Abstract
The splanchnic-chromaffin cell synapse is the site at which stimulus-secretion coupling in the adrenal medulla is regulated. However, since the discovery that acetylcholine underlies chemical signaling at this synapse, attention has been disproportionately placed on postsynaptic chromaffin cell function. As a result, the determinants of Ca2+-sensing and exocytosis from splanchnic nerves remain poorly understood. This study shows, for the first time, that a ubiquitous Ca2+-binding protein, synaptotagmin-7 (Syt7) is expressed within the neurons that innervate the adrenal medulla. In synapses that lack Syt7, evoked excitatory postsynaptic currents (EPSCs) are smaller in amplitude and decay with more rapid kinetics than wild-type synapses stimulated in an identical manner. EPSCs in Syt7-deficient adrenal slices also fail to facilitate, which is ordinarily a robust property of these synapses. These data are the first to implicate a role for Syt7 in regulating short-term synaptic plasticity in the peripheral nervous system.
Introduction
The “fight-or-flight” response encapsulates a range of physiological changes that are evolutionarily designed to prepare an individual to respond to a stressor (1, 2). This stressor may take on the form of an external, physical threat to the organism, an emotionally charged event, or even metabolic disturbances. Although the precise response to each stressor may vary, all share some basic characteristics of sympathetic activation (3). The heart rate and strength of contraction increase as blood is directed to skeletal muscles and the brain, and away from non-exigent activities such as digestion. Respiration is facilitated, and the pupils dilate to improve vision and assess the threat. Furthermore, the body mobilizes energy stores to supplement this demanding state, via the stimulation of gluconeogenesis in the liver (4, 5).
A key effector of the sympathetic nervous system and the “fight or flight” response is the adrenal medulla (3). Secretion from the adrenal medulla is dependent on input from preganglionic, sympathetic fibers which pass into the gland via the splanchnic nerves (6, 7). The fibers terminate on adrenomedullary chromaffin cells which synthesize and store catecholamines (epinephrine and norepinephrine) and vasoactive peptides in dense core granules (8). Due to their experimental accessibility, a great deal is now known about the mechanisms underlying dense core granule exocytosis from chromaffin cells (9, 10). Much of this knowledge has been derived from experiments in dissociated cells, where pharmacological and genetic manipulations are readily executed. By comparison, a paucity of information exists on the properties of splanchnic neurons, themselves. This is remarkable, considering their essential role in regulating post-synaptic chromaffin cell release, and, more broadly, in tuning blood hormone profiles to satisfy demands imposed by environmental and metabolic stressors (11–13).
The purpose of this study was to characterize the role of one synaptic protein in particular – synaptotagmin-7 (Syt7) – in regulating exocytosis at the splanchnic-chromaffin cell synapse. Syt7 belongs to a large family of proteins, numbering 17 in total, many of which couple calcium influx to vesicle fusion (14–16). Syt7 is expressed in diverse cell-types, including chromaffin cells, in which it serves as a specialized, high affinity calcium-sensor in regulated exocytosis (14, 17–22). Our interest in Syt7 at splanchnic synapses has developed as a result of recent discoveries concerning Syt7 function at central nervous system (CNS) synapses. There, Syt7 is now generally acknowledged to regulate synaptic transmission in important ways, and to be required for forms of synaptic plasticity that are driven by subtle variations in calcium levels, including asynchronous release and facilitation (18, 23–27).
Recent studies have demonstrated that the splanchnic-chromaffin cell synapse shares features common to synapses in the CNS. It exhibits forms of short- and long-term plasticity that are regulated by variations in sympathetic nervous system activity (13). Indeed, one context in which the plasticity of these synapses is important is in response to fasting-induced hypoglycemia (13, 28). Nevertheless, to date, no role has been found for any synaptotagmins in splanchnic neurons, let alone Syt7.
The experiments described in this study were performed on adrenal slices obtained from wild-type (WT) animals or animals in which the Syt7 gene had been deleted (hereafter referred to as Syt7 KO) (29). The absence of Syt7 had several readily discernible effects on splanchnic synaptic transmission, which was evoked by stimulating preganglionic input to chromaffin cells via a bipolar stimulating electrode. Specifically, evoked EPSCs in KO slices are smaller in amplitude and decayed with faster kinetics than those in WT slices. Moreover, facilitation, which is ordinarily a robust property of this synapse, was abrogated in the absence of Syt7. The magnitude of tonic or basal currents on which synchronous EPSCs ride (evoked by high frequency stimulus trains) were substantially smaller in KO synapses than in WT synapses. Despite these obvious electrophysiological differences, circulating levels of epinephrine were not different between WT and Syt7 KO animals. This suggests that under basal conditions, compensatory mechanisms must exist to sustain proper levels of catecholamine release, even in the global absence of Syt7.
