Challenging young neurons in new brain regions can reveal important insights into how the environment sculpts neuronal fate and maturation. This protocol describes a procedure to harvest interneuron precursors from specific brain regions and transplant them either homotopically or heterotopically into the brain of postnatal pups.
Neuronal fate determination and maturation requires an intricate interplay between genetic programs and environmental signals. However, disentangling the roles of intrinsic vs. extrinsic mechanisms that regulate this differentiation process is a conundrum for all developmental neurobiologists. This issue is magnified for GABAergic interneurons, an incredibly heterogeneous cell population that is born from transient embryonic structures and undergo a protracted migratory phase to disperse throughout the telencephalon. To explore how different brain environments affect interneuron fate and maturation, we developed a protocol for harvesting fluorescently labeled immature interneuron precursors from specific brain regions in newborn mice (P0-P2). At this age, interneuron migration is nearly complete and these cells are residing in their final resting environments with relatively little synaptic integration. Following collection of single cell solutions via flow cytometry, these interneuron precursors are transplanted into P0-P2 wildtype postnatal pups. By performing both homotopic (e.g., cortex-to-cortex) or heterotopic (e.g., cortex-to-hippocampus) transplantations, one can assess how challenging immature interneurons in new brain environments affects their fate, maturation, and circuit integration. Brains can be harvested in adult mice and assayed with a wide variety of posthoc analysis on grafted cells, including immunohistochemical, electrophysiological and transcriptional profiling. This general approach provides investigators with a strategy to assay how distinct brain environments can influence numerous aspects of neuron development and identify if specific neuronal characteristics are primarily driven by hardwired genetic programs or environmental cues.
Proper cortical function requires a balance between excitatory projection neurons and inhibitory GABAergic interneurons, an extremely heterogeneous population with distinct morphologies, electrophysiological properties, connectivity and neurochemical markers. Abnormal development and function of interneurons (and specific interneuron subgroups) has been linked to the pathobiology of psychiatric disorders such as schizophrenia, autism and epilepsy1,2,3. Furthermore, many genes implicated in these brain disorders are strongly enriched in young interneurons4. Thus, a greater understanding of the mechanisms that regulate interneuron fate determination and maturation is needed to understand normal development and potential etiologies of numerous brain diseases.
Forebrain interneurons are born primarily from two transient embryonic structures, the medial and caudal ganglionic eminences (MGE and CGE, respectively). These postmitotic cells (interneuron precursors) then undergo a protracted tangential migration phase to disperse throughout the telencephalon where they integrate into a wide variety of circuits. MGE-derived interneurons consist of three largely non-overlapping, neurochemically defined subgroups: fast spiking parvalbumin (PV+) interneurons, non-fast spiking somatostatin (SST+) interneurons, and late spiking neuronal nitric oxide synthase (nNOS+) interneurons that constitute hippocampal neurogliaform and ivy cells. Numerous labs have identified several mechanisms within the MGE that regulate initial fate decisions into PV+ or SST+ interneurons, including spatial gradients of morphogens, birthdate of interneuron precursors, and the mode of neurogenic division5,6,7,8,9,10. It has been proposed that interneurons initially differentiate into 'cardinal classes' and then progressively mature into 'definitive classes' as they interact with their environment11. Recent evidence indicates that some mature interneuron subtypes may be genetically hardwired as these cells become postmitotic in the ganglionic eminences, indicating that early defined intrinsic genetic programs may play a larger role than previously appreciated12,13. However, the key question of how the intrinsic genetic programs interact with environmental cues to drive differentiation into distinct interneuron subtypes remains largely unexplored.
Numerous studies have transplanted embryonic MGE cells directly into a variety of brain regions, with the consensus results that grafted cells mature and release GABA to generally inhibit the local endogenous circuitry14,15,16,17,18,19. These promising observations have generated significant interest in using human induced pluripotent stem cell (hIPSC)-derived interneurons to treat a variety of brain diseases. However, very few of these studies assess if these grafted cells mature into the expected types of mature interneurons, a critical component when one thinks about translational approaches.
