higher resolution examples of GFP+ cells imaged under 40× objective showing a mature neuronal morphology are shown at the bottom.DAT, days after transplantation; GFAP, glial fibrillary acid protein.Data are represented as mean ± SEM; one-way ANOVA *p < 0.05.

Figure 9 . 4 .Figure 10 .
Figure 9. Maturation profile for MGE-derived interneurons transplanted in hippocampus.(A) Representative magnified cell images showing GFP+ cells co-expressing PV and SST at 30 DAT in pup, juvenile, and adult hippocampus.(B) Plot showing the quantification of cells co-expressing GFP and NeuN, SST, PV, GFAP, and Oligo in the hippocampus at 30 DAT for pup, juvenile, and adult groups (n = 3-4 mice per marker).Note the reduced GFP+ cells co-expressing SST and increased GFP+ co-expressing GFAP cells in the pup group.DAT, days after transplantation.Data are represented as mean ± SEM; one-way ANOVA *p < 0.05.Abbreviations as in Fig. 4.