In sum, we demonstrate here that Syt7 is indeed expressed within the splanchnic neurons which innervate the adrenal medulla, and plays a significant role in evoked release and in facilitation at their nerve terminals. Thus, our data provide the first indication of a role of Syt7 in regulating short-term synaptic plasticity in the peripheral nervous system.
Materials and Methods
Animals
Litters of adult male and female Syt7-/- (gift of Dr. Joel Swanson; (29)) and Syt7+/+ (from a C57BL/6J background and obtained from Jackson Labs, Bar Harbor, ME) were used in these studies (17, 29). Animals were group housed (2 to 5 per ventilated cage) with 24 hr (12/12 dark/light cycle) access to food and water. All animal procedures and experiments were conducted in accordance with the University of Michigan Institutional Animal Care and Use Committee. No randomization was performed to allocate subjects in the study
In Vitro Electrophysiology Recordings and Analysis
3-4-month-old animals (male and female; 18 to 25 g) were gas anesthetized using an isoflurane drop jar technique and sacrificed by guillotine decapitation (all procedures are in accordance with approved UM IACAC protocol PRO000009265). Chromaffin cells are responsible for releasing catecholamines in response to stress, including hypoxia. Hence, isoflurane is used to induce a faster loss of consciousness compared to CO2 euthanasia (30 s to 1 minute versus several minutes) and reduce animal stress.
Adrenal glands were then quickly removed from the kidney and placed in ice cold (4°C) slicing solution containing, in mM: 62.5 NaCl, 2.5 KCl, 1.25 KH2PO4, 26 NaHCO3, 5 MgCl2, 0.5 CaCl2, 20 Glucose and 100 Sucrose (pH maintained at 7.4 by saturation with O2/CO2, 95/5% respectively) with at an osmolarity of 315 milliosmolar. Glands were subsequently embedded in 3.5% agarose block solution at 4°C. Approximately 300 µm thick sections were cut with a microtome (VF-300, CompresstomeTM; Precisionary instruments, Natick MA). Slices were transferred to a stabilization chamber where they were maintained at room temperature for 60 min in artificial cerebrospinal fluid (ACSF) containing in mM: 125 NaCl, 2.5 KCl, 1.25 KH2PO4, 26 NaHCO3, 1 MgCl2, 2 CaCl2 and 20 Glucose, pH 7.4 (with 95% O2 and 5% CO2 bubbling through the solution, 300 mOsm). Then individual slices were transferred to the microscope to the recording chamber (300 µL volume) continuously superfused with ACSF (1-2 mL/min) at room temperature.
The adrenal gland was visualized at in the microscope (Nikon Eclipse FN-1) at X10 to determine the recording and stimulation areas. The processes of preganglionic neurons were activated using a focal stimulating FHC tungsten metal bipolar electrode (2-3 MΩ). The electrode was placed in the border between the adrenal cortex and the adrenal medulla around 50 - 100 µm away from the recording electrode. Recording micropipettes were pulled (P-97; Sutter Instruments, Novato, CA) from borosilicate glass capillaries (1.5 mm O.D.; Harvard Apparatus, Holliston, MA) for a final resistance of 3-6 MΩ. Pipettes were filled with a cesium-based internal solution of the composition in mM: 135 CsCl, 4 NaCl, 0.4 GTP, 2 Mg-ATP,0.5 CaCl2, 5 EGTA and 10 HEPES pH 7.25 – 7.3 (290 mOsm). Currents were recorded with Axon Instruments, Multiclamp 700B (Axon Instruments, Union City, CA), low pass filtered at 2 kHz. Chromaffin cells in medullary slices were identified using a Nikon Eclipse FN-1 microscope with a X40 water-immersion objective and a DAGE-MTI IR-1000 video camera. Whole-cell recordings (more than 8 GΩ before break-in) were obtained in voltage-clamp configuration, acquired at 2 kHz fixing the voltage at −20 mV. Series resistance was monitored throughout the experiment and experiments were aborted if changes greater than 20% occurred. The cells were chosen according to the access resistance and visual examination of their membranes. The EPSCs were evoked by stimulating the preganglionic input at 0.1 Hz and were distinguished by their all- or-none response to presynaptic stimulation and fast kinetics (11, 13, 30). Once a “synapse” was identified, pre-ganglionic input was stimulated at 10 – 30 mV intensity (0.5 ms pulse duration) every 15 seconds at intervals of 60, 100, 200 and 500 ms or during high frequency trains (20Hz). Stimulation waveforms were introduced via a Grass S48 stimulator (Quincy, MA) that was triggered using Clampex9 software.
