equivalent neurogenic potential of wild-type and gfp

BASIC AND EXPERIMENTAL RESEARCH

Equivalent Neurogenic Potential of Wild-Type andGFP-Labeled Fetal-Derived Neural Progenitor Cells

Before and After Transplantation Into the RodentHippocampus

Guilherme Lepski,1,2,5 Cinthia E. Jannes,3 Johanna Wessolleck,1 Eiji Kobayashi,4 and Guido Nikkhah1

Introduction. The hippocampal formation is a specific structure in the brain where neurogenesis occurs throughoutadulthood and in which the neuronal cell loss causes various demential states. The main goal of this study was to verifywhether fetal neural progenitor cells (NPCs) from transgenic rats expressing green fluorescent protein (GFP) retain theability to differentiate into neuronal cells and to integrate into the hippocampal circuitry after transplantation.Methods. NPCs were isolated from E14 (gestational age: 14 days postconception) transgenic-Lewis and wild-type Sprague-Dawley rat embryos. Wild-type and transgenic cells were expanded and induced to differentiate into a neuronal lineage invitro. Immunocytochemical and electrophysiological analysis were performed in both groups. GFP-expressing cells wereimplanted into the hippocampus and recorded electrophysiologically 3 months thereafter. Immunohistochemical analysisconfirmed neuronal differentiation, and the yield of neuronal cells was determined stereologically.Results. NPCs derived from wild-type and transgenic animals are similar regarding their ability to generate neuronalcells in vitro. Neuronal maturity was confirmed by immunocytochemistry and electrophysiology, with demonstrationof voltage-gated ionic currents, firing activity, and spontaneous synaptic currents. GFP-NPCs were also able to differ-entiate into mature neurons after implantation into the hippocampus, where they formed functional synaptic contacts.Conclusions. GFP-transgenic cells represent an important tool in transplantation studies. Herein, we demonstratetheir ability to generate functional neurons both in vitro and in vivo conditions. Neurons derived from fetal NPCs wereable to integrate into the normal hippocampal circuitry. The high yield of mature neurons generated render these cellsimportant candidates for restorative approaches based on cell therapy.

Keywords: Neural progenitor cells, Transplantation, Electrophysiology, Synaptic currents.

(Transplantation 2011;91: 390–397)

The hippocampal formation represents an interestingstructure in the brain, not only because it is the site of

many degenerative processes but also because it is a structurein which new neurons are generated throughout adult life. Atthe present time, it is unclear how and to what extent thesenew neurons contribute to the learning process and storage ofnew information. A better understanding of stem-cell biologyand hippocampal neurogenesis can make inroads also to an-

other interesting explorative field toward the functional res-toration of the human mnestic center.

To date, fetal neuronal tissue is mostly used for restor-ative strategies in the brain. A vast body of evidences confirmsgreat neurogenic potential of precursors present in the unde-veloped brain, which are able to survive in the host tissue anddifferentiate into mature neurons, form synaptic contacts de-tectable by electron microscopy, release neurotransmitters,and restore functional deficits in both animal models andclinical trials for Parkinson’s disease (1–5). Although the ob-servation that fetal tissue transplantation induces functionalamelioration in behavioral tests dates back to 1979 (6), theexperimental evidence of functional integration into local cir-

This work was supported by DFG (Deutsche Forschungsgemeinschaft),DAAD (Deutscher Akademischer Austauschdienst), the German Parkin-son Foundation, and BMBF (Bundesministerium fur Bildung undForschung, Germany).

1 Department of Stereotactic and Functional Neurosurgery, Laboratory ofMolecular Neurosurgery, University of Freiburg, Freiburg, Germany.

2 Department of Neurosurgery, University of Tubingen, Tubingen, Germany.3 Laboratory of Molecular Biology LIM15, Department of Neurology School

of Medicine, University of Sao Paulo, Sao Paulo, Brazil.4 Division of Organ Replacement Research, Centre for Molecular Medicine,

Jichi Medical University, Tochigi, Japan.5 Address correspondence to: Guilherme Lepski, Ph.D., Department of Neu-

rosurgery, Eberhard-Karls-Univesity, Hoppe-Seyler-Strasse 3, 72076 Tu-bingen, Germany.

E-mail: [email protected]

G.L. participated in research design, writing of the manuscript, performanceof the research, and data analysis; C.E.J. and J.W. participated in perfor-mance of the research; E.K. contributed an experimental tool (animalmodel); and G.N. participated in research design, writing of the manu-script, and data analysis.

Received 17 September 2010.Accepted 22 October 2010.Copyright © 2011 by Lippincott Williams & WilkinsISSN 0041-1337/11/9104-390DOI: 10.1097/TP.0b013e3182063083

390 | www.transplantjournal.com Transplantation • Volume 91, Number 4, February 27, 2011

http://www.transplantjournal.com

cuits was published only in 2005 (7, 8). In this regard, becauseof technical difficulties in labeling transplanted cells specifi-cally and recognizing them in living slices, few reports havedescribed electrophysiologic maturation and integration ofprogenitor cells grafted in the brain. Considering fetal-derived neural progenitor cells (NPCs), there are only fewreports on functional maturation after transplantation intothe brain of neonatal rats (9, 10). Under these conditions, thefavorable microenvironment in the developing brain, per-missive to maturation of stem cells, represents a distinct sit-uation to that observed in the adult or in diseased brain.

