Abstract
Genetically encoded voltage indicators (GEVIs) based on microbial rhodopsins utilize the voltage-sensitive fluorescence of the all-trans retinal (ATR) cofactor, while in electrochromic (eFRET) sensors, donor fluorescence drops when the rhodopsin acts as depolarization-sensitive acceptor. We systematically assessed Arch(D95N), Archon, and QuasAr, as well as the eFRET sensors MacQ-mCitrine and QuasAr-mOrange, in C. elegans. ATR-bearing rhodopsins reported on voltage changes in body wall muscles (BWMs) and the pharynx, the feeding organ, where Arch(D95N) showed ca. 125 % ΔF/F increase per 100 mV. The ATR fluorescence is very dim, however, using the retinal analog dimethylaminoretinal (DMAR), it was boosted 250-fold. eFRET sensors provided sensitivities of 45 % to 78 % ΔF/F per 100 mV, induced by BWM action potentials (APs). All sensors reported differences in muscle depolarization induced by a voltage-gated Ca2+-channel mutant. Optogenetically evoked de-or hyperpolarization of motor neurons increased or eliminated AP activity and caused a rise or drop in BWM sensor fluorescence. Last, we could analyze voltage dynamics across the entire pharynx, showing uniform depolarization but compartmentalized repolarization of anterior and posterior parts. Our work establishes all-optical, non-invasive electrophysiology in intact C. elegans.
INTRODUCTION
Activity of excitable cells like muscles and neurons can be measured by electrophysiology, Ca2+-or voltage imaging1-7. While electrical measurements provide the highest sensitivity and temporal accuracy, imaging methods are much more versatile for applications in living animals and for recording multiple cells simultaneously. Genetically encoded Ca2+ indicators (GECIs) nowadays cover most of the visible spectrum with comparably narrow spectral width, thus enabling multiplexing with other optical tools8-10. Furthermore, GECIs were improved immensely since their first reporting in the late 1990’s, to provide several thousand-fold of fluorescence increases upon Ca2+-binding5. With differential kinetic properties, some GECIs enable detection of single action potentials (APs). They are thus widely used in neuroscience, yet in many neuron types Ca2+ imaging does not reflect spiking dynamics, and does not permit to assess high frequency APs or sub-threshold voltage fluctuations11. Also, Ca2+ is an indirect measure of neuronal activity, and neuronal depolarization is not always accompanied by a Ca2+ increase12. Finally, GECIs are less-well suited to detect a hyperpolarization of cells, as there is usually no decrease of Ca2+ below the basal cytosolic concentration. Thus, voltage imaging tools (genetically encoded voltage indicators - GEVIs) are an important addition to the toolbox of sensors for excitable cell activity11,13-18.
The development of GEVIs started later than that of the GECIs, and several different designs were explored that couple detection of voltage changes to fluorescence changes. These range from FRET sensors to circularly permuted GFP, similar as for GECIs4,5. Yet, such tools for a long time reached only a few percent in fluorescence change per 100 mV of membrane voltage change, thus making detection of single APs a challenge. More recently, microbial rhodopsins were found to exhibit a voltage-dependent fluorescence of their chromophore, retinal17,19. The fluorescence signal change was higher than for the other protein-based GEVIs (ca. 20-40 % per 100 mV), and could be improved by several mutagenesis screens13,18. However, the absolute fluorescence signals of the rhodopsins are very small. In particular, the voltage sensitivity of the fluorescence, due to the nature of the process requiring more than one photon to be absorbed20, becomes appreciable only at very high light intensities. Thus, electrochromic Förster resonance energy transfer (eFRET) sensors were developed that couple fluorescence of a normal fluorescent protein (FP) to the rhodopsin. The latter is acting as a FRET acceptor upon depolarization, thus quenching the (much stronger) fluorescence of the FP, which acts as a FRET donor14,16. Several variants of eFRET sensors have been reported, which are combinations of variants of archaerhodopsin (QuasAr18, Archon13) or other proton pumps like Mac (from Leptoshaeria maculans) or Ace2N, derived from the Acetabularia rhodopsin proton pump14,21,22. Each of these proteins is coupled with specific linkers to fluorescent proteins like mOrange, mCitrine, mNeon, or mRuby3, depending on the absorption spectrum of the respective rhodopsin, to achieve maximal FRET efficiency, and to achieve imaging in different spectral regions.
An additional way to improve voltage-dependent fluorescence of the rhodopsin may be to alter the chemical properties of the retinal chromophore. We previously used retinal analogs to reconstitute function of microbial rhodopsin optogenetic tools, and to alter their characteristics23. During these studies, we observed that some retinal analogs conferred a much higher fluorescence to the rhodopsin. Here, we surveyed a range of GEVIs based on microbial rhodopsins for use in excitable cells of the nematode Caenorhabditis elegans, an important model system in physiology, and in molecular, cellular and behavioral neurobiology. We demonstrate that eFRET sensors are robust tools allowing to analyze voltage signals with little experimental effort, and that the retinal analog dimethylaminoretinal (DMAR) strongly improves absolute fluorescence signals, however, at the cost of lower fluorescence changes. We show that direct voltage-dependent retinal fluorescence can be measured in the infrared at high excitation intensity, and that membrane voltage can be set concomitantly with rhodopsin tools for depolarization or hyperpolarization. Alterations in AP amplitude and duration could be robustly detected in gain-of-function mutants affecting the L-type Ca2+ channel EGL-19 in a spontaneously active muscular pump, the pharynx, allowing also correlation with the timing of pump events. Last, GEVIs enabled to visualize the spatiotemporal compartmentalization of voltage changes in this muscular structure.
