Grafted neural stem cells develop into functional pyramidal and integrate into host cortical circuitry

Ulrica Englund*, Anders Bjo¨ rklund*, Klas Wictorin*, Olle Lindvall†, and Merab Kokaia†‡

Sections of *Neurobiology and †Restorative Neurology, Wallenberg Neuroscience Center, BMC A-11, Lund University, S-221 84 Lund, Sweden

Communicated by Per O. Andersen, University of Oslo, Oslo, Norway, September 30, 2002 (received for review March 26, 2002) In vitro expanded neural stem͞progenitor cells can undergo success of whole-cell recordings, neonatal rats were used be- region-specific differentiation after transplantation to the devel- cause of higher numbers of GFP-labeled neurons detected as oping or adult brain, and display morphologies and markers compared with adults (12). Axonal connections of the grafted characteristic of mature neurons. Here we have used patch-clamp cells were studied by using the retrograde tracer Fluoro- techniques to explore whether grafted stem cells also can develop gold (FG). physiological properties of mature neurons and become function- Our data indicate that transplanted GFP-positive (GFPϩ) ally integrated within host neural circuitry. The immortalized RN33B cells with morphological features of pyramidal neurons neural progenitor cell line, RN33B, prelabeled with GFP by using a extend long-distance axonal projections to appropriate brain lentiviral vector, was transplanted into the cortex or regions, develop a physiologically mature neuronal phenotype, of neonatal rats. We found that the grafted GFP-positive cells and become functionally integrated into host cortical circuitry. differentiated into cells with morphological features of cortical or hippocampal pyramidal neurons, and that many of them had Materials and Methods established appropriate cortico-thalamic and contralateral hip- Cell Culture and Lentiviral Transduction. RN33B cells were ex- pocampal connections, respectively, as revealed by retrograde panded as described (4, 5). A recombinant lentiviral vector tracing. Whole-cell patch-clamp recordings from grafted cells with carrying the reporter gene for GFP was produced (refs. 10 and morphological characteristics of pyramidal neurons showed that 11; see refs. 9 and 12 for details), and added to the culture they were able to generate action potentials, and received func- medium at a multiplicity of infection of 5 (based on the trans- tional excitatory and inhibitory synaptic inputs from neighboring ducing units͞ml on 293T cells) for 48 h, after which the medium cells. These data provide evidence that grafted neural progenitors was changed. After transduction, Ϸ85% of the RN33B-cells can differentiate into morphologically mature pyramidal projec- expressed GFP. tion neurons, establish appropriate long-distance axonal projec- tions, exhibit normal electrophysiological properties, and become Transplantation. Postnatal day 1–2 (P1–P2) Sprague–Dawley rats functionally integrated into host cortical circuitry. (B&K Universal, Stockholm) were used as recipients. Surgery was performed under deep hypothermia and with the animals progenitor cells ͉ GFP ͉ electrophysiology ͉ whole-cell recording ͉ mounted in a miniaturized stereotaxic frame (for details see ref. retrograde tracing 13). All animal-related procedures were conducted in accor- dance with local ethical guidelines and approved animal care ultipotent neural stem or progenitor cell lines can differ- protocol. The cell suspensions (100,000 cells per ␮l in Hanks’ Mentiate to neuronal and glial phenotypes, both in vitro and balanced salt solution) were injected with a glass capillary (70 after transplantation into the developing or adult brain (1–3). ␮m i.d.), attached to a 2-␮l Hamilton syringe, either into Immature progenitors possess the capability to migrate within hippocampus (100,000 cells) at anterior (A) ϭϪ1.2 and Ϫ1.5, NEUROSCIENCE the host brain parenchyma, and can, at least in some cases, adopt lateral (L) ϭϪ1.5, and ventral (V) ϭϪ1.5 and Ϫ1.8; or cortex morphological features and express markers of mature neurons. (100,000 cells) at A ϭ 0.5, L ϭ 2.2, and V ϭϪ2.5 and Ϫ3.0. The It remains unclear, however, whether grafted neural stem͞ latter ventral coordinates fell within the striatum, but the cells progenitor cell lines also can develop the physiological proper- investigated here were all located in the cortex. ties of mature neurons and become functionally integrated into For retrograde tracing of cortico-thalamic and contralateral host neural circuitry. hippocampal projections, respectively, eight animals received In the present study, we have used whole-cell patch-clamp RN33B cell transplants either unilaterally (n ϭ 6) or bilaterally recording to analyze the electrophysiological properties and (n ϭ 2) into cortex, and unilaterally into the hippocampus (n ϭ functional integration of intracerebrally grafted cells derived 5). Twelve rats were grafted bilaterally into the cortex for from a conditionally immortalized neural progenitor cell line, electrophysiological recordings. RN33B, generated from embryonic rat brainstem by retroviral transduction of the temperature-sensitive simian virus 40 large- FG Injections. At 15 weeks after transplantation, FG (2% solution T-antigen (4, 5). The RN33B cell line has a remarkable neuro- in sterile saline; Molecular Probes) (14) was iontophoretically genic capacity both in neonatal and adult recipients, and can injected from a glass micropipette (20 ␮m i.d.) with a 5-␮A differentiate into -like cells with morphologies of pyra- positive current, pulsed for 7 of every 14 s over 20 min, by using midal cells and after transplantation into cortex or a constant current generator Midgard (EDCO Scientific, Chapel hippocampus, and into medium-sized densely spiny neurons Hill, NC). For tracing of cortico-thalamic projections, FG was after transplantation into the striatum (5–9). The RN33B cells were first transduced with the GFP gene by using a lentiviral vector, and then implanted into the cortex and hippocampus. Abbreviations: FG, Fluorogold; aCSF, artificial cerebrospinal fluid; EPSC, excitatory postsyn- aptic current; EPSP, excitatory ; IPSC, inhibitory postsynaptic current; The GFP marker allows visualization of the grafted cells in their GABA, ␥-aminobutyric acid; NMDA, N-methyl-D-aspartate; NBQX, 2,3-dihydroxy-6-nitro-7- entirety, including fine details of their and (9), sulfamoylbenzo[f]quinoxaline; D-AP5, D(Ϫ)-2-amino-5-phosphonopentanoic acid; PTX, and identification of grafted cells by their native fluorescence for picrotoxin. electrophysiological recordings in brain slices. To increase the ‡To whom correspondence should be addressed. E-mail: [email protected].

