EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS
1
Epilepsy Center, Institute of Clini‐ cal Sciences, Lund University Hos‐ pital, Lund, Sweden; 2 Institute of Reconstructive Neurobiology, Life & Brain Center, University of Bonn and Hertie Foundation, Bonn, Germany; 3 Department of Psychia‐ try and Behavioral Sciences, Stan‐ ford University, Stanford, CA Corresponding Author: Miss Nata‐ lia Avaliani, Epilepsy Center, Insti‐ tute of Clinical Sciences, Lund Uni‐ versity Hospital, Lund, Sweden, e‐ mail : [email protected]; * M.A. and M.K. are equivalent co‐ senior authors Received May 16, 2014; accepted for publication July 14, 2014 ©AlphaMed Press 1066‐5099/2014/$30.00/0 This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typeset‐ ting, pagination and proofreading process which may lead to differ‐ ences between this version and the Version of Record. Please cite this article as doi: 10.1002/stem.1823
Optogenetics Reveal Delayed Afferent Synap‐ togenesis On Grafted Human Induced Pluripo‐ tent Stem Cell‐Derived Neural Progenitors NATALIA AVALIANI1, ANDREAS TOFT SØRENSEN1, MARCO LEDRI1, JOHAN BENGZON1, PHILIPP KOCH2, OLIVER BRÜSTLE2, KARL DEISSEROTH3, *MY ANDERSSON1, *MERAB KOKAIA1 Key words. Human iPS cells • Neural differentiation • Synaptic integration • Electrophysiology • Epilepsy ABSTRACT Reprogramming of somatic cells into pluripotency stem cell state have opened new opportunities in cell replacement therapy and disease model‐ ing in a number of neurological disorders. It still remains unknown, how‐ ever, to what degree the grafted human induced pluripotent stem cells (hiPSCs) differentiate into a functional neuronal phenotype and if they in‐ tegrate into the host circuitry. Here we present a detailed characterization of the functional properties and synaptic integration of hiPSC‐derived neu‐ rons grafted in an in vitro model of hyperexcitable epileptic tissue, namely organotypic hippocampal slice cultures (OHSC), and in adult rats in vivo. The hiPSCs were first differentiated into long‐term self‐renewing neuroepithelial stem (lt‐NES) cells, which are known to form primarily GABAergic neurons. When differentiated in OHSCs for six weeks, lt‐NES cell‐derived neurons displayed neuronal properties such as TTX‐sensitive sodium currents and action potentials (APs), as well as both spontaneous and evoked postsynaptic currents, indicating functional afferent synaptic inputs. The grafted cells had a distinct electrophysiological profile com‐ pared to host cells in the OHSCs with higher input resistance, lower resting membrane potential and APs with lower amplitude and longer duration. To investigate the origin of synaptic afferents to the grafted lt‐NES cell‐ derived neurons, the host neurons were transduced with Channelrhodopsin‐2 (ChR2) and optogenetically activated by blue light. Simultaneous recordings of synaptic currents in grafted lt‐NES cell‐derived neurons using whole‐cell patch‐clamp technique at 6 weeks after grafting revealed limited synaptic connections from host neurons. Longer differen‐ tiation times, up to 24 weeks after grafting in vivo, revealed more mature intrinsic properties and extensive synaptic afferents from host neurons to the It‐NES cell‐derived neurons, suggesting that these cells require extend‐ ed time for differentiation/maturation and synaptogenesis. However, even at this later time‐point, the grafted cells maintained a higher input re‐ sistance. These data indicate that grafted lt‐NES cell‐derived neurons re‐ ceive ample afferent input from the host brain. Since the lt‐NES cells used in this study show a strong propensity for GABAergic differentiation, the host‐to‐graft synaptic afferents may facilitate inhibitory neurotransmitter release, and normalize hyperexcitable neuronal networks in brain diseases, e.g. such as epilepsy. STEM CELLS 2014; 00:000–000
STEM CELLS 2014;00:00‐00 www.StemCells.com
©AlphaMed Press 2014
Optogenetics reveal delayed afferent synaptogenesis on grafted human iPS cell‐derived neural progenitors
2