RESULTS
Archaerhodopsin Arch(D95N) equipped with ATR shows dim, fluctuating fluorescence in muscle, that is enhanced by DMAR
To explore the potential of analyzing muscle voltage changes via rhodopsin/retinal fluorescence, we expressed Arch(D95N)17, QuasAr (an improved mutated variant of Arch18) and Archon13, a further evolved version in body wall muscle (BWM) cells, or in pharyngeal muscle, and supplemented the animals with all-trans retinal (ATR). Under an epi-fluorescence microscope equipped with a 100W HBO lamp for excitation and a sCMOS camera, very dim fluorescence could be observed in the range of 700 nm. Using a 637 nm laser, and an EMCCD camera, fluorescence was more readily observable (Fig. 1A, B).
In comparison to Arch(D95N), the fluorescence of QuasAr and Archon was 2.4 and 2.1-fold more intense, respectively, and showed a 1.7-fold higher contrast over background fluorescence (in BWMs; Fig. 1D, E). In immobilized animals, this fluorescence showed intensity fluctuations in the range of 20-25 % ΔF/F that likely represent voltage changes (Supp. Fig. 1), possibly corresponding to APs as reported earlier for C. elegans BWM cells24,25.
When we supplemented the animals with the retinal analog DMAR23 (Fig. 1D, inset), we observed a ca. 250-fold higher absolute fluorescence (in pharyngeal muscle; Fig. 1B, D), and similarly, when we used retinal analog VI, fluorescence levels increased 81-fold (gain-corrected; note this was in Arch wt, in which this analog does not mediate function, and thus does not hyperpolarize the cell; analog VI could not be incorporated by Arch(D95N)). This was accompanied by higher signal contrast (138-fold and 44-fold higher for DMAR and analog VI, respectively; Fig. 1E). The fluorescence yield of Arch(D95N) could also be increased by using 10-fold higher ATR concentration during animal cultivation, and a 50x higher camera gain during imaging (Fig. 1B, D), resulting in fluorescence grey values that were similar to those observed for DMAR. Increased fluorescence due to DMAR or increased ATR and gain was also observed in BWM cells (Fig. 1B). Last, we also expressed eFRET sensors, QuasAr-mOrange and MacQ-mCitrine, in pharyngeal muscle or BWM (Fig. 1C). Since here, fluorescence of canonical FPs is observed, their absolute fluorescence was much higher than for any of the rhodopsin-only sensors (ca. 3000-fold and 800-fold higher fluorescence than Arch(D95N) equipped with ATR, for QuasAr-mOrange and MacQ-mCitrine, respectively; Fig. 1D), and likewise, their contrast level compared to background was much higher (460-and 211-fold higher, respectively; Fig. 1E).
Intrinsic activities in the pharynx and neuromuscular systems are accompanied by fluorescence changes of rhodopsin GEVIs
To assess if the fluorescence of the sensors is also voltage-dependent, we observed it during intrinsic activity of the muscular systems. For the pharynx, we immobilized animals such that they would still exhibit pharyngeal pumping, when pre-treated with serotonin (to mimic the presence of food). As exemplified for QuasAr-mOrange, fluorescence changes are apparent upon pharyngeal muscle contraction as a drop in the fluorescence that, importantly, was only observed when the protein was supplemented with ATR, thus yielding a functional eFRET acceptor (Fig. 2A, B, Supp. Video 1; the fluorescence of the terminal bulb (TB; grinder region) was used to visualize pharyngeal contractions as the TB lumen opens up).
Supplementary Video 1: QuasAr-mOrange in pharynx, spontaneous pumping.
For Arch(D95N) in BWM cells, we observed fluorescence fluctuations that occurred at a rapid time scale (Fig. 2C), reminiscent of BWM APs appearing spontaneously in dissected animals, that can be recorded by current clamp electrophysiology (Fig. 2E). To verify that the fluorescence fluctuations were in response to muscular depolarization, we analyzed the muscle arms of QuasAr expressing BWM cells in the head (Fig. 2D; Supp. Video 2). Here, we observed fluorescence fluctuations in a dorso-ventrally reciprocal fashion, as would be expected for the control of muscles during undulatory locomotion of the animal. As we and others showed previously, such reciprocal activity is apparent in immobilized animals26, even though at a slower time scale than expected for freely moving animals. We observed fluctuations of about 10-20 %ΔF/F for all the rhodopsin-based voltage sensors tested (Arch(D95N), Archon, QuasAr; Fig. 2F, G, H; Supp. Fig. 1; Supp. Videos 3, 4). Importantly, when we expressed and co-activated the natural anion-selective channelrhodopsin ACR227,28 in cholinergic motor neurons (MNs), fluorescence fluctuations in BWM cells expressing QuasAr were absent (Fig. 2I; Supp. Fig. 1D), verifying that the fluctuations were evoked by presynaptic cholinergic MNs that depolarized BWM cells via acetylcholine (ACh) signaling.
Supplementary Video 2: QuasAr in muscle arms, dorso-ventrally alternating intrinsic activity.
Supplementary Video 3: QuasAr in BWMs, intrinsic activity.
Supplementary Video 4: QuasAr in BWMs, intrinsic activity.