www.pnas.org͞cgi͞doi͞10.1073͞pnas.252589099 PNAS ͉ December 24, 2002 ͉ vol. 99 ͉ no. 26 ͉ 17089–17094 Downloaded by guest on September 30, 2021 injected into ipsilateral thalamus, at two sites centered on the clamp mode, depolarization current pulses of 1 s duration and of ventro-lateral and ventro-basal thalamic nuclei (A ϭϪ2.0, L ϭ increasing magnitude were delivered to the recorded cells 2.3, V ϭϪ5.4 and A ϭϪ3.0, L ϭ 2.3, V ϭϪ5.4, respectively, through patch pipettes. PTX or NBQX plus D-AP5 were always with tooth bar (TB) at Ϫ3.0), i.e., in the area of the thalamus present in the perfusion medium when EPSP͞EPSCs or IPSCs where we have previously observed dense GFPϩ terminal were recorded, respectively. projections from intracortical grafts of RN33B cells (9). Com- missural projections were traced by FG injections into the Immunohistochemistry. At 1 week after the FG injections (i.e., 16 contralateral CA3, by using two injections at A ϭϪ3.0 to Ϫ1.5, weeks after transplantation), animals were killed. Rats were an- L ϭ 3.0, V ϭϪ3.2; and A ϭϪ4.0 to Ϫ1.5, L ϭ 3.0, V ϭϪ3.2, aesthetized with an overdose of pentobarbital (70 mg͞kg, Apoteks- with TB set at Ϫ2.3. bolaget, Umea˚, Sweden) and transcardially perfused (for details see ref. 9). Briefly, coronal or sagittal sections were cut on a freezing Electrophysiology. Four to seven weeks after transplantation, rats microtome at a thickness of 40 ␮m. The primary antibodies used were anaesthetized with halothane (Fluothane, Astra Zeneca, were GFP, chicken polyclonal (1:5,000, Chemicon) and FG, rabbit London), and their brains were taken out and placed into gassed polyclonal (1:2,000, Chemicon). For 3,3-diaminobenzidine (DAB)- ͞ (95% O2 5% CO2) ice-cold artificial cerebrospinal fluid (aCSF) reacted specimens, sections were pretreated with 3% H2O2 and containing 119 mM NaCl, 2.5 mM KCl, 1.3 mM MgSO4, 2.5 mM 10% methanol, and then incubated for 24 h with the primary CaCl2, 26.2 mM NaHCO3,1mMNaH2PO4, and 11 mM glucose. antiserum. For fluorescent double labeling, sections were first We focused on the cortex because, compared with the hippocam- incubated with the two primary antibodies together and then for 2 h pus, it contained higher number of GFP-expressing neurons, with the Cy2 and Cy3 conjugated secondary antibodies (1:400, increasing our chances for successful whole-cell recordings. Jackson ImmunoResearch). Coronal cortical slices (300 ␮m thick) were cut on a Vibratome (Ted Pella, Inc., Redding, CA), placed into a submerged cham- Microscopy and Cell Counting. The number of GFPϩ cells double- ber with gassed aCSF, warmed up to 35°C for 30 min, and stored labeled with FG in the cortex (n ϭ 4 rats) and hippocampus (n ϭ at room temperature for 1–4 h. Slices were then taken into the 4 rats) was counted in every sixth section. Colocalization analysis submerged recording chamber, which was permanently perfused was performed by using the OPENLAB software (ImproVision, with gassed aCSF at room temperature. Transplanted GFPϩ Coventry, U.K.) for scanning the sections in the z plane. An cells were identified by using UV light, whereas IR light in average of 9 and 6 sections were counted for the cortical and combination with differential interference contrast microscopy hippocampal grafts, respectively, in each animal. The GFPϩ technique were used for visual approach and patch-clamping. cells with morphological features of pyramidal projection neu- Whole-cell recordings were made with glass pipettes filled with rons (angular cell bodies and apical and basal branching den- one of three different solutions: for current-clamp recordings, drites) were counted in the deep cortical layers, where the 140 mM KCl͞10 mM NaCl͞10 mM Hepes͞4mMMgATP͞0.4 majority of FG-labeled host cortico-thalamic neurons were mM GTP͞0.2 mM EGTA; for voltage-clamp and excitatory located. In the hippocampus, we counted grafted GFPϩ cells postsynaptic current (EPSC) recordings, 117.5 