Molecular and Electrophysiological Characterization of

GFP-Expressing CA1 Interneurons in GAD65-GFP Mice

Corette J. Wierenga1*, Fiona E. Müllner1, Ilka Rinke2, Tara Keck1¤, Valentin Stein2, Tobias Bonhoeffer1
1 Department of Cellular and Systems Neurobiolgy, Max Planck Institute of Neurobiology, Martinsried, Germany, 2 Synaptic Receptor Trafficking, Max Planck Institute of
Neurobiology, Martinsried, Germany

Abstract
The use of transgenic mice in which subtypes of neurons are labeled with a fluorescent protein has greatly facilitated
modern neuroscience research. GAD65-GFP mice, which have GABAergic interneurons labeled with GFP, are widely used in
many research laboratories, although the properties of the labeled cells have not been studied in detail. Here we investigate
these cells in the hippocampal area CA1 and show that they constitute ,20% of interneurons in this area. The majority of
them expresses either reelin (7062%) or vasoactive intestinal peptide (VIP; 1562%), while expression of parvalbumin and
somatostatin is virtually absent. This strongly suggests they originate from the caudal, and not the medial, ganglionic
eminence. GFP-labeled interneurons can be subdivided according to the (partially overlapping) expression of neuropeptide
Y (4263%), cholecystokinin (2563%), calbindin (2062%) or calretinin (2062%). Most of these subtypes (with the exception
of calretinin-expressing interneurons) target the dendrites of CA1 pyramidal cells. GFP-labeled interneurons mostly show
delayed onset of firing around threshold, and regular firing with moderate frequency adaptation at more depolarized
potentials.

Citation: Wierenga CJ, Müllner FE, Rinke I, Keck T, Stein V, et al. (2010) Molecular and Electrophysiological Characterization of GFP-Expressing CA1 Interneurons in
GAD65-GFP Mice. PLoS ONE 5(12): e15915. doi:10.1371/journal.pone.0015915
Editor: Izumi Sugihara, Tokyo Medical and Dental University, Japan
Received August 24, 2010; Accepted November 29, 2010; Published December 31, 2010
Copyright: ß 2010 Wierenga et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Marie Curie grants IEF #40528 and ERG #256284 (CJW), the German National Academic Foundation (FEM) and the Max
Planck Society. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: wierenga@neuro.mpg.de
¤ Current address: MRC Centre for Developmental Neurobiology, King’s College London, London, United Kingdom

Introduction mice is brightly labeled with GFP. These interneurons are found in
most brain areas and the spinal cord [11]. GFP is already
GABAergic interneurons comprise ,10–20% of the total expressed during embryonic development, which makes these
neuronal population and are essential for controlling and transgenic mice very suitable for developmental studies. In this
synchronizing the output of the principal cells [1–3]. There are study, we provide a detailed analysis of the molecular and
many different types of interneurons, executing diverse functions electrophysiological profile of GFP-labeled cells in the hippocam-
in shaping the activity of neuronal networks. It has proven difficult pal CA1 area of GAD65-GFP mice. We report that GFP-labeled
to formulate an unequivocal definition of the different interneuron cells are characterized by a high coincidence of reelin expression
types that exist in the brain [4]. Recent work describing the origin (suggesting they emanate from the CGE), axons targeting the
and development of different interneuron types has contributed dendritic layers, and regular firing properties.
greatly towards solving this issue. Cortical and hippocampal
interneurons were shown to be born outside of the cortex in the Methods
ventral telencephalon and to migrate tangentially during develop-
ment to their final location in the adult brain [1,5]. The majority All experimental procedures were carried out in compliance
of GABAergic interneurons originate from the medial ganglionic with the institutional guidelines of the Max Planck Society and the
eminence (MGE) or the caudal ganglionic eminence (CGE) [6–8]. local government (Regierung von Oberbayern; Statement of
In addition, a small fraction of interneurons are generated in the Compliance #A5132-01). All animals are sacrificed prior to the
preoptic area [9,10]. Interneurons with different origin form removal of organs in accordance with the European Commission
separate interneuron classes and display distinct cellular proper- Recommendations for the euthanasia of experimental animals
ties. A full understanding of the developmental relationship (Part1 and Part 2). Breeding and housing as well as the euthanasia
between different types of interneurons will greatly contribute to of the animal are fully compliant with the German and European
define an unambiguous interneuron classification. applicable laws and regulations concerning care and use of
In order to better understand the function of different types of laboratory animals.
interneurons, different lines of transgenic mice have been created
in which specific subsets of GABAergic interneurons are labeled. Immunohistochemistry
GAD65-GFP mice [11] are being used in numerous studies by Adult GAD65-GFP mice (P50-100) were anesthetized with
many different labs [12–15]. A subset of GABAergic cells in these Ketamine (0.21 mg/g) and Xylazine (0.015 mg/g) and perfused

PLoS ONE | www.plosone.org 1 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