ChR2-mediated depolarization and ACR2-mediated hyperpolarization of MNs enables calibration of rhodopsin voltage signals in BWMs
We sought to estimate how voltage-induced fluorescence changes compare to electrical depolarization voltages. The APs of BWM cells can be evoked by stimulated release of ACh from MNs. When recorded by current clamp, ChR2 stimulation evoked mean APs of 29.15 mV (n=8; Figs. 2E; 3A, B, I). Given the enhanced imaging possibilities using laser excitation, we tested the possibility to achieve imaging in BWM cells concomitant with photo-depolarization and -hyperpolarization of pre-synaptic MNs. To this end, we expressed ChR2 or ACR2 in cholinergic MNs, as well as three different voltage sensors in BWMs (Arch(D95N), QuasAr, and Archon). Fluorescence for all three proteins was well visible along the BWM plasma membrane of the respective transgenic animals, following 637 nm laser excitation (Fig. 1A). Imaging basal fluorescence signals (i.e. without actuator-mediated stimulation) revealed spontaneous brief increases of fluorescence of ≈15 % ΔF/F for QuasAr, and ≈10 % for Archon and for Arch(D95N) (Fig. 2F-H; Supp. Fig. 1). These lasted for ca. 50-100 ms, and were reminiscent of APs measured by electrophysiology in BWM cells (Fig. 2E). During prolonged recordings (25 s), QuasAr and Archon showed robust signals that showed little alteration in baseline, while Arch(D95N) exhibited photo bleaching (Supp. Fig. 1A, D, H).
When we used ChR2 stimulation of MNs, but not in animals lacking ChR2, mean evoked increased ΔF/F signals, corresponding to depolarization, were observed for all sensors: ≈22.7 % for QuasAr and Arch(D95N), and ≈13.3 % for Archon, for the first evoked peak (Fig. 3C-F, I; for Archon, see Supp. Fig. 1C, H, I). Interestingly, trains of APs could be observed that were overlaid on the overall increased fluorescence rise. When averaged over the 5 s stimulus period, signals were ≈21.4 % for QuasAr, ≈14.6 % for Archon, and ≈17% for Arch(D95N). When we used ACR2 for hyperpolarization of cholinergic MNs, the fluorescence of QuasAr dropped (repeatedly) by about 14.6 %, and no AP-like spikes, overlaid on the fluorescence trace, were observable (Figs. 2I, 3G-I; Supp. Fig. 1F-I). These measurements allow a rough calibration of the observed fluorescence voltage signals, namely ca. 78 % ΔF/F per 100 mV for QuasAr and Arch(D95N), ca. 46 % ΔF/F per 100 mV for Archon. This data is in a similar range as reported in the literature (ca. 45% ΔF/F per 100 mV for Arch(D95N)17, QuasAr18 and Archon13), though it appears as if the first two sensors perform better in C. elegans tissue.
Arch(D95N) shows robust fluorescence increases in pharyngeal muscle
Next, we assessed voltage-dependent changes in rhodopsin fluorescence in the pharynx. This muscular pump, used for feeding, exhibits APs and contractions. Arch(D95N), despite showing dim fluorescence, displayed strong fluorescence changes upon pharynx activity, on average ca. 122 % ΔF/F (measured across the entire organ; Fig. 4A, D; Supp. Video 5). The pharynx AP was previously reported to be between 80 and 115 mV 29-31 (Fig. 4C), on average 97 mV. Thus, in the pharynx, a 100 mV depolarization corresponds to ca. 126 % fluorescence increase, which is larger than reported for Arch(D95N) or QuasAr in mammalian cells and neurons17,18, and also largely exceeds the fluorescence increase observed in BWM cells upon excitation (Fig. 3C, D, I), for unknown reasons. The signal to noise ratio (SNR) was also very good (ca. 70; Fig. 4D). However, detecting Arch(D95N) fluorescence in BWM cells requires a strong excitation laser (we achieved up to 1800 mW/mm2 at 637 nm; in the pharynx, ca. 1/10th of this intensity was sufficient) and an EMCCD camera, i.e. equipment not available in every lab. We thus also assessed the potential of DMAR-supplemented Arch(D95N) for voltage measurement. The raw fluorescence was largely increased (Fig. 1D) and could be visualized with a common sCMOS camera and excitation through an HBO lamp (30 mW/mm2, 620 nm). This excitation, however, did not allow recording APs well. Possibly, also Arch(D95N)-DMAR fluorescence must be highly excited to become voltage sensitive. Thus these measurements were performed with laser excitation (637 nm, ca. 180 mW/mm2, and an EMCCD camera). Arch(D95N)-DMAR exhibited activity-dependent fluorescence increases of about ca. 27 % (i.e. ca. 28 % / 100 mV), much smaller than for ATR-supplemented Arch(D95N), with a SNR of ca. 35 (Fig. 4A, D; Supp. Video 6). Signals were very uniform within a pump train in the same animal, and varied more between animals (Supp. Fig. 2A, B). Thus, while the signal change per mV is ca. 5-fold lower for DMAR than for ATR, the much increased detectability of the fluorescence makes this a valid alternative. We also tested analog VI in Arch wt voltage imaging. However, the fluorescence of this retinal analog did not show any voltage dependence (data not shown).
Supplementary Video 5: Arch(D95N)-ATR in pharynx, spontaneous pumping.
Supplementary Video 6: Arch(D95N)-DMAR in pharynx, spontaneous pumping.