mM Cs glu- located in the FG-labeled pyramidal layer of the CA3 region and conate͞17.5 mM CsCl͞10 mM Hepes͞8 mM NaCl͞0.2 mM displaying pyramidal-like cellular profile with branching apical EGTA͞2mMMgATP͞0.3 mM GTP͞5 mM QX-314 Br; for and basal dendrites. voltage-clamp and inhibitory postsynaptic current (IPSC) re- cordings, 135 mM CsCl͞10 mM Hepes͞0.2 mM EGTA͞8mM Results NaCl͞2mMMgATP͞0.3 mM GTP͞5 mM QX-314 Br, pH 7.2; Neuronal Differentiation. In the dorsal fronto-parietal cortex, osmolarity 295 mOsm, pipette resistance 4–6M⍀. Before each GFPϩ cells were found scattered in an area extending 2–3 mm experiment, biocytin (final concentration of 0.5%) was added to in rostro-caudal and medio-lateral directions, primarily in the the pipette solution. This biocytin diffused into the cells during deep cortical layers (IV–VI) (Figs. 1A and 2A). A large fraction whole-cell recordings and labeled them. After each electrophys- of the differentiated cells exhibited morphologies of cortical iological experiment, slices were stained either with AMCA- pyramidal neurons, with large angular somas and rich trees of avidin (7-amino-4-methylcoumarin-3-acetic acid; Vector Labo- branching and spine-bearing apical and basal dendrites. The cells ratories) or Cy3-avidin (Jackson ImmunoResearch) conjugates had the normal vertical alignment, i.e., perpendicular to the (1:200) to visualize biocytin, and double-labeling with GFP in the cortical layers, with the apical extending toward the pial recorded RN33B cells was confirmed. In total, 12 double-labeled surface (Fig. 1A). Other GFPϩ grafted cells, albeit lower in transplanted RN33B cells and 20 control (host) cells were number, had differentiated into smaller, multipolar cells with recorded in 12 animals. The average age of the animals from -like morphologies, as well as into cells with char- which the transplanted RN33B and corresponding control cells acteristic glial morphologies, including -like cells in the were recorded was 5.2 Ϯ 0.3 and 4.6 Ϯ 0.5 weeks, respectively. cortical gray matter (Fig. 1A), and -like cells in The EPSCs, IPSCs, and excitatory postsynaptic potentials and near the corpus callosum. (EPSPs) were filtered at 2.9 kHz and sampled at 10 kHz with a In the hippocampal formation, GFPϩ cells with neuronal patch-clamp amplifier (HEKA Electronics, Lambrecht, Ger- morphologies were distributed throughout the CA3 region and many). Data were stored on a G4 Macintosh computer for off- the hilus of the (Fig. 1B). Many of these cells were line analysis. The non-N-methyl-D-aspartate (NMDA) receptor- located in the CA3 pyramidal layer with a morphology and mediated EPSCs and ␥-aminobutyric acid type A (GABAA) position closely resembling intrinsic hippocampal pyramidal receptor-mediated IPSCs were recorded at holding potential of neurons, with bipolar branching dendrites and a normal orien- Ϫ70 mV, whereas NMDA receptor-mediated EPSCs were re- tation perpendicular to the cell layer (Fig. 1B). The GFPϩ cells corded at ϩ40 mV. 2,3-dihydroxy-6-nitro-7-sulfamoylbenzo- located within the hilus exhibited morphologies of both pyra- [f]quinoxaline (NBQX; 5 ␮M), D(Ϫ)-2-amino-5-phosphonopen- midal cells and interneurons. Occasionally, GFPϩ cells with tanoic acid (D-AP5; 50 ␮M), and picrotoxin (PTX; 100 ␮M) morphological features of granule cells were found within the were added to the perfusion medium to block non-NMDA, dentate layer (data not shown). Similar to the cortex, NMDA, and GABAA receptor-mediated currents, respectively. some GFPϩ grafted cells in the hippocampus had differentiated EPSP͞EPSCs and IPSCs were induced by electrical stimulation into astrocyte-like cells. Overall, Ͻ30% of the transplanted of the cortical areas close to the recorded cells by using bipolar differentiated cells expressed astrocyte-like morphologies, and stainless steel electrodes. To induce action potentials in current- only few cells had oligodendrocyte-like appearance.