transcardially with 0.1 M phosphate-buffered saline (PBS, Molecular Devices) and Bessel filtered at 2 kHz. Pipette solutions
pH 7.3–7.4), followed by 4% paraformaldehyde in PBS. The contained the following (in mM): 150 K-methyl sulfate, 4 KCl, 4
brain was removed from the skull, postfixed overnight in the same NaCl, 4 MgATP, 0.4 Mg GTP, and 10 HEPES and 30 mM Alexa
fixative at 4uC, and then transferred to 30% sucrose in PBS for at Fluor 594. Recordings were not corrected for a liquid junction
least 2 days. Coronal sections were cut on a freezing microtome at potential (predicted value: 214 mV). Firing patterns were
30 mm thickness. Free-floating sections were rinsed 3–5 times with recorded in current clamp by injecting 1-second current pulses
PBS with 0.1% Triton X, incubated in a blocker solution of variable amplitude (DI = 1–5 pA around firing threshold; DI
containing 0.4% Triton X-100 and 10% goat serum for 2 hours = 50 or 100 pA otherwise).
at room temperature. Primary antibodies were applied overnight Resting membrane potentials (, 240 mV) were measured at
at 4uC in 0.1 M phosphate buffer with 0.4% Triton and 5% goat zero holding current soon after gaining whole cell access. Series
serum. Following extensive washing, appropriate secondary resistances ranged from 5 to 25 MOhm. Bridge-balance compen-
antibodies were applied at a concentration of 1:200. sation was applied in a subset of experiments. Input resistances
The following primary antibodies were used in this study: were measured by the steady state response to a 100 ms pulse of
chicken anti-GFP (Chemicon #06-896; 1:1000), rabbit anti- 25 mV in voltage clamp, or using a linear regression of voltage
GABA (Sigma A2052; 1:2000), mouse anti-GAD67 (Chemicon deflections in response to small current steps (between 5 pA and 50
MAB5406; 1:2000), mouse anti-reelin (MBL CR50; 1:500), rabbit pA), with equivalent results. Membrane time constants were
anti-VIP (Immunostar #20077; 1:500), mouse anti-parvalbumin determined by fitting an exponential curve to responses to the
(Swant PV235, 1:2000), rat anti-somatostatin (Chemicon same small current steps at 0.2–500 ms from current onset. Action
MAB354; 1:500), rabbit anti-calretinin (Swant #7699/3H; potential thresholds were defined as the voltage at which the slope
1:1000), rabbit anti-NPY (Immunostar #22940; 1:1000), mouse trajectory reached 10 mV/ms (subtracting residual capacitive
anti-CCK (Dr. Ohning, UCLA Cure #9303; 1:1000), rabbit anti- artifacts if necessary). Threshold values were averaged only for the
calbindin (Swant CB-38a; 1:5000). Secondary antibodies were smallest three superthreshold steps to minimize influence of
conjugated with Alexa488, Alexa633 and Cy3 (Molecular Probes, uncompensated series resistance. Action potential amplitudes were
Invitrogen). measured from threshold. Afterhyperpolarization amplitudes were
Image stacks (3756375 mm, 5126512 pixels; Dz = 1.5 mm) defined as the difference between action potential threshold and
were acquired over the entire depth of the section with a Leica the minimum membrane potential attained during the after-
Confocal microscope (LCS SP2) and analyzed with ImageJ. For hyperpolarization. Action potential parameters are given as the
each coronal brain section, 2-4 images were taken from the CA1 average over the first action potentials of a train. Saturating
area in each hippocampus. The imaged regions were chosen to frequencies were determined as maximum spike number per
maximize the number of GFP cells they contained. Expression of current step; maximal frequencies were defined as the reciprocal of
labeled proteins was determined by visual inspection of the raw or, the minimum inter-spike interval. The amplitude of the sag
in a few cases, Gaussian filtered (s = 2 pixels) image stacks and was current was determined for hyperpolarizing steps as the potential
performed for each label independently before colocalization was difference between the minimum and steady state voltages
addressed. Cells were considered positive if the somatic fluores- attained during current injection, and its time constant from a
cence was clearly (.,15%) above the local background level in monoexponential fit between minimum and sag steady state
multiple depth levels. For antibodies that do not stain the nucleus decay. Both parameters depended linearly on the steady state
(GAD67, somatostatin, NPY and reelin), brightly labeled cells with voltage upon current injection, and values were interpolated to
nuclear staining were considered false positive and discarded. The give the value at 2100 mV to allow comparison. Irregular spiking
percentage of positive cells was determined per section and values was defined by the coefficient of variation (CV) of the inter spike
are given as averages 6 SEM over all sections. intervals exceeding a threshold of 0.5. Delayed onset firing was
Most antibodies resulted in clear staining of only a few positive defined by the delay of the first spike exceeding 100 ms. Strongly
cells without background staining, clearly distinguishing positive adapting cells were defined as cells that stopped firing before the
and negative cells. GABA immunostaining showed high extracel- end of the current injection. Parameters of firing accommodation
lular background staining at the top and bottom of our sections. are averages of the five longest spike trains. For early accommo-
We therefore only analyzed the middle portion of them, where dation, the ratio of the first and the fifth spike parameter was
positive cells could be clearly distinguished from background. As taken; for late accommodation, it was the ratio between fifth and
the NPY antibody showed relatively high background staining in last spike. Statistical differences between interneuron groups were
all cells, cells were only considered NPY-positive when they were tested with an ANOVA, followed by a post-hoc Neuman-Keuls
clearly brighter than their neighboring cells. For some NPY- test. Differences were considered significant if p,0.05. Values are
sections we blindly repeated the analysis to ensure reproducibility given as mean 6 standard error.
of our results. The CCK antibody also strongly stained blood
vessels. Immunostaining with the reelin antibody resulted in Morphology
intense labeling of some cells, while other cells only had a dim After recordings, hippocampal slices were fixed in 4%
punctate labeling around their cell bodies. Only the cells with paraformaldehyde in PBS overnight and multiple high resolution
intense somatic labeling were considered reelin-positive. It was image stacks (typically 1206120 mm, 102461024 pixels; Dz =
recently described that the punctuate labeling reflects binding of 0.75 mm) covering the entire cell, were taken with a Leica
extracellular reelin by reelin-negative cells [16]. Confocal microscope (LCS SP2). Morphological reconstructions
were produced manually from multiple image stacks. Morpholog-
Firing patterns ical properties were quantified with ImageJ. Soma roundness was
Transverse hippocampal slices were prepared from juvenile defined as 4p*area/perimeter2, with values between 0 and 1 and
GAD65-GFP mice (P14–21) as described previously [17]. Whole- rounder somata closer to 1 [18]. The aspect ratio of the soma was
cell patch clamp recordings were made at room temperature using defined as the ratio between the longest and shortest axis. Somata
a Multiclamp 700B amplifier (Molecular Devices). Signals were with aspect ratio .2 are sometimes referred to as fusiform [4,18].
digitized at 5 kHz (Digidata 1440A and pClamp 10.2 Software; The dendritic tree was described by the number of principal

