Cytoplasmic nanojunctions between lysosomes and sarcoplasmic reticulum are required for specific calcium signaling
Nicola Fameli, Oluseye A. Ogunbayo, Cornelis van Breemen, A. Mark Evans
11 Cytoplasmic nanojunctions between lysosomes and sarcoplasmicreticulum are required for specific calcium signaling
Nicola Fameli , , ∗ , Oluseye A. Ogunbayo , Cornelis van Breemen , † A. Mark Evans , † , ∗ E-mail: [email protected] † co-senior authors Abstract
Herein we demonstrate how nanojunctions between lysosomes and sarcoplasmic reticulum(L-SR junctions) serve to couple lysosomal activation to regenerative, ryanodine receptor-mediated cellular Ca waves. In pulmonary artery smooth muscle cells (PASMCs) it hasbeen proposed that nicotinic acid adenine dinucleotide phosphate (NAADP) triggers in-creases in cytoplasmic Ca via L-SR junctions, in a manner that requires initial Ca release from lysosomes and subsequent Ca -induced Ca release (CICR) via ryanodinereceptor (RyR) subtype 3 on the SR membrane proximal to lysosomes. L-SR junction mem-brane separation has been estimated to be <
400 nm and thus beyond the resolution oflight microscopy, which has restricted detailed investigations of the junctional coupling pro-cess. The present study utilizes standard and tomographic transmission electron microscopyto provide a thorough ultrastructural characterization of the L-SR junctions in PASMCs.We show that L-SR nanojunctions are prominent features within these cells and estimatethat the junctional membrane separation and extension are about 15 nm and 300 nm, re-spectively. Furthermore, we develop a quantitative model of the L-SR junction using thesemeasurements, prior kinetic and specific Ca signal information as input data. Simulationsof NAADP-dependent junctional Ca transients demonstrate that the magnitude of thesesignals can breach the threshold for CICR via RyR3. By correlation analysis of live cellCa signals and simulated Ca transients within L-SR junctions, we estimate that “trig-ger zones” with a 60–100 junctions are required to confer a signal of similar magnitude. Thisis compatible with the 130 lysosomes/cell estimated from our ultrastructural observations.Most importantly, our model shows that increasing the L-SR junctional width above 50 nmlowers the magnitude of junctional [Ca ] such that there is a failure to breach the thresholdfor CICR via RyR3. L-SR junctions are therefore a pre-requisite for efficient Ca signalcoupling and may contribute to cellular function in health and disease. The importance of cytoplasmic nanojunctions to cellular signaling and thus to the modulationof cell function was recognised several decades ago [1, 2], henceforth the extent to whichcellular nanospaces may contribute to the regulation of cell function received little attention.Nevertheless there is now a growing recognition of the widespread occurrence and functionalsignificance of cytoplasmic nanospaces in cells from species across several kingdoms [3–9].In this respect, membrane-membrane junctions between lysosomes and the sarcoplas-mic reticulum (L-SR junctions) are of particular interest; not least because they have beenhypothesized to couple calcium signaling between these two organelles [10, 11]. a r X i v : . [ q - b i o . S C ] J a n That L-SR junctions may play an important role in cellular Ca signaling was uncov-ered through early studies on the Ca mobilizing messenger nicotinic acid adenine dinu-cleotide phosphate (NAADP) [12], which demonstrated that NAADP released Ca from astore other than the sarco/endoplasmic reticulum (S/ER) [13], that could then trigger fur-ther Ca release from the S/ER by Ca -induced Ca release (CICR) [14–17]. A majoradvance in our understanding was then provided by the demonstration that the NAADP-released Ca was from an acidic lysosome-related store [10,18,19] in a manner that requirestwo pore segment channel subtype 2 (TPC2) [20]. However, studies on pulmonary arterialsmooth muscle cells (PASMCs) had also identified a significant specialization, namely L-SRnanojunctions. It was hypothesized not only that these nanojunctions were necessary forcoupling between lysosomes and the SR but that they could both coordinate and restricttheir relationship to the SR by preferentially targeting ryanodine receptors while excludinginositol 1,4,5-trisphosphate (IP ) receptors [10, 11]. Importantly, NAADP-dependent Ca bursts primarily arise from lysosomes in the perinuclear region of PASMCs and appear topromote rapid, local Ca transients that are of sufficient size to activate clusters of SRresident ryanodine receptor subtype 3 (RyR3) and thus initiate, in an all-or-none manner, apropagating global Ca wave [10, 11].The specialization of the proposed L-SR junction is likely important in smooth musclecell physiology, e.g., in vasomotion, and in the recycling of organelles and programmedcell death by autophagy [21]. However, L-SR junctions may also make as yet unforeseencontributions to vascular pathologies as highlighted by the fact that Niemann-Pick diseasetype C1 results, in part, from dysregulation of lysosomal Ca handling [22] and is known toprecipitate portal hypertension [23], while other lysosomal storage diseases (e.g., Pompe andGaucher disease) accelerate pulmonary arterial hypertension [24, 25]. Moreover, observedhypertension is often associated with dysfunction of cholesterol trafficking [26], increasedplasma cholesterol levels, vascular lesion formation, atherosclerosis/thrombosis and medialdegradation [23, 27]. Therefore lysosomal Ca signaling is of considerable clinical interest.That L-SR junctions may be of further significance to pathology is also evident, for example,from the fact that in the pulmonary artery smooth muscle L-SR junctions underpin Ca waves initiated by endothelin 1, the levels of which are elevated in pulmonary hypertension,systemic hypertension and atherosclerosis [28, 29]. An understanding of how specific Ca signals are functionally initiated therefore has important translational implications.Lysosomal Ca regulation has been of current interest in several recent research andreview articles (e.g., [30–34]), yet the mechanism by which Ca signals are generated by theendolysosomal system has not yet been modeled in a truly quantitative manner. Given thelikely importance of L-SR junctions to Ca signaling in health and disease, we sought todetermine whether this nano-environment would indeed be able to effectively generate thepreviously observed NAADP-induced Ca signals.Due to the minute spatial scale of the nanojunctions generating the primary Ca signals,accurate investigation of dynamic signaling within these spaces cannot be achieved withcurrently available instrumentation. To overcome this issue, we took an integrative approachby combining our own electron microscopy of L-SR nanojunctions, existing kinetic data onthe Ca transporters and buffers, and prior knowledge of the NAADP-induced Ca signalfeatures into a quantitative model of a typical L-SR nanojunction. This model is basedon stochastic simulations of intracellular Ca diffusion by Brownian motion implementedusing the particle simulator MCell (freely available at mcell.org) [35–37]. In particular, weset out to verify the following hypotheses in PASMCs: (1) L-SR nanojunctions should beobservable in the ultrastructure of these cells, (2) NAADP induces sufficient Ca releasefrom the lysosome to initiate activation of RyR3 embedded in the junctional SR, and (3) the combined effect of activation of L-SR nanojunctions in a cytoplasmic “trigger zone”determines the threshold of global [Ca ] i for the biphasic release process.In the present manuscript we have verified the existence of L-SR nanojunctions within theultrastructure of PASMCs, and shown that lysosomes can release sufficient Ca to activateCICR via RyR3 clusters embedded in the junctional SR. Perhaps most importantly, we showthat L-SR coupling is determined both by the integrity of L-SR junction on the nanoscaleand the quantal summation of Ca release from multiple, activated junctional complexes. signals within isolated pulmonary arterysmooth muscle cells The relevant background findings that stimulated the development of the work presentedhere were first reported by Evans’ group in [17] and [10], and are summarized in figure 1.The example record in figure 1A highlights the fact that NAADP appears to activate atwo-phase Ca signal, which can exhibit an identifiable “shoulder” during the initial risingphase of the signal (figure 1A, time point 1), followed by a second faster phase of signalamplification (figure 1A, time point 2). It is notable that the delay to the initiation of thesecond phase of amplification is variable [10, 11, 17] and due to this fact the shoulder is notalways evident (see for example figure 1B, lower panel). Previous studies have demonstratedthat this two-phase response results from initiation by NAADP of Ca bursts from lysosome-related stores [10] in a manner that requires TPC2 [38] and that Ca bursts are subsequentlyamplified by CICR from the SR via clusters of RyR3 [11,17]. Panel B of this figure illustratesthe fact that prior depletion of SR stores by pre-incubation with thapsigargin (1 µ M) blocksthe amplification phase, while depletion of acidic Ca stores with bafilomycin abolishes theentire NAADP-induced Ca signal. The previous studies cited above provided calibratedestimates of the changes in intracellular [Ca ] input into the model below.Based on earlier optical microscopy work, like the data in figure 1, and immunofluores-cence results, it has been proposed that, for the lysosomal Ca release to trigger CICR,L-SR nanojunctions are required and that they consist of apposing patches of lysosomal andSR membranes separated by a narrow space of nano-scale dimension [10, 17]. These studiesled to an upper limit of 400 nm for the separation of the junctional membranes. We proposethat L-SR junctions do indeed represent cellular nanojunctions and that they might play arole of accentuating Ca gradients, akin to that of plasma membrane (PM)-SR junctionsthat are pivotal in the process of SR Ca refilling during asynchronous [Ca ] waves [39].We furthermore hypothesize that in order for these nanojunctions to appropriately regulateCa signaling, they must be separated by a distance of approximately 20 nm and be of theorder of a few hundred nm in lateral dimensions, as inferred from previous studies on PM-SRjunctions [10, 11]. To identify lysosomes, SR regions and L-SR nanojunctions, we recorded and surveyed 74electron micrographs of rat pulmonary arterial smooth muscle taken from samples preparedas described in the Materials and Methods section.The images in figure 2 provide a setof examples. Since we were aiming to detect L-SR junctions, namely close appositionsof the lysosomal and SR membranes, immuno-gold labeling of lysosomes was prohibited,since this technique compromises membranes definition by electron microscopy to the extent
Figure 1. A, B,
Upper panels show a series of pseudocolour images of the Fura-2fluorescence ratio (F340/F380) recorded in two different pulmonary artery smooth muscle cellduring intracellular dialysis of 10nM NAADP before (A) and after (B) depletion of SR storesby pre-incubation (30 min) with 1 µ M thapsigargin. Note the spatially localized ‘Ca bursts’.A, Lower panel shows the record of the Fura-2 fluorescence ratio against time corresponding tothe upper left panel of pseudocolours images; note the discrete shoulder in the rising phase ofthe F340/F380 ratio that corresponds to the initial ‘Ca bursts’. B, Lower panel showspaired responses to 10 nM NAADP under control conditions (black), following depletion of SRCa stores with 1 µ M thapsigargin (blue) and following depletion of acidic stores with 1 µ Mbafilomycin-A1 (red). Scale bars: 10 µ m. that we would be unable to assess junctional architecture. Instead, in images like those infigure 2, lysosome identification was accomplished by relying on the knowledge of lysosomalultrastructural features, which has accumulated over the past 50 years since the discovery ofthe lysosomes (see for example, [40]).In standard (2D) transmission electron microscopy (TEM) images, lysosomes typicallyappear as elliptical structures bound by a single lipid bilayer, a feature that distinguishesthem from mitochondria. Depending on the lysosomal system stage, they also tend to havea more or less uniformly electron-dense interior as compared to the surrounding cytosol [41].They can be distinguished from endosomes by their larger size and darker lumen and theydiffer from peroxisomes, since the latter usually display a geometrically distinct and markedly Figure 2.
