We used recording and stimulation electrodes in longitudinal hippocampal brain slices and longitudinally positioned recording and stimulation electrodes in the dorsal hippocampus in vivo to evoke extracellular postsynaptic potentials and demonstrate long-term synaptic plasticity along the longitudinal interlamellar CA1.
The study of synaptic plasticity in the hippocampus has focused on the use of the CA3-CA1 lamellar network. Less attention has been given to the longitudinal interlamellar CA1-CA1 network. Recently however, an associational connection between CA1-CA1 pyramidal neurons has been shown. Therefore, there is the need to investigate whether the longitudinal interlamellar CA1-CA1 network of the hippocampus supports synaptic plasticity.
We designed a protocol to investigate the presence or absence of long-term synaptic plasticity in the interlamellar hippocampal CA1 network using electrophysiological field recordings both in vivo and in vitro. For in vivo extracellular field recordings, the recording and stimulation electrodes were placed in a septal-temporal axis of the dorsal hippocampus at a longitudinal angle, to evoke field excitatory postsynaptic potentials. For in vitro extracellular field recordings, hippocampal longitudinal slices were cut parallel to the septal-temporal plane. Recording and stimulation electrodes were placed in the stratum oriens (S.O) and the stratum radiatum (S.R) of the hippocampus along the longitudinal axis. This enabled us to investigate the directional and layer specificity of evoked excitatory postsynaptic potentials. Already established protocols were used to induce long-term potentiation (LTP) and long-term depression (LTD) both in vivo and in vitro. Our results demonstrated that the longitudinal interlamellar CA1 network supports N-methyl-D-aspartate (NMDA) receptor-dependent long-term potentiation (LTP) with no directional or layer specificity. The interlamellar network, however, in contrast to the transverse lamellar network, did not present with any significant long-term depression (LTD).
The hippocampus has been widely used in cognitive studies1,2,3. The hippocampal lamellar network in the transverse axis forms the tri-synaptic circuitry that is made up of the dentate gyrus, CA3, and CA1 regions. The lamellar network is considered to be a parallel and independent unit4,5. This lamellar viewpoint has influenced the use of transverse orientation and transverse slices for both in vivo and in vitro electrophysiological studies of the hippocampus. In light of emerging research, the lamellar hypothesis is being reevaluated6 and attention is also being given to the interlamellar network of the hippocampus. With regards to the hippocampal interlamellar network, the CA3 region has long been investigated7,8,9,10, however the longitudinal CA1 hippocampal region has received relatively little attention until recently. With regards to the CA1 interlamellar network, the short-term synaptic properties along the dorsoventral longitudinal hippocampal CA1 axis of rats have been shown to vary11. Also, clusters of hippocampal cells responding to the phase and the place were found to be arranged systematically along the longitudinal axis of the hippocampus in rats, undergoing a short term memory task12. Also, epileptic seizure activities were found to be synchronized along the whole hippocampus along the longitudinal axis13.
Most studies of the longitudinal CA1 hippocampal region however, have utilized input from the CA3 to the CA1 regions11,14,15. Using a unique protocol to make longitudinal brain slices, our previous work demonstrated the associational connectivity of CA1 pyramidal neurons along the longitudinal axis and implicated its ability to process neuronal signaling effectively16. However, there is a need to determine whether the CA1 pyramidal neurons along the longitudinal axis without transverse input can support long term synaptic plasticity. This finding can add another angle into investigations of neurological issues pertaining to the hippocampus.
The ability of neurons to adapt the efficacy of information transfer is known as synaptic plasticity. Synaptic plasticity is implicated as the underlying mechanism for cognitive processes such as learning and memory17,18,19,20. Long-term synaptic plasticity is demonstrated as either long-term potentiation (LTP), which represents the strengthening of neuronal response, or long-term depression (LTD), which represents the weakening of neuronal response. Long-term synaptic plasticity has been studied in the transverse axis of the hippocampus. However, this is the first study to demonstrate long-term synaptic plasticity in the hippocampal longitudinal axis of CA1 pyramidal neurons.
Building from a protocol used by Yang et al.16, we designed the protocol to demonstrate LTP and LTD in the hippocampal longitudinal axis of CA1 pyramidal neurons. We used C57BL6 male mice with ages ranging between 5-9 weeks old for in vitro experiments and 6-12 weeks old for in vivo experiments. This detailed article shows how longitudinal hippocampal brain slices from mice were obtained for in vitro recordings and how in vivo recordings were recorded in the longitudinal axis. For in vitro recordings, we investigated directional specificity of longitudinal CA1 synaptic plasticity by targeting the septal and temporal end of the hippocampus. We also investigated layer specificity of the longitudinal CA1 synaptic plasticity by recording from the stratum oriens and stratum radiatum of the hippocampus. For in vivo recordings, we investigated the angles that best correspond to the longitudinal direction of the hippocampus.
Using both in vivo and in vitro extracellular field recordings, we observed that the longitudinally connected CA1 pyramidal neurons presented with LTP, not LTD. The transverse orientation involving both CA3 and CA1 neurons, however, supports both LTP and LTD. The distinction in the synaptic capabilities between the transverse and the longitudinal orientation of the hippocampus could speculatively signify differences in their functional connectivity. Further experiments are needed to decipher the differences in their synaptic capabilities.
