Ectopic Mossy Fiber Pathfinding in the Hippocampus Caused the Abnormal Neuronal Transmission in the Mouse Models of Psychiatric Disease

Abstract

Appropriate axonal pathfinding is a necessary step for the function of neuronal circuits. The mossy fibers (MFs) in the hippocampus of CaMKIIα heterozygous knockout (CaMKIIα-hKO) psychiatric model mice project onto not only the stratum lucidum but also the stratum oriens region in the CA3, which is a projection pattern distinct from that in normal mice. Thus, we examined the electrophysiological properties of the MF–CA3 connection in this mutant mouse on field recordings and found a lower synaptic connection. This study suggested that the phenotype of abnormal MF pathfindings could induce aberrant neuronal functions, which may link to cognition and memory.

Therapeutic interventions for psychiatric disorders such as schizophrenia (SCZ) are hampered by a lack of clear molecular signatures underlying the pathophysiology of the disorders. However, several studies point to abnormal neuronal connectivity in the hippocampus, which potentially relates to the cognitive impairments observed in SCZ patients.¹⁾ The abnormal hippocampal synaptic transmission found in SCZ patients could result from inappropriate neuronal connectivity.

Appropriate axonal guidance has to be established during development and maintained throughout life to coordinate neural circuits and their functions. Mossy fibers (MFs), which are axons of granule cells in the dentate gyrus (DG), innervate to the CA3 subregion, forming a hippocampal circuit between DG and CA3. During postnatal development, MFs extend to both the stratum lucidum (SL) and the stratum oriens (SO), forming supra-pyramidal bundles (SPB) and infra-pyramidal bundles (IPB), respectively. The formation of IPB, however, is transient and undergoes pruning during development, leaving mostly SPB upon maturation.²⁾ It is of note that some pathological conditions such as temporal lobe epilepsy result in ectopic formation of IPB.³⁾ The CaMKIIα heterozygous knockout (CaMKIIα-hKO) mouse is considered one of the psychiatric model mice as they showed hyper locomotor activity, the deficits of working memory and social interaction,^4,5) which also have the ectopic MF projection in SO.⁶⁾ In the present study, we investigated the electrophysiological properties in the MF–CA3 connection in CaMKIIα-hKO mice to characterize the functional abnormalities of the ectopic MF distribution.

MATERIALS AND METHODS

Two- to four-month-old male mice (age-matched littermates) were used for the experiments. CaMKIIα-hKO mice were generated and maintained in a C57BL6/N^tac background⁷⁾ (Please see our Supplementary Materials and figure for more details). All procedures were approved by the Institutional Animal Care and Use Committee at Astellas Research Institute of America LLC and undertaken in accordance with the Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 in the protection of animals used for scientific purposes A13 055 34. All efforts were made to minimize the animal’s suffering and the number of animals used.

Primary antibodies used were anti-ZnT3 (1 : 1000, ABIN1742412, antibodies-online, GA, U.S.A.), anti-calbindin (1 : 1000, CB-38a, Swant, Marly, Switzerland), anti-calretinin (1 : 1000, 6B3, Swant), anti-CaMKIIα (1 : 1000, ab22609, Abcam, Cambridge, U.K.), and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1 : 1000, 5174S, Cell Signaling Technology, MA, U.S.A.).

Multielectrode array (MEA) field recordings of hippocampal slices from CaMKIIα-hKO mice were conducted by Neuroservice (Domaine de Saint Hilaire, France). All data were recorded with a MEA set-up commercially available from MultiChannel Systems MCS GmbH (Reutlingen, Germany) (Details in supplementary materials and figure).

RESULTS AND DISCUSSION

Previously, we reported the ectopic projection of MFs into SO in mice with a hKO of the CaMKIIα gene (generated by the Tonegawa lab).⁷⁾ In the present study, we created a similar mouse line lacking the CaMKIIα gene by inserting a puromycin resistance gene into exon11 (Fig. 1A). Consistent with our previous report, Western blot analysis showed significantly up-regulated expression of ZnT-3 protein, a maker of mossy fiber, in CaMKIIα-hKO mice (Fig. 1B), and immunohistochemistry showed ectopic MF pathfindings (supplementary figure). Furthermore, we observed an increase in calretinin, a marker for immature neurons, and a decrease in calbindin, a marker for mature neurons, in the DG of CaMKIIα-hKO mice (Fig. 1C), as Yamasaki et al. previously reported.⁴⁾ This finding showed the face validity of CaMKIIα-hKO mice as a model of schizophrenia because this immaturity in DG was also seen in patients with schizophrenia.⁸⁾ Moreover, the formation of IPB mossy fiber could be due to the immaturation of GCs in DG because IPB formation is only seen in immature GCs.

