The presynaptic terminal makes two decisions about release, and two proteins make them

Plasticity research has spent forty years looking at the postsynaptic membrane — receptor number, receptor phosphorylation, spine morphology. But a synapse cannot be potentiated if nothing is released into it. The presynaptic terminal answers two separate questions on every action potential: how many vesicles are available, and whether any of them fuse. Synapsin I answers the first. Synaptotagmin 1 answers the second. Both answers are set by phosphorylation and Ca²⁺, which is why the phosphosites on synapsin are among the most direct readouts of presynaptic state available to a bench scientist.

Two proteins, two decisions

Synapsin was found before anyone knew what it did. In 1977 Ueda and Greengard purified an endogenous substrate for cAMP-dependent protein kinase from bovine cerebral cortex synaptic membranes, named it Protein I, and noted that it appeared specific to nervous tissue and resolved into two polypeptides of about 86 and 80 kDa ¹. The function came later: synapsin I binds and bundles F-actin in a phosphorylation-dependent manner, with F-actin binding distributed across three separate regions of the molecule ². That gave a mechanical picture — a phosphoprotein on the cytoplasmic face of synaptic vesicles, cross-linking them to the actin cytoskeleton and to each other.

The functional consequence of that tether was shown directly in the lamprey giant reticulospinal synapse. Vesicle clusters at release sites resolve into two pools — a distal, synapsin-containing pool and a proximal pool adjacent to the presynaptic membrane that lacks synapsin. Injecting synapsin antibodies presynaptically removed the distal pool and left the proximal pool intact, and depressed release evoked by 18–20 Hz stimulation while leaving 0.2 Hz release essentially unaffected ³. Synapsin, in other words, does not gate the first vesicle. It gates the reservoir that keeps a terminal going under load.

Mouse genetics agreed. In synapsin I-deficient mice, densely packed vesicles remained only in a narrow rim at active zones while the rest dispersed through the terminal; evoked glutamate release fell, and recovery after high-frequency depletion was greatly delayed ⁴. Knocking out all three synapsin genes left basal single-stimulus transmission at excitatory synapses unchanged but tripled the rate of synaptic depression during stimulus trains, with a matching loss of vesicles more than 150 nm from the active zone. At inhibitory synapses the phenotype inverted: basal transmission fell, depression kinetics did not change, and the deficit was in the readily releasable pool rather than the reserve ⁵. The same protein does different jobs at excitatory and inhibitory terminals.

More recent work reframes the tether as a phase. Purified synapsin forms a distinct liquid condensate in aqueous buffer, captures small lipid vesicles into it, and disassembles rapidly on phosphorylation by CaMKII — the in vitro counterpart of the synapsin dispersion seen at stimulated presynaptic sites ⁶. The reserve pool may be less a cross-linked lattice than a droplet with a phosphorylation switch on it.

Synaptotagmin 1 was characterized along a different line. Its cytoplasmic region contains two internal repeats homologous to the regulatory region of protein kinase C — the C2A and C2B domains — and recombinant cytoplasmic fragments bind phosphatidylserine. Although Drosophila and rat sequences are only 57% identical overall, 78% of residues within the two repeats are invariant across rat, human, and fly ⁷. That conservation pointed at a core function, and the knockout supplied it. Homozygous synaptotagmin 1 mutant mice die within 48 hours of birth. In cultured hippocampal neurons from those animals, the fast synchronous component of Ca²⁺-dependent release is severely reduced, while asynchronous release — spontaneous mEPSC frequency, hypertonic sucrose-evoked release, α-latrotoxin-evoked release — is unaffected ⁸. Synaptotagmin 1 is not required for fusion. It is required for fusion to happen on time.