17090 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.252589099 Englund et al. Downloaded by guest on September 30, 2021 Fig. 2. Schematic illustration of the distribution of GFPϩ grafted cells (green and yellow pyramids), the injection sites of the retrograde tracer FG (large red circles), and the distribution of FG-labeled host cells (red dots). (A)Inthe fronto-parietal cortex, the GFPϩ cells labeled by FG from the thalamus were found in the same layers as the FG-labeled host cells, intermingled with the intrinsic FGϩ cortico-thalamic neurons. About one-fifth of the GFPϩ pyrami-

dal-like cells (yellow pyramids) were found to contain the FG label. (B)Inthe NEUROSCIENCE hippocampus, FG was injected contralateral to the grafts in the medial and lateral CA3 region (large red circles). About one-third of the GFPϩ cells Fig. 1. Morphologies of GFPϩ grafted cells in the cortex and hippocampus possessing pyramidal neuron morphologies were GFP͞FG-double-labeled at 16 weeks after transplantation. (A) In the cortex, a large fraction of the (yellow pyramids). cc, corpus callosum; Cx, cortex; CpU, cuadate putamen; Hpc, GFPϩ cells differentiated into cells displaying morphologies of cortical pyra- hippocampus; Thal, thalamus. midal neurons (arrowheads), with large angular somas and rich trees of branching and spine-bearing apical and basal dendrites (see Inset for high magnification). The cells had the normal vertical alignment, perpendicular to Axonal Projections. To determine whether the grafted pyramidal- the cortical layers, with the apical dendrite extending toward the pial surface. ϩ In addition, some of the GFPϩ cells in the hippocampus had differentiated into like GFP neurons had established axonal connections with typical astrocyte-like cell profiles (asterisks). (B) A significant number of the their primary targets, we injected the retrograde tracer FG into GFPϩ cells had integrated into the pyramidal layer of CA3 and exhibited the ventro-lateral and ventro-basal thalamic nuclei (in animals morphology and position closely resembling the intrinsic hippocampal pyra- with cortical transplants; Fig. 2A) or into the hippocampal CA3 midal neurons (arrowhead), with bipolar branching dendrites and a normal region (in animals with hippocampal transplants; Fig. 2B). The ϭ ␮ ␮ orientation perpendicular to the cell layer. (Scale bar 65 minA and 180 m ability of the GFPϩ neurons to extend long distance projecting in B.) mol, molecular layer; py, pyramidal layer. axons is suggested by the presence of GFPϩ fibers along the internal capsule and hippocampal commissures, and networks of ϩ ϩ beaded GFP terminals in the ipsilateral thalamus and con- In addition, GFP cells with immature or undifferentiated tralateral hippocampus, as observed in animals with cortical and appearance were observed in varying numbers, often forming hippocampal cell grafts, respectively (9). Consistent with these chains of cells associated with small-sized blood vessels. As observations, a significant proportion of the GFPϩ pyramidal- reported previously (9), these cells express the progenitor mark- like cells, Ϸ20% in cortex and 35% in hippocampus, were found ers nestin and vimentin, and in a few cases also NG2. They are, to contain the FG label (Fig. 3 and Table 1). In the fronto- in most cases, not detectable at longer survival times, suggesting parietal cortex, the grafted GFPϩ cells labeled by FG from the that they constitute a pool of undifferentiated RN33B cells that thalamus were found in the same layers as the FG labeled, with time either undergo differentiation or die (9). GFP-negative host cells, and were thus intermingled with the