Electrochromic FRET sensors expressed in pharyngeal muscle report on voltage and enable concomitant analysis of muscle contraction
Last, we analyzed the eFRET sensors QuasAr-mOrange and MacQ-mCitrine in pharyngeal muscle cells during spontaneous pumping. For quantification we drew a region of interest (ROI) encircling the entire pharynx. A ROI outside the pharynx, but within the head region of the animal was analyzed for background signal. Fluorescence of both sensors could be readily visualized with standard epifluorescence microscope excitation intensities, as obtained from an HBO lamp or typical LEDs. MacQ-mCitrine and QuasAr-mOrange showed ca. −25 and ca. −10 % reductions in the fluorescence upon pharyngeal pumping, respectively (Fig. 4B; Supp. Fig. 2C-F; Supp. Videos 1 and 7). SNR was comparable for both sensors (ca. 50; Fig. 4D). Thus, despite lower overall fluorescence (Fig. 1D), MacQ-mCitrine is the more sensitive voltage reporter. The signals for both sensors were quite uniform in the same animal. However, mean signal traces between animals showed a range of amplitudes (Supp. Fig. 2C-D). Based on the reported pharyngeal AP, the signals of MacQ-mCitrine and QuasAr-mOrange correspond to a change in fluorescence of ≈26 and ≈10.4 % ΔF/F per 100 mV. For QuasAr-mOrange expressed in BWMs, we could more directly ‘calibrate’ this by assaying QuasAr-mCitrine signals observed with ChR2 stimulation in MNs, and comparing this to similarly illuminated animals lacking ChR2 expression (Fig. 4E): Although the sensor excitation light leads to a minor stimulation of ChR2, a blue-light pulse caused different fluorescence increases in the two strains (ca. 10 % smaller for animals expressing ChR2). As evoked APs are 29 mV, this corresponds to ca. −34 % ΔF/F0 per 100 mV. Thus, QuasAr-mOrange seems to be more sensitive in BWMs than in the pharynx.
Supplementary Video 7: MacQ-mCitrine in pharynx, spontaneous pumping.
Analyzing a small ROI enclosing the pharyngeal TB grinder region allowed deriving the opening of the pharynx from the fluorescence signal (Figs. 2A, B; 4B). Thus, we could correlate depolarization and the contraction of the muscle. We wrote scripts to systematically analyze the measured fluorescence traces. These involved automatic detection of relevant events, aligning, synchronizing, and averaging them. Then, parameters like amplitude, area under the peak, AP and pump duration, both defined as rise from and return to the baseline, or by using the full-width at half-maximum (FWHM), delay of pump vs. voltage, rise and drop of the voltage as τon and τon values, could be derived (Fig. 5A). This way, voltage and pump signals could be analyzed and compared for Arch(D95N) equipped with ATR and DMAR, MacQ-mCitrine, and QuasAr-mOrange (Fig. 5B-H; 8-15 animals with 2-10 APs, each).
Pumps could not be deduced for Arch(D95N)-ATR as the basal fluorescence was too dim to detect any decrease due to TB opening. A comparison between signals observed by the tested sensor-retinal combinations showed, as deduced from manual analysis, different amplitudes of the pharyngeal AP (132, 22.6, −23.5 and −11.6 % ΔF/F0 for Arch(D95N)-ATR, Arch(D95N)-DMAR, MacQ-mCitrine, and QuasAr-mOrange, respectively), while the duration of the AP was very similar (263.5, 283, 238 and 281 ms, respectively, no significant differences). For Arch(D95N)-DMAR, MacQ-mCitrine and QuasAr-mOrange, we could also compare other parameters among the sensors: The pump duration was 255, 264, and 304 ms, respectively (no significant difference); the delay from beginning of the voltage and pump FWHM was 85, 103 and 98 ms, respectively (no significant difference); voltage rise (ca. 15 ms) and decay (ca. 35 ms), as well as pump rise (erroneous for Arch(D95N), 23-39 ms for the eFRET sensors) and decay time constants (19-39 ms; Supp. Fig. 3) were not significantly different between the sensors. This demonstrates a tight coupling of voltage and contraction kinetics, observable with all of the sensors, and allows correlating voltage changes and physical contraction of this muscular structure. It should also allow to characterize aberrations of these times in mutants affecting pharyngeal muscle physiology, in ways not easily accessible by electrophysiology, as the movement of the muscles hinders precise intracellular recordings.
‘Gain-of-function’ mutants in the L-type voltage gated Ca2+ channel EGL-19 exhibit reduced voltage increases, but prolonged pump duration
To see if the rhodopsin voltage sensors can be used to quantitatively compare mutants affecting APs, we focused on the L-type voltage gated Ca2+ channel (VGCC) EGL-19. This channel carries currents that shape the plateau phase of the pharyngeal AP32,33. Loss-of-function mutants are lethal, however, several ‘gain-of-function’ (g.o.f.) alleles have been isolated and characterized, e.g. by their effects on muscular tone (EGL-19 is also expressed in BWM cells). We analyzed two g.o.f. alleles, n2368 and n582ad952. These affect amino acids close to the pore domain of the 2nd module of the VGCC (n2368 → G365R), or a double mutation in the pore of the 1st and the voltage sensor of the 3rd module (n582ad952 → S372L, R899H). We compared 8-15 animals per genotype, averaging 2-10 APs per animal. Imaging pharynx pumping with either Arch(D95N)-ATR or -DMAR or the two eFRET voltage sensors showed a significantly reduced voltage amplitude signal for n2368 (to 52-65 % of the wt; Fig. 5B-H), and a reduced signal (not significant) in the n582ad952 allele. For the MacQ-mCitrine sensor, both alleles showed a significantly prolonged AP and pump duration (Fig. 5E, F; pump duration was also increased when measured by QuasAr-mOrange, Fig. 5G, H). Thus, the term ‘g.o.f.’ is mostly valid by the consequence of these alleles on BWM contraction properties (i.e. prolonged pump), but not for the AP (which in BWMs is likewise reduced in amplitude, though also prolonged24,25). Therefore, these alleles are hypomorphic rather than g.o.f., emphasizing that voltage imaging is a valuable extension of current methods for characterization of mutations in ion channels, affecting excitable cells. In particular, imaging is a more easily applied alternative to electrophysiology, which is experimentally demanding and not available in many labs.