Recently, Kobayashi and coworkers (11) developed aninbred transgenic Lewis rat strain that express green fluores-cent protein (GFP) ubiquitously. In this strain, the reportergenes were driven under a ubiquitous cytomegalovirus en-hancer/chicken �-actin promoter. This rat model representsa valuable tool in the context of cell-based therapies and cel-lular transplantation studies. For instances, the GFP is anideal marker because it simplifies genotyping and can be vi-sualized directly under ultraviolet light or by immunohisto-chemical enhancement. However, it remains to be clarifiedwhether the neurogenic potential of GFP-labeled NPCs issimilar to that of wild-type cells, and whether GFP expressionis stable through all developmental stages.

Based on the background outlined, the main scope ofthis study was to evaluate the neurogenic potential of ex-panded GFP-labeled NPCs in comparison with wild-type

ones and the ability of those cells to generate functional neu-rons after implantation into the adult hippocampus, wherethey might be integrated into local neuronal circuitries.

RESULTS

Differentiation of Cells Derived From GFP-Expressing Transgenic Fetal Tissue

After expansion in medium containing fibroblastgrowth factor b and epithelial growth factor, aggregates ofNPCs are formed, so-called neurospheres, which are stronglypositive for markers of immature neuroepithelial cells such asnestin (Fig. 1A). At this stage, cultures derived from wild-typeand transgenic animals were stained for nestin and glialfibrilar acidic protein (GFAP), without statistical difference(��5%, P for nestin 0.8675 and for GFAP 0.9132, n�5 ex-periments). In both groups, the percentage of microtubule-associated protein 2 (MAP2)-positive cells at this stage was 0(P�0.9119; Fig. 2A).

By cultivating the cells onto polyornithine-coated coverslips in the presence of brain-derived neurotrophic factor and3-isobutyl-1-methylxanthin, phosphodiesterase inhibitor, cellu-lar elongations grew from the spheres, and neuroblasts wereformed, which then differentiated into neurons during their mi-gration process away from the spheres, as described previously(Fig. 1B–D) (12). After 1 week, expression of neuronal markerswas observed, such as MAP2, isoforms a and b (MAP2ab), �III-

FIGURE 1. In vitro neuronal dif-ferentiation of green fluorescentprotein (GFP)-labeled fetal-derivedrodent neural stem cells. (A) Con-focal microscopy of a typical neu-rosphere strongly positive fornestin (red), in addition to the GFPsignal (green), formed after long-term expansion in medium contain-ing fibroblast growth factor b andepithelial growth factor. (B–D) Cul-ture appearance after plating ontopoly-L-ornithine-coated cover slipsmaintained in differentiation me-dium for 7 days; microtubule-associatedprotein2(MAP2)-positiveneurons (red), also GFP positive,can be observed migrating awayfrom the core of the spheres.(E–H) Triple fluorescence show-ing the morphology of a matureneuronal cell, which was re-corded from and filled with biocy-tin during recording (MAP2, red;GFP, green; and biocytin, blue).(I) Confocal microscopy illustrat-ing a MAP2, isoforms a andb-positive neuron (red) in a wild-type culture. All scale bars ad-justed to 40 �m.

© 2011 Lippincott Williams & Wilkins 391Lepski et al.

tubulin (which marks also young neuronal cells), neuron-specific nuclear protein (NeuN), and neurofilament 200 kDa(NF200). The comparison between wild-type and transgeniccultures revealed no statistical difference, for � of 5% (P values:MAP2, 0.5643; �III-tubulin, 0.6782; NeuN, 0.6734; and NF200,0.7719; Fig. 2B).

Comparison Between Transgenic and Wild-TypeCultures Regarding Functional Maturation InVitro

At the seventh differentiation day in vitro, GFP-expressing and wild-type cells were compared regarding their

electrophysiologic properties. Voltage-clamp recordings re-vealed high-amplitude Na and K currents, with voltage-dependent activation identical to that of mature neuronalcells. In Figure 3, a typical recording from a neuron derivedfrom GFP-NPCs is shown. At this stage, we could elicit trendsof action potentials in current clamp, and spontaneous syn-aptic currents could be recorded on both cell groups in acomparable extent. The peak amplitudes of Na and K cur-rents were analyzed in both groups, and no statistical differ-ence was found (Fig. 2C, P�0.4378).