INTRODUCTION Induced pluripotent stem (iPS) cells are somatic cells, reprogrammed by the addition of transcription factors to restore pluripotency [1]. The opportunity to create pluripotent stem cells from human fibroblasts has ex‐ panded the field of regenerative medicine and holds promise for the development of patient‐specific cell transplantation therapies [2, 3]. In temporal lobe epi‐ lepsy (TLE), a severe and often pharmacoresistant neu‐ rological condition, which is characterized by hyperexcitability and recurring seizures, patients often display hippocampal sclerosis with extensive loss of GABAergic inhibitory interneurons [4‐6]. Koch and col‐ leagues [7] have previously reported that following growth factor withdrawal, human neural stem cell lines, derived from embryonic stem cells and induced pluripo‐ tent stem cells generated in their laborato‐ ry, preferentially differentiate into GABAergic neurons by default [7, 8]. These cell lines could therefore be considered as suitable candidates for transplantation in TLE patients to replenish lost GABAergic neurons and thereby potentially counteract seizures. However, our knowledge on how transplanted hiPSC‐ derived neural progenitors differentiate in the brain, and whether they functionally integrate establishing appropriate afferent‐ efferent synaptic connections to the host neurons, is rather limited. In preclinical studies, grafting has been performed in animal models, for example of Parkinson’s disease and stroke, demonstrating the ability of such cells to differentiate into neuronal phenotype and seemingly contribute to improved functional outcomes [9‐12]. Transplanted long‐term self‐renewing neuroepithelial stem (It‐NES) cell‐derived neurons have been shown to acquire functional synaptic afferents [10, 12, 13]. However, it is not clear whether these af‐ ferents originate from the graft‐derived neurons or neurons from the host brain. This is an important ques‐ tion, since synaptic communication between grafted hiPSC‐derived neurons and the host neurons may be a crucial element for ameliorating impaired function due to the neurodegenerative diseases. To explore the func‐ tional neuronal properties and host‐to‐graft synaptic integration of GABAergic neurons generated from hiPSC‐derived lt‐NES cells in hyperexcitable tissue, we first grafted these cells into organotypic hippocampal cultures in vitro [14, 15]. In a second series of experi‐ ments, we transplanted It‐NES cells in rat hippocampus in vivo to allow for longer survival and differentiation time of up to 6 months. To explore host‐to‐graft synap‐ tic connections, we used an optogenetic approach. While expressing the light‐sensitive cation channel Channelrhodopsin‐2 (ChR2) in host neurons and selec‐ tively activating them by blue‐light, we recorded synap‐ tic currents from the grafted It‐NES cell‐derived neurons (see cartoon in Fig. 4 and 6; and Suppl. Fig. 3 and 4). Grafted It‐NES cell‐derived neurons in vitro hippo‐ campal tissue required over six weeks to differentiate www.StemCells.com
into a neuronal phenotype, and still did not fully resem‐ ble the electrophysiological characteristics of host neu‐ rons. Some limited host‐to‐graft synaptic afferents were also observed in It‐NES cell‐derived neurons. Longer survival time after transplantation in vivo allowed for further maturation of neuronal properties of It‐NES‐ derived neurons (although they still maintained a high input resistance) and significantly improved their affer‐ ent connectivity from host neurons.
METHODS
Animals Balb/C, ChR2‐YFP (Thy1‐ChR2/EYFP, Jackson laborato‐ ries) and GAD65‐EGFP (Glutamate Acid Decarboxylase 65‐Enhanced Green Fluorescent protein [16]) mice pups at postnatal day 6‐8 and male immune deficient nude rats (NIH Nude rat, Charles River), weighing 250‐300 g corresponding to 7‐9 week old age, used for the exper‐ iments, were housed under a 12/12‐h light cycle with ad libitum access to water and food. Mice pups were housed in standard cages with the mother, while nude rats were housed in individually ventilated cages. The Lund Ethical Committee for Experimental Animals ap‐ proved all experimental procedures.