Fig. 1. The Immature Dentate Gyrus Displayed an Ectopic Mossy Fiber Distribution

(A) The image of Western blot showed a reduction of CaMKIIα protein in the hippocampus of CaMKIIα-hKO mice. * p<0.05; Student’s t-test (n=6). (B) The image of Western blot showed an increase of Znt-3 protein in the hippocampus of CaMKIIα-hKO mice. *** p<0.001; Student’s t-test (n=6). (C) The image of calretinin and calbindin expression in the dentate hippocampus showed an increase of calretinin and a decrease of calbindin in CaMKIIα-hKO mice. WT; wild type mouse, Het; CaMKIIα-hKO mouse, GCL; granule cell layer.

Next, we examined the functional properties of the aberrancies of the MF distribution via field recording (Fig. 2A). First, we evaluated the Input (MF stimulation)/Output (evoked response in SL) relationship of MF synapses in transverse hippocampal slices of CaMKIIα-hKO mice. Overall, the amplitude of the evoked responses in response to the MF stimulation was significantly lower in the CA3 of CaMKIIα-hKO mice (Fig. 2B). We next assessed the oscillatory activity in the hippocampus under kainate exposure because of preferential localization of kainate receptor GluK5 on MFs,⁹⁾ (Figs. 2C, D). As shown in Fig. 2D, 100 nM kainate rapidly triggered network oscillations, both in control wild type (WT) and CaMKIIα-hKO mice. However, the strength of the kainate-induced oscillations was much larger in WT mice than in CaMKIIα-hKO mice (Figs. 2D, E). One hundred micromole Ketamine strongly decreased the strength of gamma oscillations in WT mice hippocampal slices, and the effect was in the same range in CaMKIIα-hKO mice. Next, we checked the coherence of the oscillatory activity between SL and SO by comparing the activity detected electrode hit on SL with the hit on SO, to examine the effects of IPB on the oscillatory activity (Fig. 2F). As illustrated in the individual slices cross-correlation graph in WT mice (Fig. 2G), the oscillations displayed opposite directions (main frequency values close to −1) between SO and SL. On the contrary, in CaMKIIα-hKO mice, the oscillations globally did not display opposite directions (Fig. 2G) with the exception of one single slice out of the 9 tested (the main frequency displayed positive values, except for one slice). The cross-correlation index at the end of the kainate-exposure period (cross-correlation value at t=0 ms) was plotted and was found to significantly differ between WT and KO mice (Fig. 2H).

Fig. 2. The Electrophysiological Properties of the Mossy Fiber–CA3 Connection in the Hippocampal Slices of CaMKIIα-hKO Mice

(A) Representative trace of fEPSP in the CA3 region of WT and CaMKIIα-hKO mice. (B) Comparison of I/O properties in the SL layer of the CA3 region of WT and CaMKIIα-hKO mice. The mean values of raw evoked-response amplitudes are plotted±S.E.M. versus the stimulation intensity. Data averaged from 27 electrodes from 8 slices prepared from 6 WT mice and data averaged from 39 electrodes from 9 slices prepared from 6 CaMKIIα-hKO mice. * p<0.05, ** p<0.01, *** p<0.001; repeated measures two-way ANOVA followed by a post hoc Bonferroni test. (C) Representative network oscillations in the SL of WT and CaMKIIα-hKO mice. The oscillatory activity presented above was low pass filtered at 80 Hz. (D) Evaluation of 100 µM Ketamine on kainate-induced gamma oscillations in the CA3 hippocampal region of WT and CaMKIIα-hKO mice. After a 20-min control period, hippocampal network oscillations were elicited by 100 nM kainate for 60 min. Network oscillations were then monitored in the presence of 100 µM Ketamine, co-applied with 100 nM kainate for 60 min. Data averaged from 27 electrodes, from 8 slices prepared from 6 WT mice and data averaged from 39 electrodes from 9 slices prepared from 6 CaMKIIα-hKO mice. (E) Kainate-induced gamma oscillations in the CA3 hippocampal region of WT and in CaMKIIα-hKO mice. The graph displays the AUC of the power spectrum in the gamma range (30–80 Hz) after a 60-min exposure to 100 nM kainate in the SL layer of WT mice and CaMKIIα-hKO mice. * p<0.05; Mann–Whitney test (n=8–9 slices from 6 animals). (F) Topological recording site of SL and SO in the hippocampal slice. (G) Cross-correlation of oscillations between SL and SO for each slice (after a 60-min exposure to 100 nM kainate). The x-axis represents the lag period (ms). The y-axis represents the cross-correlation function estimate for each lag number. (H) Cross-correlation index in WT and CaMKIIα-hKO mice hippocampal slices at the end of the kainate-exposure period. The cross-correlation value at a time 0 ms is presented in the graph above for each individual slice from WT and CaMKIIα-hKO mice. *** p<0.001; Student’s t-test (n=8–9 slices from 6 animals). WT; wild type mouse, Het; CaMKIIα-hKO mouse, SL; stratum lucidum, SO; stratum oriens, fEPSP; field excitatory postsynaptic potential.