It was previously shown that the dihydropyridine analog nemadipine A (NemA) is a potent inhibitor of EGL-19, or may accelerate its desensitization34,35. We analyzed whether nemadipine could influence the voltage-dependent fluorescence signals of the GEVIs tested in the pharynx. We compared the data from animals incubated in NemA to animals incubated in just the DMSO-containing vehicle (Supp. Fig. 4). As expected, NemA decreased wild type (wt) voltage signals, and essentially restored the egl-19 g.o.f. alleles to DMSO-control levels in wt. This was true for all sensors tested (Arch(D95N)-DMAR, QuasAr-mOrange, and MacQ-mCitrine), although to different degrees.
Imaging spatially compartmentalized voltage signals in excitable cell ensembles
Voltage recording can be done by electrophysiology with high accuracy. Yet, in electrically coupled cell systems, it is not possible to assess potentially compartmentalized electrical properties of the cell ensemble. This is expected to be the case for the pharynx, where anterior and posterior portions act differently, as was suggested for the different events visible in electropharyngeogram (EPG), i.e. the extracellular recording of pharyngeal electrical activities36-38. EPGs are current recordings with a complex, yet stereotypic structure (Fig. 6C, inset). They include currents corresponding to pharyngeal muscle depolarization (‘outward’ currents, as they correspond to current flowing from the extracellular space into the muscle) and repolarization events (i.e. ‘inward currents’). Also sometimes observed are currents corresponding to the activity of inhibitory neurons that terminate the pharyngeal AP; these show up as small inward currents, and may thus correspond to negative charge rushing into the muscle. Such events are not spatially resolvable by EPG recordings, and have not been addressed by sharp electrode recordings from different sections of the pharynx (only the TB has been recorded from)29-32,39. We thus explored if the imaging approach allows visualizing compartmentalized activity in the pharynx. To this end, we assessed a MacQ-mCitrine recording of a pump train with uniform single events (Supp. Video 7). To focus on changes in voltage, rather than absolute signals, we calculated difference videos, by deducing the value of every pixel from the value of the same pixel in the preceding video frame. After temporally aligning and averaging of twelve events, we could clearly observe depolarization and repolarization events (Fig. 6A; Supp. Video 8). The spatiotemporal differences in voltage change could be timed to single video frames and assessed in a kymograph along the longitudinal axis of the pharynx (Fig. 6A, B). The entire pharynx synchronously depolarized, while repolarization, following ca. 144 ms after the depolarization onset, occurred first in the anterior portion of the pharynx (corpus), and was followed 50 ms later by repolarization in the isthmus and TB. In the region connecting corpus and isthmus was a small section showing an additional depolarization between anterior and posterior repolarization phases (at 161 ms), and an additional repolarization phase following repolarization of the TB and isthmus. This region could match the connection between pm5 and pm4 muscle cells. Also, here are multiple inputs from pharyngeal neurons, thus possibly the minor signals occur from neuronal input to the pm5 muscle. It will be interesting to visualize this region in more detail, ideally with markers for pharyngeal neuron cell types.
Supplementary Video 8: MacQ-mCitrine in pharynx, spontaneous pumping, difference video, 12 events aligned, averaged, 10x slowed.
The EPG measures currents from the extracellular medium into pharyngeal muscle. It thus represents changes in muscle polarization. We thus wondered if the mean voltage difference events, averaged across the whole pharynx, would correspond to the EPG. When we calculated the inverse mean signal change of the entire pharynx, it matched the time course of a typical EPG recording well (Fig. 6C, see inset for a typical EPG record). A single, major depolarization spike (corresponding to ‘E’ or ‘E2’) was followed by two repolarization spikes, corresponding to the anterior and posterior parts of the pharynx (‘R1’ and ‘R2’)36,38-41. Thus, the optical measurement matches the EPG measurement, but outperforms it with respect to spatial information.
DISCUSSION
In this work, we surveyed a number of microbial rhodopsin-based optical voltage sensors in different muscular tissues of C. elegans. These include the rhodopsin variants Arch(D95N)(equipped with ATR and the retinal analog DMAR), QuasAr, and Archon, as well as the eFRET sensors MacQ-mCitrine and QuasAr-mOrange. We show that all of them can be used to detect action potentials with robust fluorescence changes of up to 126 % per 100 mV in the pharynx, and 78 % in body wall muscles. These values were larger than previously reported for cells and neurons of other organisms; it may have to do with the lipid composition of the C. elegans cell membranes, which might influence the properties of the Arch protein in a beneficial manner. Making these high ΔF/F0 values accessible, however, requires the use of a high-power excitation laser (up to 1800 mW/mm2 637 nm light), that are achieved only in a small field of view, and an EMCCD camera, to enable recording at high frame rate despite the low absolute fluorescence. We note that the addition of ATR to the culture medium caused increased background fluorescence upon blue light stimulation, thus low concentrations of ATR are recommended. Lower excitation light intensities (180 mW/mm2) were possible by using DMAR as an alternative to ATR. This strongly boosted the absolute fluorescence levels, yet reduced the fluorescence change per voltage change by a factor of 5. Possibly, a similar branched photocycle as in Arch(D95N)-ATR20 is present with DMAR, requiring more than one photon to be absorbed. Nonetheless, it was feasible to use this sensor/retinal analog combination to robustly measure parameters of pharyngeal pumping that were comparable to those measured with the eFRET sensors, and to compare VGCC g.o.f. mutants to wt.