A point mutation knocked into the endogenous locus that lowers the Ca²⁺ affinity of synaptotagmin 1 roughly twofold lowers the Ca²⁺ sensitivity of release roughly twofold, without changing spontaneous release or readily releasable pool size ⁹ — about as clean a demonstration that a protein is the sensor as this field has produced. Crystal structures of the primed pre-fusion assembly show a tripartite interface between synaptotagmin 1, the SNARE complex, and complexin, with a second synaptotagmin 1 molecule bound through a separate primary interface. Ca²⁺ binding releases that locked assembly, permits full SNARE zippering, and triggers fusion ¹⁰. Complexin activates and clamps the core machinery, and RIM proteins position primed vesicles next to the Ca²⁺ channels that feed them, so Ca²⁺ reaches synaptotagmin within a fraction of a millisecond of channel opening ¹¹.

Phosphorylation is the control layer, and the site numbering is a trap

Synapsin is a multi-kinase substrate, and the site-to-kinase assignments matter more than the site names. What is well sourced:

Ser9 (site 1), domain A. The only phosphorylation site shared by all synapsins, and a substrate of both cAMP-dependent protein kinase and CaM kinase I. Phosphorylation here dissociates synapsin from synaptic vesicles; domain A binds phospholipids, and phosphorylation inhibits that binding, making Ser9 a direct phospho-switch for vesicle association ¹². Ser9 is the cAMP-dependent protein kinase substrate site in the intact hippocampus as well ¹³.

Ser566 and Ser603, domain D — CaMKII. Ser603 is designated site 3 and is described as the pivotal CaMKII target; it is also phosphorylated by p21-activated kinases 1, 2, and 3 in a Cdc42-dependent manner, so a Ser603 signal is not by itself proof of CaMKII activity ¹⁴. Ser566 is likewise a CaMKII site, and sits in a stretch of domain D flanked by O-GlcNAc modifications that modestly alter its Kₘ ¹⁵. In vivo, Ser603 behaves as the calmodulin kinase II site ¹³.

Ser62, Ser67, and Ser549 — MAPK/ERK. MAP kinase stoichiometrically phosphorylates synapsin I at these three residues. BDNF and NGF increase phosphorylation at them in cerebrocortical neurons and PC12 cells respectively, while KCl depolarization decreases it, and MAPK-dependent phosphorylation reduces the ability of synapsin I to promote G-actin polymerization and to bundle actin filaments ¹⁶. Ser62/67 reads as the Erk1/2 substrate site in hippocampus ¹³.

Ser549 and Ser551 (sites 6 and 7) — Cdk5. Cdk5 phosphorylation of synapsin I enhances its binding to F-actin and sets the ratio between resting and recycling vesicle pools; site 7 alone is sufficient to reproduce the deletion phenotype, identifying it as the central switch for Cdk5-mediated homeostatic scaling ¹⁷.

Note that Ser549 appears in two lists. It is a MAP kinase site by in vitro stoichiometry ¹⁶ and site 6 for Cdk5 ¹⁷; that is a real convergence, not a bookkeeping error. Note also what is not stated above: the site numbers 2, 4, and 5 are in common circulation but are used inconsistently across papers and species, and we have not asserted them here. If a figure legend or a catalog entry assigns a number, check it against the residue, and report the residue.

The practical upshot for antibody work is the same as for any signaling protein. Total synapsin tells you how much presynaptic machinery is present. Phospho-synapsin tells you what the kinases have been doing to it. Neither is interpretable without the other. In one study that measured both against behavior, hippocampal synapsin I and phospho-synapsin I both tracked spontaneous alternation performance in control rats, while a thiamine-deficiency model showed a selective ~30% reduction in phospho-synapsin I ¹⁸.

The fields that buy these reagents

Epilepsy. Synapsin I-deficient mice show a markedly increased electrographic and behavioral seizure response to electrical stimulation ⁴. Synapsin triple-knockout mice are more revealing still: they transition abruptly in early adulthood from a seizure-free latent state to consistent sensory-induced seizures, with compensatory upregulation of GAD67 during the latent period — epileptogenesis with a visible before and after ¹⁹.