Representative electron micrographs of rat pulmonary artery SMC regionscontaining lysosomes (L), several SR cisterns, and including several examples of L-SRjunctions (arrows). Also indicated are nuclei (N), Golgi apparatus (G), mitochondria (M), amultivesicular body (MVB) and extra-cellular space (ECS). Scale bars = 500 nm.Magnifications: A,C 80,000 × , B, 60,000 × , D, 70,000 × . darker structure called “crystalloid” in their lumen. Moreover, it would appear that perox-isomes are found far more frequently in liver, kidney, bronchioles and odontoblasts than inother cell types (see, for example, [40,42,43]). Occasionally, organellar remnants are still vis-ible inside these ovals, a characteristic that identifies them as multi-vesicular bodies (“MVB”in figure 2A). As it is at times questionable whether MVB’s are late endosomes or endosome-lysosome hybrids (compare, for example, [40] and [44]), we have excluded organelles (3 intotal) with such characteristic from our statistical count.From each of the relevant smooth muscle regions surveyed, we obtained high-resolutionimages of areas containing lysosomes and L-SR junctions in order to quantitatively charac-terize them (arrows in figures 2 and 3A). Using a software graphics editor (inkscape.org) andthe image scale bar as a calibration gauge, we measured the lysosome size, as the length ofthe major and minor axes of their elliptical 2D projections (in orange and grey, respectively,in figure 3A), the L-SR widths, that is the distance between lysosomal and SR membranesat places where the two were about 30 nm or closer to each other (in purple in figure 3A), and the L-SR extensions as a percentage ratio between the junctional SR and the lysosomalmembranes (in turquoise in figure 3A). From these measurements, we extrapolated the 3Djunctional SR extension, both as a percentage of the lysosomal surface and as a length innm. The histograms displayed in figure 3B–D visually summarize the data collected fromthe image analysis. The mean and standard deviation values of the measured parametersare reported in table 1. Figure 3. A,
High magnification (150,000 × ) electron micrograph of a region of Figure 2Bcontaining 3 L-SR junctions (arrows); coloured tracings as shown were used to measurelysosome dimensions, L-SR widths and extensions. Scale bar = 100 nm. B–D,
Histogramsshowing distribution of several relevant lysosomal and L-SR junctional parameters, used tocharacterize the junctions and inform the quantitative model. B, lysosomal dimensions asmajor and minor axes of oval shape in micrographs; C, L-SR junctional width; D, percentageapposition between junctional SR and lysosome perimeter as projected in 2D micrographs.
Estimates of the various parameters gathered in this phase of the study were used as abasis to build a 3D software object to represent a typical L-SR junction. This reconstructionwas then used to design the simulations mimicking Ca diffusion in the L-SR nanojunctions,as is described below. To gather more direct information on the 3D morphology of L-SR junctions, we acquired a setof tomograms of those regions from the same sample blocks used to obtain the images in figure2. In figure 4, we report snapshots from one of the tomograms; in these stills, we have alsotraced out parts of one lysosome and the closely apposed SR region that together form a L-SRjunction. These tomograms are very helpful in clarifying the detailed morphology of L-SR
Table 1. L-SR junction characterization parameters parameter mean ± SD noteslysosome minor axis (325 ±
65) nm n = 29lysosome major axis (398 ±
91) nm n = 29L-SR width (16 ±
7) nm n = 41L-SR overlap (23 ±
13) % n = 29L-SR lateral extension (262 ± ±
15) % extrapolated from 2D measurementsMean and standard deviation values of L-SR junction parameters, calculated from data as inFigure 3B–D. junctions and informing on the spatial variability of the SR network. For example, while theSR segment shown in a single 2D tomographic scan (figure 4A) would appear to be continuousand part of a large SR compartment, it actually branches out into narrower cisterns asrevealed by 3D tomographic reconstruction (figure 4B). Furthermore, it is interesting toobserve the fact that one extension of the SR appears to couple with multiple organelles.Thus, the 3D views generated by tomography are paramount for demonstrating the presenceof a true junctional complex and for the design of a prototypical L-SR environment asa software mesh object, on which we may simulate the NAADP-mediated localized Ca release. Figure 4. A,
Snapshot from a TEM tomogram of a L-SR region of rat pulmonary arterysmooth muscle, illustrating, among other things, a single SR extension apparently formingjunctions with several lysosomes. Magnification = 62 , × . B, Same snapshot shown in A,but with a lysosome (orange) and a portion of SR (turquoise) partially traced out in 3D. Thesepseudo-colour tracings underscore how the SR can appear as a large cistern in a given plane,but can actually branch out in different directions when viewed in 3D. Scale bars ≈
100 nm.
The model aims to verify whether NAADP-induced Ca release from the lysosomal systemcould be responsible for the localized Ca signal preceding the global Ca wave (see figure 1and [10]), which triggers a propagating wave by CICR via RyR3s localized at L-SR junctions,as predicted by earlier observations [10]. bursts To understand the generation of the signal “shoulder”, such as that observed at time point 1in the lower panel of figure 1A, by Ca bursts within L-SR nanojunctions, let us note that itsmagnitude corresponds to the difference in [Ca ] between the resting level of ≈
100 nM [17]prior to NAADP stimulation (up to the point, at which NAADP enters the cytoplasm underthe whole-cell configuration (WC) in figure 1) and the value of ≈
400 nM [17], correspondingto [Ca ] i at time point 1 in the example record shown in figure 1A, lower panel (theseconcentration values were obtained via a standard calibration procedure [17]). This leadsto a ∆[Ca ] shoulder of approximately 300 nM [17]. Let us now estimate the number oflysosomes required to generate a ∆[Ca ] shoulder of such magnitude in a PASMC, given thefollowing assumptions:1. That the luminal [Ca ] of a lysosome, [Ca ] lys , is in the range of 400–600 µ M, asdetermined in mouse macrophages [45] and that it is homogeneous across the lysosomalpopulation;2. That the Ca release rate during bursts is ≈ ions/s (based on findings in [46], butsee next section for a detailed analysis on the rate time-variation) and that it becomesnegligible for values of [Ca ] lys below ≈ µ M, as suggested by the single channelkinetics of TPC2 signaling complexes in lipid bilayer studies [46];3. That lysosomes are spheres with a radius of ≈
180 nm, as gathered from our EMcharacterization (see figures 2, 3, 4, and table 1), and hence that their volume is V lys = (4 π/ . × − m) = 2 . × − L;4. That a smooth muscle cell cytosolic volume can be calculated as V cyt = 2 . × − Lby modeling a cell as a 130- µ m-long cylinder, 6 µ m in diameter [47], and accountingfor nuclear, SR, mitochondrial and lysosomal volumes.With these assumptions accepted, from points 1. and 2. above we gather that a lysosomecan release a potential ∆[Ca ] lys = (400 to 600) µ M − µ M = 3 . × − M to 5 . × − Minto the cytosol (as we elaborate in the next section, this should take ≈ .