Paired-pulse ratios (PPRs) of EPSCs (2 – 4 events per cell) were calculated by dividing the amplitude of the second EPSC by that of the first (PPR = EPSC2/EPSC1). For experiments involving the application of hexamethonium (hexane-1,6-bis (trimethylammonium bromide)) a non-depolarizing nicotinic receptor (nAChR) antagonist at concentration of 100µM was added to the bath, and stimulation was conducted at 0.1 Hz (unless otherwise indicated) the slice was perfused with ACSF (1-2 mL/min) in presence of the drug for 5 minutes before washout and for before 5 min to obtain a baseline response. Peak current amplitudes were searched identified manually using pCLAMP 10 (Molecular Devices, San Jose, CA), and visually monitored to exclude the erroneous noise. The current response was fit by a single exponential equation to obtain the time constant (tau) of decay. Basal current (during high frequency stimulation) is measured as the difference between the sustained currents reached during the train and the overall baseline current of the record (26). Data in all figures represent the mean ± standard error of the mean (SEM). Unless stated otherwise, statistical significance was assessed by unpaired two-tailed Student’s t-test, or one-way ANOVA followed by Tukey’s post-hoc test.
Reverse transcription and quantitative PCR
Reverse transcription was performed on mouse adrenal medullas dissected from adrenal glands and homogenized, as published in(17). Adrenal medullas from two animals in each WT and Syt7 KO group are considered as one experiment. Specifically, four adrenal medullas from two animals were homogenized in one 1.5-ml Eppendorf tube on ice for 45 seconds with motorized pestle mixer (Argos Technologies, Inc, Vernon Hills, IL). More than three experiments were performed for each target. RNeasy Mini (Qiagen, Valencia, CA) was used to isolate the RNAs. The first strand cDNA synthesis was performed with 400 ng of RNAs using the qScript cDNA SuperMix kit (Quanta Biosciences, Beverly, MA). The reverse transcription product was kept at −20°C until qPCR was performed. qPCR primers for the target genes were designed with online tools (GenScript PCR Primer Design and NCBI primer designing tool). The forward (Fw) and reverse (Rv) primer sequences are as follows (listed 5’ to 3’):
TH Fw GCGCCGGAAGCTGATTGCAG
TH Rv CCGGCAGGCATGGGTAGCAT
α3 Fw GCCGCTGTCCATGCTGATGC
α3 Rv ACAGGCGGTGCTCAGCTTCC
β4 Fw GGCCTGGAACAGCTCCTGCT
β4 Rv AGCCACTGGATGCTGCCGTT
Immunofluorescence
3-4 month-old mice (male and female) were administered a ketamine/xylazine mixture (80 mg/kg body weight ketamine and 10 mg/kg xylazine) via intraperitoneal injection. The mice were perfused by cardiac perfusion of PBS followed by 4% paraformaldehyde (PFA). Adrenal glands were immediately removed and fat and connective tissues surrounding the glands were trimmed off. Glands were then kept in 4% PFA for 2 hours and then immersed in series of sucrose solutions (10%, 15% and 30% in 1x PBS) until the tissues sink. After that, they were carefully transferred to a mold containing OCT media without any bubbles and immediately frozen using dry ice. Slices were then sectioned in a cryostat. Upon staining, adrenal gland slices were fixed with 4% PFA for 10 minutes, and washed 3x with PBS. Following fixation, slices were permeabilized using 0.2% triton X-100 in PBS for 20 minutes. Slices were then blocked with 5% BSA and 0.1% tween in TBS for one hour, before incubating with primary antibodies (rabbit anti-Syt7 (SySy, Göttingen, Germany), mouse anti-TH (Sigma Aldrich, St. Louis, MO), and goat anti-ChAT (MilliporeSigma, Burlington, MA)) at a 1:400 dilution and at 4°C overnight. On the following day, slices were washed 3x for 15 minutes each with TBST and incubated with secondary antibodies diluted in blocking buffer for 1hr at room temperature. Fluorescently conjugated secondary antibodies Alexa 488, 568, and 647 (1:250, ThermoFisher, Waltham, MA) were used. Secondary antibodies were washed three times with TBST for 15 minutes each and mounted on glass slides with Prolong Gold.