Englund et al. PNAS ͉ December 24, 2002 ͉ vol. 99 ͉ no. 26 ͉ 17091 Downloaded by guest on September 30, 2021 Fig. 3. Double immunofluorescent stainings of GFP and FG in cortex and hippocampus. (A–C) A significant number of the GFPϩ cells (A, green), found in deeper cortical layers, were double-labeled with FG (B, red; merged image in C). Arrowhead in C shows a clear example of a double-labeled cell. (D–F) High magnification of a GFP-expressing cell (D, arrowhead) with axonally retrogradelly transported FG granules (E, arrowhead) distributed in the cell body (F, merged image, arrowhead) and in the proximal part of the apical dendrite (F, see Inset). (G–I) GFP͞FG double-labeled grafted cell (G, GFP, green; H, FG, red; merged image in I) located in the pyramidal layer of the CA3 region and fulfilling the morphological criteria of a pyramidal neuron. (Scale bar ϭ 80 ␮minA,35␮minB, and 100 ␮minC.)

intrinsic FG-labeled cortico-thalamic neurons. In the hippocam- 4B1). These properties of transplanted RN33B cells were similar pus, the FG͞GFP double-labeled cells occurred both in the CA3 to those of control cells, i.e., cortical pyramidal neurons in layers pyramidal layer and in the dentate hilus, intermingled with host IV–VI recorded from the same group of animals (data not pyramidal neurons traced retrogradely from the contralateral shown). The average resting membrane potential (72.3 Ϯ 1.6 hippocampus. mV, n ϭ 12) and input resistance (106 Ϯ 13 MOhm, n ϭ 12) of RN33B cells did not differ from those of host control neurons Physiological Properties and Functional Integration. In this analysis, (73.6 Ϯ 1.2 mV, n ϭ 20; 113 Ϯ 15 MOhm, n ϭ 11, respectively). we only included grafted pyramidal-like cells, which showed Spike amplitudes, as measured between depolarization pulse and co-localization of GFP autofluorescence and biocytin staining spike peaks, were also not different between the host (72.0 Ϯ 3.6 (Fig. 4A). All recorded RN33B cells were located in deep cortical mV) and transplanted RN33B (67.8 Ϯ 3.8 mV) cells. layers. Next we explored whether transplanted RN33B cells received First, we tested the ability of these cells to generate action excitatory synaptic inputs from host neurons. In support of this potentials. In all recorded RN33B cells, depolarizing current hypothesis, electrical stimulation of the surrounding cortex pulses elicited action potentials, which exhibited pronounced evoked EPSPs in the RN33B cells (Fig. 4B2). Because GABAA frequency accommodation and afterhyperpolarizations (Fig. receptor-mediated inhibition was blocked by PTX, EPSPs were

17092 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.252589099 Englund et al. Downloaded by guest on September 30, 2021 Table 1. Numbers of GFP؉ neurons with pyramidal morphology in cortex and hippocampus retrogradely labeled with FG No. of No. of GFP͞FG %ofFG- Rat no. GFP-labeled cells* double-labeled cells* labeled cells

Cortex 1671421 2381437 3811114 4 112 13 12 Average 75 13 21 Hippocampus 1311342 217424 322836 4301447 Average 25 10 37

*Cells with pyramidal neuronal morphology were counted in every sixth section throughout the region containing both GFPϩ cells and FG-labeling, in the pyramidal layer of the CA3 region in the hippocampus and deeper layers in the cortex. Morphological criteria were angular cell bodies and apical and basal branching dendrites.