The MacQ-mCitrine eFRET sensor, despite 75 % lower absolute fluorescence intensities, showed ca. 3-fold higher ΔF/F0 values than QuasAr-mOrange. Parameters of pharyngeal APs and pumping were similar for all sensors, emphasizing that they did not alter the properties of the pharynx relative to each other. Electrophysiological measurements are demanding and do not allow to analyze muscular movement and contraction parameters concomitantly, while the imaging approach shown here does. Since the parameters we compared between wt and egl-19 g.o.f. alleles did not differ between the three sensors, it can be expected that also other mutations affecting pharyngeal physiology and pumping will be accessible using the methods described here. Furthermore, we showed that nemadipine A, a VGCC blocker, similarly affected - actually restored - pharyngeal AP characteristics in egl-19 g.o.f. mutants, as shown earlier by electrophysiology34. The signals we determined for pharyngeal voltage fluorescence fluctuations were quite comparable in pump trains of single animals, but differed across comparisons between different animals. This may be due to the differences in immobilization of each individual, and this may be improved by more reproducible immobilization conditions as in a microfluidic device.
The imaging approach also allows following fluorescence fluctuations in a spatiotemporally defined manner across an ensemble of electrically coupled cells. This confirmed earlier assumptions made from EPG recordings, i.e. that the terminal bulb repolarization is delayed relative to the corpus repolarization. Yet, this was never simultaneously measured by sharp electrodes in the two parts of the pharynx, emphasizing the new possibilities provided by the optical measurement. Here, we observed not only the functional electrical compartmentalization of the two major parts of the pharynx, but also observed smaller sub-compartments at the interface of these two parts of the pharynx with potentially distinct electrical activities during distinct phases of the pharyngeal excitation-repolarization cycle. This could be further assessed, e.g. in gap junction mutants or cell-specific knock-down animals. Evaluating voltage change rates across the pharynx allowed to generate an ‘optical EPG’ that recapitulated most of the characteristic features of the classical EPG.
The combination of voltage imaging in the infrared spectral range and blue-light stimulation of ChR2 in excitable cells enables all-optical electrophysiology approaches for C. elegans, and will greatly facilitate such measurements without the need for complicated electrophysiological approaches42. This should enable to analyze mutants affecting neuromuscular junction function that are too small to allow dissection needed for patch-clamp electrophysiology. Additionally, the fact that voltage imaging is performed in intact animals is a distinct advantage over measurements in dissected animals, where the composition of the extracellular and pipette solutions may alter the endogenous physiology. Future work will use the methods established here for C. elegans neurons, which are even more complicated to address by electrophysiology. In sum, we compared a range of useful tools for all-optical interrogation of muscular tissues and ensembles, in response to intrinsic activities, as well as optogenetically evoked neurotransmission in the nematode.
METHODS
C. elegans culture and transgenic animals
Worms were cultivated on nematode growth medium (NGM) plates, seeded with E. coli OP50-1 strain, in 6 cm petri dishes, as described previously42. Generation of transgenic animals was performed by microinjection of plasmid DNA, at varying concentrations43. For transgenic strains ZX1917, ZX1918 and ZX1958, 3 ng/µL of the respective voltage sensor plasmid DNA was injected into the gonads of mother animals. For strains ZX1920, ZX1953, ZX2476, ZX2479 and ZX2483, 10 ng/µL of the respective voltage sensor plasmid DNA were used in combination with either 1.5 ng/µL of the co-expression marker pmyo-2::CFP or 30 ng/µL of pelt-2::GFP or pmyo-3::CFP. Additionally, for strain ZX2483 50 ng/µL of punc-17::ACR2::eYFP were injected.
Transgenic C. elegans strains
The following strains were used or generated: DA95: egl-19(n582ad952), MT6129: egl-19(n2368), ZX1907: zxEx922[pmyo2::Arch::2xMyc; pmyo-3::CFP], ZX1917: zxEx942[pmyo-2::QuasAr::mOrange; pmyo-3::CFP], ZX1918: zxEx943[pmyo-2::MacQ::mCitrine; pmyo-3::mCherry], ZX1920: zxIs121[pmyo-3::QuasAr::mOrange; pmyo-3::CFP], ZX1951: egl-19(n2368); zxEx942[pmyo-2::QuasAr::mOrange; pmyo-3::CFP], ZX1952: egl-19(n2368); zxEx942[pmyo-2::MacQ-mCitrine; pmyo-3::mCherry], ZX1953: zxIs120[pmyo-3::ArchD95N::2xmyc-tag; pmyo-3::CFP], ZX1954: zxIs6; zxIs120[pmyo-3::ArchD95N::2xmyc-tag; pmyo-3::CFP], ZX1955: egl-19(n582ad952); zxEx942[pmyo-2::QuasAr::mOrange; pmyo-3::CFP], ZX1956: egl-19(n582ad952); zxEx942[pmyo-2::MacQ-mCitrine; pmyo-3::mCherry], ZX1958: zxEx944[pmyo-2::ArchD95N::2xmyc-tag; pmyo3::CFP], ZX1959: egl-19(n582ad952); zxEx944[pmyo-2::ArchD95N; pmyo-3::CFP], ZX1960: zxIs5[punc-17::ChR2(H134R); lin-15+]; zxIs121[pmyo-3::QuasAr::mOrange; pmyo-3::CFP], ZX2476: zxEx1139[pmyo-3::QuasAr; pmyo-2::CFP], ZX2479: zxIs5[punc-17::ChR2(H134R); lin-15+]; zxEx1142[pmyo-3::Archon::GFP; pmyo-2::CFP], ZX2482: zxIs5[punc-17::ChR2(H134R); lin-15+]; zxEx1145[pmyo-3::QuasAr; pmyo-2::CFP], ZX2483: zxEx1146[punc-17::ACR2::eYFP; pmyo-3::QuasAr; pelt-2::GFP].