All animals were treated in accordance with the guidelines and regulations from the Animal Care and Use of Laboratory of National Institute of Health. All methods described here have been approved by the Institutional Animal Care and Use Committee (IACUC) of City University of Hong Kong and Incheon National University.
1. In vivo field recording
2. In vitro field recording
We explored long-term synaptic plasticity of longitudinal CA1 pyramidal neurons of the hippocampus using extracellular field recordings both in vivo and in vitro. LTP and LTD are facets of long-term synaptic plasticity that have been demonstrated in the transverse axis of the hippocampus to be unidirectional.
We showed here that using longitudinal hippocampal brain slices, there is LTP in the CA1 longitudinal axis of the hippocampus. We prepared longitudinal slices of the hippocampus along the septotemporal axis, which is perpendicular to the transverse slices (Figure 1). Using recordings from the CA1 region of the hippocampus, we showed the presence of LTP that was not direction specific. There were no statistically significant differences in the recordings from the septal or temporal (Figure 2) side of the longitudinal hippocampal brain slice. We also showed the presence of LTP that was not layer specific; thus, recordings from both stratum radiatum and stratum oriens (Figure 2) showed successfully induced LTP in the longitudinal brain slice. We used D-AP5, an NMDAR antagonist to demonstrate that the LTP induced was dependent on NMDA receptors (Figure 3). What happens in vitro does not necessarily reflect in vivo conditions, so we investigated LTP in vivo. Figure 4a shows a schematic diagram of the stimulation and recording electrode positioned in the dorsal hippocampus along the longitudinal axis of CA1 region in vivo. The position of the electrodes used for the recording and stimulation was verified by lesion marks and crystal violet staining (Figure 4a). We demonstrated the presence of LTP in vivo in the longitudinal CA1 region (Figure 4b).
Using already established protocols for inducing LTD, we failed to successfully induce LTD both in vivo and in vitro (Figure 5).
Figure 1. A schematic drawing of transverse and longitudinal hippocampal brain slices. This figure is adapted and modified from Sun et al. 201821. Please click here to view a larger version of this figure.
Figure 2. LTP in longitudinal slices. Synaptic responses at S.R. (a) or S.O. (b) in longitudinal slices are potentiated right after tetanus stimulation with both temporal and septal inputs (S.R./temporal (n = 12, c), S.R./septal (n = 12, c), S.O./temporal (n = 10, d), S.O./septal (n = 9, d).
The n stands for the number of slices. Error bars represent SE. This figure is adapted and modified from Sun et al. 201821. Please click here to view a larger version of this figure.
Figure 3. NMDAR-dependent LTP in longitudinal slices. (a,b) LTP induction in temporal and septal direction is blocked by 50 μM D-AP5 (temporal, n = 6, a) (septal, n = 5, b). (c,d) LTP induction in temporal and septal direction is also blocked by D-AP5. The n stands for slices. Error bars represent SE. This figure is adapted and modified from Sun et al. 201821. Please click here to view a larger version of this figure.
Figure 4. In vivo LTP in the interlamellar network. (a) A schematic drawing of recording and stimulation electrodes in anesthetized animals. The loci of recording (on the septal side of CA1) and stimulating electrodes (on the temporal side of CA1) were identified by lesion marks. (b) LTP is induced in the interlamellar connection by 100 Hz high frequency stimulation (HFS) (n = 10 mice). Color traces: before (black) and after (red) HFS. Error bars represent SE. This figure is adapted and modified from Sun et al. 201821. Please click here to view a larger version of this figure.
Figure 5. Absence of in vivo and in vitro LTD in Interlamellar CA1 network. (a) 1 Hz-pp LFS does not induce in vivo LTD. (b) 1 Hz pp-LTP, (c) 5 Hz LFS, and (d) 1 Hz LFS do not produce LTD on either the temporal or septal sides of longitudinal brain slice. while LTD is induced by 1 Hz pp-LFS in transverse slices: temporal (n = 8), septal (n = 11) and transverse (n = 6) with 1 Hz pp-LFS; temporal (n = 3) and septal (n = 3) with 5 Hz LFS; temporal (n = 3) and septal (n = 3) with 1 Hz LFS. The n stands for slices. Error bars represent SE. This figure is adapted and modified from Sun et al. 201821. Please click here to view a larger version of this figure.
Figure 6. Input-output curve presenting fEPSP slope in response to increasing stimulus input in hippocampal brain slice. Please click here to view a larger version of this figure.
Figure 7. Surgical tools used for hippocampal isolation during in vitro brain slicing. Please click here to view a larger version of this figure.
Figure 8. A longitudinal brain slice ready for recording. Stimulation electrode and recording pipette are inserted in the stratum radiatum. Please click here to view a larger version of this figure.
Figure 9. A transverse hippocampal brain slice ready for recording. Stimulation electrode is inserted at Schaffer collateral CA3 region and recording pippette is inserted at CA1 region. Please click here to view a larger version of this figure.