03 s, a time framethat ensures that released Ca can be considered unbuffered). Each lysosome contributionto [Ca ] i , ∆[Ca ] i , lys , can then simply be calculated by taking the lysosome-to-cell volumeratio into account: ∆[Ca ] i , lys = ∆[Ca ] lys V lys V cyt ≈ ] shoulder as ∆[Ca ] shoulder ∆[Ca ] i , lys ≈
60 to 100 lysosomes (2)In summary, between 60 and 100 lysosomes would be necessary (and possibly sufficient) toprovide a ∆[Ca ] shoulder of 300 nM, which is typically observed during the localized Ca release phase of the NAADP-induced Ca signals. How many lysosomes do we actually expect to be in a PASM cell of our sample tissue?We can obtain a rough estimate of this number from the TEM imaging we performed forthis study. In each 80-nm-thick TEM sample section, we see 5–10 lysosomes/cell. Lysosomesare predominantly localized to the perinuclear region of the cytoplasm, but seldom in thesubplasmalemmal area, consistent with previous observations by optical microscopy [10, 11].If we simplify the geometry of a typical smooth muscle cell to a 130- µ m-long cylinder withradius of 6 µ m, and considering that lysosome radii are around 180 nm, as mentioned above,it is reasonable to assume that separate sets of lysosomes would be observable in TEM imagestaken at distances into the sample of about 180 nm from each other. Neglecting the sliceswithin one lysosomal diameter of the cylinder surface—given the near total lack of observedlysosomes in those subvolumes—then our images suggest that we can expect a total of about130 lysosomes/cell. It is encouraging that we obtain from this count a higher number thanthe 60–100 lysosome range we derived in equation (2), in that on the one hand it is plausibleto think that not all of the lysosomes in a cell may be activated in synchrony, nor maythey all be involved in NAADP-mediated signaling, and on the other hand experience tellsus that evolution has built in some redundancy of function in order to provide a thresholdand also a margin of safety for the generation of this type of Ca signals. Moreover, theestimated number of junctions is based on a value for [Ca ] lys determined in macrophages,and it is plausible that the total releasable [Ca ] lys in a PASMC may differ from that valueand that it may also vary over the lysosome population. Lastly, the uncertainty in thenumber of lysosomes/cell evidenced from the electron micrographs as described above mayalso contribute to this discrepancy. signal In the hypothesized model outlined in [10,11], Ca bursts activate SR resident RyR3 chan-nels within L-SR junctions and thus initiate a propagating Ca wave by CICR. The stochas-tic simulations developed here attempt to reproduce the phenomenon of the generation of[Ca ] transients within individual L-SR junctions, considering the Ca release kinetic re-quirements for lysosome-resident TPC2 signaling complexes and the rate of Ca capture bythe SERCA2a localized on the neighbouring SR membrane [48], and to link these junctionaltransients to the observed bursts.The thorough quantitative image analysis described in the previous section yields criticalinformation for our first modeling phase, in which we built a dimensionally accurate virtuallysosome and a portion of the SR system, closely apposing the lysosome (figure 5), so as toreproduce a NAADP-triggered Ca signal within the nanospace of a representative L-SRjunction as faithfully as possible.From the available literature we estimated the number of TPC2 and SERCA2a likely dis-tributed on the lysosome and SR membranes, respectively. We obtained the former numberby dividing the macroscopic whole-lysosome conductance, calculated from the current valuesreported in a recent study on TPC2-mediated Ca current in isolated lysosomes [49], bythe single channel conductance determined in [46], thus estimating that a typical lysosomalmembrane may contain ≈
20 TPC2. To obtain this value, we used the experimental condi-tion data provided in [49] to extract values for the ionic potential, E ion , across the lysosomalmembrane. We then employed the authors’ current-voltage ( I - V m ) data to compute whole-lysosome conductance values ( g WL ) as a function of the applied membrane potential, V m ,from Ohm’s law: I = g WL ( V m − E ion ).Moreover, we have estimated the density of SERCA2a on the SR within L-SR junctions tobe equivalent to that previously predicted for PM-SR junctions as approximately 6250 /µ m [39]. Other input data for the model are the estimates of lysosomal volume and of the Figure 5. A and B,
3D software reproduction of a lysosome closely apposed to a portion ofSR, thereby forming an ≈ ] NJ transients like the ones reported in figure 7. Scale bar ≈
100 nm.The model geometry and code files are available from the corresponding author. [Ca ] lys , from which we calculated the actual number of ions in the lysosome prior to thebeginning of Ca release (see previous section). This, in turn, was used in the extrapolationof the TPC2 complex Ca release rate as a function of time, as follows.A recent electrophysiological study of the Ca conductance of TPC2 signaling complexprovides valuable information regarding its biophysical properties and, importantly, high-lights the fact that, for a given relatively low activating concentration of NAADP (10 nM, asin figure 1), the channel open probability appears to depend on [Ca ] lys (this is likely due toa partial neutralization of the electrochemical potential across the lysosomal membrane) [46].We used the Ca conductivity measured in [46] and values of the lysosome membrane poten-tial ( [50]) to calculate the maximal Ca current, I max , as 2 . × ions/s. We then applieda weighted quadratic fit to the channel’s open probability ( P o ) as a function of the [Ca ] lys data in [46] with constraints that the P o would tend to zero at [Ca ] lys = 80 µ M, basedon the observation that below [Ca ] lys = 100 µ M essentially no single channel openingswere observed [46]. Another constraint for the fit was that the curve be within the standarddeviation value of P o at the highest reported [Ca ] lys = 1 mM. From the quadratic fit,we then obtained our own P o -vs-[Ca ] lys table (plotted in figure 6A) and assumed that at the beginning of Ca release, the Ca current would be P o , max × I max at the maximalluminal concentration until the luminal concentration decreased to the next point in the P o -vs-[Ca ] lys relationship. At this point, the release rate decreases to a new (lower) P o × I max until the luminal concentration reaches the next lower point in the table, and so on until P o = 0 at [Ca ] lys = 80 µ M. The Ca release rate as a function of time obtained in thismanner is shown in figure 6B. Figure 6. A, open probability P o of the TPC2-dependent Ca conductance reproduced froma quadratic fit to the data reported in [46]. B, Ca release rate calculated as explained in thetext, based on the P o in A. To implement an approximated SERCA pump action, we used a simplified version ofa multi-state kinetic model developed in [51], which was further informed by more recentisoform-specific studies [52, 53]. In particular, since we are interested in simulating theSERCA2a Ca uptake only in terms of its influence on the shaping of the junctional [Ca ]transient, we have opted not to implement the steps of the multistate model that deal with theCa unbinding from the SERCA on the SR luminal side of the pump. In brief, the reactionsSERCA2a undergo are: (1) binding/unbinding of the first Ca ; (2) binding/unbinding ofthe second Ca .The Ca diffusivity was obtained from studies, in which it was concluded that given theknown kinetics of typical Ca buffers, the range of free Ca after it enters a cell can be upto 200 nm. Therefore, due to the nano-scale of our system, the trajectories of Ca releasedby TPC2 signaling complexes in the simulations are governed by the measured diffusivity offree Ca (2 . × − m / s; [54, 55]). Once Ca are buffered they acquire the measureddiffusivity of the buffers (8 . × − m / s; [56]).We summarize the quantitative model input data in table 2.Using MCell as a stochastic particle simulator, we ran a number of simulations to rep-resent the NAADP-mediated Ca release that is supposed to occur at L-SR junctions inPASM experiments. Released Ca is assumed to undergo Brownian motion in the surround-ing space, including the L-SR nanospace. SERCA2a placed on the neighbouring SR surfacemay capture Ca according to our approximation of their known multistate model. Todetermine how this regenerated cellular environment can shape a Ca transient we “mea-sured” the junctional [Ca ] by counting the ions within the L-SR volume at any given timeand dividing the number by the volume. The snapshots in figure 5 are part of the visualoutput of this phase of the work. The data to determine the simulated Ca transients inthe L-SR junctions were collected in a measuring volume placed between the lysosomal and Table 2. Quantitative model input data quantity value source∆[Ca ] shoulder