Imaging was performed using a confocal microscope (LSM880, Zeiss, Germany) with a 63x oil immersion objective in Airyscan mode. Excitation was accomplished using 405-, 488-, 561-, and 633-nm lasers. All images were further processed in Adobe Photoshop CS6 software. Plot profiles were generated with ImageJ.
Mass spectrometry
Plasma epinephrine levels were quantified using a liquid chromatography-mass spectrometry method modified from P. Song, et al. (31). In brief, 9 µL of plasma was spiked with 1 µL of 12.5 mM ascorbic acid to prevent epinephrine oxidation as well as 1 µL of 1 µM D6-ephinephrine for determination of extraction efficiency. Proteins were precipitated by adding 39 µL of cold acetonitrile and centrifuging for 10 min at 12,100 × g. 20 µL of supernatant was removed and immediately derivatized with benzoyl chloride. For the reaction, 10 µL of 100 mM sodium carbonate, 10 µL of benzoyl chloride (2% in acetonitrile, v/v), and 10 µL of sulfuric acid (1% in 4:1 water/acetonitrile, v/v) were added sequentially, briefly vortexing between each addition.2 Calibration standards were prepared from 0.1-50 nM in artificial cerebrospinal fluid to approximate plasma salt composition. Samples were analyzed in triplicate (5 µL/injection) on a Phenomenex Kinetex C18 column (100 × 2.1 mm, 1.7µm) by a Vanquish UHPLC system (ThermoFisher Scientific, Gemering, Germany) and detected on a TSQ Quantum Ultra triple quadrupole mass spectrometer (Thermo Fisher Scientific, San Jose, CA) operated in multiple-reaction monitoring mode. The flowrate was 600 µL/min. Mobile phase A consisted of 10 mM ammonium formate and 0.15% (v/v) formic acid in water. Mobile phase B was pure acetonitrile. Peak integration was performed using XCalibur 3.0 software.
Enzyme-linked immunosorbent assay (ELISA) for epinephrine
Adrenal glands were dissected from 3-4 month-old male mice and weighed. Glands were then lysed in 300 µL of 0.01 N HCl and 0.15 mM EDTA using handheld, battery-operated homogenizer. Supernatant was diluted 500-fold and 10 µL of the diluted sample was used for ELISA following the manufacturer’s protocol (Epinephrine ELISA kit KA1882, Abnova, Taipei, Taiwan). Briefly, 10 µL of samples, standard and controls were loaded onto cis-diol-specific affinity-based gel plate and extracted, acylated and then converted enzymatically. 100 µL of the extracted epinephrine was then transferred to a microtiter plate for competitive ELISA. Quantification was performed by measuring absorbance at 450 nm against the reference wavelength of 630 nm and analyzed from reference curve generated using known standard concentrations.
Western blotting
Adrenal glands were dissected from 3 month-old mice and the cortex tissue surrounding the adrenal medulla was removed carefully using fine scissors. Adrenal medulla was then homogenized in urea-based tissue lysis buffer. Equal volume of the lysate was loaded onto 4-20% tris-glycine mini gels (Novex, Invitrogen). Proteins were transferred onto a membrane and stained using Ponceau-S to check or normalize for equal protein load. The membrane was blocked with a 5% milk solution. The blot was incubated with primary antibody (anti-TH, 1:1000, Sigma Aldrich, Catalog. No. T2928) and followed by secondary antibody incubation and then developed using ECL reagent in an imager.