Neurodevelopmental disorders. A nonsense SYN1 mutation (Q555X) segregates with epilepsy and autism spectrum disorder in a large French-Canadian family, and further SYN1 mutations were found in 1.0% of French-Canadian individuals with autism and 3.5% of those with epilepsy. Most cluster in the proline-rich D domain that carries the kinase sites, and none rescued the vesicle-pool phenotype of synapsin I knockout neurons ²⁰. On the synaptotagmin side, a de novo SYT1 variant (I368T) in a child with an early-onset dyskinetic movement disorder, severe motor delay, and profound cognitive impairment localized normally to nerve terminals but slowed vesicle fusion and altered endocytosis kinetics — a dominant-negative presynaptic disease ²¹.

Neurodegeneration. Synapse number is the strongest structural correlate of cognitive decline in Alzheimer's disease. In frontal cortex biopsies from patients with mild to moderate disease, synapse counts in lamina III of area 9 correlated significantly with Mini-Mental State scores ²². In a prospective series with multivariate analysis, a model built on midfrontal and inferior parietal synapse density plus parietal plaque counts reached a correlation coefficient of 0.96 with the Mattis Dementia Rating Scale, with plaque density contributing only 26% of that strength ²³. That result is why synaptic markers, rather than amyloid burden, anchor a great deal of quantitative neuropathology.

What the antibodies actually let you do

Synapsin and synaptotagmin are the workhorse presynaptic markers because they are abundant, vesicle-associated, and give crisp punctate staining in fixed tissue and culture. The standard experiment is puncta quantification: image a defined neuropil volume, threshold, segment, and count objects within a size range, reported per unit area or per unit dendrite length. The refinement that turns a marker count into a synapse count is colocalization — pair a presynaptic marker with a postsynaptic one and count only the coincident objects. Colocalization-based detection substantially outperforms single-channel counting, and super-resolution imaging shows that puncta scored as structurally complete synapses contain significantly more scaffold protein clusters than incomplete ones ²⁴.

The limitations belong in the methods section, not the discussion. Confocal resolution is roughly 200 nm laterally and worse axially, while a cortical synapse is a few hundred nanometers across and the neuropil packs them densely. Two puncta that overlap in a confocal voxel are apposed, not necessarily in contact, and in dense tissue a fraction of colocalization events will be coincidental. Array tomography — repeated immunostaining and imaging of ultrathin serial sections — was developed in part to address this, and offers better spatial resolution, particularly along the z axis, together with depth-independent immunofluorescent staining ²⁵. Where that is not available, a second orthogonal measure helps: a Western blot of the same marker on matched tissue, a second marker pair, or an electrophysiological readout of release. A change in puncta count that survives two independent measurements is a finding; one that does not is a staining artifact until proven otherwise.

Phospho-specific antibodies of this type are typically used to read activation state rather than abundance, which imposes its own discipline: phosphatase inhibitors from the moment of collection, fixation conditions validated for the phospho-epitope specifically, and a total-protein antibody run on the same material, because a phospho-signal is only interpretable as a ratio. Lambda phosphatase treatment of a parallel sample is the cheapest specificity control available for this class of reagent.

Caveat — A drop in synapsin or synaptotagmin immunoreactivity is not by itself evidence of synapse loss. Both proteins are dynamically regulated, and synapsin in particular disperses from the terminal on stimulation ⁶ and dissociates from vesicles on Ser9 phosphorylation ¹². A terminal that has just been driven hard can look depleted of synapsin while remaining structurally intact. Distinguishing "fewer terminals" from "the same terminals in a different state" requires a structural marker that is not phosphoregulated, or a second modality.

Choosing and reporting the reagent

The five-pillar framework from the International Working Group for Antibody Validation — genetic, orthogonal, independent-antibody, tagged-protein, and immunocapture/mass-spectrometry strategies — remains the reference standard, applied in an application-specific way ²⁶. For presynaptic markers the genetic pillar is unusually accessible: synapsin knockout and synaptotagmin 1 knockout tissue both exist, and a knockout-validated image is worth more than a page of dilution tables. Report the catalog number, lot, host species, fixation and antigen retrieval, and the validation evidence you relied on.