300 nM [17] and this article[Ca ] lys µ M [45]lysosomal average radius 180 nm this article, figure 3(cylindrical) SMC radius 6 µ m [47]SMC length 130 µ m [47]SMC cytosolic volume 2 . × − L this articleTPC2 complexes per lysosome 20 [46, 49] and this articleTPC2 complex Ca release rate variable this article, figure 5SERCA2a surface density 6250 /µ m [39]SERCA2a first Ca binding (unbinding) rate 5 × M − s − (60 s − ) [51–53]SERCA2a second Ca binding (unbinding) rate 4 × M − s − (60 s − ) [51–53]free Ca diffusivity, D free . × − m / s [54, 55]Ca buffer diffusivity, D buffer . × − m / s [56]Summary of input parameters used in various phases of the quantitative model, and referencesto the origin of their values. SR membranes in the virtual L-SR junction (rust-coloured box in figure 5B).Previous experimental findings about VSMC PM-SR junctions indicate that disruptionof nanojunctions can have profound consequences on Ca signaling properties of the cells.For example, when calyculin A was used to separate superficial SR portions from the plas-malemma of the rabbit inferior vena cava, it was observed that [Ca ] i oscillations wouldcease [57]. This was later corroborated by a quantitative model of the PM-SR junction’s rolein the refilling of SR Ca during oscillations [39]. In another study by the van Breemenlaboratory, it was observed that the mitochondria-SR junctions of airway SMC displayeda variable average width as a function of the state of rest or activation of the cell [58]. Itmakes sense then to study the effect of changes in the junctional geometry on the transientsgenerated by our simulations. Therefore, we ran several simulations, in which the separationbetween the lysosomal and SR membranes was increased from 10 nm to 100 nm in steps of10 nm. In figure 7A, we report three sample transients from this set of simulations obtainedusing three different junctional membrane separations, as indicated in the inset legend. Thevalue of each of the points graphed in this plot is the average value of 100 simulations, ineach of which the random number generator within MCell is initiated with a different seed(see Materials and Methods). We report representative error bars as 3 × the standard error,to convey the >
99% confidence interval of the data. As one would expect, the transientnanojunctional (NJ) Ca concentration, [Ca ] NJ , decreases in magnitude, as the junc-tional L-SR membrane-membrane separation increases, simply because the released Ca has a larger junctional volume available over which to spread. However, it is important tomake a quantitative comparison between this change in [Ca ] NJ and the [Ca ] i require-ments to activate the putative RyR3 population of the junctional SR. This can be attainedby analyzing the time scale of both the recorded and simulated Ca signals. In the final step of the development of our model, we analyzed the relationship between thetime scale of the [Ca ] NJ transients resulting from our model (figure 7) and that of theobserved Ca signal in figure 1. Figure 7. A, calculated nanojunctional [Ca ] transient, [Ca ] NJ , “measured” inside thevolume of the recreated L-SR nanojunction shown in figure 5. To show the effect of changes inthe junctional geometry, we report three transients calculated using different junctional widthsof 20, 50 and 100 nm. B, [Ca ] NJ vs width of junction, concentration values are temporalaverages of the transients as in panel A, calculated over an interval of 0.065 s (solid circles)and 0.038 s (empty circles); see text for explanation. The shaded area indicates theapproximate threshold values for CICR at RyR3s [59]. Let us note that the typical duration of the simulated transients, t transient , is about 0.06s and recall that these represent ∆[Ca ] above the resting [Ca ]. On the other hand, thebuild up to the maximum value of ∆[Ca ] shoulder takes about 5 s (we refer to this time as t shoulder ; figure 1). This interval is about two orders of magnitude larger than the duration ofthe simulated individual transients (figure 7A). One way to reconcile the hypothesis that L-SR junctions are at the base of the observed NAADP-induced Ca signals such as the onesin figure 1 and in particular that the signal shoulder emerges from lysosomal Ca releaseat L-SR junctions, is to bring forward the assumption that the signal shoulder may be theresult of a sequential summation effect over many L-SR junctions, each working accordingto an all-or-none mechanism of Ca release, and that the “firing” of one junction causes acascading effect across the set of junctions that produce the shoulder. Then the duration ofthe shoulder upstroke can be expressed as: t shoulder = N NJ × t transient (3)where N NJ is the number of L-SR nanojunctions that yield ∆[Ca ] shoulder . Since we calcu-lated above that Ca release from N NJ ≈ t transient , the duration the [Ca ] NJ transient, by reversing equation (3) and obtaining t transient =0.05–0.08 s. It isnoteworthy that this range of values is obtained in a manner completely independent of oursimulation results, which yielded a similar range of values.To gain quantitative insight into the effects of varying the junctional width, we thencalculated the temporal average of the [Ca ] NJ over t transient (using the middle value of therange calculated via equation (3)) and graphed it as a function of the junctional width. Theresult of this analysis is reported in figure 7B (solid blue circles). As we have anticipated atthe end of the previous section, the decrease in magnitude of these data is to be expected,however in this plot we also indicated the range of [Ca ] i values (shaded area) over whichmaximum SR Ca release via RyR3 is reported to occur in skeletal muscle [59]. Thiscomparison underscores the important constraint played by the width of the L-SR junctionsand indicates that, unless the membrane separation is kept below about 30 nm, it is notpossible for [Ca ] NJ to breach the threshold for RyR3 Ca release. Let us also point outthat the [Ca ] NJ data in figure 7B would shift upward, toward concentration values thatwould make the junction more prone to promote RyR3 release, if the temporal average weretaken over a shorter transient time, t (cid:48) transient < t transient around the [Ca ] NJ peak. However,in that case equation (3) indicates that a larger N NJ (than 80, picked as the middle of the60–100 range) would have to contribute to the signal summation that results in a 5-second t shoulder . Interestingly, this possibility agrees with our lysosome count from TEM images (130lysosomes/cell) and with the argument of natural redundancy we contemplated to explainthe discrepancy between the calculated and observed lysosome number estimates. As anexercise we have recalculated the [Ca ] NJ time-averaged over t (cid:48) transient obtained using 130lysosomes in equation (3) (empty purple circles in figure 7B), and this indeed shows thatRyR3 Ca release threshold would be cleared more readily. These observations suggest thatactivation of RyR3s in the L-SR junctions not only depends on the concentration of Ca near them, but also on their exposure time to this concentration. We have recently introduced the concept of the “pan-junctional SR”, which states thatCa release and uptake at a family of specific nanojunctions connected by a continuousbut variable SR lumen integrates cellular control over multiple functions [4]. The lysosome-sarco/endoplasmic reticulum (L-S/ER) junction is the most current junction to be consideredin this context and exhibits perhaps the highest degree of plasticity of the family of nano-junctions of the SR; the mechanism and function of lysosomal Ca signaling is currentlyhotly debated [60].By means of a thorough ultrastructural study in rat pulmonary artery smooth mus-cle, we have observed and characterized L-SR nano-junctions, which had been previouslyhypothesized on the basis of optical measurements of Ca signals and optical immunocy-tochemistry [10, 11]. Our observations corroborate the previously reported finding (in [10]and [17]) that lysosomes in PASMCs tend to cluster in the perinuclear region, as is evidentin our micrographs (e.g., figure 2A). We find that L-SR junctions are on average 15 nm inwidth (equivalent to our preliminary reports [8, 61] and to recent observations in culturedfibroblasts [33]) and extend approximately 300 nm in lateral dimensions, thereby involvingabout 15% of the lysosomal membrane (table 1).5
99% confidence interval of the data. As one would expect, the transientnanojunctional (NJ) Ca concentration, [Ca ] NJ , decreases in magnitude, as the junc-tional L-SR membrane-membrane separation increases, simply because the released Ca has a larger junctional volume available over which to spread. However, it is important tomake a quantitative comparison between this change in [Ca ] NJ and the [Ca ] i require-ments to activate the putative RyR3 population of the junctional SR. This can be attainedby analyzing the time scale of both the recorded and simulated Ca signals. In the final step of the development of our model, we analyzed the relationship between thetime scale of the [Ca ] NJ transients resulting from our model (figure 7) and that of theobserved Ca signal in figure 1. Figure 7. A, calculated nanojunctional [Ca ] transient, [Ca ] NJ , “measured” inside thevolume of the recreated L-SR nanojunction shown in figure 5. To show the effect of changes inthe junctional geometry, we report three transients calculated using different junctional widthsof 20, 50 and 100 nm. B, [Ca ] NJ vs width of junction, concentration values are temporalaverages of the transients as in panel A, calculated over an interval of 0.065 s (solid circles)and 0.038 s (empty circles); see text for explanation. The shaded area indicates theapproximate threshold values for CICR at RyR3s [59]. Let us note that the typical duration of the simulated transients, t transient , is about 0.06s and recall that these represent ∆[Ca ] above the resting [Ca ]. On the other hand, thebuild up to the maximum value of ∆[Ca ] shoulder takes about 5 s (we refer to this time as t shoulder ; figure 1). This interval is about two orders of magnitude larger than the duration ofthe simulated individual transients (figure 7A). One way to reconcile the hypothesis that L-SR junctions are at the base of the observed NAADP-induced Ca signals such as the onesin figure 1 and in particular that the signal shoulder emerges from