Analysis of Functional Maturity of GFP-Expressing Cells After Implantation Into theRodent Hippocampus

Of the surviving cells expanded in vitro, 97.4% ex-pressed high levels of GFP just before stereotactic implanta-tion. Three months after implantation, the hippocampus waspopulated by GFP-positive cells, most of which were also pos-itive for mature neuronal markers. Figure 4 depicts examplesof cells positive for MAP2ab (A–C), NF200 (D–F), and NeuN(G–I). Synthesis of neurotransmitters typically encounteredin the hippocampal formation, such as �-aminobutyric acid(GABA)-amino-transporter 1 (J–L) and glutamate (M–O),was also demonstrated by immunohistochemistry. Stereo-logical quantifications revealed 4514�317 GFP-positive cells,and 3528�219 NeuN/GFP-positive neurons derived fromthe implanted cells, which corresponds to 3.0% and 2.4% ofthe implanted cells.

The functional maturation of the implanted cells was alsoassessed by patch clamp (Fig. 5). The voltage-clamp protocolrevealed large inward Na� and outward K� currents, as con-firmed after pharmacologic blocking. These current-voltage re-lationships depicted a clear voltage dependence. We were alsoable to elicit trends of action potentials in current clamp mode,by injection of �20- to �70-pA current during 1000 msec froma holding potential of �80 mV (Fig. 5E). Excitatory postsynapticpotentials were recordable in current clamp mode (Fig. 5F);blockade of the GABAergic component with bicuculline wasshown to reduce the frequency and amplitude of synaptic events,whereas total blockade was achieved when the glutamate �-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate, glutamatereceptor blocker 6-cyano-7-nitroquinoxaline-2,3-dione,AMPA/kainate receptor antagonist was added to bicuculline.Taken together, our data indicate that neurons generatedfrom grafted cells were indeed being incorporated into localhost neuronal networks.

DISCUSSIONIn this study, we report complete morphologic matura-

tion of GFP-labeled NPCs in vitro and after implantation intonormal nonlesioned hippocampus, evidenced (1) by expressionof mature neuronal markers in immunohistochemical staining(NeuN, MAP2, and NF200) and (2) through synthesis of GABAand glutamate, demonstrated by immunohistochemistry. In ad-dition, electrophysiology confirmed neuronal maturity, detect-ing large Na� and K� voltage-dependent currents, firing oftrends of action potentials, and the presence of spontaneousGABAergic and glutamatergic excitatory synaptic potentials.Moreover, the functional maturation achieved by GFP-labeledcells was similar to that of wild-type cells, and GFP expression

FIGURE 2. Comparison between immuncytochemicaland electrophysiologic properties of cultures derived fromtransgenic and wild-type animals. (A) Quantification ofcells positive for markers of immaturity after three pas-sages (at least 3 weeks) in expansion medium, beingcounted the cells that migrated away from the neuro-spheres and grew adherent to the covers slip surface,in relation to the total 4�,6-diamidine-2-phenylindoledihydrochloride-positive nuclei. (B) Quantification of positiv-ity for markers of immature (�III-tubulin) and mature neuronalcells (microtubule-associated protein 2, isoforms a and b[MAP2ab], neuron-specific nuclear protein [NeuN], and neu-rofilament 200 kDa [NF200]) after 7 days in differentiation me-dium. (C) Results of electrophysiologic recordings from cellspatched at the seventh differentiation day, being shown thepeak conductances for Na and K currents. Overall, no statisticsignificant difference was observed for the compared groupsat a significance level of 5% (with Man-Whitney U test). GFAP,glial fibrilar acidic protein.



was stable throughout the differentiation process. The func-tional integration of transplanted cells reported herein supportsthe notion that the activity of neurons derived from the graftedcells can be regulated by host neurons. Precisely how these cellsinfluence the overall behavioral performance in animals remainunclear and should be the focus of future investigations.

Fetal neural tissue has been used extensively for restorativepurposes. These important reports provided the first evidencethat primary fetal cells were able to survive within the brain of thehost and generate mature neurons, which in turn grew long ax-ons that could reach relatively distal structures, organize newsynapses, and release neurotransmitters such as dopamine (6,13–17). Several clinical trials in humans with Parkinson’s diseasesubmitted to fetal mesencephalus implants into the striatumhave shown significant improvements in motor symptoms andF18-fluordopa uptake in the striatum (1, 18). Postmortem anal-ysis revealed survival of tyrosine hydroxylase immunore-active cells, together with evidence of neosynaptogenesison electron microscopy, and integrity of the blood-brainbarrier, proving that functional integration of the graft hadtaken place in the host (2, 4). Nevertheless, how implantedcells integrate functionally with the adult host brain re-mains an intriguing and unexplained issue.