Molecular biology
Plasmid pAB4 (punc-17::ACR2::eYFP) was already described previously (Ref. 27).
Plasmid pAB16 (pmyo-3::QuasAr) was generated via Gibson Assembly based on pmyo-3::QuasAr::mOrange and pPD96.52 (pmyo-3, Fire Lab Vector Kit, Addgene plasmid #1608), using restriction enzymes BamHI and XbaI and primers QUASAR_fwd (5’-cccacgaccactagatccatATGGTAAGTATCGCTCTGCAG-3’) and QUASAR_rev (5’-gtcctttggccaatcccgggCTCGGTGTCGCCCAGAATAG-3’). Plasmid pAB19 (pmyo-3::Archon::GFP) was generated via subcloning from rig-3::wArchon1::GFP (a gift from Steven Flavell) into pPD96.52, using restriction enzymes KpnI und SacI.
Plasmids pNH13 (pmyo-2::QuasAr::mOrange) and pNH12 (pmyo-2::MacQ::mCitrine) were generated by subcloning of plasmids #59173 and #48762 (Addgene) into pPD132.102 (pmyo-2, #1662, Addgene) via restriction with BamHI and EcoRI. A site-directed mutagenesis of plasmid pNH10 (pmyo-2::Arch::2xMycTag) resulted in plasmid pNH11 (pmyo-2::Arch(D95N)::2xMycTag), using primers oNH7 (5’TATGCCAGGTACGCCAACTGGCTGTTTACCACC3’) and oNH8 (5’-GGTGGTAAACAGCCAGTCGGCGTACCTGGCATA-3’).
Voltage imaging of immobilized animals
One day prior to the experiments, transgenic animals at L4 stage were transferred onto NGM plates seeded with OP50-1 lawns, grown from bacterial suspensions supplemented with ATR (Sigma-Aldrich, USA) or the respective ATR-analogues DMAR or VI (final ATR concentration: 0.01 mM for experiments with ChR2 or ACR2 blue-light stimulation, 1 mM for imaging experiments or Arch(D95N) in the pharynx, and 0.1 mM for imaging eFRET sensors); note that for analog VI we used Arch wt, an HBO lamp (30 mW/mm2 excitation at 620 nm; analog DMAR could be visualized well with HBO illumination, however, APs were observed clearly only with 637 nm laser excitation, 180 mW/mm2). For imaging, worms were placed onto 10 % agarose pads (dissolved in M9 buffer) on microscope slides and immobilized with polystyrene beads (0.1 µm diameter, at 2.5% w/v, Sigma-Aldrich). Imaging was performed on an inverted microscope (Zeiss Axio Observer Z1), equipped with an 40x oil immersion objective (Zeiss EC Plan-NEOFLUAR 40x/ N.A. 1.3, Oil DIC ∞ / 0.17), a 470 nm LED light source (KSL 70, Rapp OptoElectronic, Hamburg, Germany) or monochromator (PolychromeV, Till Photonics) for photostimulation, a 637 nm laser (OBIS FP 637LX, Coherent), a galilean beam expander (BE02-05-A, Thorlabs) for excitation of voltage sensors, and an EMCCD Camera (Evolve 512 Delta, Photometrics). Voltage sensor fluorescence was imaged at 637 nm excitation (1800 mW/mm2 for direct imaging of QuasAr and Archon in BWMs and ca. 180 mW/mm2 for pharynx measurements and direct imaging of Arch(D95N)-DMAR in BWMs) and 700 or 747 nm emission (700/75 ET Bandpass filter, integrated in Cy5 filter cube, AHF Analysentechnik) for ATR, and at 780 nm (780 LP ET Longpass Filter) for DMAR. eFRET sensors were imaged with 545/30 or 472/30 nm excitation (HBO100 100W mercury lamp, Zeiss, at 30 mW/mm2 for 545/30 nm and 10 mW/mm2 for 472/30 nm) and 610/75 or 520/35 nm emission (472/30 ET Bandpass, 610/75 ET Bandpass and 520/35 ET Bandpass Filters, AHF Analysentechnik), respectively. Stimulation of optogenetic actuators (ChR2 or ACR2) was performed with 300 µW/mm2 at 470 nm. Acquisition times, frame rates and gain used for imaging rhodopsin voltage sensors were 1 or 2 ms, 158 or 190 fps and gain 1 (for DMAR) or gain 10 or 50 (for ATR). For eFRET sensors, it was 1 ms exposure, ca. 190 fps and gain 10, if not otherwise stated.
To induce pharyngeal pumping under imaging conditions, the animals were incubated in 3 µl serotonin (20 mM, in M9 buffer) for 3 minutes immediately prior to the experiments. For drug testing, animals were exposed to either 30 nM nemadipine A, dissolved in 0.2 % DMSO in M9 or 0.2 % DMSO (vehicle control, in M9 buffer) respectively.
Image acquisition, processing and automated data analysis of pharyngeal voltage and pump events
Video acquisition (with frame rates of 160 or ca. 190 fps, at 1 ms exposure time) was performed by using the Photometrics PVCAM Device Adapter for µManager44. For BWM experiments, raw image sequences were analyzed via ROI (region of interest) selection of signal and background and multi-measure function in ImageJ (National Institutes of Health, USA; https://imagej.nih.gov/ij/index.html). Changes in fluorescence (ΔF/F0, or ΔF/Fmean, where Fmean (or F, for simplicity) is the mean fluorescence (Fmean) of the ROI, calculated for the entire imaging period) were calculated as follows:
Contrast was calculated as the ratio between mean fluorescence of a ROI across the pharyngeal or BWM tissue expressing the voltage sensor, and the mean fluorescence of a ROI within the animal, but chosen in tissue not expressing sensors and showing no autofluorescence.