Electron microscopy
After anesthetizing 2-month-old mice with 2% isoflurane in 100% O2, transcardial perfusion-fixation was performed on a downdraft table equipped with a chemical waste collection system. Immediately upon cannulation of the left ventricle, perfusion was initiated with 200ml of 2% glutaraldehyde in an "artificial mammalian Ringer’s solution" with elevated buffering capacity (100mM NaCl, 30mM HEPES buffer pH 7.4 and 2mM CaCl2). Thereafter, adrenals were dissected from the fixed mice and maintained in this same aldehyde fixative until post-fixation. This was performed by first washing away the fixative with 100mM cacodylate buffer (always containing the same 2 mM Ca2+ to optimize membrane structure, in this and all subsequent steps, so this buffer will henceforth be termed “cacodylate-Ca”). After this wash, the adrenals were sliced into 500um thick sections with a homemade tissue-chopper and post-fixed for 30min with 0.25% OsO4 and 0.25% potassium ferrocyanide in cacodylate-Ca buffer, made fresh by mixing the two immediately before application. After washing away this post-fixative with more cacodylate-Ca buffer, the adrenals were then “mordanted” for 30min with 0.5% tannic acid (Polysciences mw 1700) plus 0.5% saponin (Mallinck-rodt) again in cacodylate-Ca buffer. After washing this away with more cacodylate-Ca buffer and lowering the pH to 5.2 with a brief wash in acetate buffer, the adrenals were finally “block stained” for 30min with 0.5% uranyl acetate in pH 5.2 acetate buffer. Thereafter, the UA was washed out briefly with acetate buffer and the adrenals were dehydrated in ethanol, starting with 50% ethanol in water, on up through 75% and 95%, and finally 100% ethanol for many hours or overnight. Finally the adrenal slices were embedded in Araldite 502 epoxy resin (so-called “English Araldite”) by exchanging ethanol with propylene oxide for 30min, then with a mixture of 2/3rds epoxy and 1/3rd propylene oxide for 2 hours, and then into 100 Araldite epoxy overnight (all with continuous rotation to optimize plastic-infiltration).
After a final exchange into freshly prepared Araldite epoxy, the resin was polymerized in a 70ºC vacuum-oven for 24-48 hrs. Thereafter, the embedded adrenals were sectioned with a Reichert-Jung Ultracut-E ultramicrotome using a Diatome diamond knife, first as 0.5 micron “semi-thins” for viewing in the LM after staining with 1% toluidine blue and 1% sodium borate in water for 15 sec on a hotplate, in order to choose the proper regions for examination, then further thin-sectioned at 40nm for the absolutely crispest membranes, at 90nm for best general overviews, and at 200-250 nm for obtaining 3-D information about overall nerve-orientation within the adrenals. Thin sections were retrieved onto high-transmission fine-hexagonal 200-mesh copper grids (made in England by Guilder, LTD, and sold in the US by Ladd Industries, cat. no. G200HHC), after the grids were coated with a silver-thin film of Formvar and then carbon-coated with 10 sec of evaporated carbon, for maximum specimen stability. Finally, the thin-sections were stained for 5 min on a drop of 1% lead citrate in a closed dish with NaOH pellets around to prevent CO2-precipitation of the lead, and washed with a gentle stream of water from a squirt bottle. They were examined with a standard transmission electron microscope (JEM1400 TEM) operated at 80KV and mounted with the smallest available objective aperture, for maximum contrast (and especially, for maximum removal of chromatic aberration from the quarter-micron thick sections). Images were obtained with the highest resolution digital camera available, the AMT ‘BioSprint’ 29 Megapixel Camera, which produces 6.5k × 4.5k images. The digital images thus obtained were processed with Photoshop, taking special advantage of its “high-pass” filter when very dark features happened to be located next to very light areas in the images, which made otherwise made details hard to resolve.
Results
Synaptotagmin-7 is present in preganglionic, cholinergic axons that innervate the adrenal medulla
Chromaffin cells of the adrenal medulla are innervated by preganglionic sympathetic fibers that pass into the gland by way of the splanchnic nerves (7). An image of an adrenal section containing axons and, ostensibly, nerve terminals, in addition to chromaffin cells, is shown in Figure 1A. The section has been immunolabeled with antibodies for choline acetyltransferase (ChAT) and tyrosine hydroxylase (TH) to label ACh-producing neurons and chromaffin cells, respectively. Some axons were observed to encircle chromaffin cells as they pass through the medulla (Figure 1A). Viewed at a higher magnification in Figures 1B-C, ChAT-positive fibers display occasional varicosities in which Syt7 fluorescence is also enriched, which suggests that they are forming en passant synapses with the chromaffin cells. A region in Figure 1C (dotted black line) was subjected to a line fluorescence intensity analysis, which compared the brightness of Syt7 with that of TH and ChAT. The corresponding graph in Figure 1D shows that coincident with an increase in ChAT fluorescence, Syt7 fluorescence also increases. On the other hand, TH fluorescence is highest in areas from which ChAT is excluded.