Organotypic Cultures Organotypic hippocampal slice cultures (OHSC) were prepared as 250 µm thick sections of postnatal day 6–8 mice hippocampus as previously described in [17]. After decapitation, brains were removed and the two hemi‐ spheres were separated. Hippocampi were removed from each hemisphere and embedded in agar to offer mechanical support while slicing the sections in +3°C modified artificial cerebrospinal fluid containing in mM: Sucrose 195, KCl 2.5, NaH2PO4 1.25, NaHCO3 28, CaCl2 0.5, L‐ascorbic acid 1, pyruvic acid 3, glucose 7, and MgCl2 7 (all from Sigma‐Aldrich) equilibrated with carbogen (95% O2/5% CO2). After slicing, sections were kept for 15 min in ice‐cold washing medium containing Hanc’s Balanced Salt Solution (HBSS) with HEPES 20 mM, glucose 17.5 mM, NaOH 0.88 mM and penicil‐ lin/streptomycin (all from Gibco) before placing individ‐ ual slices on membrane inserts (Millipore, PICM01250) in 240 µl culturing medium in 24‐well dishes. The cultur‐ ing medium contained 50% MEM, 25% horse serum, 18% HBSS and 2% B27 supplemented with 0.5% penicil‐ lin/streptomycin solution (Life Technologies), glutamine 2 mM, glucose 11.8 mM, sucrose 20 mM. Slices were cultured as interface cultures at 37°C, 5% CO2 and am‐ bient O2 in 90% humidity. Medium was changed on day one of culturing and three times per week thereafter. Concentration of B27 in the medium was decreased to 0.1% after one week. ©AlphaMed Press 2014
3
Optogenetics reveal delayed afferent synaptogenesis on grafted human iPS cell‐derived neural progenitors
Channelrhodopsin expression Channelrhodopsin‐2 (ChR2) was expressed in the Balb/c mice hippocampal slices using the lentiviral construct LV‐Syn‐hChR2(H134R)‐EYFP (Addgene plasmid #20945). Viral vectors, produced by Lund University Vector Core facility following conventional protocols, was applied directly on the top of the hippocampal slice cultures on the day of culture preparation (Titer: 3.1x10^7 parti‐ cle/ml, 3μl/slice) and 24h before cell grafting. For expressing ChR2 in Nude rat hippocampus, AAV5‐hSyn‐hChR2(H134R)‐EYFP (Addgene, plasmid # 26973) was injected bilaterally in isoflurane anaesthe‐ tized rats. The total amount of virus suspension injected in each hippocampus was 4.5 μl with the titer of 2*10^12 particle/ml. With the rats’ head fixed in a ste‐ reotaxic frame, viral vector was injected through a glass capillary at 0.1 μl/min rate in the following coordinates: anterior‐posterior (AP) ‐6.2mm, medial‐lateral (ML) ±5.2mm, dorsal‐ventral (DV) ‐6.0, ‐4.8 and ‐3.6 mm, 1.5 μl at each location in DV plane. The reference points were bregma for the AP, midline for the ML and dura for the DV axis. The glass capillary was left in each DV point for 5 minute after injection to prevent the back‐ flow of the viral particles through the injection track.
Generation of hiPSC‐derived lt‐NES cells Human iPSC‐derived lt‐NES cells were produced as pre‐ viously described [7]. Long‐term expandable cell cul‐ tures were maintained as monolayer rosettes plated on Poly‐L‐Ornithine+Laminin (Sigma‐Aldrich) coated T25 flasks in proliferation medium, based on DMEM‐F12 supplemented with N2 supplement (1:100), L‐glutamine (1:100), glucose (1.6g/l) and penicillin/streptomycin (1:100) (all from Invitrogen). The growth factors FGF (10ng/ml), EGF (10ng/ml) (both from R&D systems) and B27 (1:1000) (Invitrogen) were added to the medium every day. Cells were passaged at a ratio of 1:2 to 1:3 every second to third day, using trypsin (Sigma‐Aldrich).
Cell grafting and transplantation For grafting on organotypic tissue cultures, or intrahippocampal transplantation in nude rats, hiPSC‐ derived It‐NES cells transduced with lentivirus carrying enhanced GFP, or retrovirus carrying RFP (both pro‐ duced by Lund University Vector Core facility following conventional protocols), were trypsinized and spun down at 300g. The cells were resuspended in DMEM/F12 medium to reach a concentration of 5 x 106 cells/ml and applied on the hippocampal slices with a pipette in the amount of 10 000 cells/slice. Grafting of the cells was done a few hours after the hippocampal slice cultures were prepared, except for when lentivirus was used for ChR2 expression, then the cells where grafted 24h after virus application. For intrahippocampal transplantation, cells were re‐ suspended in cytocon buffer to reach a concentration of 100 000 cells/μl and injected stereotaxically in both hippocampi in the same coordinates as the virus, 3μl in www.StemCells.com
total per hippocampus (1 μl at each DV coordinate). This was performed one week after AAV virus injection, to ensure extracellular virus clearance [18].
Electrophysiology Electrophysiological recordings in hippocampal organotypic cultures were performed one, three and six weeks in vitro after It‐NES cell grafting. The cultures were excised on their culturing membrane and trans‐ ferred to the recording chamber.