PLoS ONE | www.plosone.org 2 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

dendrites, the verticality (ratio between radial and tangential For a first characterization of GFP-labeled interneurons in
dendritic span) and the orientation of its major axis with respect to GAD65-GFP mice, we performed immunohistochemistry for the
the pyramidal cell layer (0u reflect a radial orientation similar to inhibitory neurotransmitter GABA and the GABA-producing
pyramidal cells, 90u is tangential). The axon was distinguished enzyme GAD67 in hippocampal sections of adult GAD65-GFP
from the dendrites based on its thinner and smoother appearance. mice. GFP-positive neurons were present in all layers of the
In some cases the axon was tortuous with ramifications and hippocampal CA1 area, with the highest density in the upper
boutons could be discerned. dendritic layers (Fig. 1B). Virtually all GFP-labeled cells were
immuno-positive for GAD67 and/or GABA (Fig. 1C, D),
Results confirming their identity as GABAergic interneurons [11]. This
was also confirmed by the recordings of inhibitory currents in pairs
This study was prompted by the observation that in acute of presynaptic GFP-labeled cells and postsynaptic CA1 pyramidal
hippocampal slices and slice cultures [19] of GAD65-GFP mice, neurons (Müllner, Bonhoeffer and Wierenga, unpublished data).
the GFP-labeling was clearly non-uniform. We noticed denser
labeling in the upper dendritic layers, stratum radiatum and GFP-labeled interneurons contain reelin or VIP
stratum lacunosum-moleculare (Fig. 1A), suggesting that GFP- Previous reports have suggested that the majority of GFP-
labeled interneurons in GAD65-GFP mice target preferentially labeled interneurons in GAD65-GFP mice originates from the
dendrites of CA1 pyramidal cells. We therefore set out to caudal ganglionic eminence (CGE) [11,20]. It was recently shown
characterize GFP-labeled interneurons in the hippocampal CA1 that most interneurons originating from the CGE express either
area in detail. reelin or vasoactive intestinal peptide (VIP) (Miyoshi et al., 2010),

Figure 1. GFP-positive GABAergic interneurons. A: Maximal projection image illustrating the distribution of GFP-labeled profiles in the CA1
layers. Calbindin (red), labeling a fraction of pyramidal cells, is only shown to facilitate recognition of the layers. B: Distribution of GFP-labeled
interneurons in the hippocampal CA1 area in GAD65-GFP mice. C: Percentage of GFP-labeled cells that expressed GABA (blue), GAD67 (red) or both
(purple). Data from 268 GFP cells; 10 sections. D: Example of triple immunostaining for GFP (green), GABA (blue) and GAD67 (red). Scale bars are
30 mm. Abbreviations of CA1 layers: Or - oriens; Pyr – pyramidale; Rad – radiatum; LM – lacunosum moleculare.
doi:10.1371/journal.pone.0015915.g001

PLoS ONE | www.plosone.org 3 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