lysosomal Ca releaseat L-SR junctions, is to bring forward the assumption that the signal shoulder may be theresult of a sequential summation effect over many L-SR junctions, each working accordingto an all-or-none mechanism of Ca release, and that the “firing” of one junction causes acascading effect across the set of junctions that produce the shoulder. Then the duration ofthe shoulder upstroke can be expressed as: t shoulder = N NJ × t transient (3)where N NJ is the number of L-SR nanojunctions that yield ∆[Ca ] shoulder . Since we calcu-lated above that Ca release from N NJ ≈ t transient , the duration the [Ca ] NJ transient, by reversing equation (3) and obtaining t transient =0.05–0.08 s. It isnoteworthy that this range of values is obtained in a manner completely independent of oursimulation results, which yielded a similar range of values.To gain quantitative insight into the effects of varying the junctional width, we thencalculated the temporal average of the [Ca ] NJ over t transient (using the middle value of therange calculated via equation (3)) and graphed it as a function of the junctional width. Theresult of this analysis is reported in figure 7B (solid blue circles). As we have anticipated atthe end of the previous section, the decrease in magnitude of these data is to be expected,however in this plot we also indicated the range of [Ca ] i values (shaded area) over whichmaximum SR Ca release via RyR3 is reported to occur in skeletal muscle [59]. Thiscomparison underscores the important constraint played by the width of the L-SR junctionsand indicates that, unless the membrane separation is kept below about 30 nm, it is notpossible for [Ca ] NJ to breach the threshold for RyR3 Ca release. Let us also point outthat the [Ca ] NJ data in figure 7B would shift upward, toward concentration values thatwould make the junction more prone to promote RyR3 release, if the temporal average weretaken over a shorter transient time, t (cid:48) transient < t transient around the [Ca ] NJ peak. However,in that case equation (3) indicates that a larger N NJ (than 80, picked as the middle of the60–100 range) would have to contribute to the signal summation that results in a 5-second t shoulder . Interestingly, this possibility agrees with our lysosome count from TEM images (130lysosomes/cell) and with the argument of natural redundancy we contemplated to explainthe discrepancy between the calculated and observed lysosome number estimates. As anexercise we have recalculated the [Ca ] NJ time-averaged over t (cid:48) transient obtained using 130lysosomes in equation (3) (empty purple circles in figure 7B), and this indeed shows thatRyR3 Ca release threshold would be cleared more readily. These observations suggest thatactivation of RyR3s in the L-SR junctions not only depends on the concentration of Ca near them, but also on their exposure time to this concentration. We have recently introduced the concept of the “pan-junctional SR”, which states thatCa release and uptake at a family of specific nanojunctions connected by a continuousbut variable SR lumen integrates cellular control over multiple functions [4]. The lysosome-sarco/endoplasmic reticulum (L-S/ER) junction is the most current junction to be consideredin this context and exhibits perhaps the highest degree of plasticity of the family of nano-junctions of the SR; the mechanism and function of lysosomal Ca signaling is currentlyhotly debated [60].By means of a thorough ultrastructural study in rat pulmonary artery smooth mus-cle, we have observed and characterized L-SR nano-junctions, which had been previouslyhypothesized on the basis of optical measurements of Ca signals and optical immunocy-tochemistry [10, 11]. Our observations corroborate the previously reported finding (in [10]and [17]) that lysosomes in PASMCs tend to cluster in the perinuclear region, as is evidentin our micrographs (e.g., figure 2A). We find that L-SR junctions are on average 15 nm inwidth (equivalent to our preliminary reports [8, 61] and to recent observations in culturedfibroblasts [33]) and extend approximately 300 nm in lateral dimensions, thereby involvingabout 15% of the lysosomal membrane (table 1).5 signaling In an effort to achieve quantitative understanding of the phenomenon of NAADP-mediatedCa transients and verify the proposal that these may be generated in L-SR junctions[10,17], we focused on one of the prominent features of these Ca signals, namely the local-ized Ca bursts that precede the propagating Ca wave (figure 1), which we refer to as thesignal shoulder (∆[Ca ] shoulder ). In the first instance, we have estimated the potential con-tribution of local bursts of Ca release from individual lysosomes to the elevation of global[Ca ] i observed during this shoulder in the experimental records. From this, and using thedimensions of a typical smooth muscle cell, we calculated that 60–100 lysosomes would berequired to cause an elevation of comparable magnitude to the signal shoulder. This is lowerthan the estimate for the total number of lysosomes per cell, 130, we obtained from ourultrastructural study, but comparable in order of magnitude. We have already mentionedabove a number of factors in favour of observing a greater number of lysosomes/cell thanthe estimated number required to generate the signal shoulder. In addition, several otherelements contribute a degree of uncertainty to those estimates, such that their discrepancymay not be as large. We need to consider that P o data for the Ca conductance associatedwith the TPC2 signaling complex published in [46], on which we based our TPC2 rate table,show some variability according to the standard deviation bars, which, in turn, implies anuncertainty in the interpolated P o (figure 5A). Moreover, we cannot exclude the possibilitythat these data reflect a contribution from multiple channels ( N P o ) rather than a purelysingle channel P o . Lastly, the standard deviation of the simulated transient (only the stan-dard error is shown in figure 7A) and the variability in the experimental determination ofthe [Ca ] sensitivity of RyR may allow for some uncertainty in the estimated number ofnanojunctions.To take this study further and understand whether the observed L-SR junctions couldgive rise to [Ca ] i transients of appropriate magnitude to trigger Ca release from RyR3channels on the junctional SR, we developed a quantitative stochastic model of Ca dy-namics in the junctional nanospaces. We have previously published a simplistic version ofthis model, which nonetheless captured the essential features of the problem and yielded anindication that such Ca transients in the L-SR junctions would be possible [4]. However,the simulated transients we obtained displayed unphysiological features, such as an abruptonset and decay. We show here that this was largely due to lack of a faithful representationof the open probability for the Ca conductance of the TPC2 signaling complex. We havenow combined experimental information on the biophysical properties of conductance andopen probability (from [46]) and on the luminal [Ca ] of the lysosomes (from [45]) to im-plement a more realistic Ca release rate model in the simulations. As a consequence, weare able to output a physiologically meaningful junctional transient profile (figure 7A), andobserve that the [Ca ] NJ transients generated within our model junctions—in turn, basedon those observed in our TEM images—reach peaks of about 20 µ M. While we are still notable to measure these transients in individual L-SR junctions (peri-lysosomal Ca probesare only recently becoming available—see for example [62]—and so far have not been used insmooth muscle cells, vascular or otherwise), it is worth noticing that the values we find arecomparable to those recently measured in so-called Ca hot spots in the mitochondria-ERjunctions of neonatal ventricular cardiomyocytes (rat culture) [63] and in RBL-2H3 and H9c2cells earlier [64]. These results therefore suggest that the hypothesis presented for the roleof L-SR junctions in cellular Ca signaling is certainly plausible. Although our simulations were successful, comparison of the Ca transients in figures 1 and7 not surprisingly reveals a striking difference between the simulated Ca kinetics of a singleL-SR junction and those of the whole cell. This illustrates important aspects related to theconcept of lysosome-SR “trigger zone” previously introduced by one of us (AME, [10]) andrecently proposed as a facilitating factor in lysosomal-ER signaling in reverse, whereby Ca release from ER compartments enables NAADP-mediated activation of acidic organelles [32].Briefly, this concept suggests that in PSMCs clusters of lysosomes and closely apposedSR regions containing sets of RyRs may act together to form specialized intracellular com-partments that transform NAADP-stimulated localized lysosomal Ca release into cell-wideCa signals via RyR-supported CICR. If the firing of one junction were sufficient for sub-sequent initiation of the CICR across the entire SR, the experimentally observed threshold(figure 1) would be much lower and the rate of junctional coupling by CICR faster. In otherwords, due to the high [Ca ] i threshold of about 10 µ M for Ca activation of RyR3, CICRengendered by a single L-SR junction is likely to die out, unless reinforced by a process ofquantal releases by other L-SR junctions within a cluster [65]. Therefore it seems more likelythat multiple L-SR junctions work in concert in a process characterized by both additiveand regenerative elements to provide the necessary threshold and margin of safety requiredto ensure the propagation by CICR of the global wave via the more widely distributed