Acute hippocampal slice preparation Acute hippocampal slices were prepared from It‐NES cell transplanted Nude rats at five and six months after grafting. The Nude rats were anaesthetized and decapi‐ tated, the brains where removed from the skull and rapidly immersed in ice‐cold sucrose‐based cutting solu‐ tion, containing (in mM): sucrose 75, NaCl 67, NaHCO3 26, glucose 25, KCl 2.5, NaH2PO4 1.25, CaCl2 0.5, MgCl2 7 (pH 7.4, osmolarity 300‐305 mOsm). The cerebellum was removed and the two hemispheres were divided with a razor blade, positioned on the medial side and a ‘’magic cut’’ [19] was performed on the dorsal cortex. The hemispheres where then glued onto a cutting ped‐ estal with ‘’magic cut’’ side down and transferred to the cutting chamber containing sucrose‐based cutting solu‐ tion maintained at 2‐4°C and constantly oxygenated with carbogen (95 % O2/5 % CO2). Transverse slices of 300 μm thickness comprising hippocampus as well as entorhinal cortex were cut on a vibratome (VT1200S, Leica Microsystems) and immediately transferred to an incubation chamber containing artificial cerebrospinal fluid (aCSF) containing (in mM): NaCl 119, KCl 2.5, MgSO4 1.3, NaHCO3 26.2, NaH2PO4 1, glucose 11, CaCl2 2.5 (300 mOsm, pH7.4), oxygenated with carbogen and maintained at 34°C for 20 minutes. After cutting, slices were allowed to rest for one hour at room temperature before whole‐cell patch‐clamp recordings were per‐ formed. Individual slices were then placed in the re‐ cording chamber, continuously perfused at a rate of 3 ml/min with carbogen‐equilibrated aCSF.
Acute Human Neocortical slices Human neocortical resections were obtained from De‐ partment of Neurosurgery at Lund University Hospital. Informed consent was obtained from the patients, in accordance to the Declaration of Helsinki, and local Eth‐ ical Guidelines. Tissue was immediately placed in carbogen bubbled sucrose solution, containing (in mM): 200 sucrose, 21 NaHCO3, 10 glucose, 3 KCl, 1.25 NaH2PO4, 1.6 CaCl2, 2 MgCl2, 2 MgSO4. The tissue was also cut in the bubbled sucrose solution, in coronal slic‐ es of 500 um thickness on a Leica Vibratome. Slices were allowed to rest for 3 hours in oxygenated human artificial cerebrospinal fluid (haCSF) at 34°C, containing (in mM): 129 NaCl, 21 NaHCO3, 10 glucose, 3 KCl, 1.25 NaH2PO4, 2 MgSO4, 1.6 CaCl2, (300 mOsm and 7.4 pH), and then transferred to the recording chamber per‐ ©AlphaMed Press 2014
4
Optogenetics reveal delayed afferent synaptogenesis on grafted human iPS cell‐derived neural progenitors
fused with oxygenated haCSF at a flow rate of 3 ml/minute for whole‐cell patch‐clamp recordings.
Whole‐cell patch clamp recordings GFP‐ and RFP‐expressing cells in the slices were identi‐ fied using a wide‐band excitation filter and visualized for whole‐cell patch‐clamp recordings using infrared differential interference contrast video microscopy (BX50WI; Olympus). Recordings were performed at 32‐ 34°C using a glass pipette filled with solution containing (in mM): K‐gluconate 122.5, KCl 17.5, NaCl 8, KOH‐ HEPES 10, KOH‐EGTA 0.2, MgATP 2 and Na3GTP 0.3 (295 mOsm, pH 7.2; all from Sigma‐Aldrich), except for re‐ cordings for inhibitory connectivity in OHSCs, where the pipette solution contained (in mM): 135 CsCl, 8 NaCl, 10 HEPES, 0.2‐1 EGTA, 2 MgATP, and 0.2 GTP. Average pi‐ pette tip resistance was between 3–5 MΩ. Pipette cur‐ rent was corrected online before gigaseal formation while fast capacitive currents were compensated for during cell‐attached configuration. Only experiments with series‐resistance values of less than 20 MΩ were selected for analysis. Biocytin was included in the pi‐ pette solution at 0.5–1 mg/ml to retrospectively identify recorded cells. All recordings were done using a HEKA EPC10 amplifier (HEKA Elektronik, Germany) and sam‐ pled at 10 KHz.