In our view, neural stem cells from fetal brain tissueremain a potential source of cells for brain repair because theycan be clonally expanded in vitro for prolonged periods. Wehave recently reported that long-term expanded humanNPCs retain the ability to generate morphologic mature neu-rons in vitro (19) and give rise to young neuronal cells aftergrafting into the normal central nervous system (20). In thisreport, we counted more than 3000 surviving mature neuronsderived from the implanted GFP-NPCs in the rodent hip-pocampus. In a previous report, other authors (10) workingwith a progenitor cell line RN33B implanted into the hip-pocampus and neocortex of neonatal rats observed 25 pyra-midal cells in the hippocampus and 75 in the cortex of 4animals. Extrapolating these values for each sixth brain sliceto the entire brain gives a total of 150 cells in the hippocampusand 450 in the neocortex, which are significantly lower valuesthan those reported in this study. One possible explanationfor this discrepancy might be the enriched environment, inwhich our animals were maintained after surgery. The impor-tance of the environmental conditions for hippocampalneurogenesis was emphasized by other authors (21–23). Inaddition to possible technical differences in the culture meth-ods and implantation techniques, it might be also reasonedpossible that pyramidal neuronal cells indeed represent asmall subgroup of the neuronal cells being generated fromneural stem cells. In fact, the interneuron cell type seemed topredominate in our slices.

The recent development of transgenic animals that ex-press ubiquitously cell markers such as LacZ or GFP has per-mitted an enormous advance on the investigation of the fateof transplanted neuronal progenitors into the normal or dis-eased brain (11). However, the genetic modification is as-sumed to cause enhanced toxic and immunogenic effects,which in turn might impair proper neuronal maturation andintegration into the host tissue (11, 24). In this study, wereport similar functional properties of neurons derived fromboth transgenic and wild-type fetal neural tissue. In addition,it has been described that adult neuronal cells initially labeled

FIGURE 3. In vitro functional maturation of green fluo-rescent protein-labeled neural stem cells derived fromrodent fetal neuronal tissue. (A) Phase contrast micros-copy showing the patching process of a cell in culture.(B) Voltage clamp traces showing absolute inward andoutward currents from cell patched in A; holding poten-tial �80 mV, pulse duration 100 msec, intensities from�70 to �70 mV at 10 mV increment; leak and capacitivecurrents were subtracted by a p/-4 protocol (for detailssee text), traces represent average of five consecutivesweeps. (C) Voltage-dependent activation of Na� cur-rents (I–V relationship). Experimental data were fittedwith a Boltzmann function multiplied with a driving forcederived from the Nernst equation in the formI(V)�([V�E]Gmax)/(1�exp[�(V�V1⁄2)/k]), see text.Best-fit parameters were for the represented cell V1/2��8.7�0.5 mV, slope�7.6�0.4 mV, gmax�39.6�1.6 nS,and ENa�101.4�2.4 mV. (D) I–V relationships for out-ward currents (K� currents), fitted with the same theo-retical model described for Na�. Best-fit parameterswere V1/2�15.5�0.8 mV, slope�11.0�0.3 mV,gmax�29.8�1.9 nS, and EK��89.5�10.1 mV. (E) Trendsof action potentials elicited in current clamp after injec-tion of a 70-pA current applied during 100 msec. (F)Spontaneous synaptic currents recorded from the samecell.


with GFP do not continue expressing GFP by the end of thematuration process (25). Our data do not corroborate theseprevious observations but underline the value of GFP-labelingmethods on studying neuronal differentiation after trans-plantation. Moreover, in further studies on dopaminergic

differentiation of GFP-labeled NPCs from the same rat strainin a rat model of Parkinson’s disease, we could verify that upto 80% of striatal tyrosine hydroxylase-positive cells on thelesion side coexpress GFP 15 weeks after transplantation,proving a remarkable stability of the marker (26).

FIGURE 4. Mature neu-ronal phenotype after im-plantation of neural stemcells in the hippocampus.Confocal pictures of im-munostained hippocam-pal slices 3 months afterimplantation of fluorescent-labeled neural stem cells.Positivity for mature neu-ronal markers such asmicrotubule-associated pro-tein 2, isoforms a and b(MAP2, A–C), neurofilament200 kDa (NF200, D–F), andneuron-specific nuclear pro-tein (NeuN, G–I) is shown,and synthesis of neurotrans-mitters �-aminobutyric acid(�-aminobutyric acid-amino-transporter 1 positivity, J–L)and glutamate (M–O) is dem-onstrated. Green representsunenhanced green fluores-cent protein (GFP)-signal,and red represents Alex-aFluor 594. Scale bars, 100�m.



Another central aspect of this study investigates thegeneration and integration of new neuronal cells derivedfrom implanted stem cells into the hippocampus. It has beenrecently shown that young granular cells in the dentate gyrusdiffer substantially to their older neighboring counterparts.For instance, they have high input resistance, which leads toenhanced excitability and generation of action potentials inresponse to even weak excitatory stimuli, in addition to beingmore suitable to long-term potentiation, which indicates anenhanced synaptic plasticity (27–31). In general, an adequatelearning process is dependent on proliferation of local neuralstem cells, survival of newborn neurons, and also apoptosis ofrelatively immature cells that have not established learning-related synaptic connections (32, 33). In fact, the hypothesisthat inhibition of neurogenesis may cause learning impair-ment, while enhanced neurogenesis may improve learning,has recently been confirmed by various authors (21, 34 –37).It has been demonstrated by mathematical simulations thatincorporation of new neurons facilitates both clearance of oldmemories and storage of new information (38 – 40). This pat-tern seems to represent a complex balance where an increasednumber of new granule cells can facilitate epileptiform activ-ities (41). Conversely, the role of artificially implanted stemcells in this system, with the possibility of synaptic integrationand formation of new circuits, and the behavioral significanceof these changes have been largely unknown and represent aninteresting constellation, both from a basic science and a clin-ical point of view.