RyR2along extra-junctional SR, once the Ca release wave escapes an RyR3-enriched SR region,as previously suggested in [11].In this respect, our results also allowed us to establish a time interval for the duration ofthe transient ( t transient ) that would be compatible with the hypothesis that sequential sum-mation of all-or-none lysosomal Ca release events from a set of individual L-SR junctionsis responsible for generating ∆[Ca ] shoulder . Remarkably, this value is comparable to theaverage duration of the simulated transients, which was determined on the basis of ultra-structural details and completely independently of the summation effect hypothesis. Theassumption of summation may imply that the role of the NAADP as a stimulus is limited tothe initial lysosomal Ca release, while the recruitment of subsequent L-SR junctions maybe governed by further lysosomal calcium release, by SR Ca release via Ca activatedjunctional RyR3, and/or propagation and combination of calcium signals via inter-junctionalclusters of RyR3. Therefore, from our model we envision the intriguing possibility of an im-portant regulatory role of lysosomal Ca content by SR Ca , in such a way that summationof calcium signals at multiple junctional complexes may give rise to the shoulder. While onlyfurther experiments can verify this, the quantitative corroboration provided by our findingssets on firmer ground the conclusion from earlier studies that Ca bursts from lysosomes areindeed responsible for initiating the first phase of a cell wide Ca wave via L-SR junctions. Carrying the signal time-scale analysis further and using the calculated t transient to take atemporal average of the simulated [Ca ] NJ at different junctional widths, we find that abovea width of about 30 nm the transients would be unable to trigger Ca release from the RyR3s(figure 7B). Thus it is possible that heterogeneity and plasticity are controlled by a variablewidth of the junctional nanospace. For example, in atrial myocytes it has been proposed thatNAADP evokes Ca release from an acidic store, which enhances general SR Ca releaseby increasing SR Ca load and activating RyR sites [34]. This functional variant may beprovided by either: (1) An increase in junctional distance such that [Ca ] NJ is insufficientto breach the threshold for activation of RyR2, yet sufficient to allow for increases in luminal Ca load of the SR via apposing SERCA2 clusters; or (2) L-SR junctions in cardiac muscleformed between lysosome membranes and closely apposed regions of the SR which possessdense SERCA2 clusters and are devoid of RyR2. Further ultrastructural studies on cardiacmuscle and other cell types may therefore provide a greater understanding of how L-SRjunctions may have evolved to provide for cell-specific modalities within the calcium signalingmachinery.Under metabolic stresses, such as hypoxia, lysosomal pathways have been proposed toprovide for autophagic glycogen metabolism via acid maltase in support of energy supply,and significant levels of protein breakdown during more prolonged metabolic stress. More-over, although controversial, in some cell types it has also been suggested that lysosomesmay contribute to the energy supply by providing free fatty acids for beta-oxidation by mi-tochondria [66]. It may be significant, therefore, that TPC2 gating and thus autophagymay be modulated by mTOR [31, 67]. This may well speak to further roles for lysosomalcalcium signaling and L-SR junctional plasticity during hypoxic pulmonary vasoconstric-tion and even during the development of hypoxia pulmonary hypertension [68], not leastbecause AMP-activated protein kinase has been shown to modulate autophagy through thephosphorylation and inhibition of mTOR (see for example, [69]).The process of autophagy serves not only to regulate programmed cell death, but alsoto recycle organelles, such as mitochondria, through a process of degradation involving lyso-somal hydrolases [70]. For this to occur L-SR junctions would likely be disrupted and thusselect for local rather than global Ca signals in order to facilitate fusion events amonglysosomes, endosomes, autophagosomes and amphisomes [71], which follows from the factthat Ca plays a pivotal role in vesicle trafficking and fusion [72]. Active non-synchronousmovements of TPC2-expressing vesicles have been detected in live-cell imaging experimentsusing GFP-tagged proteins [20]. Thus, lysosomal Ca signaling may modulate plasticityin the manner required at all stages of multiple membrane fusion events that are dependenton Ca for the effective formation of the SNARE complexes [72]. In short, transport ofproteins between lysosomes, Golgi apparatus, and plasma membrane via lysosomes and en-dosomes may be coordinated at different stages by both global Ca signals and spatiallyrestricted Ca release from acidic stores in a manner that is in some way determined byL-SR junctional integrity.Loss of integrity of L-SR junctions may also contribute to disease, given that lysosomalCa release both depletes luminal Ca and causes intraluminal alkalinization [73]. There-fore, disruption of L-SR junctions may modulate the activity of pH-sensitive hydrolyticlysosomal enzymes, such as glucocerebrosidase and acid sphingomyelinase, which exhibita marked loss of function at pH > We have determined that L-SR junctions, about 15 nm in width and extending to ≈ signals and strong support for the proposal that these Ca signals are generated at L-SR junctions. Even within thevariability of our recorded values of L-SR junctional widths and extension, our results suggestthat localized [Ca ] transients due to junctional Ca release can without fail reach valuesrequired to breach the threshold for CICR from junctional RyR3s. Perhaps most significantly,disruption of the nanojunctions decreases [Ca ] NJ below the value for CICR via junctionalRyR3s. Therefore, consistent with previous studies on the PM-SR membrane [39], we haveestablished that L-SR junctions are required to allow such signals to be generated and thatthere is a 30–50 nm limit on junctional width, above which there is loss of junctional integrityand inadequate control of ion movements within the junctional space. This suggests that theobserved L-SR junctions in PAMSCs are not only capable of delivering localized Ca burstsof the required magnitude, but are also necessary if lysosomes are to fulfill this identifiedrole in Ca signaling. In other words, L-SR nanojunctions are a necessary and sufficientcondition for generating local Ca bursts essential for NAADP-induced Ca waves. Mostimportantly, however, this study demonstrates the importance of junctional architecture onthe nanoscale to the capacity for coupling across cytoplasmic nanospaces, tight regulationof ion transport and thus signal transduction. In turn, this highlights the possibility thatalterations in the dimensions and architecture of intracellular nanojunctions lead to celldysfunction and hence disease. All the experiments and procedures were carried out in accordance with the guidelines of theUniversity of British Columbia Animal Care Committee and in accordance with the UnitedKingdom Animals (Scientific Procedures) Act 1986.
Male Wistar rats weighing 220–250 g were anesthetized with 3 mL of Thiotal. The thoraciccavity was opened and flooded with warm physiological saline solution (PSS) containing 145mM NaCl, 4 mM KCl, 1 mM MgCl , 10 mM HEPES, 0.05 mM CaCl , and 10 mM glucose(pH 7.4). After extraction of the heart and lungs and their placement in HEPES buffer,rings from the primary and secondary branches of the pulmonary artery were dissectedand immediately immersed in fixative solution. The primary fixative solution contained2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer at room temperature. The arteryrings were then washed three times in 0.1 M sodium cacodylate (30 min in total). In theprocess of secondary fixation, the tissue rings were fixed with 1% OsO in 0.1 M sodiumcacodylate buffer for 1 h followed by three 10-minute washes with distilled water and enbloc staining with 2% uranyl acetate. The samples were then dehydrated in increasingconcentrations of ethanol (25, 50, 75, 80, 90, and 95%). In the final process of dehydration,the samples underwent 3 washes in 100% ethanol. The artery rings were then resin-infiltratedin increasing concentrations (30, 50, and 75% in ethanol) of a 1:1 mix of Epon and Spurr’sresins. The infiltration process was completed by three passages in 100% resin. All ofthe ethanol dehydration and resin infiltration steps were carried out by using a laboratorymicrowave oven. The blocks were finally resin-embedded in molds and polymerized overnightin an oven at 60 ◦ C. For standard (2D) electron microscopy imaging, 80-nm sections were cut from the embeddedsample blocks on a Reichert Ultracut-E microtome using a diamond knife and were collectedon uncoated 100- and 200-mesh copper grids (hexagonal or square meshes). The sectionswere post-stained with 1% uranyl acetate and Reynolds lead citrate for 12 and 6 minutes,respectively. Electron micrographs at various magnifications were obtained with a Hitachi7600 transmission electron microscope at 80 kV.Lysosomes in these images were identified according to their well established appearancefeatures: they are single lipid bilayer membrane-bound, with a granular, more or less uniformluminal matrix that is more electron dense than the surrounding cytosol. Secondary lyso-somes may also contain less granular structures within the finer matrix. Moreover, lysosomesare normally distinguishable from endosomes by their larger size, hence we set a threshold“diameter” of >
200 nm for acceptance of a lysosome, below which all vesicles were excluded.