often manifested as so-called paroxysmal depolarizing shifts (PDSs; for example, see ref. 15) with superimposed, partially inactivated action potentials (Fig. 4B2). We then investigated whether the excitatory on the transplanted RN33B cells could express short-term plasticity, and whether they used the neurotransmitter glutamate and contained non-NMDA and NMDA receptors. All recorded cells exhibited paired-pulse facilitation of EPSCs, a form of short- term (16), at holding membrane potential of Ϫ70 mV (Fig. 4C1, black trace). These EPSCs were blocked by Fig. 4. Transplanted RN33B cells with neuronal morphology generate action ϩ the selective non-NMDA receptor antagonist NBQX (Fig. 4C1, potentials and are synaptically integrated into the host brain. (A) GFP cell red trace). In the presence of NBQX, NMDA receptor-mediated (Left) filled with biocytin (Right). Note another transplanted GFP cell (Left) that is not filled with biocytin. (Scale bar ϭ 25 ␮m.) (B) Current-clamp mode. EPSCs could still be induced on stimulation at holding potential ϩ Depolarization pulses induce action potentials (B1), whereas cortical stimula- of 40 mV (Fig. 4C2, black trace) and were effectively blocked tions induce EPSPs (B2) with superimposed action potentials in the double- by the selective NMDA receptor antagonist D-AP5 (Fig. 4C2, labeled transplanted RN33B cell. Resting membrane potential ϭϪ66 mV; PTX red trace). We also found that these excitatory synapses were is present in the aCSF. (C) Voltage-clamp mode. (C1) Paired-pulse facilitation active even without stimulation by recording spontaneous of non-NMDA receptor-mediated EPSCs (black trace) recorded at holding EPSCs (Fig. 4C3), which were blocked by addition of NBQX and potential of Ϫ70 mV from the transplanted double-labeled cell (same as in A). D-AP5 (Fig. 4C4). Addition of NBQX into the aCSF bocks EPSCs (red trace). (C2) NMDA receptor-

ϩ NEUROSCIENCE Finally, we examined whether transplanted RN33B cells also mediated EPSC (black trace) at holding potential of 40 mV in the presence of NBQX in the aCSF. Addition of D-AP5 blocks EPSC (red trace). All traces are an had inhibitory GABAergic synaptic inputs from host neurons. In average of three responses. (C3) Spontaneous͞miniature EPSCs are blocked the presence of NBQX and D-AP5 in the aCSF, paired stimu- (C4) by NBQX and D-AP5 addition. PTX is present in the aCSF. (D) Voltage- lations induced IPSCs, which exhibited paired-pulse depression clamp mode with holding potential of Ϫ70 mV. (D1) Paired-pulse depression (Fig. 4D1). These IPSCs were effectively blocked by addition of of IPSCs recorded from the transplanted double-labeled cell. (D2) IPSCs are PTX to the aCSF (Fig. 4D2). Spontaneous IPSCs (Fig. 4D3) blocked by addition of PTX to the aCSF. All traces are an average of three were also blocked (Fig. 4D4) by PTX application, providing responses. (D3) Spontaneous͞miniature IPSCs are blocked (D4) by PTX addi- further evidence for the presence of GABAergic inhibitory tion. NBQX and D-AP5 are present in the aCSF. Duration of the depolarizing synapses on the grafted RN33B cells. pulse in B is 1 s. Calibration for C 3 and 4 is the same as for D 3 and 4. All excitatory and inhibitory synaptic responses of the RN33B cells closely resembled those of host control pyramidal neurons Most importantly, however, we show that the graft-derived, recorded in the same animals and cortical layers (data not pyramidal-like cortical cells exhibit physiological properties of shown). mature intrinsic pyramidal neurons and become functionally Discussion integrated into host neural circuitry. These morphological and Previous studies have shown that RN33B cells, grafted into electrophysiological observations are quite remarkable consid- neonatal rat cortex and hippocampus, can generate cells with ering the fact that the generation of pyramidal neurons, both in morphological characteristics of intrinsic pyramidal neurons, cortex and hippocampus, is a prenatal process in the rat (17). including elaborated apical and basal dendritic arbors, richly Thus, at the time when the cells were implanted (P1–P2), cortical equipped with spines, and vertical alignment within appropriate and hippocampal neurogenesis of the host pyramidal cells has cortical and hippocampal layers (6–8). We demonstrate here been completed and the major efferent projections have already that a large fraction of the graft-derived neurons in both cortex been established. and hippocampus establish axonal projections with the two This study relies on the unambiguous identification of and principal targets of the intrinsic pyramidal cells, i.e., the ventral electrophysiological recording from the transplanted RN33B thalamic nuclei and contralateral CA3 region, respectively. cells. High resolution of the imaging equipment, and strong and