In summary, our study provides novel evidence that therepopulation of the adult hippocampus with new, geneticallyaltered, implanted neuronal cells is feasible and results in afunctional integration of the donor and derived neurons.This may also raise the possibility that in the future, certaindiseases primarily affecting the hippocampal formation maybe successfully treated by novel cell-based transplantationapproaches.

MATERIALS AND METHODS

Neural Progenitor Cells Isolation and CultureInbred GFP-transgenic Lewis rats used in this study were kindly provided

by Kobayashi. These rats previously described by Inoue et al. (11) expressGFP under the control of a cytomegalovirus/�-actin promoter. Time-pregnant animals were killed by intraperitoneal lethal injection of ketaminein accordance with the national and institutional guidelines for animal ex-perimentation, after approval of the Research Ethics Committee of theAlbert-Ludwig University Freiburg, Germany. The method for extractionand isolation of NPCs from the telencephalic vesicles of E14 (gestational age:14 days postconception) rat embryos has been described previously (42).Proliferation medium consisted of Dulbecco’s modified eagle medium/Ham’s nutrient medium at a ratio of 3:1, 1% mixture of penicillin, strepto-mycin and amphotericin, 2% B27 supplement (all purchased from Invitro-gen, Darmstadt, Germany), fibroblast growth factor b (R&D Systems,Wiesbaden-Nordenstadt, Germany), and epithelial growth factor (PeproTech, Hamburg, Germany) at a concentration of 20 ng/mL each, and 5�g/mL heparin (Sigma-Aldrich, Munich, Germany). The medium waschanged every 5 to 6 days and passages performed once a week by lightmechanical dissociation of the formed spheres.

Induction of Neuronal Phenotype In VitroAfter at least three passages, cells were plated at high densities on poly-

L-ornithine (Sigma)-coated cover slips in a medium composed of minimumessential medium (Sigma),1% N2 serum supplement, 1% sodium pyruvate, 2

FIGURE 5. Functional maturation of neurons derived fromimplanted neural stem cells. (A) Infrared differential interfer-encecontrastmicroscopyshowingthepatchingprocessofacell,after its identification by fluorescence. (B) Voltage-clamp tracesshowing absolute inward and outward currents from cellpatched in A; holding potential �80 mV, pulse duration 100msec, intensities from �70 to �70 mV at 10 mV increment; leakand capacitive currents were subtracted by a p/-4 protocol (fordetails, see text), traces represent average of five consecutivesweeps. (C) I–V relationship of Na� currents. Curve fitting withthe same equation shown in Figure 2. Best-fit parameters werefor the represented cell V1/2��25.1�0.8 mV, slope�3.7�0.5mV, gmax�67.8�3.7 nS, and ENa�70.7�2.5 mV. (D) I–V rela-tionships foroutwardcurrents(K�currents), fittedwith thesametheoretical model described for Na�. Best-fit parameters wereV1/2��3.3�1.0 mV, slope�9.5�0.6 mV, gmax�19.8�1.3 nS,and EK��102.7�10.6 mV. (E) Action potentials elicited in cur-rentclampafterinjectionofa50-pAcurrentappliedduring1sec.(F) Spontaneous excitatory postsynaptic potentials without phar-macologic blocking in the first trace; in the middle trace,GABAergic currents were blocked after application of 10-�Mbicuculline; and in the lower trace, �-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate, glutamate receptor-mediatedglutamatergic currents were additionally blocked by applica-tion of 6-cyano-7-nitroquinoxaline-2,3-dione, AMPA/kainate re-ceptor antagonist 10 �M.


mM L-glutamine, 4 mM glucose, 0.1% bovine serum albumin, 1% mixture ofpenicillin, streptomycin and amphotericin (all from Invitrogen), supple-mented with brain-derived neurotrophic factor 25 ng/mL (R&D Systems)and 0.5 mM 3-isobutyl-1-methylxantine (3-isobutyl-1-methylxanthin,phosphodiesterase inhibitor, Sigma-Aldrich). Medium was changed everyother day. Cells were fixed and immunostained immediately after electro-physiologic recording.

Stereotactic ImplantationTo minimize the trauma related to the stereotactic implantation of the cell

solution, we used a microtransplantation technique as described previously(43). Twenty-six–week-old female Sprague-Dawley rats (approximatelyweighing 190 g) were used as graft hosts and were therefore fixed to a stereo-tactic frame Cunningham (Stoelting Co., Dublin, Ireland) under generalanesthesia with ketamine (Pfizer, Karlsruhe, Germany). The coordinatesused for implants into the dentate gyrus of the hippocampus were anterior-posterior (stereotactic coordinate), �5.4 mm; lateral (stereotactic coordi-nate), �4.2; vertical (stereotactic coordinate), �6.5 to �4.5 for 10 separatedeposits made 0.2 mm apart. The cells were resuspended at 150,000 cells/�L.Each deposit (10 per animal) constituted 0.2 �L of cell suspension implantedinto the posterior aspect of the hippocampal formation.