To address how environment influences interneuron differentiation and maturation, a strategy was devised to transplant immature interneuron precursors into new brain environments in order to examine whether grafted interneurons adopt features of the host environment or retain features from the donor environment20. MGE transplants are not suitable to address this question because the MGE contains a mixed population of interneuron and GABAergic projection cells that disperse throughout numerous brain regions21. Without knowing where these MGE cells would have migrated, one can not fully assess how these transplantations are affected by the brain environment. By harvesting interneuron precursors at early postnatal timepoints, this problem is circumvented by obtaining immature cells that have completed their migration and reached their target brain region but have minimal interaction with the environment. By focusing on specific features of interneurons that are differentially expressed between distinct brain regions, one can then determine how the host environment changes interneuron properties. The general approach outlined in this protocol should be applicable to any investigator that wants to examine how young neurons behave when challenged in a new environment.
All experimental procedures were conducted in accordance with the National Institutes of Health guidelines and were approved by the NICHD Animal Care and Use Committee (ACUC). The protocol described below utilizes Nkx2.1-CreC/+;Ai9+/- pups to harvest MGE-derived interneuron precursors, but can be performed on any desired fluorescent reporter mouse line. Both male and female early postnatal mice (P0-P2) were used indiscriminately for donor and host tissue.
1. Solution Preparation
Reagent | Molecular Weight | Concentration (mM) | Grams/500 mL |
Sodium chloride | 58.44 | 87 | 2.54 |
Sodium bicarbonate | 84.01 | 26 | 1.09 |
Potassium chloride | 74.55 | 2.5 | 0.09 |
Sodium phosphate monobasic | 119.98 | 1.25 | 0.08 |
Glucose | 180.16 | 10 | 0.9 |
Sucrose | 342.3 | 75 | 12.84 |
2. Dissection Preparation
3. Transplantation Preparation
4. Removal of P0-P2 mouse brain
Figure 1: Schematic and images for brain dissection, technique #1
Dissection technique described in Steps 5.1-5.10. If striatal tissue is desired, place P1 brain in brain matrices ventral side up. Place razor blades into matrice slots through anterior brain to obtain coronal sections through striatum. Remove striatal pieces from both hemispheres, repeat for all sections containing striatum that are anterior to hippocampus. If striatal tissue is not desired, simply hemisect the brain, place the hemispheres medial side up in the dish, and remove the ventral-medial brain structures (thalamus, basal ganglia, etc.) to expose the hippocampus. Use tweezers to pinch of hippocampus, then flip hemisphere over and use tweezers to dissect out a chunk of cortical tissue. The vertical black lines through the schematic hemispheres where the cuts would be to remove the striatal sections. Scale bar = 500μm. Please click here to view a larger version of this figure.
5. Harvest Striatum, Hippocampus and Cortex, Technique #1
Figure 2: Schematic and images for brain dissection, technique #2
Dissection technique described in Steps 6.1-6.7. Pin brain onto a dissecting dish dorsal side up. After peeling the cortex forward, the hippocampus and striatum are visible and can be removed, then a section of cortex can be removed as described in previous technique. Depending on transgenic mouse line, striatum can be cleaned up to remove globus pallidus and other tissue. In Nkx2.1Cre;Ai9 mouse line, the globus pallidus has a significantly higher density of tomato+ cells compared to the striatum. Scale bar = 500 μm. Please click here to view a larger version of this figure.
6. Harvest Striatum, Hippocampus and Cortex, Technique #2
7. Generating Single Cell Dissociations
8. Preparing FACS-purified Cell Solutions for Transplantation
9. Transplantation Into P0-2 WT Pups
Figure 3: Schematic and images for transplantation
(A) Pictures of injection setup. Note that lambda is clearly visible through the pup's skull and should be used for zeroing the micropipette. Scale bar = 1 inch. (B) Schematic depicting the injection procedure to target the hippocampus. Please click here to view a larger version of this figure.
This protocol demonstrates how to harvest specific brain regions from early postnatal brains (Figure 1-2), collect single cell dissociations of interneuron precursors, and transplant these cells into various brain regions in naive WT postnatal pups (Figure 3). For posthoc analysis, brains that received interneuron precursor grafts were harvested between P30-35 to characterize cell morphology, neurochemical markers and electrophysiological properties. These types of assays are often carried out between P21-P30 in normal mice, but since the maturation of transplanted cells might be slightly delayed due to the dissection/dissociation procedure, waiting an additional 5-10 days is recommended to compensate for this delayed maturation. The type of analysis to be performed will dictate the proper strategy to harvest the brain. Notably, we did not observe preferential cell death of specific interneuron subgroups that could bias for or against certain subtypes20.
