Neurexin-2 restricts synapse numbers and restrains the presynaptic release probability by an alternative splicing-dependent mechanism

Significance Previous in vivo studies showed that neurexin-2 unexpectedly regulates synaptic connectivity in hippocampal circuits not by inducing, but by repressing excitatory synapses. The present study demonstrates that as assessed by heterologous synapse formation experiments, a commonly used in vitro synaptogenesis assay, neurexin-2 powerfully induces synapse formation. However, neurexin-2 deletions nevertheless increased the number and the release probability of synapses in cultured hippocampal neurons, indicating a synapse-restricting function consistent with the in vivo results. Rescue experiments revealed that the two phenotypes of the neurexin-2 deletion were differentially reversed by distinct combinations of neurexin-2βsplice variants, suggesting a dual restrictive mechanism of neurexin-2 in synaptic connectivity that differs from that of other neurexins and that is regulated by alternative splicing of neurexin-2.


SUPPORTING FIGURES and LEGENDS
Fig. S1 | Independent validation of the finding that the Nrxn2 deletion induces an increase in the excitatory synapse density and the frequency of mEPSCs in cultured hippocampal neurons, but has no effect on the inhibitory synapse density or the frequency of mIPSCs A & B. The Nrxn2 deletion produces an increase in the excitatory synapse density (A, representative confocal images of hippocampal cultured neurons double-immunolabelled for an excitatory synapse (vGluT1) and a dendritic (MAP2) marker; B, summary graphs of the density (left) and staining intensity (right) of vGluT1 + puncta quantified adjacent to MAP2-positive dendrites).
C & D. The Nrxn2 deletion does not alter the inhibitory synapse density (C, representative confocal images of hippocampal cultured neurons double-immunolabelled for an inhibitory synapse (vGAT) and a dendritic (MAP2) marker; D, summary graphs of the density (left) and staining intensity (right) of vGAT + puncta).
E & F. The Nrxn2 deletion causes an increase in the frequency but not amplitude of mEPSCs (E, representative mEPSC traces; F, cumulative probabiliyt plots of the interevent interval (left) or amplitude (right) of mEPSCs, with insets depicting summary graphs of the mEPSC frequency (left) or amplitude (right)). mEPSCs and mIPSCs were recorded in 1 μM TTX and 50 μM picrotoxin The Nrxn2 deletion has no effect on the frequency of amplitude of mIPSCs (G, same as E but for mIPSCs recorded in 1 μM TTX, 10 μM CNQX and 50 μM AP-5; H, same as F but for mIPSCs).
Numerical data shown are means ± SEMs; numbers of cells/sets of independent cultures analyzed are listed for each condition in the graphs. Statistical assessments were performed by the Mann-Whitney test comparing a condition to the ΔCre controls or one-way ANOVA followed by post hoc Tukey's test. (**, P<0.01, *** P<0.001).

Fig. S2 | Independent validation of the observation that the Nrxn2 deletion produces an increase in excitatory synaptic strength due to an enhanced presynaptic release probability in cultured hippocampal neurons, but has no effect on inhibitory synaptic strength
A-C. The Nrxn2 deletion increases the amplitude of evoked AMPAR-and NMDAR-EPSCs (A, representative traces (left) and summary graph of AMPAR-EPSCs (right); B & C, two independent sets of experiments monitoring NMDAR-EPSCs (left representative traces; right, NMDAR-EPSC amplitude summary graphs).

D.
The Nrxn2 deletion does not alter the amplitude of IPSCs (left, representative traces; right, summary graph of the IPSC amplitude).

E.
The elevated amplitude of NMDAR-EPSCs induced by the Nrxn2 deletion is associated with an enhancement in paired-pulse depression, suggesting an increase in release probability (left, representative traces of NMDAR-EPSCs evoked by two closely spaced stimuli; right, summary plot of the paired-pulse ratio as a function of the interstimulus interval).
Numerical data are means ± SEMs; numbers of cells/sets of independent cultures analyzed are listed for each condition. Statistical assessments were performed by the Mann-Whitney test comparing a condition to the ΔCre controls or one-way ANOVA followed by post hoc Tukey's test. (*, P<0.05; **, P<0.01).

Fig. S3 | Demonstration that Nrxn2b rescue proteins containing various combinations of SS4 and SS5 splice variants are efficiently transported to the neuronal cell surface
Representative images of Nrxn2 cKO neurons that were co-infected with lentiviruses expressing eGFP-Cre with a nuclear localization signal (green) and the various HA-tagged Nrxn2 rescue constructs containing the indicated Nrxn2b splice variants. Neurons were stained on the surface with HA antibodies and then with MAP2 after permeabilization. The green nuclear stain is due to the eGFP-Cre that contains a nuclear localization signal.