In this study, we report on the in vivo
consequences of late embryonic deletion of Mef2a
, and Mef2a/c/dTKO
selectively in the brain of mice. In the nervous system, MEF2 transcription factors have emerged as regulators of activity-dependent neuronal survival and differentiation 
. Here, using behavioral, electrophysiological and synapse morphogenesis analysis, we show that the loss of MEF2A and MEF2D in Mef2a/dDKO
does not impact learning and memory, long-term potentiation, or synapse number but does affect short-term synaptic plasticity in the CA1 region of the hippocampus. Generating triple MEF2 knockouts we found the Mef2a/c/dTKO
mice have dramatically increased apoptosis in the brain, suggesting that MEF2A, C, and D redundantly regulate neuronal survival.
Previous knockdown of MEF2, using a dominant negative MEF2 mutant, was shown to cause apoptosis in cerebellar granule neurons in vitro
. The brain-specific Mef2aKO
mice generated for this study were useful in determining specific, as well as redundant, roles of MEF2 isoforms in controlling neuronal survival in vivo
mice are normal in brain size and do not show evidence of neuronal apoptosis, whereas Mef2a/c/dTKO
mice show smaller brain size and increased apoptotic cell death compared to littermate controls. Interestingly, brain-specific deletion of MEF2C alone does not cause any abnormalities in brain size or apoptosis 
, suggesting the deficits in the Mef2a/c/dTKO
are the result of functional redundancy of the MEF2 family members to regulate neuronal survival in vivo
In this study, we found that embryonic deletion of Mef2a and Mef2d in the brain using the Mef2a/dDKO mice impaired motor coordination, but did not affect other behaviors including hippocampal-dependent learning and memory. Mef2a and Mef2d are highly expressed in the CA1 subregion of the hippocampus, thus we focused our analysis of spine number on CA1 pyramidal neurons, LTP analysis by stimulation of the Schaffer collateral pathway, and synaptic plasticity measurements in this subregion of the hippocampus.
Recent work has suggested that MEF2 is a critical regulator of excitatory synapse number and function. MEF2A and MEF2D have been reported to negatively regulate excitatory synapse number of hippocampal neurons during synaptic development in vitro
based on RNAi experiments 
. Surprisingly, the Mef2a/dDKO
mice generated in this study had no changes in the density of dendritic spines, a proxy measure of excitatory synapses, of the CA1 neurons using z-series of Golgi-stained tissue. To more closely examine whether this apparent disparity is due to in vitro
versus in vivo
differences, we cultured primary hippocampal neurons from floxed Mef2a/d
mice and then infected them with a high-titer lentivirus expressing Cre recombinase and monitored Mef2a/d
deletion. We found no significant difference in the frequency of mEPSCs of the Mef2a/d
null neurons compared to GFP-infected neurons, suggesting that MEF2A and D are not important regulators of excitatory synapse function and likely not regulators of excitatory synapse number.
In contrast to our findings regarding synapse number in the CA1 region of the Mef2a/dDKO
mice, the conditional deletion of Mef2c
in brain suppresses the number of excitatory synapses in the dentate gyrus and regulates basal and evoked synaptic transmission 
. MEF2C is highly expressed in the dentate gyrus with only low-level expression in other subregions of the hippocampus 
. Our previous study on the role of MEF2C in vivo, focused on the analysis of synapse number and function in the dentate gyrus. In the present study, we focused instead on the CA1 subregion because of the high level expression of Mef2a
in this region of the hippocampus. The lack of an alteration in synapse number and the Mef2a/dDKO
suggests that MEF2A and MEF2D are not key regulators of synapse number. Our current findings that Mef2a/c/dTKO
mice have no changes in the density of dendritic spines of the CA1 neurons, is consistent with the low level of expression of MEF2C in this region of the hippocampus, and suggests that in the surviving neurons, at least structurally, synapses are present at normal levels.
To investigate the functional consequences of MEF2A and D loss, we used field potential recordings to activate Schaffer collateral fibers in the CA1 region to assess synaptic connectivity and found that the Mef2a/dDKO
mice had no significant change in input-output curve, suggesting that the number of synapses activated by action potential stimulation is not changed. In contrast, the Mef2a/c/dTKO
mice have a reduction in the input-output curve. However, this may be due to these mice having a decrease in the number of neurons because of the ongoing apoptosis. We also investigated presynaptic function by measuring PPF, a short-term enhancement of synaptic activity in response to the second of two paired stimuli, believed to be due to residual calcium in the presynaptic terminal after the initial stimulus 
. The Mef2a/dDKO
mice have an increase in PPF suggesting a decrease in release probability, a measure of the likelihood of neurotransmitter release following an action potential and an important determinant of synaptic strength. The Mef2a/c/dTKO
mice also show an increase in PPF suggesting a decrease in release probability, but it should be noted that this could also be an indicator of synaptic degeneration triggered by ongoing apoptosis 
The cell death observed in the Mef2a/c/dTKO mice raises complications in the evaluation of synaptic electrophysiology. Nevertheless, the similar number of spines, normal LTP, and increase in PPF similar to that observed in the Mef2a/dDKO suggest that there is no additive effect of loss of Mef2c in the CA1 region, and that synapses in the Mef2a/c/dTKO mice exhibit normal plasticity following an LTP induction protocol. These data suggest that MEF2A and D play only a subtle role in regulating synaptic function. In contrast, the previous reports on the conditional MEF2C knockouts suggest that this isoform has a significant effect in regulating excitatory synapse structure and function in dentate gyrus. Our data suggests that while the roles of the MEF2 isoforms are distinct in their impact on synaptic number and function, they have overlapping effects in cellular maintenance and survival. Collectively, these results highlight the specific roles for the MEF2A, C and D isoforms in regulation of synaptic function and structure in vivo.