Englund et al. PNAS ͉ December 24, 2002 ͉ vol. 99 ͉ no. 26 ͉ 17093 Downloaded by guest on September 30, 2021 stable GFP expression in combination with retroactive biocytin lations were applied, the IPSCs expressed paired-pulse depres- labeling make us absolutely confident that all recordings made sion. This effect is thought to be mediated by activation of from the grafted RN33B cells are genuine. presynaptic GABAB receptors (refs. 21 and 22, but see ref. 23), All grafted RN33B cells with pyramidal-like neuronal mor- resulting in a feedback inhibition of GABA release. This obser- phology, which were recorded in current-clamp mode 4–7 weeks vation, therefore, may suggest that the inhibitory synapses after transplantation, were able to generate action potentials. formed by host neurons on the RN33B cells also contain The amplitude of action potentials, as well as input resistance presynaptic GABAB receptors. Taken together, inhibitory syn- and resting membrane potential of these cells closely resembled apses on the transplanted RN33B cells closely resemble those those of surrounding host cortical pyramidal neurons, indicating present on host cortical pyramidal neurons. that at this time point, the transplanted RN33B cells had reached Our data show the ability of in vitro expanded neural stem͞ a degree of maturation similar to that of host neurons. In line progenitor cells to develop into functional neurons and become with our observations, van Praag et al. (18) have reported that synaptically integrated into host cortical circuitry after grafting new granule cells formed from endogenous progenitors in the into the of neonatal rats. Whether similar integration subgranular zone of the rat dentate gyrus were able to generate would occur in adult brain remains to be elucidated. However, action potentials in response to depolarizing pulses at 4 weeks of at the stage of development when RN33B cells were trans- age. In the study of Auerbach et al. (19), neuronal precursors planted, cortical neurogenesis is complete (17), whereas synap- from cultured (for 6 days) embryonic tissue transplanted into the togenesis continues well into the third postnatal week in this hippocampus of 18-day-old rat embryos did not exhibit action region (24). Although the RN33B cells have a high intrinsic potentials earlier than 4 weeks after implantation, and only two propensity to form neurons, both in vitro and in vivo (4, 5, 7), the cells with neuronal morphology could generate spikes at this extent of neurogenesis and the types of neurons formed differ time point. Taken together, these data indicate that a time period greatly between different brain regions (see ref. 9). The ability of at least 4 weeks is needed to allow progenitors to develop a of the grafted progenitors to differentiate into diverse, function- physiologically mature neuronal phenotype. ally mature neuronal phenotypes and form appropriate long- The generation of EPSPs and PDSs in the RN33B cells distance axonal connections is thus likely to depend on signals induced by stimulation of host tissue strongly suggests that coming from the local tissue environment. The factors governing surrounding neurons make excitatory synapses on the trans- differentiation and functional integration of transplanted stem͞ planted cells. It seems less likely that this response was caused by progenitor cells remain largely unknown. Interestingly, in adult self-innervation, because action potentials generated by depo- hosts subjected to focal excitotoxic lesions, Shihabuddin et al. larizing pulses in the transplanted cells never induced EPSPs or (25) and Lundberg et al. (9) have observed that the grafted PDSs in the same cell (for example, see ref. 20). RN33B cells differentiate into fully mature neurons only in areas Whole-cell recordings of the transplanted RN33B cells in the where the intrinsic neurons are intact or partially spared, and voltage-clamp mode showed that they expressed paired-pulse that cells that migrate into neuron-depleted (i.e., gliotic) areas facilitation of EPSCs, which was similar to that recorded in host remain undifferentiated. These data strongly suggest that neu- pyramidal neurons, and is a characteristic feature of central ronal differentiation, and perhaps also functional integration, of excitatory synapses (for example, see ref. 16). These EPSCs were the grafted RN33B cells are controlled by the presence of as yet blocked by the selective non-NMDA receptor antagonist NBQX, unidentified local cues, and that direct cell-to-cell interactions, supporting the nature of these synapses. More- rather than diffusible molecules, are involved. over, excitatory synapses on the transplanted cells contained The demonstration that grafted stem͞progenitor cell lines can functional NMDA receptors, which were blocked by the NMDA produce functionally integrated neurons, even after host neu- receptor antagonist D-AP5. Together with the observation of rogenesis is complete and in brain areas outside the classical spontaneous EPSCs in the transplanted RN33B cells, our data neurogenic regions, suggest that such cells could be useful for indicate that excitatory synapses on these cells are functional, neuronal replacement and brain repair. express short-term plasticity and have a pharmacological profile similar to that of host pyramidal neurons. We thank Birgit Haraldsson, Ulla Jarl, Anneli Josefsson, Monica Lun- The transplanted RN33B cells exhibited both stimulation- dahl, Anna-Karin Olde´n, and Bengt Mattsson for excellent technical induced and spontaneous IPSCs, which were blocked by PTX, a assistance. This study was supported by the Swedish Research Council common GABAA receptor antagonist. This finding implies that (04X-3874 and 33X-08666), the Medical Faculty at the University of host neurons make not only excitatory but also inhibitory, Lund, and the Elsa and Thorsten Segerfalk, Crafoord, Knut and Alice GABAergic synapses on the grafted cells. When paired stimu- Wallenberg, and So¨derberg Foundations.