Compound | Slicing Solution(mM) | ACSF (mM) |
CaCl2.2H2O | 0.5 | 2 |
Glucose | 25 | 25 |
KCl | 2.5 | 2.5 |
MgCl2.6H2O | 7 | 1 |
NaCl | 87 | 125 |
NaH2PO4 | 1.3 | 1.3 |
NaHCO3 | 25 | 25 |
Sucrose | 75 |
Table 1: Concentrations of compounds in brain slice and artificial cerebrospinal fluid solutions.
The protocol demonstrates the method to induce long-term synaptic plasticity in vivo as well as from brain slices in the longitudinal CA1-CA1 axis of the hippocampus in vitro. The steps outlined give enough details for an experimenter to investigate LTP and LTD in a longitudinal hippocampal CA1-CA1 connection. Practice is needed to hone the skills required to successfully record field excitatory potentials.
In addition to needing practice, there are several critical steps that are essential to obtaining good results. First, it was shown previously that the angle to which the brain slices were made could either truncate or preserve the longitudinal projections of the pyramidal neurons in the CA1-CA1 region of the hippocampus16. The longitudinal pyramidal neurons project from the transverse neurons at an angle that is nearly perpendicular. As the CA1 neurons propagate in diverse angles within the hippocampus, the longitudinal connection between them lays out along the dorsoventral axis of the hippocampus. Thus, for in vitro recording, the experimenter must keep this in mind to accurately target the CA1-CA1 hippocampal neurons along the longitudinal axis by cutting the isolated hippocampus tissue along the dorsal-ventral axis. Also, for in vivo recordings, the angle at which the stimulation and the recording electrodes are positioned determines whether the results obtained are representative of the longitudinal axis or a mixture of both the transverse and longitudinal axis. Further investigations utilizing CRISPR-Cas9 can be done to confirm whether the evoked response is solely from the CA1 region since it could be a mixture of responses from both the CA1 and the CA3 regions.
Secondly, for in vitro experiments, the experimenter must ensure that the brain slicing solution, ACSF, work bench and all equipment or instruments that come in contact with the brain slice are free from contaminants. Any form of contamination will lead to the deterioration of the integrity or death of the brain slice. Maintaining a clean electrode surface will ensure good and stable recordings for both in vitro and in vivo experiments.
We have shown that the longitudinal hippocampal CA1 network exhibits NMDAR-dependent LTPs, but not LTDs. The trisynaptic circuit, however, presents with both LTP and LTD22,23. This implies that the longitudinal CA1 network and the tri-synaptic circuitry have unique features. Our protocol makes use of only electrophysiological recordings and therefore is limited in finding the difference between these two networks.
The search for a cure for brain diseases such as schizophrenia continues. Decline or deformity of CA1 hippocampal subregions have been linked with some schizophrenic symptoms24,25. The application of our protocol, though basic, has brought to light a unique synaptic capability of the CA1 longitudinal hippocampal subregion. This knowledge is useful in designing experiments that can further investigate this debilitating brain disease along the longitudinal CA1 axis of the hippocampus.
The authors have nothing to disclose.
This work was supported by Incheon National University (International Cooperative) Research Grant. We will like to thank Ms. Gona Choi for assisting with some data collection.
Atropine Sulphate salt monohydrate, ≥97% (TLC), crystalline | Sigma-Aldrich | 5908-99-6 | Stored in Dessicator |
Axon Digidata 1550B | |||
Calcium chloride | Sigma-Aldrich | 10035-04-8 | |
Clampex 10.7 | |||
D-(+)-Glucose ≥ 99.5% (GC) | Sigma-Aldrich | 50-99-7 | |
Eyegel | Dechra | ||
Isoflurane | RWD Life Sciences | R510-22 | |
Magnesium chloride hexahydrate, BioXtra, ≥99.0% | Sigma-Aldrich | 7791-18-6 | |
Matrix electrodes, Tungsten | FHC | 18305 | |
Multiclamp 700B Amplifier | |||
Potassium chloride, BioXtra, ≥99.0% | Sigma-Aldrich | 7447-40-7 | |
Potassium phosphate monobasic anhydrous ≥99% | Sigma-Aldrich | 7778-77-0 | Stored in Dessicator |
Pump | Longer precision pump Co., Ltd | T-S113&JY10-14 | |
Silicone oil | Sigma-Aldrich | 63148-62-9 | |
Sodium Bicarbonate, BioXtra, 99.5-100.5% | Sigma-Aldrich | 144-55-8 | |
Sodium Chloride, BioXtra, ≥99.5% (AT) | Sigma-Aldrich | 7647-14-5 | |
Sodium phosphate monobasic, powder | Sigma-Aldrich | 7558-80-7 | |
Sucrose, ≥ 99.5% (GC) | Sigma-Aldrich | 57-50-1 | |
Temperature controller | Warner Instruments | TC-324C | |
Tungsten microelectrodes | FHC | 20843 | |
Urethane, ≥99% | Sigma-Aldrich | 51-79-6 | |
Vibratome | Leica | VT-1200S | |
Water bath | Grant Instruments | SAP12 |