Intrinsic properties Resting membrane potential (RMP) was recorded in current clamp mode at 0 pA immediately after estab‐ lishing whole‐cell configuration. Series resistance and input resistance (Ri) was calculated from a 5 mV pulse and monitored throughout the experiment. Action po‐ tential (AP) threshold was determined by 500 ms square current step injections at RMP, with 5 pA increments for It‐NES cell‐derived neurons in OHSCs, 20 pA increments for It‐NES cells‐derived neurons in acute slices of rat hippocampus and 50 pA increments for the host cells of the OHSCs and human cortical neurons. Ramp injection of 1 s current was used to determine action potential threshold in addition to step depolarization. AP ampli‐ tude was measured from threshold to peak and dura‐ tion was measured as the width at the threshold. The duration of the afterhyperpolarization (AHP) was meas‐ ured as the entire duration of the potential below RMP, immediately following the action potential. Whole‐cell currents were measured in voltage‐clamp mode at a holding potential of ‐70mV and voltage steps were de‐ livered in 10 mV (200 ms) increments. Voltage‐gated sodium channels were blocked with 1 μM tetrodotoxin (TTX, Tocris).
Spontaneous and evoked synaptic currents Spontaneous excitatory postsynaptic currents (sEPSCs) in OHSCs were measured at –70 mV in the presence of the GABAA‐receptor blocker picrotoxin (PTX, 100 mM, Tocris). Evoked EPSCs in OHSCs were elicited with two square‐wave pulses (100 μs, 150–300 μA, 20 Hz) deliv‐ www.StemCells.com
ered by a DS3 Constant Current Isolated Stimulator (Digitimer Ltd) connected to a single silver‐silver chlo‐ ride wire inserted in a glass pipette filled with aCSF (1.5‐ 2 MΩ tip resistance). MiniAnalysis software (Synaptosoft Inc.) was used for detection and analysis of spontaneous postsynaptic currents. Their rise‐time was measured as the time between 10 and 90 % of the max‐ imum amplitude, while the decay time was first fitted with a single exponential and then measured as above. N‐methyl‐ D‐aspartic acid (NMDA) and a‐amino‐3‐ hydroxy‐5‐methyl‐4‐isoxa‐ zolepropionic (AMPA) recep‐ tors were blocked using 50 μM (2R)‐amino‐5‐ phosphonovaleric acid D‐AP5 and 5 μM 2,3‐dihydroxy‐ 6‐nitro‐7‐sulfamoyl‐benzo[f]quinoxaline‐ 2,3‐dione (NBQX, Abcam Biochemicals), respectively.
Optogenetics For optogenetic depolarization of ChR2 expressing cells, blue light was applied at 460 nm wavelength with a LED light source (Prizmatix, Modiin Ilite, Israel) and deliv‐ ered through a water immersion 40x microscope objec‐ tive. Stimulation of ChR2 expressing cells was done ei‐ ther by continuous application of the blue light, for 5‐ 15s or 1ms pulses, paired or as a train of 10, separated by 100 ms intervals. Spontaneous postsynaptic current frequencies in It‐NES‐derived neurons was analyzed before, during and after optical activation of the host tissue to evaluate synaptic connectivity between host and grafted cells.
Immunohistochemistry and imaging All slice preparations were fixed in 4% paraformalde‐ hyde solution in PB buffer (for 12 to 24 h) after patch‐ clamp recordings, then rinsed with KPBS and stored in Walter’s antifreeze medium at ‐200C. Concentrations of the primary antibodies applied overnight at room tem‐ perature was: Mouse anti‐RFP (1:1000), Rabbit anti‐GFP (1:10 000), Mouse anti‐Human Nuclei (1:300), Rabbit anti‐GAD65/67 1:1000. Secondary antibodies were used in the concentration of: Cy2 anti‐rabbit 1:400, Cy3 anti‐ mouse 1:400, Cy5 streptavidin 1:300, Cy3 streptavidin 1:400. Hoechst was applied at a concentration of 1:1000 in the last PBS rinsing solution before mounting. Images were acquired either by confocal microscopy (Fig.1g; Suppl. Fig.5 A, D and Suppl. Fig. 6; Inverted Ni‐ kon Eclipse Ti microscope Csi) or by epifluorescence microscopy (all other figures; Olympus BX61).
Data analysis Analysis of electrophysiological data was performed using Fitmaster (HEKA elektronik), Igor Pro (Wavemetrics) and MiniAnalysis (Synaptosoft Inc.). Sta‐ tistical analysis of the data was done by Student’s t‐test or ANOVA, followed by Tukey‐Cramer post‐hoc analysis, and the level of significance was set at p
