No effect on postsynaptic recruitment of NMDAR


Splice variant of neuroligin 1 (lacking insert in B)

Increases mostly size of dendritic spines and Presynaptic terminals


Abbreviations: GABA, y-aminobutyric acid; GAD65, glutamic acid decarboxylase-65; NMDAR, N-methyl-D-aspartate receptors; NR1/2, NMDA receptor subunit 1/2; PSD-95, postsynaptic density-95; VGLUT1, vesicular glutamate transporter 1; VGAT, vesicular GABA transporter.

when so-called silent synapses become conductive during synapse development, through activation by a long-term potentiation-like process involving calmodulin--dependent protein kinase II (CaMKII)43. Consistent with this proposal, transfection of these neurons with constitutively active CaMKII also resulted in the recruitment of AMPAR. It can therefore be concluded from these studies that neuroligin/P-neurexin-induced synapses are able to recruit NMDAR subunits, and that AMPAR subunits can be recruited subsequently depending upon the activation status of the cell which resembles developmental processes43,44.

4.2. The Neuroligin/P-Neurexin Complex and Synapse Function

Following the demonstration that neuroligin and P-neurexin are capable of triggering formation of both pre- and postsynaptic specializations, it was necessary to show that these resulting synapses are not just synaptic protein aggregates but rather represent functional synapses. This was first achieved by demonstrating that neuroligin-induced presynaptic elements in chimeric cultures were capable of undergoing synaptic vesicle turnover following depolarization in a similar manner to endogenous synapses34. Subsequently, electrophysiological measurements of these artificial synapses confirmed that the synapses were actually capable of transmission, using HEK293 cells co-transfected with neuroligin and subunits of NMDAR and AMPAR that allowed direct measurement of currents in the heterologous cells38. P-Neurexin was first shown to be essential for synapse function through the application of soluble P-neurexin in neuronal cultures, which disrupted the formation of synapsin-positive clusters at chimeric and endogenous synapses34. This study further demonstrated that the synapse-inducing effects of neuroligin were mediated through its P-neurexin partner. In another approach to disrupt the complex, neuroligin 2 was overexpressed in neurons at very high levels such that it mislocalized to the entire dendritic surface37, resulting in disrupted postsynaptic receptor clustering of PSD-95, gephyrin, and NR1, and a reduction of synaptic transmission. Finally, the role of the complex was also explored using small inhibitory RNA (siRNA) against different neuroligin isoforms, whereby a reduction in neuroligin expression resulted in disrupted synaptic transmission40.

An alternative strategy to study their role at synapses has been to employ mutated versions of P-neurexin and neuroligin. Extracellularly, the AChE domain of neuroligin 1 was shown to be necessary for both P-neurexin binding and the induction of presynaptic terminal formation17,58. In agreement with this, upon truncating the extracellular domain of neuroligin, El-Husseini and colleagues demonstrated a loss in its ability to induce presynaptic clustering in neurons41. Likewise for P-neurexin, removal of the LNS domain resulted in an impaired ability to induce postsynaptic differentiation37. The glycosylation-rich domain was also found to be important for P-neurexin function since its deletion disrupted the ability to instigate synapse formation, however, it could not induce synaptogenesis by itself, suggesting that it may be required for structural positioning of the LNS domain. The synaptogenic activity of the neuroligin/P-neurexin complex appears to result from trans-synaptic aggregation, since aggregation of neuroligin by application of P-neurexin attached to beads, or by antibodies, results in the same clustering as co-culturing37. P-Neurexin may therefore exert its effects through induced clustering of neuroligin. As described above, oligomerization of neuroligin is essential in this process16 since mutants which are still capable of binding P-neurexin, but can no longer oligomerize with each other, fail to induce presynaptic terminal formation17. In addition, loopexchange mutants of neuroligin 1 with acetylcholinesterase demonstrated that the capacity to bind P-neurexin is necessary for its ability to induce synapse formation58. Intracellularly, the PDZ-recognition sequence at the C-terminus of neuroligin is responsible for the interaction with PSD-95. Thus, truncation of the C-terminal region resulted in a failure to recruit PSD-95 to postsynaptic clusters39-41, reflecting the loss of interaction between neuroligin and PSD-95 but surprisingly did not affect the targeting of neuroligin itself to synapses45. This mutant also resulted in a loss of recruitment of AMPAR subunits and in reduced excitatory transmission, whilst inhibitory transmission was unaltered39. In addition, a mutant in which the intracellular and transmembrane domains were replaced by a GPI anchor (allowing membrane sequestering without any direct intracellular signalling) was still able to induce presynaptic specializations34, demonstrating that the extracellular domain was sufficient. However, its activity was less than for the full-length neuroligin, suggesting that there may therefore still be a need for a feedback system, or cross-communication.

Taken together, the discrepancies observed on postsynaptic differentiation are most likely due to differences in the levels of overexpression of these proteins. Despite these inconsistencies, however, these results hint at a mechanism by which induction of both neurexin and neuroligin clustering results in recruitment of additional pre- and post-synaptic proteins. Future studies on the time course of recruitment and clustering of endogenous neurexins and neuroligins at the synapse may help clarify some of these issues.