Animal CareAll animals were immunosuppressed, commencing 1 day before transplanta-

tion, by daily intraperitoneal injections of cyclosporine A (Sandimmun, 10 mg/kg, Novartis Pharma, Nuerenberg, Germany). Immunosuppression was com-bined with prophylactic oral administration of antibiotics (Sulfadoxin andTrimethoprin, Bayer, Germany) in the drinking water. Animals were housed ina temperature-controlled enriched environment under a 12-hr light/dark cyclewith access to food and water ad libitum.

ElectrophysiologyStandard whole-cell patch clamp recording methods were used to exam-

ine the physiologic properties of implanted fetal-derived neuronal stem cells.To this end, the animals were anesthetized with isoflurane introduced to theinspiration airflow (4%–5%, Abbott, Ludwigshafen, Germany) and killed bydecapitation, in accordance with national and institutional guidelines.Transverse 300-�m–thick slices were cut from the hippocampus of the im-planted animals, 12 weeks after grafting, using a custom-built vibrating mi-crotome (44). Slices were maintained at 35°C for 30 min after slicing and thenstored at room temperature in sucrose-based solution containing (in milli-molar): 87 NaCl, 25 NaHCO3, 2.5 KCl, 1.25 NaH2PO4, 75 sucrose, 0.5 CaCl2,7 MgCl2, and 25 glucose, equilibrated with 95% O2 and 5% CO2. Trans-planted cells were first identified by their fluorescence, excited by a 405-nmdiode fluorescence lamp, and visualized with a charged coupled device cam-era (“charged coupled device,” Zeiss, Jena, Germany). After cell identifica-tion, a patch clamp was established under visual control using infrareddifferential interference contrast videomicroscopy (Zeiss). A detailed de-scription of the recording conditions, solutions, technique, and off-line dataprocessing was provided in a previous publication (45).

ImmunostainingTwelve weeks after implantation, animals were killed, and the brains were

cryosectioned at 40 �m in the coronal plane. The immunohistochemical andimmunocytochemical procedures were carried out as described previously (20,45, 46). The following primary antibodies were used: anti-GFAP (polyclonal,produced in rabbits, purchased from Millipore, Temecula, CA used at 1:600),antidoublecortin (polyclonal, guinea pig, Santa Cruz, Heidelberg, Germany,1:3000), anti-�III tubulin (monoclonal, mouse, Sigma, 1:400), anti-MAP2ab(monoclonal, mouse, Millipore, Temecula, CA, 1:200), anti-NeuN (monoclo-nal, mouse, Millipore, Temecula, CA, 1:250), anti-NF200 (monoclonal, mouse,Millipore, Temecula, CA, 1:200), anti-GABA-amino-transporter 1 (polyclonal,rabbit, Millipore, Temecula, CA, 1:500), and antiglutamate (monoclonal,mouse, Millipore, Temecula, CA, 1:5000). The secondary antibodies were anti-rabbit, anti-mouse or anti-guinea pig AlexaFluor 594 (Molecular Probes,Karlsruhe, Germany), used at 1:200. The nuclei were stained with 4�,6-diami-

dine-2-phenylindole dihydrochloride (Sigma, 1:10,000). The original GFP signalwas not enhanced with specific antibodies. Biocytin staining was revealed byDyLight 649 Streptavidin (1:200, Vector Laboratories, Burlingame, CA).

In Vitro and In Vivo Quantifications andStatistical Analysis

In vitro quantifications of marker positivity in fixed cultures were per-formed under epifluorescence. Therefore, at least 10 cover slips from 3 con-secutive experiments were analyzed. Mean comparisons between groupswere performed by using the Mann-Whitney U statistical test, because dis-tribution of the residues was not Gaussian. All data are reported asmean�standard error of the mean, and the significance level (P) is indicatedin the text. For computation, we used SPSS version 13.0 Software (SPSS Inc.,Chicago, IL).

To quantify the neuronal differentiation in vivo, we used one of the fiveseries of slices to stain with NeuN. Coexpression of NeuN and GFP wasanalyzed under an epifluorescence microscope BX61 (Olympus Europe,Hamburg, Germany), equipped with a high-sensitivity digital camera DP70and stereology system computer assisted stereological toolbox (Olympus Eu-rope, Hamburg, Germany). Cells expressing both markers were counted byscanning the slide with an automatic microcator (Heidenhain, Traunreut,Germany) at 40� magnification and with an optic frame of 50�50 �m,taking care to focus over the entire slice thickness (40 �m) for each frame.Finally, the entire hippocampus of each animal was scanned.

Pictures shown were obtained in a Leica (Munique, Germany) TCS SP2confocal system (405-nm diode, ArKr 488 nm, Ar 594 nm, and Xn 633-nmlasers) and were digitalized at 2048�2048 pixels.

