the position of mitochondria and er in relation to that of ... · mitochondria were identified by...
TRANSCRIPT
Jour
nal o
f Cel
l Sci
ence
RESEARCH ARTICLE
The position of mitochondria and ER in relation to that of thesecretory sites in chromaffin cells
Jose Villanueva1, Salvador Viniegra1, Yolanda Gimenez-Molina1, Virginia Garcıa-Martinez1,Giovanna Exposito-Romero1, Maria del Mar Frances1, Javier Garcıa-Sancho2 and Luis M. Gutierrez1,*
ABSTRACT
Knowledge of the distribution of mitochondria and endoplasmic
reticulum (ER) in relation to the position of exocytotic sites is
relevant to understanding the influence of these organelles in tuning
Ca2+ signals and secretion. Confocal images of probes tagged to
mitochondria and the F-actin cytoskeleton revealed the existence of
two populations of mitochondria, one that was cortical and one that
was perinuclear. This mitochondrial distribution was also confirmed
by using electron microscopy. In contrast, ER was sparse in the
cortex and more abundant in deep cytoplasmic regions. The
mitochondrial distribution might be due to organellar transport,
which experiences increasing restrictions in the cell cortex. Further
study of organelle distribution in relation to the position of SNARE
microdomains and the granule fusion sites revealed that a third
of the cortical mitochondria colocalized with exocytotic sites
and another third located at a distance closer than two vesicle
diameters. ER structures were also present in the vicinity of
secretory sites but at a lower density. Therefore, mitochondria and
ER have a spatial distribution that suggests a specialized role in
modulation of exocytosis that fits with the role of cytosolic Ca2+
microdomains described previously.
KEY WORDS: Exocytosis, Chromaffin cell, Mitochondria,
Endoplasmic reticulum, Confocal microscopy
INTRODUCTIONAdrenomedullar chromaffin cells are widely used as a model to
study exocytosis of dense vesicles. After stimulation of the
splacnic nerve, these cells release catecholamines in response to
the elevation of the cytosolic Ca2+ concentration ([Ca2+]c) in a
process involving the transport of granules, translocation to the
plasma membrane, docking at the secretory sites and fusion of
membranes with extrusion of soluble contents (Burgoyne et al.,
1993). Major cellular structures, such as the cortical cytoskeleton,
play fundamental roles in different stages of the secretory cascade
(Gutierrez, 2012), whereas organelles, such as the endoplasmic
reticulum (ER) and mitochondria, control and shape [Ca2+]c
elevations at the subplasmalemmal region (Garcıa et al., 2006;
Garcıa-Sancho et al., 2012). In addition, chromaffin granules
could be acting as a Ca2+ reservoir influencing vesicular transport
and exocytosis (Camacho et al., 2008). According to the actual
vision, voltage-operated Ca2+ channels (VOCCs) of the plasma
membrane trigger fast Ca2+ elevations in the cytosol that are
amplified by the ER, thus generating high-Ca2+ microdomains in
the proximity of secretory sites (Garcıa et al., 2006). Upon
exocytosis, nearby mitochondria take up this Ca2+, thus
restricting the levels of the messenger that reach the cell core
(Villalobos et al., 2002). Therefore, triads composed of plasma
membrane channels, ER and mitochondria optimize Ca2+ signals
in a subtle way (Montero et al., 2000; Garcıa et al., 2006; Garcıa-
Sancho, 2012) and avoid large elevations of Ca2+ in the cell core,
which could otherwise trigger cell excitotoxicity (Duchen, 2012).
This view of the role of subcellular organelles in exocytosis is
mostly based on measurements of Ca2+ concentrations in the
interior of organelles and secretory behaviour during activation of
chromaffin cells (Nunez et al., 2007). However, there is a lack of
structural studies that have directly assessed association between
the location of secretory sites and organelle distribution.
In the present work, by using high-magnification confocal
microscopy, we have characterized the populations of
mitochondria and ER elements in cultured bovine chromaffin
cells in relation to their distances to the secretory apparatus or to
the exocytotic sites. Our results support the idea of functional
triads proposed previously and, in addition, we find distinct
subpopulations of organelles located in the immediate vicinity of
secretory sites, which are in position to generate both the
localized Ca2+ signals and the energy needed to sustain
exocytosis.