Catalog no.	Antibody	Host
P40027	Synapsin I	Rabbit
P40030	Phospho-Synapsin I (Ser549)	Rabbit
P40032	Synaptotagmin 1	Rabbit
P40028	Phospho-Synapsin I (Ser9)	Rabbit
P40029	Phospho-Synapsin I (Ser62/67)	Rabbit
P40031	Phospho-Synapsin I (Ser603)	Rabbit

References

1. Ueda T, Greengard P. Adenosine 3':5'-monophosphate-regulated phosphoprotein system of neuronal membranes. I. Solubilization, purification, and some properties of an endogenous phosphoprotein. J Biol Chem. 1977;252:5155–5163. PMID 194903
2. Bähler M, Benfenati F, Valtorta F, et al. Characterization of synapsin I fragments produced by cysteine-specific cleavage: a study of their interactions with F-actin. J Cell Biol. 1989;108:1841–1849. doi:10.1083/jcb.108.5.1841 · PMID 2497104
3. Pieribone VA, Shupliakov O, Brodin L, et al. Distinct pools of synaptic vesicles in neurotransmitter release. Nature. 1995;375:493–497. doi:10.1038/375493a0 · PMID 7777058
4. Li L, Chin LS, Shupliakov O, et al. Impairment of synaptic vesicle clustering and of synaptic transmission, and increased seizure propensity, in synapsin I-deficient mice. Proc Natl Acad Sci USA. 1995;92:9235–9239. doi:10.1073/pnas.92.20.9235 · PMID 7568108
5. Gitler D, Takagishi Y, Feng J, et al. Different presynaptic roles of synapsins at excitatory and inhibitory synapses. J Neurosci. 2004;24:11368–11380. doi:10.1523/JNEUROSCI.3795-04.2004 · PMID 15601943
6. Milovanovic D, Wu Y, Bian X, De Camilli P. A liquid phase of synapsin and lipid vesicles. Science. 2018;361:604–607. doi:10.1126/science.aat5671 · PMID 29976799
7. Perin MS, Johnston PA, Ozcelik T, et al. Structural and functional conservation of synaptotagmin (p65) in Drosophila and humans. J Biol Chem. 1991;266:615–622. PMID 1840599
8. Geppert M, Goda Y, Hammer RE, et al. Synaptotagmin I: a major Ca2+ sensor for transmitter release at a central synapse. Cell. 1994;79:717–727. doi:10.1016/0092-8674(94)90556-8 · PMID 7954835
9. Fernández-Chacón R, Königstorfer A, Gerber SH, et al. Synaptotagmin I functions as a calcium regulator of release probability. Nature. 2001;410:41–49. doi:10.1038/35065004 · PMID 11242035
10. Zhou Q, Zhou P, Wang AL, et al. The primed SNARE-complexin-synaptotagmin complex for neuronal exocytosis. Nature. 2017;548:420–425. doi:10.1038/nature23484 · PMID 28813412
11. Südhof TC. Neurotransmitter release: the last millisecond in the life of a synaptic vesicle. Neuron. 2013;80:675–690. doi:10.1016/j.neuron.2013.10.022 · PMID 24183019
12. Hosaka M, Hammer RE, Südhof TC. A phospho-switch controls the dynamic association of synapsins with synaptic vesicles. Neuron. 1999;24:377–387. doi:10.1016/s0896-6273(00)80851-x · PMID 10571231
13. Tallent MK, Varghis N, Skorobogatko Y, et al. In vivo modulation of O-GlcNAc levels regulates hippocampal synaptic plasticity through interplay with phosphorylation. J Biol Chem. 2009;284:174–181. doi:10.1074/jbc.M807431200 · PMID 19004831
14. Sakurada K, Kato H, Nagumo H, et al. Synapsin I is phosphorylated at Ser603 by p21-activated kinases (PAKs) in vitro and in PC12 cells stimulated with bradykinin. J Biol Chem. 2002;277:45473–45479. doi:10.1074/jbc.M206673200 · PMID 12237306