1. Gage, F. H. (2000) Science 287, 1433–1438. 12. Englund, U., Ericson, C., Rosenblad, C., Mandel, R. J., Trono, D., Wictorin, 2. Svendsen, C. & Caldwell, M. (2000) Prog. Brain Res. 127, 13–34. K. & Lundberg, C. (2000) NeuroReport 11, 3973–3977. 3. Martinez-Serrano, A., Rubio, F., Navorro, B., Bueno, C. & Villa, A. (2001) 13. Cunningham, M. G. & McKay, R. D. (1993) J. Neurosci. Methods 47, 105–114. Curr. Gene Ther. 1, 279–299. 14. Schmued, L. C. & Fallon, J. H. (1986) Brain Res. 377, 147–154. 4. Whittemore, S. R. & White, L. A. (1993) Brain Res. 615, 27–40. 15. Jones, R. S. & Lambert, J. D. (1990) Neuroscience 34, 657–670. 5. Onifer, S. M., Whittemore, S. R. & Holets, V. R. (1993) Exp. Neurol. 122, 16. Zucker, R. S. & Regehr, W. G. (2002) Annu. Rev. Physiol. 64, 355–405. 130–142. 17. Bayer, S. & Altman, J. (1995) in The Rat , ed. Paxinos, G. 6. Shihabuddin, L. S., Hertz, J. A., Holets, V. R. & Whittemore, S. R. (1995) (Academic, San Diego), pp. 1041–1078. J. Neurosci. 15, 6666–6678. 18. van Praag, H., Schinder, A. F., Christie, B. R., Toni, N., Palmer, T. D. & Gage, 7. Shihabuddin, L. S., Brunschwig, J. P., Holets, V. R., Bunge, M. B. & F. H. (2002) Nature 415, 1030–1034. Whittemore, S. R. (1996) J. Neurocytol. 25, 101–111. 19. Auerbach, J. M., Eiden, M. V. & McKay, R. D. (2000) Eur. J. Neurosci. 12, 8. Lundberg, C., Field, P. M., Ajayi, Y. O., Raisman, G. & Bjorklund, A. (1996) 1696–1704. Brain Res. 737, 295–300. 20. Lessmann, V. & Heumann, R. (1998) Neuroscience 86, 399–413. 9. Lundberg, C., Englund, U., Trono, D., Bjorklund, A. & Wictorin, K. (2002) Exp. 21. Davies, C. H. & Collingridge, G. L. (1993) J. Physiol. 472, 245–265. Neurol. 175, 370–387. 22. Pitler, T. A. & Alger, B. E. (1994) J. Neurophysiol. 72, 2317–2327. 10. Zufferey, R., Nagy, D., Mandel, R. J., Naldini, L. & Trono, D. (1997) Nat. 23. Jensen, K., Lambert, J. D. & Jensen, M. S. (1999) J. Neurophysiol. 82, 42–49. Biotechnol. 15, 871–875. 24. Sutor, B. & Luhmann, H. J. (1995) Perspect. Dev. Neurobiol. 2, 409–419. 11. Zufferey, R., Donello, J. E., Trono, D. & Hope, T. J. (1999) J. Virol. 73, 25. Shihabuddin, L. S., Holets, V. R. & Whittemore, S. R. (1996) Exp. Neurol. 139, 2886–2892. 61–72.

17094 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.252589099 Englund et al. Downloaded by guest on September 30, 2021