Based on our data, the ectopic MF projection was related to abnormal electrophysiology of DG-CA3 connectivity in CaMKIIα-hKO mice. Although MFs increased their subfield, MF–CA3 synaptic strength was functionally immature in CaMKIIα-hKO mice. In contrary to our expectation, fEPSP in SL was decreased in the mutant and looked different from the findings of Yamasaki et al., which demonstrated increased basal MF synaptic strength.⁴⁾ There are several differences in experimental conditions between Yamasaki et al. and ours,⁴⁾ such as stimulation site and frequency, age of animals, and temperature during recording, which are all known to affect neuronal activity.^10–12) Also, as shown in Figs. 2A and B, the amplitude of fEPSP depends on stimulation intensity, yet Yamasaki et al. did not mention how much intensity they gave.⁴⁾ Also, the most prominent difference could be that Yamasaki et al. examined the fEPSP/fiber volley (not fEPSP itself) as the MF synaptic strength.⁴⁾ Because the fiber volley reflects the activation of presynaptic fibers and CaMKIIα activates GluK5 which is preferentially localized on presynaptic MFs,¹³⁾ it is possible that the fiber volley could be decreased in CaMKIIα-hKO mice. So, if the amplitude of fiber volley is dramatically decreased, the ratio of the peak fEPSP amplitude to fiber volley amplitude could be increased.

One possibility is that the weak MF function observed in this study may have resulted from the greater number of branch points of MFs in CaMKIIα-hKO mice. Starting from the initial discussion by Grossman that axonal propagation might fail at branch points,¹⁴⁾ several studies have indicated that propagation is perturbed by axonal branch points in a number of neurons.^15–21) In addition, a lower release probability of glutamate from the presynaptic terminals could also contribute to the smaller fEPSP. To test this, the fiber volley, frequency facilitation, or spontaneous excitatory postsynaptic current in CA3 should be examined in a future study.

In addition, the gamma oscillation elicited by kainate was also decreased in CaMKIIα-hKO mice (Figs. 2D, E). Gamma oscillation plays an important role for learning and memory, and CaMKIIα-hKO mice actually showed the cognitive deficits in addition to the impaired spatial learning previously reported,⁷⁾ which are observed in SCZ.²²⁾ Also, the ectopically wider distribution of MFs in CaMKIIα-hKO mice may induce the same direction of oscillations between SL and SO (Figs. 2G, H). In contrast, WT mice elicited the oscillation with a time-lag. This suggests that IPB may also generate the oscillations at SO and possibly interfere with the oscillations in SL. To strongly clarify the involvement of MFs in oscillation activity, spontaneous excitatory or inhibitory postsynaptic current and firing at CA3 post-synaptic neurons including interneurons in SL and SO should be examined in the future. In conclusion, we identified the lower synaptic connection between ectopic MFs and CA3. The results of our study suggest that the phenotype of abnormal MF pathfindings could induce aberrant neuronal oscillations, which may link to cognition and memory. For the future study, the identity of MF origin and fiber volley should also be pharmacologically validated in each slice, which strengthens our findings.

Acknowledgment

We outsourced electrophysiology measurements in the slices to Neuroservice (France) and appreciate their input to the figures.

Conflict of Interest

All authors are employees of Astellas Pharma.

Supplementary Materials

The online version of this article contains supplementary materials.