ACKNOWLEDGMENTSThe authors thank Prof. Josef Bischofberger and Dr.

Christoph Schmidt-Hieber for the supervision and assistance inelectrophysiology.

REFERENCES1. Freed CR, Greene PE, Breeze RE, et al. Transplantation of embryonic

dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001;344: 710.

2. Kordower JH, Goetz CG, Freeman TB, et al. Dopaminergic transplantsin patients with Parkinson’s disease: Neuroanatomical correlates ofclinical recovery. Exp Neurol 1997; 144: 41.

3. Lindvall O, Brundin P, Widner H, et al. Grafts of fetal dopamine neu-rons survive and improve motor function in Parkinson’s disease.Science 1990; 247: 574.

4. Mendez I, Sanchez-Pernaute R, Cooper O, et al. Cell type analysis offunctional fetal dopamine cell suspension transplants in the striatumand substantia nigra of patients with Parkinson’s disease. Brain 2005;128: 1498.

5. Wernig M, Zhao JP, Pruszak J, et al. Neurons derived from repro-grammed fibroblasts functionally integrate into the fetal brain and im-prove symptoms of rats with Parkinson’s disease. Proc Natl Acad SciUSA 2008; 105: 5856.

6. Bjorklund A, Stenevi U. Reconstruction of the nigrostriatal dopaminepathway by intracerebral nigral transplants. Brain Res 1979; 177: 555.

7. Sorensen AT, Thompson L, Kirik D, et al. Functional properties andsynaptic integration of genetically labeled dopaminergic neurons inintrastriatal grafts. Eur J Neurosci 2005; 21: 2793.

8. Uchida K, Momiyama T, Okano H, et al. Potential functional neuralrepair with grafted neural stem cells of early embryonic neuroepithelialorigin. Neurosci Res 2005; 52: 276.

9. Aguirre AA, Chittajallu R, Belachew S, et al. NG2-expressing cells in thesubventricular zone are type C-like cells and contribute to interneurongeneration in the postnatal hippocampus. J Cell Biol 2004; 165: 575.

10. Englund U, Bjorklund A, Wictorin K, et al. Grafted neural stem cellsdevelop into functional pyramidal neurons and integrate into host cor-tical circuitry. Proc Natl Acad Sci USA 2002; 99: 17089.

11. Inoue H, Ohsawa I, Murakami T, et al. Development of new inbredtransgenic strains of rats with LacZ or GFP. Biochem Biophys Res Com-mun 2005; 329: 288.



12. Svendsen CN, ter Borg MG, Armstrong RJ, et al. A new method for therapid and long term growth of human neural precursor cells. J NeurosciMethods 1998; 85: 141.

13. Bjorklund A, Stenevi U, Schmidt RH, et al. Intracerebral grafting of neu-ronal cell suspensions. II. Survival and growth of nigral cell suspensionsimplanted in different brain sites. Acta Physiol Scand Suppl 1983; 522: 9.

14. Dunnett SB, Bjorklund A, Schmidt RH, et al. Intracerebral grafting ofneuronal cell suspensions. IV. Behavioural recovery in rats with unilateral6-OHDA lesions following implantation of nigral cell suspensions in dif-ferent forebrain sites. Acta Physiol Scand Suppl 1983; 522: 29.

15. Gage FH, Bjorklund A, Stenevi U, et al. Intracerebral grafting of neu-ronal cell suspensions. VIII. Survival and growth of implants of nigraland septal cell suspensions in intact brains of aged rats. Acta PhysiolScand Suppl 1983; 522: 67.

16. Nikkhah G, Bentlage C, Cunningham MG, et al. Intranigral fetal dopa-mine grafts induce behavioral compensation in the rat Parkinsonmodel. J Neurosci 1994; 14: 3449.

17. Nikkhah G, Cunningham MG, Cenci MA, et al. Dopaminergic micro-transplants into the substantia nigra of neonatal rats with bilateral6-OHDA lesions. I. Evidence for anatomical reconstruction of the ni-grostriatal pathway. J Neurosci 1995; 15: 3548.

18. Olanow CW, Goetz CG, Kordower JH, et al. A double-blind controlledtrial of bilateral fetal nigral transplantation in Parkinson’s disease 1.Ann Neurol 2003; 54: 403.

19. Maciaczyk J, Singec I, Maciaczyk D, et al. Combined use of BDNF,ascorbic acid, low oxygen, and prolonged differentiation time gener-ates tyrosine hydroxylase-expressing neurons after long-term in vitroexpansion of human fetal midbrain precursor cells. Exp Neurol 2008;213: 354.

20. Maciaczyk J, Singec I, Maciaczyk D, et al. Restricted spontaneous invitro differentiation and region-specific migration of long-term ex-panded fetal human neural precursor cells after transplantation intothe adult rat brain. Stem Cells Dev 2009; 18: 1043.

21. Kempermann G, Kuhn HG, Gage FH. More hippocampal neurons inadult mice living in an enriched environment. Nature 1997; 386: 493.