For immunohistochemical analysis, mice were perfused with 4% paraformaldehyde and the brains were removed. 50 μm vibratome slices were prepared through the targeted brain region and stored in antifreeze solution and/or processed for immunostaining as previously described20. Some brains did not contain any tomato+ cells, which could be due to improper targeting (e.g., injection too deep into the ventricle), cells lost or undergoing apoptosis during the grafting procedure, or rejection of transplanted cells by the host. Based on final cell counts, it is estimated that only 2-5% of grafted cells survive20, which is in line with other transplantation procedures22,23.
Not surprisingly, there was significant variability in the total number of tomato+ cells in successful transplants, ranging from dozens to several thousand tomato+ cells (Figure 4A). Grafted cells were localized in the correct regions, with many displaying interneuron morphologies and well-characterized interneuron neurochemical markers (Figure 4B). Similar cell survival numbers and maturation profiles were observed even when cells were grafted into new environments in heterotopic transplantations (Figure 4C).
In addition to immunohistochemical analysis, electrophysiological analysis on grafted cells was performed to confirm that they have integrated into brain circuitry and display expected intrinsic and firing properties. Brains were harvested from P30-35 mice and slices prepared for physiological recordings as previously described20. The grafted interneurons presented adult-like physiological properties and distinct firing patterns could be characterized that were representative of well-characterized interneuron subtypes (Figure 5A), suggesting that grafted interneurons were able to properly mature in the host environment. To verify that transplanted cells were integrated in the neuronal network, sEPSCs were also recorded (Figure 5B). In addition, a subset of transplants were performed with interneurons expressing ChR2 followed by recording from pyramidal cells localized near transplanted interneurons. These data demonstrated that postsynaptic GABAergic currents are evoked by blue light (Figure 5C-D).
Figure 4: Grafted interneuron precursors populate host brain regions Representative sections from P30 WT mice that were transplanted with tomato+ interneuron precursors at P1. (A) In homotopic cortex-to-cortex transplantations, the grafted cells populate all cortical layers and display morphologies that mimic endogenous interneurons. Selected images highlight the variability in cell numbers from different transplants, with the left image having a much greater number of tomato+ cells per section compared to the transplant on the right side. (B) Low magnification (left) and high magnification (right) representative sections from homotopic hippocampus-to-hippocampus grafts. Note that the majority of tomato+ cells in the stratum oriens (SO) express SST (likely O-LM cells) whereas many tomato+ cells in the stratum pyramidale (SP) express PV (likely basket cells), similar to endogenous hippocampal interneurons. (C) Example of a heterotopic transplantation (Cortex-to-Striatum) with tomato+ present in the striatum. Scale bars = 200 μm in A in low mag panel in B, 50 μm in C and high power mag in B. Please click here to view a larger version of this figure.
Figure 5: Grafted interneurons are electrophysiologically mature and integrate in the host neuronal network
(A) Representative examples of the highest firing frequencies recorded from grafted interneurons. Left, Fast Spiking interneuron from a Hip-to-Ctx graft, injected current steps: -100 pA and 520 pA; right, Non-Fast Spiking interneuron from a Ctx-to-Ctx graft, injected current steps: -100 pA and 360 pA. (B) Example of sEPSCs recorded in a Late Spiking interneuron from a Hip-to-Hip transplant. (C) Representative image displaying Nkx2.1-Cre;Ai32 cells (YFP) from a Ctx-to-Ctx transplant with a biocytin-filled (red) pyramidal cell. Scale bar = 50 μm. (D) Example of a GABAergic postsynaptic currents evoked by blue light pulses recorded in pyramidal cells, recorded with a [145 mM] Cl-. In black, average traces; in red, average response recorded in the presence of Gabazine. Please click here to view a larger version of this figure.