The images referenced in Figure 1 suggest that there to be a close morphological relationship between splanchnic neurons and chromaffin cells, which is similar in form to the relationship between pre- and post-synaptic cells within the brain (7). This supposition was readily verified by electron microscopy (EM), on slices of adrenal medulla from mice that were perfusion-fixed for optimal preservation of structure (see methods). Throughout all these slices, abundant examples of vesicle-laden nerve terminals were found, with almost every single chromaffin cell showing its own synaptic contact. These contacts displayed all the features that are found in synapses throughout the CNS and PNS, including typical pre- and postsynaptic densities, striking accumulations of small synaptic vesicles, both round and flat, plus variable accumulations of somewhat larger dense-cored synaptic vesicles, abundant mitochondria, etc. Representative electron micrographs of such medullary synapses are shown in Figure 2. They are observed to house abundant collections of electron-lucent, pleomorphic vesicles, in addition to some dense-cored vesicles (DCVs). Based on images obtained from 6 different glands we prepared for EM, DCVs comprise only a small proportion of the total number of vesicles in these splanchnic synapses, and their numbers vary dramatically from one synapse to the next (compare DCVs in Figure 2A vs. 2B). Likewise, the number of mitochondria varied widely from one to the next, as again can be seen by comparing Figure 2A and B. Careful vesicle size-measurements were made on these EMs and are shown in Figure 2C and D. They indicate a mean of 50nm for the electron lucent vesicles, versus a mean of 80nm for the DCVs – values that are wholly consistent with previously published electron microscopic studies (6, 7).
Synaptotagmin-7 is a calcium sensor for exocytosis in splanchnic nerve terminals
Having determined that Syt7 is present in the neurons that innervate the adrenal medulla, our next goal was to assess its role in splanchnic synaptic transmission. To do so, a bipolar tungsten electrode was placed at the border between the cortex and medulla in an adrenal section prepared as described in the Methods (Figure 3A). Pulses of between 10 – 30 mV (0.5 ms duration) were applied to stimulate splanchnic nerve endings while recording from chromaffin cells, which were voltage-clamped in the whole-cell configuration. Excitatory postsynaptic currents (EPSCs), evoked by splanchnic stimulation, were measured in glands obtained from both WT and Syt7 KO animals (Figure 3B, C). These currents are reversibly inhibited by the nicotinic receptor antagonist, hexamethonium (Figure 3C) (13). The amplitude of evoked EPSCs (Figure 3D) is significantly smaller in slices that lack Syt7 (Figure 3E) compared to WT slices, and decay with faster kinetics (Figure 3F).
Paired-pulse facilitation is eliminated in synapses lacking Syt7
Syt7 has been reported to function as a specialized calcium sensor that mediates synaptic facilitation in several types of synapses in the brain (32, 33). Accordingly, we tested whether it functions in a similar capacity in splanchnic nerve terminals. Data, presented in Figure 4A, show that facilitation is indeed a property of synapses within the adrenal medulla. This was demonstrated by applying two successive depolarizing pulses and calculating the paired-pulse ratio or PPR (the amplitude of the second EPSC divided by the amplitude of the first). Interstimulus intervals (ISIs) ranging from 60 ms to 200 ms consistently resulted in PPRs above 1 (Figure 4A). On the other hand, PPRs above 1 were never noted at synapses lacking Syt7, irrespective of the ISI (Figure 4A). To rule out that this was not a consequence of the first pulse releasing so much transmitter that the terminals were already partly exhausted, we reduced evoked release by lowering extracellular calcium from 2.0 mM to 0.5 mM (18) (Figure 4B; see also Figure S1). Even under these conditions of low release probability, facilitation was not observed in any splanchnic synapses from Syt7-deficient mice.