22. Nilsson M, Perfilieva E, Johansson U, et al. Enriched environment in-creases neurogenesis in the adult rat dentate gyrus and improves spatialmemory. J Neurobiol 1999; 39: 569.

23. van PH, Kempermann G, Gage FH. Running increases cell prolifera-tion and neurogenesis in the adult mouse dentate gyrus. Nat Neurosci1999; 2: 266.

24. Stripecke R, Carmen VM, Skelton D, et al. Immune response to greenfluorescent protein: Implications for gene therapy. Gene Ther 1999; 6:1305.

25. Francis JS, Olariu A, Kobayashi E, et al. GFP-transgenic Lewis rats as acell source for oligodendrocyte replacement. Exp Neurol 2007; 205: 177.

26. Krause M, Ganser C, Papazoglou A, Kobayashi E, Nikkhah G. Greenfluorescent protein transgenic rats: a new tool for neural transplanta-tion. Manuscript in preparation.

27. Schmidt-Hieber C, Jonas P, Bischofberger J. Enhanced synaptic plas-ticity in newly generated granule cells of the adult hippocampus 4.Nature 2004; 429: 184.

28. Ambrogini P, Lattanzi D, Ciuffoli S, et al. Morpho-functional charac-terization of neuronal cells at different stages of maturation in granulecell layer of adult rat dentate gyrus. Brain Res 2004; 1017: 21.

29. Esposito MS, Piatti VC, Laplagne DA, et al. Neuronal differentiation inthe adult hippocampus recapitulates embryonic development. J Neu-rosci 2005; 25: 10074.

30. Ge S, Goh EL, Sailor KA, et al. GABA regulates synaptic integration ofnewly generated neurons in the adult brain. Nature 2006; 439: 589.

31. Tozuka Y, Fukuda S, Namba T, et al. GABAergic excitation promotesneuronal differentiation in adult hippocampal progenitor cells. Neuron2005; 47: 803.

32. Dobrossy MD, Drapeau E, Aurousseau C, et al. Differential effects oflearning on neurogenesis: Learning increases or decreases the numberof newly born cells depending on their birth date. Mol Psychiatry 2003;8: 974.

33. Epp JR, Spritzer MD, Galea LA. Hippocampus-dependent learningpromotes survival of new neurons in the dentate gyrus at a specific timeduring cell maturation. Neuroscience 2007; 149: 273.

34. Cao L, Jiao X, Zuzga DS, et al. VEGF links hippocampal activity withneurogenesis, learning and memory. Nat Genet 2004; 36: 827.

35. Dupret D, Revest JM, Koehl M, et al. Spatial relational memory re-quires hippocampal adult neurogenesis. PLoS One 2008; 3: e1959.

36. Saxe MD, Battaglia F, Wang JW, et al. Ablation of hippocampal neuro-genesis impairs contextual fear conditioning and synaptic plasticity inthe dentate gyrus. Proc Natl Acad Sci USA 2006; 103: 17501.

37. Snyder JS, Hong NS, McDonald RJ, et al. A role for adult neurogenesisin spatial long-term memory. Neuroscience 2005; 130: 843.

38. Aimone JB, Wiles J, Gage FH. Computational influence of adult neu-rogenesis on memory encoding. Neuron 2009; 61: 187.

39. Chambers RA, Potenza MN, Hoffman RE, et al. Simulated apoptosis/neurogenesis regulates learning and memory capabilities of adaptiveneural networks. Neuropsychopharmacology 2004; 29: 747.

40. Deisseroth K, Singla S, Toda H, et al. Excitation-neurogenesis couplingin adult neural stem/progenitor cells. Neuron 2004; 42: 535.

41. Meltzer LA, Yabaluri R, Deisseroth K. A role for circuit homeostasis inadult neurogenesis. Trends Neurosci 2005; 28: 653.

42. Singec I, Knoth R, Meyer RP, et al. Defining the actual sensitivity andspecificity of the neurosphere assay in stem cell biology. Nat Methods2006; 3: 801.

43. Nikkhah G, Olsson M, Eberhard J, et al. A microtransplantation ap-proach for cell suspension grafting in the rat Parkinson model: A de-tailed account of the methodology. Neuroscience 1994; 63: 57.

44. Geiger JR, Bischofberger J, Vida I, et al. Patch-clamp recording in brainslices with improved slicer technology. Pflugers Arch 2002; 443: 491.

45. Lepski G, Jannes CE, Maciaczyk J, et al. Limited Ca(2�) and PKA-pathway dependent neurogenic differentiation of human adult mesen-chymal stem cells as compared to fetal neuronal stem cells. Exp Cell Res2010; 316: 216.

46. Lepski G, Jannes CE, Strauss B, et al. Survival and neuronal differenti-ation of mesenchymal stem cells transplanted into the rodent brain aredependent upon microenvironment. Tissue Eng Part A 2010; 16: 2769.


equivalent neurogenic potential of wild-type and gfp

Documents