To obtain electron microscopic tomograms, we cut 200-nm-thick sections from the samesample blocks used for standard imaging. The sections were then collected on Formvar coatedslot copper grids and post-stained with 1% uranyl acetate and Reynolds lead citrate for 20and 10 minutes, respectively. We surveyed the sample sections using a FEI Tecnai G2 200 kVtransmission electron microscope and identified regions of interest containing L-SR junctions.We then acquired tomograms of several of those regions by taking 2D scans through thesample every 5 ◦ of inclination as it was tilted between − ◦ and +60 ◦ with respect tohorizontal. The scans were reconstructed with the Inspect3D software tools and structuresof interest, primarily lysosomal and SR membranes in the same cellular neighbourhood, weretraced out using Amira software. The images of the samples were further processed using GIMP (GNU Imaging ManipulationProgram, open source, available at gimp.org) to enhance membrane contrast in the interestof improving our characterization of the L-SR junctions.The SR and lysosomal membranes were outlined, highlighted and measured in pixelsusing the Inkscape software (open source, available at inkscape.org), converting the pixelmeasurements to nm using the scale bar in the recorded micrographs. By modifying theInkscape script for measuring lengths, we were able to output the measurements directlyinto a text file, which we used to produce the histograms in Figure 3B–D. We used thepackage Gnuplot (open source, available at gnuplot.info) to produce the histograms andplots presented herein. imaging Single arterial smooth muscle cells were isolated from second-order branches of the pulmonaryartery. Briefly, arteries were dissected out and placed in low Ca solution of the followingcomposition (mM): 124 NaCl, 5 KCl, 1 MgCl , 0.5 NaH PO , 0.5 KH PO, 15 NaHCO , 0.16CaCl , 0.5 EDTA, 10 glucose, 10 taurine and 10 Hepes, pH 7.4. After 10 min the arterieswere placed in the same solution containing 0.5 mg/ml papain and 1 mg/ml bovine serumalbumin and kept at 4 ◦ C overnight. The following day 0.2 mM 1,4-dithio-DL-threitol wasadded to the solution, to activate the protease, and the preparation was incubated for 1 h at room temperature (22 ◦ C). The tissue was then washed at 3 ◦ C in fresh low Ca solutionwithout enzymes, and single smooth muscle cells were isolated by gentle trituration with afire-polished Pasteur pipette. Cells were stored in suspension at 4 ◦ C until required.PASMCs were incubated for 30 min with 5 µ M Fura-2-AM in Ca -free PSS in anexperimental chamber on a Leica DMIRBE inverted microscope and then superfused withFura-2 free PSS for at least 30 min prior to experimentation. Intracellular Ca concentrationwas reported by Fura-2 fluorescence ratio (F340/F380 excitation; emission 510 nm). Emittedfluorescence was recorded at 22 ◦ C with a sampling frequency of 0.5 Hz, using a Hamamatsu4880 CCD camera via a Zeiss Fluar 40 × , 1.3 n.a. oil immersion lens and Leica DMIRBEmicroscope. Background subtraction was performed on-line. Analysis was done via Openlabimaging software (Improvision, UK).NAADP was applied intracellularly in the whole-cell configuration of the patch-clamptechnique, and in current clamp mode ( I = 0) as described previously [17]. The pipettesolution contained (in mM): 140 KCl, 10 Hepes, 1 MgCl and 5 µ M Fura-2, pH 7.4. The sealresistance, as measured using an Axopatch 200B amplifier (Axon Instruments, Foster City,CA), was ≥ ≤
10 MΩ and ≤ ≈ ◦ C).
The main stages of the quantitative modeling approach are:1. the design of 3D software mesh objects (nets of interconnected triangles by which sur-faces can be represented in computer graphics) representing a typical L-SR region,including a whole lysosome and a portion of neighbouring SR network. These ob-jects are built to-scale following the ultrastructural characterization of the L-SR junc-tional regions as it results from our electron microscopy image analysis; this phase wascarried out using the “3D content creation suite” Blender (open source, available atblender.org);2. the positioning of the relevant transporters on the reconstructed membranes (TPC2complexes on the lysosome, SERCA2a and RyR3 on the SR) according to informationgathered from the literature on their typical membrane densities and the implementa-tion of the transporter known kinetics and multistate models and of the ion diffusivities(Ca and mobile Ca buffers);3. the simulation of molecular Brownian motion in the cytosol by random walk algorithms;this phase was performed by writing appropriate code for the stochastic particle simu-lator MCell (freely available at mcell.org) [35–37]. In a nutshell, MCell reproduces therandomness of the molecular trajectories, of the ion transporter flickering and of therelevant chemical reactions by probabilistic algorithms governed by random numbergenerators (iterative mathematical algorithms, which produce a random sequence ofnumbers once initiated by a given number called seed). This enables the simulation ofa number of microphysiological processes, all stochastically different from one another.The average outcome of the processes, to mimic the instrumental output during ex-perimental measurements, is obtained by taking the average of a desired quantity, e.g.,[Ca ], over a large number of simulations all initiated with a different seed;4. the measurement of simulated [Ca ] in the L-SR junctions from the process of Ca release via TPC2-related signaling complex on the lysosome and SR Ca uptake bythe SERCA2a pumps, and static as well as dynamic visualization of the simulations;this stage is part of MCell’s output. Acknowledgements
We are very grateful to Garnet Martens and the University of British Columbia BioimagingFacilty for their assistance. We also acknowledge the help of David Walker and ArashTehrani with TEM image analysis and acquisition. This research has been enabled by theuse of computing resources provided by WestGrid and Compute/Calcul Canada (westgrid.ca,computecanada.ca).
References [1] Van Breemen C (1977) Calcium requirement for activation of intact aortic smooth mus-cle. J Physiol 272: 317-29.[2] van Breemen C, Saida K, Yamamoto H, Hwang K, Twort C (1988) Vascular smoothmuscle sarcoplasmic reticulum. function and mechanisms of ca2+ release. Ann N YAcad Sci 522: 60-73.[3] Malmersj¨o S, Meyer T (2013) Inside-out connections: the er meets the plasma mem-brane. Cell 153: 1423-4.[4] van Breemen C, Fameli N, Evans AM (2013) Pan-junctional sarcoplasmic reticulum invascular smooth muscle: nanospace ca2+ transport for site- and function-specific ca2+signalling. J Physiol 591: 2043-54.[5] Moore EDW, Wasteneys GO (2012) Nanospace biophysics. editorial. Protoplasma 249Suppl 1: S1.[6] Carrasco S, Meyer T (2011) Stim proteins and the endoplasmic reticulum-plasma mem-brane junctions. Annu Rev Biochem 80: 973-1000.[7] (2010) Nanospace biophysics exploratory workshop — peter wall institute for advancedstudies. URL .[8] (2011) Vascular and smooth muscle physiology themed meeting, nanojunctions andcalcium signalling in smooth muscle cells symposium. Univeristy of Edinburgh, ProcPhysiol Soc, vol 25.[9] (2013) Nanojunctions and calcium signalling in smooth muscle cells: From con-traction and relaxation to gene expression — iups 2013. URL .[10] Kinnear NP, Boittin FX, Thomas JM, Galione A, Evans AM (2004) Lysosome-sarcoplasmic reticulum junctions. a trigger zone for calcium signaling by nicotinic acidadenine dinucleotide phosphate and endothelin-1. The Journal of biological chemistry279: 54319–54326.[11] Kinnear NP, Wyatt CN, Clark JH, Calcraft PJ, Fleischer S, et al. (2008) Lysosomes co-localize with ryanodine receptor subtype 3 to form a trigger zone for calcium signallingby NAADP in rat pulmonary arterial smooth muscle. Cell calcium 44: 190–201.[12] Lee HC, Aarhus R (1995) A derivative of nadp mobilizes calcium stores insensitive toinositol trisphosphate and cyclic adp-ribose. J Biol Chem 270: 2152-7.[13] Aarhus R, Dickey DM, Graeff RM, Gee KR, Walseth TF, et al. (1996) Activation andinactivation of ca2+ release by naadp+. J Biol Chem 271: 8513-6.[14] Lee HC, Aarhus R (2000) Functional visualization of the separate but interacting cal-cium stores sensitive to naadp and cyclic adp-ribose. J Cell Sci 113 Pt 24: 4413-20. [15] Cancela JM, Churchill GC, Galione A (1999) Coordination of agonist-induced ca2+-signalling patterns by naadp in pancreatic acinar cells. Nature 398: 74-6.[16] Churchill GC, Galione A (2001) NAADP induces ca2+ oscillations via a two-pool mech-anism by priming IP3- and cADPR-sensitive ca2+ stores. The EMBO journal 20: 2666–2671.