while interneurons that originate from the medial ganglionic dendritic regions stratum radiatum and stratum lacunosum-
eminence (MGE) are known to express either parvalbumin or moleculare (94/117; 80%; Fig. 3C). CCK-positive GFP cells in
somatostatin [8,21]. To examine which interneurons express GFP stratum radiatum had mostly multipolar morphologies (31/46;
in GAD65-GFP mice, we performed immunohistochemistry 67%) and ,50% of them co-expressed calbindin (24/46; 52%). In
staining for reelin, VIP, parvalbumin and somatostatin in the upper dendritic layers (at the border between stratum
hippocampal sections. We found that 6962% (n = 16 sections; radiatum and stratum lacunosum-moleculare and within stratum
548 GFP cells) of GFP-positive interneurons in the CA1 area lacunosum-moleculare), dendrites and somata of CCK-positive
expressed reelin (Fig. 2A, C). The percentage of GFP cells that GFP cells were oriented tangentially (24/48; 50%) or multipolar
expressed reelin differed between layers (Fig. 2D). In particular, (17/48; 35%). These interneurons correspond well with previously
.80% of GFP-expressing cells in stratum lacunosum-moleculare described CCK-positive interneuron types that are known to
showed strong reelin expression. Furthermore, 1562% of GFP- innervate CA1 pyramidal dendrites [25,26]. They are often
positive interneurons expressed VIP (n = 9 sections; 313 cells; divided according to their main targeting area and referred to as
Fig. 2A, C), most of which were located close to or in the Schaffer collateral-associated, perforant path-associated and apical
pyramidal layer (Fig. 2D). In agreement with the previous report dendrite innervating cells [27]. In addition, a small group of CCK-
by Miyoshi and coworkers [6], reelin and VIP expression did not positive but calbindin-negative GFP cells (12/102; 12%) with
overlap (161%; n = 9 slices; 313 GFP cells). Virtually no GFP cells mostly radially oriented morphology was located close to the
were found that expressed parvalbumin (261%; n = 14 slices; 437 pyramidal cell layer. This last group fits the description of CCK-
GFP cells) or somatostatin (361%; n = 27 slices; 727 GFP cells), positive basket cells [13,28], which are known to target pyramidal
markers that are typical for interneurons derived from the MGE cell somata and dendrites [25].
(Fig. 2B, C). The high occurrence of reelin and (to a lesser extent) Expression of calbindin was found in 2062% of GFP cells
VIP indicates that the vast majority of GFP-labeled interneurons (Fig. 3A; n = 20 sections; 518 GFP cells). Calbindin was also
in GAD65-GFP mice originates from the CGE. expressed by a subset of CA1 pyramidal neurons, which made it
The fraction of GFP-interneurons expressing these four cellular impossible to unambiguously distinguish between calbindin-
markers only accounts for up to 90% of the total population. The positive GFP-negative interneurons and principal cells. We
remaining 10% might simply reflect unavoidable imperfections of therefore could not determine what fraction of calbindin-positive
immunohistochemistry experiments and analysis. Alternatively, interneurons were labeled by GFP in our sections. Calbindin-
these cells could reflect a separate class of CGE-derived expressing GFP cells were mostly located around the pyramidal
interneurons for which a marker is presently lacking [6]. layer and in stratum radiatum (70/98; 71%) and had multipolar
morphologies (Fig. 3D). Calbindin-expressing interneurons have
Molecular profiles been shown to specifically target the dendrites of pyramidal cells
We next wanted to examine the molecular profile of GFP- [29]. A few (5/84; 6%) large, tangentially oriented GFP cells in
positive cells in the hippocampal CA1 area in more detail. We stratum oriens were also calbindin-positive. These interneurons
examined the expression of the classical interneuron markers have been described to have long-range projections to the septum
neuropeptide Y (NPY), CCK, calretinin and calbindin. We found, [29,30]. They also make local projections to pyramidal cell
in agreement with earlier reports on CGE-derived interneurons, dendrites [31,32] and possibly interneurons [33].
that GFP-labeled cells expressed a variety of interneuron markers. Of all CCK-expressing cells, 4064% co-expressed calbindin.
The largest subgroup consisted of GFP cells that expressed This percentage was similar in CCK-expressing GFP cells (37%;
NPY. 4263% of GFP cells expressed NPY (Fig. 3A; n = 18 p = 0.27, x2-test; n = 20 sections; 518 GFP cells), suggesting no
sections; 368 GFP cells), and 2262% of NPY-positive interneu- specific preference. GFP cells co-expressing calbindin and CCK
rons were labeled with GFP. Although almost 30% of all NPY- were located in stratum radiatum and had mostly multipolar or
positive interneurons co-expressed somatostatin (166/595), the radially oriented dendrites.
subset of GFP-expressing cells did not (304/310; 98%). NPY- Finally, 2062% of GFP cells expressed calretinin (Fig. 3A), and
expressing GFP cells were present in all layers of the hippocampal 1161% of all calretinin-expressing cells were GFP-labeled (n = 21
CA1 area and they had mostly multipolar morphologies (73/123; sections; 557 GFP cells). Calretinin-expressing GFP cells were
59%; Fig. 3B). NPY-expressing GFP cells (31/123; 25%) with located mostly in or close to the pyramidal layer (Fig. 3E). Half of
tangentially oriented morphologies were mostly located in the the calretinin-expressing GFP cells (46/92; 50%) had radially
upper dendritic layers stratum lacunosum-moleculare and stratum oriented, bipolar morphologies. These interneurons have been
radiatum. The labeled cell types correspond well to previously described before and are known to often co-express VIP and
described NPY-positive multipolar cells, often referred to as Ivy specifically target dendrites of other interneurons in stratum
cells and neurogliaform cells [20,22,23]. NPY-positive CA1 radiatum and stratum oriens [22,34,35]. The other calretinin-
interneurons mainly target the shafts of pyramidal cell dendrites containing GFP cells had mostly multipolar morphologies (38/92;
[22,24]. 41%). Although in the neocortex a subset of calretinin-expressing
In a few sections, we performed co-staining of NPY and reelin. interneurons is known to target dendrites of principal neurons
We found that reelin expression was slightly higher in NPY- [36,37], it is not known whether such interneurons also exist in the
positive GFP cells compared to the general population of NPY- hippocampus.
positive interneurons (75% vs. 64% reelin-positive cells; p,0.05, The few GFP cells that expressed parvalbumin were located
x2-test; n = 3 sections). One possible explanation for this finding is around the pyramidal layer (8/8; 100%) and had typical radial or
that different populations of NPY-expressing cells exist [20] and multipolar morphologies, as has been described before for
that those populations are not equally represented in the GFP- parvalbumin-positive basket cells [2,38]. The somatostatin-
labeled interneurons. expressing GFP cells were concentrated in stratum oriens and
CCK was expressed in 2563% of GFP cells (Fig. 3A) and GFP- showed mostly tangential morphologies (11/16; 69%), also in
labeled interneurons accounted for 3563% of CCK-expressing agreement with previous reports [2,38].
cells in the hippocampal CA1 area (n = 20 sections; 518 GFP cells). These immunohistochemistry results indicate that GFP-labeled
The vast majority of CCK-positive GFP cells were located in the interneurons in GAD65-GFP mice display a variety of classical

PLoS ONE | www.plosone.org 4 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

Figure 2. The majority of GFP-positive interneurons contain reelin or VIP. A: Immunohistochemistry for reelin (red) and vasoactive intestinal
peptide (VIP; blue). Most GFP-positive interneurons (green) contained either reelin (yellow arrowheads) or VIP (blue arrowheads). Very few GFP-
positive cells were lacking both (green arrowheads). B: Immunohistochemistry for parvalbumin (PV; red) and somatostatin (SOM; blue), showing
minimal overlap with GFP-positive interneurons (green arrowheads). Red and blue arrowheads point to GFP-negative parvalbumin- and
somatostatin-positive interneurons. C: Summary for all GFP-positive interneurons. D: Distribution of reelin- and VIP-containing GFP-labeled
interneurons in the layers of the CA1 area. Overlap between both markers are indicated with purple. Scale bars are 30 mm.
doi:10.1371/journal.pone.0015915.g002

interneuron markers and morphologies. However, despite this Firing properties

variation the emerging common theme is that the vast majority of To further characterize the GFP-labeled interneurons, we
GFP-labeled interneurons (the exception being soma-targeting recorded firing patterns in whole-cell patch clamp mode in acute
CCK-positive basket cells and interneuron-targeting calretinin- slices of GAD65-GFP mice. The firing patterns of the majority of
positive bipolar cells) innervate the dendrites of CA1 pyramidal GFP-labeled interneurons were characterized by delayed onset
neurons. firing around threshold and continuous, regular firing at higher

PLoS ONE | www.plosone.org 5 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

PLoS ONE | www.plosone.org 6 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

Figure 3. Molecular profile of GFP-positive interneurons. A: Percentage of GFP-labeled interneurons that contained neuropeptide Y (NPY),
cholecystokinin (CCK), calbindin (CB) and calretinin (CR). Data from 18–21 sections and 350–550 GFP cells were examined for each label; error bars
represent standard error. B–D: Left: Examples of GFP-labeled interneurons, containing NPY (B), CCK (C), calbindin (D; CB) or calretinin (E; CR). Right:
Distribution of positive GFP cells over the CA1 layers. Scale bars are 10 mm.
doi:10.1371/journal.pone.0015915.g003

firing frequencies with moderate frequency adaptation (‘adapting’). We grouped the recorded cells (n = 48) in five subclasses
However, these three features were not perfectly overlapping and according to their firing patterns. Delayed onset, adapting
mixed firing patterns were also found (Fig. 4A, B). interneurons (n = 34) were subdivided into three subclasses:

Figure 4. Firing properties of GFP-positive interneurons. A: Representation of all recorded GFP-labeled interneurons, arranged according to
characteristics of their firing patterns and their classification in 5 groups. Each segment represents a single interneuron. Inner ring: adapting (Ad; dark
blue) and strongly adapting (S-Ad; dark red) cells. Second ring: cells showing delayed onset (DO; blue) and immediate onset (IO; red) firing. Third ring:
cells displaying regular (reg; light blue) and irregular (irr; light red) firing. Interneurons were divided in five groups (1–5) as indicated with yellow and
green colors. The letters correspond to example cells (a–g) in B and C. B: Examples of firing patterns of example cells a–g (as indicated in A). Upper
traces show firing around threshold, middle traces show responses to hyperpolarizing steps (100 pA step size) and intermediate firing and lower
traces show maximal firing. C: Morphology of example cells a–g (same as in B). Dendrites are shown in black, axons in red.
doi:10.1371/journal.pone.0015915.g004

PLoS ONE | www.plosone.org 7 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

regular firing interneurons (21/34; 62%; group 1 in Fig. 4A, Ba– hippocampus and neocortex, but could also indicate that in the
c), fast spiking interneurons (6/34; 18%; group 2 in Fig. 4A, Bd), GAD65-GFP mice GFP is expressed in only a subset of CGE-
and irregular spiking interneurons (7/34; 21%; group 3 in Fig. 4A, derived interneurons with a bias towards reelin-expressing
Be). Fast spiking interneurons were defined as having firing rates interneurons. This is further supported by our electrophysiolog-
.50 Hz. They had shorter action potential widths and frequently ical recordings, where we only find a subset of firing pattern
displayed rebound action potentials after hyperpolarizing current classes previously described for CGE-derived interneurons [6].
injections (Table 1). Irregular spiking interneurons were defined The vast majority of the recorded cells showed a slowly adapting
by their large variance in inter-spike intervals (coefficient of firing pattern with delayed firing around threshold. Indeed, such
variation .0.5). They tended to have slightly more hyperpolar- firing patterns were previously described as typical for reelin-
ized resting membrane potentials, but were otherwise very similar expressing CGE-derived interneurons [6]. Intrinsic bursting
to group 1 interneurons (Table 1). In most delayed onset cells interneurons, which were reported to represent .20% of CGE-
(27/34; 79%), the delayed firing remained when the cell fired 2 derived cells, were not found in our study.
or 3 action potentials (corresponding to the LS1 type of Miyoshi We found that less than 5% of GFP-labeled interneurons in
et al. [6]). The fourth group of interneurons did not exhibit GAD65-GFP mice expressed parvalbumin or somatostatin.
delayed onset firing at threshold and displayed mostly irregular These cells were brightly labeled and their location and
firing patterns (7/48; 15%; group 4 in Fig. 4A, Bf). They morphologies were consistent with the well-described character-
distinguished themselves by a fast membrane time constant, low istics of parvalbumin- and somatostatin-positive interneurons
input resistance and low action potential threshold (Table 1). [2,38]. It is therefore unlikely that these were false positives of the
Finally, the fifth group consisted of strongly adapting interneu- immunohistochemistry. Labeling of a few parvalbumin and
rons (7/48; 15%; group 5 in Fig. 4A, Bg), which stopped firing somatostatin interneurons could indicate that a small minority
prematurely before the end of the current injection. They had of MGE-derived interneurons also express GFP in the GAD65-
small action potential amplitudes and fast sag current kinetics GFP mice. It was recently shown that the preoptic area also
(Table 1). None of the recorded GFP-labeled interneurons contributes to the generation of cortical and hippocampal
showed intrinsic burst firing. interneurons [10], albeit for only 5–10%. Preoptic area derived
interneurons express mostly reelin (but not VIP), but also some
Interneuron morphology parvalbumin- and somatostatin-positive interneurons were re-
We also quantified morphological parameters of the recorded ported to origin from the preoptic area [9]. Therefore, an
GFP cells, but we were unable to detect a strict correlation alternative explanation for our findings would be that interneu-
between electrophysiological and morphological properties rons from the preoptic area also express GFP in GAD65-GFP
(Fig. 4C and Table 2). However, 4 of 6 cells that had morphologies mice. This would also contribute to the bias towards labeling of
similar to those described for CCK-positive basket cells (i.e. soma reelin-expressing interneurons. Specific markers for preoptic area
location close to pyramidal layer, axon targeting pyramidal layer, derived interneurons will be necessary to distinguish between
radial orientation; e.g. Fig. 4Ce) showed irregular firing [39]. The these possibilities in the future.
exceptions were a strongly adapting and a regular firing The majority of the GFP-labeled interneurons were located
interneuron. In addition, 10 of 11 interneurons with morphologies around the border of stratum radiatum and lacunosum-molecu-
similar to neurogliaform cells (soma location in deep dendritic lare. It has been reported previously that interneurons in that
layers, compact multipolar morphology, axon ramifying close to location mainly project to pyramidal cell dendrites [26]. Based on
soma; Fig. e.g. 4Ca,b) displayed delayed onset firing. The one our immunohistochemical and morphological findings, we con-
exception was a strongly adapting interneuron. clude that GFP-labeled interneurons include several previously
Although we were not able to reconstruct the entire axonal described subtypes, which are known to mainly project to
arbors, in most interneurons we could detect some axonal dendrites [27]. The largest class consists of NPY-positive Ivy and
ramifications and/or boutons, indicative of the axonal projection neurogliaform cells [20,22]. GFP-labeled interneurons further
area. In agreement with our immunostaining findings, the include CCK-positive interneurons targeting the somata (i.e.
majority of GFP-labeled interneurons (24/32; 75%) had axons basket cells) or dendrites (e.g. Schaffer-associated, perforant path-
ramifying in the dendritic layers (stratum radiatum and stratum associated cells) of CA1 pyramidal cells [25,40,41], and calbindin-
lacunosum-moleculare). positive cells [29]. Despite the fact that the anatomy and
immunohistochemistry of these dendritically-targeting interneu-
Discussion rons have been described in detail, their precise function is
generally more elusive [27]. In particular, it is not known if
We characterized GFP-labeled interneurons in the hippocampal interneurons expressing different chemical markers execute
CA1 area of GAD65-GFP mice. We found that the majority of different functions in the CA1 circuitry. The finding that most
labeled interneurons most likely have their origin in the CGE. GFP-labeled interneurons target pyramidal dendrites makes
While different subtypes of interneurons are labeled in this mouse GAD65-GFP mice well suited for studies focused on dendritic
line, we found a strong bias towards reelin-expressing, dendriti- inhibition. At the same time, these mice also allow selection of
cally targeting interneurons with firing patterns showing delayed specific interneuron subtypes based on known morphology and/or
onset firing around threshold and moderate firing adaptation at firing properties (e.g. CCK-positive basket cells [13,28] or reelin-
higher firing frequencies. positive neurogliaform cells at the radiatum/lacunosum-molecu-
The CGE was found to generate ,30–40% of interneurons in lare border).
the somatosensory cortex and reelin- and VIP-expressing The main exception to the dendrite-targeting rule is formed by
interneurons in a ,1:1 ratio [6,7]. In the present study, we GFP-labeled calretinin-expressing interneurons, which have been
found that GFP-labeled interneurons in the GAD65-GFP mice shown to preferentially target other CA1 interneurons and avoid
form only ,20% of CA1 interneurons and showed a strong bias pyramidal cells [35]. In the neocortex, a subset of calretinin-
towards reelin-expressing interneurons (ratio reelin:VIP was positive interneurons preferentially innervates principal cells
,4:1–5:1). This could reflect regional differences between [36,37], but such cells have not been described for the