A reduction in the EPSCs evoked by stimulating splanchnic input to chromaffin cells could be due, in theory, to a reduction in postsynaptic nicotinic receptor expression. We previously showed that nicotinic currents in dissociated WT and Syt7 KO chromaffin cells, evoked by local perfusion of ACh, are not discernibly different, rendering such a possibility unlikely (Bendahmane et al., J Neurochem, 2020). Moreover, mRNA levels for α3 and 4 subunits – which form the channel through which the majority of synaptic current activated by ACh passes (34–36) – are essentially unchanged (Figure S2).
An asynchronous mode of release was not evident at the splanchnic-chromaffin cell synapse
Because facilitation and asynchronous release both rely on the same submicromolar Ca2+ signal, reflect a transient increase in release probability, and exhibit similar decay kinetics, it follows that they may utilize the same mechanism (14). In fact, Syt7 has been shown to mediate either facilitation (18), asynchronous release (23, 25), or both (32), in various locations in the brain. Before addressing whether Syt7 has a role in mediating asynchronous release in the adrenal gland, it was first necessary to determine whether this mode of transmission is even a feature of the splanchnic synapse. In WT adrenal slices, splanchnic input to chromaffin cells was stimulated with a single depolarizing pulse via a bipolar electrode, at voltages ranging from 30 – 150 mV. Asynchronous EPSCs, should they occur, would appear as a delayed, or slower component of the evoked current that is not time-locked to stimulus application (14, 37). No events that fit that definition were evident in multiple trials performed at splanchnic synapses (Figure S3).
High frequency trains result in facilitating EPSCs in WT but not Syt7 KO synapses
Chromaffin cells of the adrenal medulla vary dramatically in the rate at which they fire (12, 38). The variation in firing rate, in turn, may reflect the changing demands placed on them as effectors of the sympathetic stress response (39). In experimental paradigms mimicking strong sympathetic activation of chromaffin cell secretion, splanchnic input has been stimulated at rates that ranged from 10 – 20 Hz (30, 38, 40). We sought to determine whether the splanchnic synapse facilitates even at these high firing rates, or whether it might undergo depression. An exemplar current trace in which the splanchnic input is stimulated at 20 Hz for 2.5 is shown in Figure 5A. WT synapses facilitate over the time course of stimulation, with successive EPSCs being larger than the one which preceded them. This is also evident in the graph shown Figure 5C, in which normalized EPSC amplitudes from several independent trials (n = 7 WT; n = 9 Syt7 KO) are plotted against stimulus number. Syt7 KO synapses, on the other hand, fail to facilitate, even at these high stimulation frequencies (Figures 5B, C, and E).
A second notable difference between WT and KO synapses, is that the basal or tonic current, on top of which synchronous EPSCs ride, is markedly smaller in the absence of Syt7 (Figure 5D). The interpretation of the basal current is not straightforward. It has been attributed previously to an asynchronous component of release (26). However, the fact that asynchronous currents were not observed at splanchnic synapses (Figure S3) would argue against this interpretation.
Epinephrine levels in blood plasma and in the adrenal medulla are normal in the Syt7 KO
Syt7 has a role in exocytosis in the presynaptic neurons that innervate postsynaptic chromaffin cells (this study) and also in postsynaptic chromaffin cells proper (17, 41, 42). It might be reasonably inferred from these observations that basal epinephrine output ought to be disrupted animals that lack Syt7. To test the validity of this idea, plasma epinephrine was measured by LC-MS (31) in whole-body blood collected from WT and Syt7 KO animals. Surprisingly, no significant difference in plasma epinephrine concentration was noted between the two groups (Figure 6A). We next investigated the possibility that tissue epinephrine content is elevated in the Syt7 KO as a means to compensate for the overall reduction in release efficacy in the adrenomedullary system. However, no significant change in the expression level of tyrosine hydroxylase transcript (Figure S3), tyrosine hydroxylase protein (Figure S4), or epinephrine content (Figures 6B – 6D) was detected in tissue harvested from WT and Syt7 KO animals.
Discussion
The splanchnic-chromaffin cell synapse is the site at which stimulus-secretion coupling in the adrenal medulla is regulated. However, since the discovery that ACh underlies chemical signaling at this synapse (43), attention has been disproportionately placed on identifying mechanisms of postsynaptic chromaffin cell function (9). As a result, the molecular determinants of Ca2+-sensing and synaptic release at splanchnic nerve endings are largely unexplored. In this study, we show for the first time that a ubiquitous Ca2+-binding protein, synaptotagmin is expressed within the neurons that innervate the adrenal medulla. These data are the first to implicate a role for any synaptotagmin, let alone Syt7, in regulating neurotransmission at these synapses.