Signal to noise ratio (SNR) was calculated as the ratio of the mean fluorescence of all peaks (= APs) observed during a pump train, and the standard deviation of the fluorescence in the ROI during resting phases.
Automated analysis of voltage and pump signals in the pharynx
To register the voltage signals in the pharyngeal muscles, a ROI was defined around the whole pharynx (Fig. 2A), whereas a smaller circular ROI was positioned over the terminal bulb lumen with the grinder, to track grinder movements and contractions with the accompanying lumen opening (observed as a lowering of the fluorescence in this ROI, as fluorescing tissue moves radially away during TB contraction). For background correction, another circular ROI was defined in an arbitrary dark region outside of the worm, or, for contrast measurements, ‘inside’ the worm, but covering tissue that is not expressing the respective voltage sensor (and avoiding the autofluorescing gut). ΔF/F was calculated in Excel (Microsoft). A custom workflow in KNIME 3.6.2 (KNIME AG, Zurich, Switzerland45) was used to synchronize pump and voltage events across animals by calling an R (R 3.5.1, The R Foundation for Statistical Computing) script to process each of the Excel tables. The R script proceeds to fit a spline curve to the input data with as many anchor points as data points present. The fit curve is sampled at constant 200 data points per second to account for the variable camera timing. Pumps are found as local minima with a centered time window of 625 ms. A manual input of indices can be used to correct this peak registration. The signals that passed manual control are synchronized first to the pump minimum. The voltage signal is then analyzed for the peak onset by searching for the minimum of the scaled difference with a lag of 50 ms of a centered moving average with window size of 105 ms. The events are subsequently synchronized to the onset of the voltage peak and grouped per animal, or analyzed individually, where appropriate. The signals are further analyzed to extract start and end of peak by searching for the first and last data point to cross a line at 10 % of the distance from the baseline to the peak minimum, respectively. Analogously, the full width at half maximum (FWHM) is calculated by searching for the two data points that cross the 50 % line between baseline and peak minimum. Duration is defined as the time difference between start and end of a peak. The kinetic parameters τON and τOFF are modelled each with a mono-exponential curve fit accounting only for the time frames from (−50 ms, FWHMstart) and (FWHMend, 305 ms), respectively. The area (voltage integral) is calculated as the integral from start to end of the peak from the baseline. The amplitude is calculated as the peak minimum to the maximum value at negative time after synchronization in the depicted time frame. The delay is reported as the time difference between the FWHMstart of the voltage and FWHMstart of the pump peaks.
Electrophysiology
Recordings were conducted from dissected BWM cells on the ventral side of the body anterior to the vulva as described previously42. Animals were immobilized with Histoacryl glue (B. Braun Surgical, Spain) and a lateral incision was made to access neuromuscular junctions along the ventral nerve cord. The basement membrane overlying muscles was enzymatically removed by incubation in 0.5 mg/ml collagenase for 10 s (C5138, Sigma-Aldrich, Germany). BWMs were patch-clamped in whole-cell mode at 22°C using an EPC10 amplifier with head stage connected to a standard HEKA pipette holder for fire-polished borosilicate pipettes (1B100F-4, WPI, USA) of 4-7 MΩ resistance. The extracellular (bath) solution contained: NaCl 150 mM; KCl 5 mM; CaCl2 5 mM; MgCl2 1 mM; glucose 10 mM; sucrose 5 mM; HEPES 15 mM, pH7.3 with NaOH, ≈330 mOsm. The pipette solution contained: K-gluconate 115 mM; KCl 25 mM; CaCl2 0.1 mM; MgCl2 5 mM; BAPTA 1 mM; Hepes 10 mM; Na2ATP 5 mM; Na2GTP 0.5 mM; cAMP 0.5 mM; cGMP 0.5 mM, pH7.2 with KOH, ≈320mOsm. Recordings were conducted in current clamp mode using Patchmaster software (HEKA, Germany), as described previously27. Light activation was performed using a LED lamp (KSL-70, Rapp OptoElectronic, Hamburg, Germany; 470 nm, 1 mW/mm2) and controlled by the EPC10 amplifier. Data were analyzed by Patchmaster software.
Software
Knime 3.6.2 (KNIME AG, Zurich, CH), R 3.5.1 (The R Foundation for Statistical Computing)
DATA AVAILABILITY
The original data, as videos or fluorescence grey values of the relevant ROIs, as well as the KNIME and R scripts used for the data analysis, will be made available through a publicly accessible server.
AUTHOR CONTRIBUTIONS
Conceived experiments and analyzed data: NAH, ACFB, RS, WSC, JL, AG; planned and conceptualized research: NAH, ACFB, RS, WSC, JL, AG; designed and programmed software: WSC; wrote the paper, with the help of the other authors: AG; acquired funding: AG.
ACKNOWLEDGEMENTS
We thank Steven Flavell and Adam Cohen for reagents, and Alexander Hirschhäuser, Heike Fettermann, Mona Hoeret, Regina Wagner and Franziska Baumbach for expert technical assistance. We are grateful to Christina Schüler for providing an original EPG record. DMAR and retinal Analog VI were gifts by Lars Kattner (Endotherm company). Some strains were provided by the Caenorhabditis Genetics Center, which is funded by the NIH - Office of Research Infrastructure Programs (grant P40 OD010440). This work was funded by Goethe University and the Deutsche Forschungsgemeinschaft (DFG), grants SFB807-P11 and EXC115 (Cluster of Excellence Frankfurt - Macromolecular Complexes) to A.G., and by an IMPReS PhD stipend to A.C.F.B.