RESULTSMitochondria and ER distribute differently in chromaffin cellsIn order to study mitochondria distribution, cultured bovine
chromaffin cells expressing RFP-tagged LifeAct, a 17-amino-
acid peptide that binds to F-actin but does not interfere with
cytoskeletal dynamics (Riedl et al., 2008), were also labeled with
Mitotracker Green and visualized by confocal microscopy with a
high-numeric aperture objective (1006; NA 1.45). Sequential
slices were obtained every 0.3 mm along the z-axis from the cell
limits, marked by the presence of the cortical F-actin (left image
in Fig. 1A) to the equatorial plane (right image in Fig. 1A; 11
images taken along 3 mm). Mitochondria (in green) were
abundant in the cortical zone, less were present in the zone
defined by the middle images and their abundance increased
again in the region adjacent to the nucleus. The existence of these
two different populations was further supported when averaging
the normalized fluorescence intensity in experiments performed
in many cells, as depicted in Fig. 1B. From these measurements
the presence of a peripheral population of mitochondria present in
the cortical region (0–1 mm of the cell limit) is evident, and an
1Instituto de Neurociencias, Centro Mixto Universidad Miguel Hernandez-CSIC,Cra de Valencia S/N, Sant Joan d’Alacant, 03550 Alicante, Spain. 2Instituto deBiologıa y Genetica Molecular (IBGM), Universidad de Valladolid and CSIC, c/Sanz y Fores, 3, 47003 Valladolid, Spain.
*Author for correspondence ([email protected])
Received 22 July 2014; Accepted 25 September 2014
� 2014. Published by The Company of Biologists Ltd | Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5105
Jour
nal o
f Cel
l Sci
ence
internal, perinuclear population, marked by increasing Mitotrackerfluorescence 2–3 mm deeper than the plasma membrane, is alsoapparent. Using a different approach, the number of mitochondria
was counted after assigning particles that were above a thresholdlevel and placing them in groups according to their distance fromthe plasma membrane (0–1, 1–2, and 2–3 mm; Fig. 1C). Again the
presence of a cortical population accounting for 40% of the totalmitochondria and a perinuclear population of a similar percentagewas apparent. The region between these two populations had a
smaller proportion of mitochondria (less than 17% of total;Fig. 1C). The total fluorescence intensity values shown in Fig. 1B,also gave very similar percentage figures to those for the number ofparticles: 38% in the 0–1 mm z-range, 15% in the 1–2-mm region
and 47% in the 2–3-mm region. The distribution of mitochondriacontrasts with the distribution of RFP–LifeAct fluorescence, whichdecreases monotonically from the cortical area to the equatorial
area (Fig. 1B).As an alternative method, the distribution of mitochondria was
analyzed in equatorial images displaying the nuclear position, as
shown in Fig. 2. In this case, the usual position of the nucleus inone pole of the cell provided evidence that a population ofmitochondria was positioned between the nucleus and the closest
plasma membrane, and therefore could be considered as bothcortical and perinuclear (CP, Fig. 2D). Estimation of thispopulation in several cells, based on fluorescence determination,provided evidence that CP mitochondria made up 12%61
(6s.e.m.) of the total mitochondria (Fig. 2E), Therefore, themitochondria population associated with the cellular cortex wasdetermined as being between 28% and 38% depending on the
criteria used (whether CP mitochondria were assigned to theperinuclear or to the cortical population, respectively).
A similar approach was used to investigate the distribution of
ER, which was labeled with ERtracker Green. In this case, we
observed a much more uniform distribution in the z-plane, with aslight increase in the ER fluorescence intensities when movingfrom the external cortical region to the internal perinuclear area
(Fig. 3A–C). Averaging ER fluorescence intensities (Fig. 3D) orcounting discrete ER particles (Fig. 3E) showed that the amountof ER increased gradually from the cortical zone to the cell
interior to reach its maximal value in the regions adjacent to thenucleus. As above, LifeAct fluorescence decreased monotonicallywith displacement towards the equatorial region (Fig. 3D).
In a morphometric analysis, the size and density ofmitochondria and ER elements present in the perinuclear andthe cortical areas were quantified and compared (Table 1). Bothorganelles were significantly smaller in size in the cortical region
than in the perinuclear region (see Area column). Interestingly,the density of mitochondria was higher in the cortical area,whereas the ER density appeared to be similar in the two regions.
Therefore, the ER and mitochondrial distributions were alsoshown to be markedly different using this approach: ERdistribution was relatively homogeneous, whereas mitochondria
showed two different subpopulations.The existence of different mitochondrial populations in
chromaffin cells was further characterized using electron
microscopy (supplementary material Fig. S1A,B). In this case,mitochondria were identified by observation of the characteristicinternal cristae (supplementary material Fig. S1C) and its size anddistance to the plasma membrane studied in ten cells.