One critical aspect of this protocol is maximizing the survivability of the cells. Ensuring that the tissue and cells are always in ice cold carboxygenated sACSF is necessary to promote cell survival. This requires an efficient dissection and dissociation strategy to minimize the length of time that the cells spend in various solution and outside of the brain environment. Depending on the number of brain regions being dissected and transplanted, it can be beneficial to have a partner aid in the dissection and/or transplantation steps to decrease the length of the experiment. To ensure the health and survival of the pup, it is also critical to minimize the ice anesthesia time (< 4 min) and keep the pup heated with light during the transplantation procedure.
There can be significant variability in the transplanted brains, some may not contain any grafted cells while other brains will contain thousands. Combining several P0-P2 litters for dissection can increase the number of cells harvested for grafting, which allows one to either increase the transplant cell density and/or increase the number of grafted pups, both of which will improve the likelihood of successful transplants. To target the proper brain region, it is critical that the pup's head is stabilized and does not move when lowering the micropipette into the brain. Any shifting of the head or bending of the micropipette shaft can result in mistargeting and inaccurate injections (such as loss of cells into the lateral ventricle). Also, performing multiple injection sites per pup (either unilaterally or bilaterally) can improve success rates in case one injection site is mistargeted. In the future, it would be optimal to develop a more efficient stereotaxic strategy to stabilize the pup and provide more accurate and consistent injections.
Th P0-P2 time frame was chosen for both scientific and technical reasons. At this age, most interneuron precursors have migrated to their terminal sites but have minimal environmental interactions. This time frame may not hold true for other neuronal cell types and brain regions, so one may want to adjust these timepoints for their specific purposes. For example, it would be interesting to compare the therapeutic potential of these P0-P2 harvested interneuron precursors with MGE-harvested cells: one age may have more benefits and/or less unwanted side effects than the other. Additionally, it could be interesting to perform heterochronic transplants (e.g., collecting cells at P0-P2 and transplanting into P7-10 brains, or vice versa) to identify how temporal changes in the environment could affect cell fate and maturation. The advantage of performing injections at P0-2 is that the skull is relatively thin and can be easily penetrated with a sharp micropipette. Any injections over P5 will require removing or thinning the skull.
The analysis described in this protocol is limited to specific characteristics of interneuron subtypes that are differentially expressed between brain regions. In the future, one could utilize the power of single cell sequencing to characterize the full transcriptome of grafted cells. This unbiased approach could identify specific genes (or larger signaling cascades) that are strongly enriched or depleted in transplanted cells compared to controls, which would provide insight into candidate environmental cues that would influence fate determination or maturation.
The authors have nothing to disclose.
This research was supported by the National Institutes of Health (K99MH104595) and the NICHD intramural research program to T.J.P. We thank Gord Fishell, in whose lab this approach was originally established.
Sodium chloride | Sigma | S7653 | |
Sodium bicarbonate | Sigma | S6297 | |
Potassium chloride | Sigma | P9541 | |
Sodium phosphate monobasic | Sigma | S0751 | |
Calcium chloride | Sigma | C5080 | |
Magnesium chloride | Sigma | M2670 | |
Glucose | Sigma | G7528 | |
Sucrose | Sigma | S7903 | |
Brain Matrices | Roboz | SA-2165 | Only needed if harvesting striatum |
Fine point Dumont Forceps | Roboz | RS-4978 | |
Microdissecting scissors | Roboz | RS-5940 | |
Razor blades | ThermoFisher | 12-640 | |
Pasteur pipettes | ThermoFisher | 1367820C | |
Nanoject III | Drummond | 3-000-207 | |
Manual Manipulator w/ stand | World Precision Instruments | M3301R/M10 | |
5 ml round bottom plastic tubes | ThermoFisher | 149591A | |
60 mm Petri dishes | ThermoFisher | 12556001 | |
100 mm Petri dishes | ThermoFisher | 12565100 | |
Pronase | Sigma | 10165921001 | |
Fetal Bovine Serum (FBS) | ThermoFisher | 16140063 | |
DNase I | Sigma | 4716728001 | |
Celltrics 50um filters | Sysmex | 04-0042327 | |
Trypan blue | ThermoFisher | 15-250-061 | |
Hemocytometer | ThermoFisher | 02-671-6 |