15. Cole RN, Hart GW. Glycosylation sites flank phosphorylation sites on synapsin I: O-linked N-acetylglucosamine residues are localized within domains mediating synapsin I interactions. J Neurochem. 1999;73:418–428. doi:10.1046/j.1471-4159.1999.0730418.x · PMID 10386995
16. Jovanovic JN, Benfenati F, Siow YL, et al. Neurotrophins stimulate phosphorylation of synapsin I by MAP kinase and regulate synapsin I-actin interactions. Proc Natl Acad Sci USA. 1996;93:3679–3683. doi:10.1073/pnas.93.8.3679 · PMID 8622996
17. Verstegen AMJ, Tagliatti E, Lignani G, et al. Phosphorylation of synapsin I by cyclin-dependent kinase-5 sets the ratio between the resting and recycling pools of synaptic vesicles at hippocampal synapses. J Neurosci. 2014;34:7266–7280. doi:10.1523/JNEUROSCI.3973-13.2014 · PMID 24849359
18. Resende LS, Ribeiro AM, Werner D, et al. Thiamine deficiency degrades the link between spatial behavior and hippocampal synapsin I and phosphorylated synapsin I protein levels. Behav Brain Res. 2012;232:421–425. doi:10.1016/j.bbr.2012.04.004 · PMID 22507301
19. Ketzef M, Kahn J, Weissberg I, et al. Compensatory network alterations upon onset of epilepsy in synapsin triple knock-out mice. Neuroscience. 2011;189:108–122. doi:10.1016/j.neuroscience.2011.05.030 · PMID 21621590
20. Fassio A, Patry L, Congia S, et al. SYN1 loss-of-function mutations in autism and partial epilepsy cause impaired synaptic function. Hum Mol Genet. 2011;20:2297–2307. doi:10.1093/hmg/ddr122 · PMID 21441247
21. Baker K, Gordon SL, Grozeva D, et al. Identification of a human synaptotagmin-1 mutation that perturbs synaptic vesicle cycling. J Clin Invest. 2015;125:1670–1678. doi:10.1172/JCI79765 · PMID 25705886
22. DeKosky ST, Scheff SW. Synapse loss in frontal cortex biopsies in Alzheimer's disease: correlation with cognitive severity. Ann Neurol. 1990;27:457–464. doi:10.1002/ana.410270502 · PMID 2360787
23. Terry RD, Masliah E, Salmon DP, et al. Physical basis of cognitive alterations in Alzheimer's disease: synapse loss is the major correlate of cognitive impairment. Ann Neurol. 1991;30:572–580. doi:10.1002/ana.410300410 · PMID 1789684
24. Dzyubenko E, Rozenberg A, Hermann DM, Faissner A. Colocalization of synapse marker proteins evaluated by STED-microscopy reveals patterns of neuronal synapse distribution in vitro. J Neurosci Methods. 2016;273:149–159. doi:10.1016/j.jneumeth.2016.09.001 · PMID 27615741
25. Micheva KD, Smith SJ. Array tomography: a new tool for imaging the molecular architecture and ultrastructure of neural circuits. Neuron. 2007;55:25–36. doi:10.1016/j.neuron.2007.06.014 · PMID 17610815
26. Uhlen M, Bandrowski A, Carr S, et al. A proposal for validation of antibodies. Nat Methods. 2016;13:823–827. doi:10.1038/nmeth.3995 · PMID 27595404

Sourcing note. Primary literature located through PubMed; every quantitative figure was taken from the source article’s own record. Findings are reported at the grain of the study that produced them and those grains are not interchangeable. Product specifications are from the current Pel-Freez datasheets.