Immunohistochemical studies of neuroligin 1 in brain tissue have revealed a localization at excitatory synapses, based on its co-localization with GluR2/3 receptors, rather than GABAa receptors5. It has been proposed that neuroligin 2, in turn, is localized exclusively to inhibitory GABAergic synapses46, leading to the hypothesis that different neuroligin isoforms may determine which type of synapse is formed. Analyses of endogenous neuroligin in neuronal cultures have revealed a similar picture, i.e., neuroligin 1 is almost always associated with excitatory synapse markers37,42, whereas neuroligin 2 is predominantly associated with inhibitory synaptic markers such as gephyrin37 or vesicular GABA transporter40,42. This distribution may change, however, when cultured neurons overexpress neuroligins37,40-42, indicating that presumably all neuroligin isoforms have the potential to associate at both excitatory and inhibitory synapses under certain experimental conditions. Although this behavior makes interpretation difficult, a common picture emerges that the neuroligin/p-neurexin complex may be responsible for the ratio between excitatory and inhibitory synapses47.

A possible mechanism for determining to which type of synapses neuroligins are localized came from an investigation of PSD-95. PSD-95 co-expression with neuroligin 1 was found to enhance the size and number of postsynaptic clusters, and resulted in a change of neuroligin localization from both excitatory and inhibitory synapses to solely excitatory41. In support of this, siRNA was used to reduce PSD-95 expression within the neuron, resulting in an increase in VGAT-positive presynaptic contacts, presumably through a shift of neuroligin 1 to inhibitory synapses, indicating that the level of PSD-95 determines neuroligin 1 localization. Surprisingly, the same mechanism seems to regulate neuroligin 2 location: co-expression of PSD-95 with neuroligin 2 also promoted neuroligin 2 to associate predominantly with excitatory synapses, instead of with both excitatory and inhibitory synapses in the case of overexpressed neuroligin 2 alone42, or instead of solely inhibitory synapses in the case of endogenous neuroligin 237. In line with these data, a C-terminally truncated neuroligin 1 resulted in impaired excitatory synaptic transmission, whereas inhibitory transmission was unaffected39. Although all neuroligin isoforms contain a PDZ-domain recognition motif necessary for PSD-95 binding, it may be that neuroligin 1 has a greater affinity for PSD-95 than neuroligin 2, or alternatively, there may be other intracellular binding partners that compete with PSD-95 for the C-terminus of neuroligin. The different approaches and resultant effects of the neuroligin/p-neurexin complex on excitatory versus inhibitory synapses are summarized in Table 2.

P-Neurexin expression has also been detected in both excitatory and inhibitory neurons8. More recently, Graf et al. found that P-neurexin-expressing non-neuronal cells induce both excitatory and inhibitory postsynaptic specializations at chimeric synapses37. Moreover, this process involved the recruitment of different neuroligins, i.e., neuroligins 1 and 2, presumably to mediate the specificity of induced PSD-95 or gephyrin clusters. Antibody-induced aggregation of neuroligin 1 was sufficient to co-aggregate PSD-95 (excitatory postsynaptic clusters) but not gephyrin (inhibitory postsynaptic clusters). Conversely, direct aggregation of neuroligin 2 clustered predominantly gephyrin, although PSD-95 clustering could also be found37. Application of soluble P-neurexin, in turn, decreased the number of inhibitory synapses induced by overexpression of neuroligins 1 and 2, while increasing the number of PSD-95 co-clustering with neuroligin42, indicating that P-neurexin binding induces the aggregation of PSD-95 with both neuroligins 1 and 2.

Interfering with the neuroligin/P-neurexin complex in vitro not only alters the clustering and recruitment of binding partners but also appears to change dramatically the balance between excitatory and inhibitory synaptic transmission. Application of soluble P-neurexin resulted in an increased overall ratio of excitatory to inhibitory miniature postsynaptic currents42. Since this affected mini frequencies, and not amplitudes, the effect was most likely caused by disruption of the presynaptic machinery, rather than postsynaptic differentiation. Suppression of any one of the three neuroligins by siRNA treatment, in turn, reduced both excitatory and inhibitory synapse numbers40. In contrast, suppression of all three neuroligin isoforms together resulted in the disruption of mainly inhibitory synaptic transmission, with only a moderate effect on excitatory transmission. Since suppression of a single isoform had such a dramatic effect, it supports the idea that deletion of any of the neuroligins severely disrupts the balance between excitation and inhibition. Synaptic activity was subsequently shown to be completely restored by the addition of only one human neuroligin isoform40, suggesting that its overexpression is sufficient to restore a full and functional neuroligin population. It has to be emphasized, however, that all these studies relied on measuring asynchronous spontaneous events rather that action potential-evoked synaptic transmission. Evoked postsynaptic responses are more difficult to record in primary neuronal cultures but such recordings are mandatory before definitive conclusions can be made on the functional aspects of the synaptic contacts induced by the neuroligin/neurexin complex in vitro.


Disturbances in the ratio of excitatory and inhibitory synapses may have implications for neurodevelopmental disorders such as autism and mental retardation48. Indeed, shortly after the human neuroligin family was cloned49, revealing the existence of a fourth human neuroligin, the genes encoding neuroligins 3 and 4 were found to be present within X-linked loci (Xp22.3 and Xq13) associated with autism50. Upon screening for neuroligin mutations within families with autistic members, a frame shift mutation was identified in neuroligin 4 at position 396 (D396X), leading to a truncation, and a substitution mutation within neuroligin 3 (R451C). Both mutations were absent from unaffected family

Table 2. Differential Effects of the Neuroligin/P-Neurexin Complex on Excitatory versus Inhibitory Synapses.




In situ/in vitro localization of

Neuroligin 1

Excitatory synapses in situ in vitro


Neuroligin 2

Inhibitory synapses in situ in vitro


Localization after overexpression of

Defeat Drugs Death

Defeat Drugs Death

This Book Is One Of The Most Valuable Resources In The World When It Comes To Helpful Info On Avoiding And Beating A Fatal Drug Addiction!

Get My Free Ebook

Post a comment