REFERENCES

• 1) Tamminga CA, Stan AD, Wagner AD. The hippocampal formation in schizophrenia. Am. J. Psychiatry, 167, 1178–1193 (2010).
• 2) Liu XB, Low LK, Jones EG, Cheng HJ. Stereotyped axon pruning via plexin signaling is associated with synaptic complex elimination in the hippocampus. J. Neurosci., 25, 9124–9134 (2005).
• 3) Römer B, Krebs J, Overall RW, Fabel K, Babu H, Overstreet-Wadiche L, Brandt MD, Williams RW, Jessberger S, Kempermann G. Adult hippocampal neurogenesis and plasticity in the infrapyramidal bundle of the mossy fiber projection: I. Co-regulation by activity. Front Neurosci., 5, 107 (2011).
• 4) Yamasaki N, Maekawa M, Kobayashi K, Kajii Y, Maeda J, Soma M, Takao K, Tanda K, Ohira K, Toyama K, Kanzaki K, Fukunaga K, Sudo Y, Ichinose H, Ikeda M, Iwata N, Ozaki N, Suzuki H, Higuchi M, Suhara T, Yuasa S, Miyakawa T. Alpha-CaMKII deficiency causes immature dentate gyrus, a novel candidate endophenotype of psychiatric disorders. Mol. Brain, 1, 6 (2008).
• 5) Hagihara H, Takao K, Walton NM, Matsumoto M, Miyakawa T. Immature dentate gyrus: an endophenotype of neuropsychiatric disorders. Neural Plast., 2013, 318596 (2013).
• 6) Nakahara S, Miyake S, Tajinda K, Ito H. Mossy fiber mis-pathfinding and semaphorin reduction in the hippocampus of α-CaMKII hKO mice. Neurosci. Lett., 598, 47–51 (2015).
• 7) Silva AJ, Paylor R, Wehner JM, Tonegawa S. Impaired spatial learning in alpha-calcium-calmodulin kinase II mutant mice. Science, 257, 206–211 (1992).
• 8) Walton NM, Zhou Y, Kogan JH, Shin R, Webster M, Gross AK, Heusner CL, Chen Q, Miyake S, Tajinda K, Tamura K, Miyakawa T, Matsumoto M. Detection of an immature dentate gyrus feature in human schizophrenia/bipolar patients. Transl. Psychiatry, 2, e135 (2012).
• 9) Schmitz D, Mellor J, Frerking M, Nicoll RA. Presynaptic kainate receptors at hippocampal mossy fiber synapses. Proc. Natl. Acad. Sci. U.S.A., 98, 11003–11008 (2001).
• 10) Mizunuma M, Takahashi N, Usami A, Matsuki N, Ikegaya Y. High-temperature, but not high-pressure, conditions alter neuronal activity. J. Pharmacol. Sci., 110, 117–121 (2009).
• 11) Villers A, Ris L. Improved preparation and preservation of hippocampal mouse slices for a very stable and reproducible recording of long-term potentiation. J. Vis. Exp., 76, 50483 (2013).
• 12) Goodell DJ, Benke TA, Bayer KU. Developmental restoration of LTP deficits in heterozygous CaMKIIα KO mice. J. Neurophysiol., 116, 2140–2151 (2016).
• 13) Carta M, Opazo P, Veran J, Athané A, Choquet D, Coussen F, Mulle C. CaMKII-dependent phosphorylation of GluK5 mediates plasticity of kainate receptors. EMBO J., 32, 496–510 (2013).
• 14) Grossman Y, Parnas I, Spira ME. Differential conduction block in branches of a bifurcating axon. J. Physiol., 295, 283–305 (1979a).
• 15) Parnas I. Differential block at high frequency of branches of a single axon innervating two muscles. J. Neurophysiol., 35, 903–914 (1972).
• 16) Van Essen DC. The contribution of membrane hyperpolarization to adaptation and conduction block in sensory neurones of the leech. J. Physiol., 230, 509–534 (1973).
• 17) Yau KW. Receptive fields, geometry and conduction block of sensory neurons in the CNS of the leech. J. Physiol., 263, 513–538 (1976).
• 18) Grossman Y, Parnas I, Spira ME. Mechanisms involved in differential conduction of potentials at high frequency in a branching axon. J. Physiol., 295, 307–322 (1979b).
• 19) Smith DO. Mechanisms of action potential propagation failure at sites of axon branching in the crayfish. J. Physiol., 301, 243–259 (1980a).
• 20) Smith DO. Morphological aspects of the safety factor for action potential propagation at axon branch points. J. Physiol., 301, 261–269 (1980b).
• 21) Ducreux C, Reynaud JC, Puizillout JJ. Spike conduction properties of T-shaped C neurons in the rabbit nodose ganglion. Pflugers Arch., 424, 238–244 (1993).
• 22) Siemerkus J, Irle E, Schmidt-Samoa C, Dechent P, Weniger G. Egocentric spatial learning in schizophrenia investigated with functional magnetic resonance imaging. Neuroimage Clin., 1, 153–163 (2012).