PLoS ONE | www.plosone.org 8 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

Table 1. Electrophysiological properties of GFP-labeled interneurons.

Delayed onset, adapting

Irregular Immediate Strongly

Regular spiking Fast spiking spiking onset, adapting adapting
(n = 12) (n = 5) (n = 4) (n = 6) (n = 5)

group 1 group 2 group 3 group 4 group 5

Resting membrane potential (mV) 24961 25162 25562 25862 25162

group 4, group 1,2,5 (p = 0.002)
Input resistance (MV) 468630 377654 401650 253650 372659
group 4, group 1 (p = 0.014)
Membrane time constant (ms) 6367 36613 57612 21612 34614
group 4, group 1 (p = 0.028)
Action potential threshold (mV) 23561 23562 23562 24262 23562
group 4, group 1–3,5 (p = 0.007)
Action potential amplitude (mV) 7862 7564 27664 28064 6564
group 5, group 1,2,4 (p = 0.041)
Action potential halfwidth (ms) 1.660.1 1.260.1 1.760.1 1.260.1 1.660.1
group 2, group 1 (p = 0.028)
Afterhyperpolarization (mV) 21961 22262 21762 21662 21962

CV of inter spike interval 0.360.05 0.260.1 0.660.1 0.760.1 0.460.1

group 3,4. group 1,2,5 (p,0.001)
Saturating Frequency (Hz) 3062 5964 1963 2163 18(4
group 2. group 1,3–5; group 1.3,5 (p,0.001)
Maximal Frequency (Hz) 7965 12069 8769 8969 6669
group 2. group 1,3–5 (p = 0.002)
Delay to first spike (ms) 496640 525680 513668 29673 70678
group 4, group 1–3; group 5, group 1–3 (p,0.001)
Rebound spikes (% of cells) 57610 83619 14618 14618 67619
group 2. group 3,4 (p = 0.023)
Sag amplitude at 2100 mV (pA) 1762 1964 1364 1064 2264

Sag decay time constant at 2100 mV (ms) 225620 158635 238639 271632 104639
group 5, group 1,3,4 (p = 0.017)
Early frequency accommodation (%) 5564 6768 5267 5967 6568

Late frequency accommodation (%) 6664 7167 6266 8266 8467

Early amplitude accommodation (%) 7363 7965 7165 7765 7065

Late amplitude accommodation (%) 8463 8265 8664 8164 8365

Early AP halfwidth accommodation (%) 16667 143614 195613 126613 157614

group 3. group 2,4 (p = 0.006)
Late AP halfwidth accommodation (%) 11663 12565 11764 12164 10965

Statistical differences with between groups were tested with ANOVA, followed by a post-hoc Neuman-Keuls test. Significant differences with corresponding p-values
(ANOVA) are indicated in red. Values are given as mean (standard error. Abbreviations: CV – coefficient of variation, AP – action potential.
doi:10.1371/journal.pone.0015915.t001

hippocampus. It has been suggested for some interneurons that the adult brain [32]. It would be interesting to examine whether
they only transiently innervate interneurons during postnatal something similar occurs in (a subset of) calretinin-expressing
development, before switching to dendrites of pyramidal cells in interneurons.

PLoS ONE | www.plosone.org 9 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

Table 2. Morphological properties of GFP-labeled interneurons.

Delayed onset, adapting

Immediate
Regular spiking Fast spiking Irregular spiking onset, adapting Strongly adapting
(n = 12) (n = 5) (n = 4) (n = 6) (n = 5)

group 1 group 2 group 3 group 4 group 5

Roundness soma 0.660.2 0.860.04 0.660.1 0.660.1 0.760.2

Aspect ratio soma 2.260.8 1.560.3 2.161.1 2.060.7 1.660.3
Verticality of dendrites 0.860.4 1.160.3 1.160.5 1.460.6 0.860.3
Orientation (degrees)* 59631 61627 37637 24625 71619
Number of primary dendrites 4.161.2 4.261.6 5.562.6 4.460.9 4.260.4

*Tangential cells have an orientation close to 90 degrees; 0 degrees reflect radially oriented cells (similar to pyramidal cells).
No significant differences in morphological parameters between groups were detected (ANOVA). Values are given as mean 6 standard error.
doi:10.1371/journal.pone.0015915.t002