[17] Boittin FX, Galione A, Evans AM (2002) Nicotinic acid adenine dinucleotide phos-phate mediates ca2+ signals and contraction in arterial smooth muscle via a two-poolmechanism. Circ Res 91: 1168-75.[18] Churchill GC, Okada Y, Thomas JM, Genazzani AA, Patel S, et al. (2002) NAADPmobilizes ca(2+) from reserve granules, lysosome-related organelles, in sea urchin eggs.Cell 111: 703–708.[19] Yamasaki M, Masgrau R, Morgan AJ, Churchill GC, Patel S, et al. (2004) Organelleselection determines agonist-specific ca2+ signals in pancreatic acinar and beta cells.The Journal of biological chemistry 279: 7234–7240.[20] Calcraft PJ, Ruas M, Pan Z, Cheng X, Arredouani A, et al. (2009) Naadp mobilizescalcium from acidic organelles through two-pore channels. Nature 459: 596-600.[21] Schneider PB, Kennedy EP (1967) Sphingomyelinase in normal human spleens and inspleens from subjects with niemann-pick disease. J Lipid Res 8: 202-9.[22] Lloyd-Evans E, Morgan AJ, He X, Smith DA, Elliot-Smith E, et al. (2008) Niemann-pickdisease type c1 is a sphingosine storage disease that causes deregulation of lysosomalcalcium. Nat Med 14: 1247–1255.[23] Tassoni JP Jr, Fawaz KA, Johnston DE (1991) Cirrhosis and portal hypertension in apatient with adult niemann-pick disease. Gastroenterology 100: 567-9.[24] Noori S, Acherman R, Siassi B, Luna C, Ebrahimi M, et al. (2002) A rare presentationof pompe disease with massive hypertrophic cardiomyopathy at birth. J Perinat Med30: 517-21.[25] Jmoudiak M, Futerman AH (2005) Gaucher disease: pathological mechanisms and mod-ern management. Br J Haematol 129: 178-88.[26] Carstea ED, Morris JA, Coleman KG, Loftus SK, Zhang D, et al. (1997) Niemann-pickc1 disease gene: homology to mediators of cholesterol homeostasis. Science 277: 228-31.[27] Ron I, Horowitz M (2008) Intracellular cholesterol modifies the erad of glucocerebrosi-dase in gaucher disease patients. Mol Genet Metab 93: 426-36.[28] Shao D, Park JES, Wort SJ (2011) The role of endothelin-1 in the pathogenesis ofpulmonary arterial hypertension. Pharmacol Res 63: 504-11.[29] Davie NJ, Schermuly RT, Weissmann N, Grimminger F, Ghofrani HA (2009) The scienceof endothelin-1 and endothelin receptor antagonists in the management of pulmonaryarterial hypertension: current understanding and future studies. Eur J Clin Invest 39Suppl 2: 38-49.[30] Wang X, Zhang X, Dong XP, Samie M, Li X, et al. (2012) Tpc proteins arephosphoinositide- activated sodium-selective ion channels in endosomes and lysosomes.Cell 151: 372-83.[31] Cang C, Zhou Y, Navarro B, Seo YJ, Aranda K, et al. (2013) mtor regulates lysosomalatp-sensitive two-pore na(+) channels to adapt to metabolic state. Cell 152: 778-90. [32] Morgan AJ, Davis LC, Wagner SKTY, Lewis AM, Parrington J, et al. (2013) Bidirec-tional ca signaling occurs between the endoplasmic reticulum and acidic organelles. JCell Biol 200: 789-805.[33] Kilpatrick BS, Eden ER, Schapira AH, Futter CE, Patel S (2012) Direct mobilisationof lysosomal ca2+ triggers complex ca2+ signals. J Cell Sci .[34] Collins TP, Bayliss R, Churchill GC, Galione A, Terrar DA (2011) Naadp influencesexcitation-contraction coupling by releasing calcium from lysosomes in atrial myocytes.Cell Calcium 50: 449–458.[35] Stiles JR, Van Helden D, Bartol J T M, Salpeter EE, Salpeter MM (1996) Miniatureendplate current rise times less than 100 microseconds from improved dual recordingscan be modeled with passive acetylcholine diffusion from a synaptic vesicle. Proc NatlAcad Sci USA 93: 5747–5752.[36] Stiles JR, Bartol TM (2001) Monte carlo methods for simulating realistic synaptic mi-crophysiology using MCell. In: Computational Neuroscience: Realistic Modeling forExperimentalists, CRC Press. pp. 87–127.[37] Kerr RA, Bartol TM, Kaminsky B, Dittrich M, Chang JCJ, et al. (2008) FAST MONTECARLO SIMULATION METHODS FOR BIOLOGICAL REACTION-DIFFUSIONSYSTEMS IN SOLUTION AND ON SURFACES. SIAM J Sci Comput 30: 3126.[38] Agbani EO, Ogunbayo OA, Parrington J, Galione A, Ma J, et al. (2011) Nicotinicacid adenine dinucleotide phosphate evokes global calcium signals in mouse pulmonaryarterial smooth muscle cells by activating two pore segment channel 2. Proc PhysiolSoc 25, SA06, volume 25.[39] Fameli N, van Breemen C, Kuo KH (2007) A quantitative model for linking Na+/Ca2+exchanger to SERCA during refilling of the sarcoplasmic reticulum to sustain [ca2+]oscillations in vascular smooth muscle. Cell Calcium 42: 565–575.[40] Rhodin JAG (1974) Histology. Oxford University Press.[41] Robinson JM, Okada T, Castellot JJ Jr, Karnovsky MJ (1986) Unusual lysosomes inaortic smooth muscle cells: presence in living and rapidly frozen cells. J Cell Biol 102:1615-22.[42] Ovalle WK (2013) Netter’s Essential Histology. Elsevier/Saunders.[43] URL .[44] Holtzman E (1989) Lysosomes. Plenum Press.[45] Christensen KA, Myers JT, Swanson JA (2002) ph-dependent regulation of lysosomalcalcium in macrophages. J Cell Sci 115: 599-607.[46] Pitt SJ, Funnell TM, Sitsapesan M, Venturi E, Rietdorf K, et al. (2010) Tpc2 is anovel naadp-sensitive ca2+ release channel, operating as a dual sensor of luminal phand ca2+. J Biol Chem 285: 35039–35046.[47] Somlyo AP (1975) Structural characteristics, mechanisms of contraction, innervationand proliferation of smooth muscle cells. ultrastructure and function of vascular smoothmuscle. Adv Exp Med Biol 57: 1-80.[48] Clark JH, Kinnear NP, Kalujnaia S, Cramb G, Fleischer S, et al. (2010) Identificationof functionally segregated sarcoplasmic reticulum calcium stores in pulmonary arterialsmooth muscle. The Journal of biological chemistry 285: 13542–13549. [49] Schieder M, R¨otzer K, Br¨uggemann A, Biel M, Wahl-Schott CA (2010) Characterizationof two-pore channel 2 (tpcn2)-mediated ca2+ currents in isolated lysosomes. J BiolChem 285: 21219–21222.[50] Koivusalo M, Steinberg BE, Mason D, Grinstein S (2011) In situ measurement of theelectrical potential across the lysosomal membrane using fret. Traffic 12: 972-82.[51] L¨auger P (1991) Electrogenic Ion Pumps. Sunderland, Mass., USA: Sinauer Associates.[52] Dode L, Andersen JP, Leslie N, Dhitavat J, Vilsen B, et al. (2003) Dissection of thefunctional differences between sarco(endo)plasmic reticulum ca2+-atpase (serca) 1 and2 isoforms and characterization of darier disease (serca2) mutants by steady-state andtransient kinetic analyses. J Biol Chem 278: 47877-89.[53] Raeymaekers L, Vandecaetsbeek I, Wuytack F, Vangheluwe P (2011) Modeling ca2+dynamics of mouse cardiac cells points to a critical role of serca’s affinity for ca2+.Biophys J 100: 1216-25.[54] Allbritton NL, Meyer T, Stryer L (1992) Range of messenger action of calcium ion andinositol 1,4,5-trisphosphate. Science 258: 1812–1815.[55] Allbritton NL, Meyer T (1993) Localized calcium spikes and propagating calcium waves.Cell Calcium 14: 691–697.[56] Keller DX, Franks KM, Bartol TM Jr, Sejnowski TJ (2008) Calmodulin activation bycalcium transients in the postsynaptic density of dendritic spines. PLoS One 3: e2045.[57] Lee CH, Kuo KH, Dai J, Leo JM, Seow CY, et al. (2005) Calyculin-a disrupts subplas-malemmal junction and recurring ca2+ waves in vascular smooth muscle. Cell Calcium37: 9-16.[58] Dai J, Kuo KH, Leo JM, van Breemen C, Lee CH (2005) Rearrangement of the closecontact between the mitochondria and the sarcoplasmic reticulum in airway smoothmuscle. Cell Calcium 37: 333-40.[59] Takeshima H, Yamazawa T, Ikemoto T, Takekura H, Nishi M, et al. (1995) Ca(2+)-induced ca2+ release in myocytes from dyspedic mice lacking the type-1 ryanodinereceptor. EMBO J 14: 2999-3006.[60] Lam AKM, Galione A (2013) The endoplasmic reticulum and junctional membranecommunication during calcium signaling. Biochim Biophys Acta 1833: 2542-59.[61] van Breemen A Mark Evans NFC (2013) Lysosome-sarcoplasmic reticulum junctions:trigger zones for calcium waves in vascular smooth muscle. Biophys J, vol 105, issue 6,volume 105.[62] McCue HV, Wardyn JD, Burgoyne RD, Haynes LP (2013) Generation and character-ization of a lysosomally targeted, genetically encoded ca(2+)-sensor. Biochem J 449:449-57.[63] Drago I, De Stefani D, Rizzuto R, Pozzan T (2012) Mitochondrial ca2+ uptake con-tributes to buffering cytoplasmic ca2+ peaks in cardiomyocytes. Proc Natl Acad Sci US A 109: 12986-91.[64] Csord´as G, V´arnai P, Golen´ar T, Roy S, Purkins G, et al. (2010) Imaging interorganellecontacts and local calcium dynamics at the er-mitochondrial interface. Mol Cell 39:121-32.[65] Zhu MX, Ma J, Parrington J, Calcraft PJ, Galione A, et al. (2010) Calcium signalingvia two-pore channels: local or global, that is the question. Am J Physiol Cell Physiol298: C430-41.5