As is the case in the CNS, the functions of Syt7 in the periphery are closely tied to a property that sets it apart from the other Ca2+-binding members of the synaptotagmin family – its exceptionally high affinity for Ca2+ (44, 45). What situations might demand such a Ca2+ sensor? Although not measured in this study, it is known that action potentials cause brief, large Ca2+ elevations in the immediate vicinity of opened voltage-gated Ca2+ channels at synaptic terminals – termed Ca2+ nanodomains (46–48). With inactivation of Ca2+ channels, and subsequent collapse of nanodomains, a modest amount (possibly submicromolar) of residual free Ca2+ persists in the presynaptic terminal before being buffered, sequestered into intracellular organelles, or extruded (49). It is in this setting of residual Ca2+, which is insufficient to activate low affinity sensors such as Syt-1 or Syt-2, that a requirement for Syt7 is manifested.
Two closely related forms of synaptic plasticity are thought to rely on submicromolar Ca2+ – asynchronous release and facilitation (50, 51). While our experiments did not uncover evidence for asynchronous release, facilitation is clearly a robust property of the splanchnic-chromaffin cell synapse (Figure 4). And, consistent with published studies in central synapses (18, 32), facilitation is abolished in the absence of Syt7, whether it is driven by a pair of closely-spaced depolarizing pulses, or a high frequency stimulus train.
WT and Syt7 KO synapses are not only different with respect to the synchronous component of the evoked EPSCs. The basal current, on top of which synchronous EPSCs ride, was also substantially reduced in synapses that lacked Syt7. Although basal current has frequently been attributed to an asynchronous component of release (i.e., where secretion is not time-locked to the arrival of an action potential; (26)), its origins at the splanchnic-chromaffin cell synapse are not immediately obvious. No evidence was found here for delayed or slow release of neurotransmitter that persists after the end of a single action potential (see Figure S3), which in principle, could account for the basal current. Thus, basal current is more likely attributable to other sources such as non-canonical (e.g., GPCR-dependent), slow postsynaptic currents. Splanchnic neurons are also known to house and secrete a multitude of peptide cargos (12). We cannot yet account for the various ways, subtle or otherwise, in which peptidergic neurotransmission contributes to the phenomena measured here.
Overall, this study provides strong evidence that basic functions of synaptic operation depend on Syt7, including a form of synaptic plasticity termed facilitation. It has been suggested that facilitating synapses function as high pass filters, encoding information in high firing rates that greatly enhance effects on postsynaptic cells (52). However, the physiological role of facilitation has remained elusive. In the context of the sympatho-adrenal system, facilitation may have a role in amplifying epinephrine discharge from chromaffin cells during conditions that increase sympathetic tone, including hypoglycemia (13, 53). The resulting increase in circulating epinephrine would then be expected to increase blood glucose via multiple metabolic pathways (13, 53, 54). Our data show that in Syt7 KO animals fed ad libitum, epinephrine content in the adrenal medulla, as well as in the blood, are not different from WT animals (Figure 6). However, a hypothesis, which future studies should test, is that regulated physiological responses to metabolic stressors (e.g., fasting) require release driven by splanchnic Syt7. Such studies may have to wait until Syt7 expression can be abrogated solely in the periphery, and in a tissue specific-manner. In fact, tissue-specific deletion of Syt7 will be necessary to definitively disentangle its functions in controlling CNS drive of the sympathetic nervous system, from pre- and post-synaptic functions of Syt7 in the adrenal medulla. While these sorts of efforts will not be trivial, the data presented here encourage deeper investigations into the molecular mechanisms of release at these and other autonomic synapses about which very little is known.
Supplemental Information
ACKNOWLEDGEMENTS
A.A. is supported by NIH grant R01 GM111997. R.T.K. is supported by NIH grant R01 EB003320. J.M.P. has been supported by a Pharmacological Sciences Training Program (T32GM007767) fellowship and is currently supported by a Charles W. Edmunds Predoctoral Fellowship (from the University of Michigan, Department of Pharmacology).
Footnotes
Corrected author list.