The majority of GFP-labeled interneurons in GAD65-GFP Acknowledgments

mice, especially those in the dendritic layers, express reelin. Reelin
is essential for the proper layering of principal cells during The authors thank Claudia Huber for excellent technical assistance with
immunohistochemistry and mouse care and Volker Staiger for help with
development, but the function of reelin in the adult brain is more
cryo-sectioning. The transgenic mice were kindly provided by Gábor
enigmatic [42,43]. Reelin function in the adult brain has been Szabó (Budapest, Hungary). The antibody against CCK was provided by
linked to cytoskeleton stabilization [44,45], as well as maintenance Dr. Ohning (UCLA Cure).
of NMDA receptors [16] and spines [46,47]. Interestingly, the
description of GFP-labeled interneurons in GAD65-GFP mice has
Author Contributions
a remarkable overlap with interneurons that express the serotonin
receptor 5HT3 [48]. Reelin and 5HT3 receptors are components Conceived and designed the experiments: CJW TK. Performed the
of a common pathway controlling the postnatal development of experiments: CJW FEM IR. Analyzed the data: CJW FEM. Contributed
reagents/materials/analysis tools: VS. Wrote the paper: CJW TB.
apical dendrites of neocortical principal cells [49]. The GAD65-
GFP mice could be a useful tool to investigate the role of reelin
produced by adult interneurons.

References
1. Danglot L, Triller A, Marty S (2006) The development of hippocampal 14. Fiszman ML, Barberis A, Lu C, Fu Z, Erdélyi F, et al. (2005) NMDA receptors
interneurons in rodents. Hippocampus 1060: 1032–1060. increase the size of GABAergic terminals and enhance GABA release. Journal of
2. Freund TF, Buzsáki G (1996) Interneurons of the hippocampus. Hippocampus Neuroscience 25: 2024–2031.
6: 347–470. 15. Galarreta M, Erdélyi F, Szabó G, Hestrin S (2008) Cannabinoid sensitivity and
3. Klausberger T, Somogyi P (2008) Neuronal diversity and temporal dynamics: synaptic properties of 2 GABAergic networks in the neocortex. Cerebral Cortex
the unity of hippocampal circuit operations. Science 321: 53–57. 18: 2296–2305.
4. Ascoli GA, Alonso-Nanclares L, Anderson SA, Barrionuevo G, Benavides- 16. Campo CG, Sinagra M, Verrier D, Manzoni OJ, Chavis P (2009) Reelin
Piccione R, et al. (2008) Petilla terminology: nomenclature of features of secreted by GABAergic neurons regulates glutamate receptor homeostasis. PloS
GABAergic interneurons of the cerebral cortex. Nature Reviews Neuroscience 9: One 4: e5505–e5505.
557–568. 17. Rinke I, Artmann J, Stein V (2010) ClC-2 voltage-gated channels constitute part
5. Anderson SA, Eisenstat DD, Shi L, Rubenstein JLR (1997) Interneuron of the background conductance and assist chloride extrusion. Journal of
migration from basal forebrain to neocortex: dependence on Dlx genes. Science Neuroscience 30: 4776–4786.
278: 474–476. 18. Karagiannis A, Gallopin T, David C, Battaglia D, Geoffroy H, et al. (2009)
6. Miyoshi G, Hjerling-Leffler J, Karayannis T, Sousa VH, Butt SJB, et al. (2010) Classification of NPY-expressing neocortical interneurons. Journal of Neurosci-
Genetic fate mapping reveals that the caudal ganglionic eminence produces a ence 29: 3642–3659.
large and diverse population of superficial cortical interneurons. Journal of 19. Wierenga CJ, Becker N, Bonhoeffer T (2008) GABAergic synapses are formed
Neuroscience 30: 1582–1594. without the involvement of dendritic protrusions. Nature Neuroscience 11:
7. Nery S, Fishell G, Corbin JG (2002) The caudal ganglionic eminence is a source 1044–1052.
of distinct cortical and subcortical cell populations. Nature Neuroscience 5: 20. Tricoire L, Pelkey KA, Daw MI, Sousa VH, Miyoshi G, et al. (2010) Common
1279–1287. origins of hippocampal ivy and nitric oxide synthase expressing neurogliaform
8. Xu Q, Cobos I, Estanislao, Rubenstein JL, Anderson SA (2004) Origins of cells. Journal of Neuroscience 30: 2165–2176.
cortical interneuron subtypes. Journal of Neuroscience 24: 2612–2622. 21. Butt SJ, Fuccillo M, Nery S, Noctor S, Kriegstein A, et al. (2005) The temporal
9. Gelman DM, Marı́n O (2010) Generation of interneuron diversity in the mouse and spatial origins of cortical interneurons predict their physiological subtype.
cerebral cortex. European Journal of Neuroscience 31: 2136–2141. Neuron 48: 591–604.
10. Gelman DM, Martini FJ, Nóbrega-Pereira S, Pierani A, Kessaris N, et al. (2009) 22. Fuentealba P, Begum R, Capogna M, Jinno S, Márton LF, et al. (2008) Ivy cells:
The embryonic preoptic area is a novel source of cortical GABAergic a population of nitric-oxide-producing, slow-spiking GABAergic neurons and
interneurons. Journal of Neuroscience 29: 9380–9389. their involvement in hippocampal network activity. Neuron 57: 917–929.
11. López-Bendito G, Sturgess K, Erdélyi F, Szabó G, Molnár Z, et al. (2004) 23. Price CJ, Cauli B, Kovacs ER, Kulik A, Lambolez B, et al. (2005) Neurogliaform
Preferential origin and layer destination of GAD65-GFP cortical interneurons. neurons form a novel inhibitory network in the hippocampal CA1 area. Journal
Cerebral Cortex 14: 1122–1133. of Neuroscience 25: 6775–6786.
12. Betley JN, Wright CVE, Kawaguchi Y, Erdélyi F, Szabó G, et al. (2009) 24. Price CJ, Scott R, Rusakov DA, Capogna M (2008) GABA(B) receptor
Stringent specificity in the construction of a GABAergic presynaptic inhibitory modulation of feedforward inhibition through hippocampal neurogliaform cells.
circuit. Cell 139: 161–174. Journal of Neuroscience 28: 6974–6982.
13. Daw MI, Tricoire L, Erdélyi F, Szabó G, McBain CJ (2009) Asynchronous 25. Cope DW, Maccaferri G, Márton LF, Roberts JDB, Cobden PM, et al. (2002)
transmitter release from cholecystokinin-containing inhibitory interneurons is Cholecystokinin-immunopositive basket and Schaffer collateral-associated inter-
widespread and target-cell independent. Journal of Neuroscience 29: neurones target different domains of pyramidal cells in the CA1 area of the rat
11112–11122. hippocampus. Neuroscience 109: 63–80.

PLoS ONE | www.plosone.org 10 December 2010 | Volume 5 | Issue 12 | e15915

 GFP-Expressing Interneurons in GAD65-GFP Mice

26. Vida I, Halasy K, Szinyei C, Somogyi P, Buhl EH (1998) Unitary IPSPs evoked 38. Jinno S, Kosaka T (2006) Cellular architecture of the mouse hippocampus: a
by interneurons at the stratum radiatum — stratum lacunosum-moleculare quantitative aspect of chemically defined GABAergic neurons with stereology.
border in the CA1. Journal of Physiology 506.3: 755–773. Neuroscience Research 56: 229–245.
27. Klausberger T (2009) GABAergic interneurons targeting dendrites of pyramidal 39. Galarreta M, Erdelyi F, Szabo G, Hestrin S (2004) Electrical coupling among
cells in the CA1 area of the hippocampus. European Journal of Neuroscience 30: irregular-spiking GABAergic interneurons expressing cannabinoid receptors.
947–957. Journal of Neuroscience 24: 9770–9778.
28. Ledri M, Andreas, Erdélyi F, Szabó G, Kokaia M (2010) Tuning Afferent 40. Klausberger T, Marton LF, O’Neill J, Huck JHJ, Dalezios Y, et al. (2005)
Synapses of Hippocampal Interneurons by Neuropeptide Y. Hippocampus: in Complementary roles of cholecystokinin- and parvalbumin-expressing GABAer-
press. gic neurons in hippocampal network oscillations. Journal of Neuroscience 25:
29. Gulyás AI, Freund TF (1996) Pyramidal cell dendrites are the primary targets of 9782–9793.
calbindin D28k-immunoreactive interneurons in the hippocampus. Hippocam- 41. Pawelzik H, Hughes DI, Thomson AM (2002) Physiological and morphological
pus 6: 525–534. diversity of immunocytochemically defined parvalbumin- and cholecystokinin-
30. Tóth K, Freund TF (1992) Calbindin D28k-containing nonpyramidal cells in the positive interneurones in CA1 of the adult rat hippocampus. Journal of
rat hippocampus: their immunoreactivity for GABA and projection to the Comparative Neurology 443: 346–367.
medial septum. Neuroscience 49: 793–805. 42. Förster E, Bock HH, Herz J, Chai X, Frotscher M, et al. (2010) Emerging topics
31. Jinno S, Klausberger T, Marton LF, Dalezios Y, Roberts JDB, et al. (2007) in Reelin function. European Journal of Neuroscience 31: 1511–1518.
Neuronal diversity in GABAergic long-range projections from the hippocampus. 43. Soriano E, Rı́o JAD (2005) The cells of Cajal-Retzius: still a mystery one century
Journal of Neuroscience 27: 8790–8804. after. Neuron 46: 389–394.
32. Takács VT, Freund TF, Gulyás AI (2008) Types and synaptic connections of 44. Chai X, Förster E, Zhao S, Bock HH, Frotscher M (2009) Reelin stabilizes the
hippocampal inhibitory neurons reciprocally connected with the medial septum. actin cytoskeleton of neuronal processes by inducing n-cofilin phosphorylation at
European Journal of Neuroscience 28: 148–164. serine3. Journal of Neuroscience 29: 288–299.
33. Gulyás AI, Hájos N, Katona I, Freund TF (2003) Interneurons are the local 45. Frotscher M (2010) Role for Reelin in stabilizing cortical architecture. Trends in
targets of hippocampal inhibitory cells which project to the medial septum. Neurosciences 33: 407–413.
European Journal of Neuroscience 17: 1861–1872. 46. Niu S, Yabut O, D’Arcangelo G (2010) The reelin signaling pathway promotes
34. Acsády L, Görcs TJ, Freund TF (1996) Different populations of vasoactive dendritic spine development in hippocampal neurons. Journal of Neuroscience
intestinal polypeptide-immunoreactive interneurons are specialized to control 28: 10339–10348.
pyramidal cells or interneurons in the hippocampus. Neuroscience 73: 317–334. 47. Pujadas L, Gruart A, Bosch C, Delgado L, Teixeira CM, et al. (2010) Reelin
35. Gulyás AI, Hájos N, Freund TF (1996) Interneurons containing calretinin are regulates postnatal neurogenesis and enhances spine hypertrophy and long-term
specialized to control other interneurons in the rat hippocampus. Journal of potentiation. Journal of Neuroscience 30: 4636–4649.
Neuroscience 16: 3397–3411. 48. Vucurovic K, Gallopin T, Ferezou I, Rancillac A, Chameau P, et al. (2010)
36. Gonchar Y, Burkhalter A (2000) Connectivity of GABAergic calretinin- Serotonin 3A receptor subtype as an early and protracted marker of cortical
immunoreactive neurons in rat primary visual cortex. Cerebral Cortex 9: interneuron subpopulations. Cerebral Cortex 20: 2333–2347.
683–696. 49. Chameau P, Inta D, Vitalis T, Monyer H, Wadman WJ, et al. (2009) The N-
37. Meskenaite V (1997) Calretinin-immunoreactive local circuit neurons in area 17 terminal region of reelin regulates postnatal dendritic maturation of cortical
of the cynomolgus monkey, Macaca fascicularis. Journal of Comparative pyramidal neurons. Proceedings of the National Academy of Sciences of the
Neurology 379: 113–132. United States of America 106: 7227–7232.

PLoS ONE | www.plosone.org 11 December 2010 | Volume 5 | Issue 12 | e15915