Interestingly, this experiment provided evidence that the twopopulations of mitochondria with cortical and perinuclearlocalization each comprised 30–40% of the total number of
mitochondria (supplementary material Fig. S1D). Again, theseestimations depend on the size of the pool of mitochondria thatare assigned as being both perinuclear and cortical. This CP
population was found to constitute ,15% of the total.
Fig. 1. Two populations of mitocondria are detected by confocal microscopy. (A) Representative confocal images from experiments performed in culturedchromaffin cells expressing RFP–LifeAct (red channel) and labeled with 100 nM Mitotracker Green (green channel). Images of five confocal planes separated by0.75 mm, starting in the basal cortical region (left) and ending in the equatorial plane (right) are shown. Scale bar: 3 mm. (B) Mean6s.e.m. values of normalizedfluorescence for the different confocal planes determined for RFP–LifeAct (n5104 images) and Mitotracker fluorescence (n5135 images), representedas a function of the distance measured from basal cortical to the equatorial plane (3 mm in total). (C) Percentage (mean6s.e.m.) of measured mitochondrialparticles with respect to the total population, corresponding to the cortical population, extending from 0 to 1 mm from the cortical plane (41%64 of the totalparticles), the intermediate population, located at between 1 and 2 mm (17%64), and the perinuclear population, present at 2 to 3 mm from the celllimits (42%63).
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5106
Jour
nal o
f Cel
l Sci
ence
Fig. 2. Distribution of mitochondria in chromaffin cells with a polarized nucleus. (A) Representative confocal image from the equatorial plane of a cell withthe nucleus localized off-center (polarized nucleus). (B,C) Fluorescence profiles obtained from the same cell as A and showing that there are subpopulations ofcortical and perinuclear mitochondria separated by transition areas (labeled with 100 nM Mitotracker Green). (D) In order to determine the mitochondrialdensity in different part of the cells, fluorescence intensity was determined in the cortical (C, extending 1 mm from the cell limits), the transition (T) and theperinuclear zones (P). In one pole of the cell, the mitochondria can be considered as forming part of the cortical and perinuclear populations (CP population).(E) Fluorescence determined for the different populations of mitochondria, expressed as a percentage of the total fluorescence (mean6s.e.m. from experimentsperformed in 10 cells). Scale bars: 3 mm.
Fig. 3. Distribution of the ER incultured chromaffin cells.(A–C) Confocal fluorescence imagesdepicting the distribution of ERtrackerGreen (100 nM) in cells expressingRFP–LifeAct (red fluorescence).Shown are panels covering fiveconfocal planes taken every 0.75 mmfrom the basal cell cortex (left) to theequatorial perinuclear area (right) anddepicting, F-actin labeling (A), theERtracker distribution (B), andcombining both channels(C). (D) Mean6s.e.m. values ofnormalized fluorescence for thedifferent confocal planes determinedfor RFP–LifeAct (n5202 images), andERtracker (n5184 images)fluorescence, represented as afunction of the distance measuredfrom the cortical basal to the equatorialplane (3 mm in total). (E) Percentage(mean6s.e.m., n5184 images) of totalER particles calculated for the corticalpopulation extending from 0 to 1 mm(13%62 of the total ER particles), theintermediate population locatedbetween 1 and 2 mm (23%63), andthe perinuclear population present at 2to 3 mm from the cell limits (64%65).Scale bars: 3 mm.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5107
Jour
nal o
f Cel
l Sci
ence
Interestingly, the size estimated for the averaged perinuclear
mitochondria particle is 0.11660.004 mm2 (mean6s.e.m.;n5184), whereas this value decreased markedly for corticalmitochondria (0.04960.002 mm2, n5201).
The two mitochondrial subpopulations have differentmobilityWhy do chromaffin cells display two different subpopulations of
mitochondria? Distinctive transport of organelles and retention inthe cortical region could play a role, as previously proposed forvesicle distribution in chromaffin and PC12 cells (Neco et al.,
2003; Rudolf et al., 2003; Gutierrez, 2012). To test thishypothesis, we measured the motion of mitochondria and theER in different areas of the cytoplasm of chromaffin cells. Forthis purpose, particles were created by thresholding of images,
and the centroid of such particles was tracked at 1-s intervals in
experiments performed at room temperature (21–22 C). Motionwas computed from the slope of the plots of the mean squaredisplacement (MSD) against time for a large number ofmitochondria and ER particles, and the data were averaged to
generate Fig. 4D. In Fig. 4A–C, sequential images separated by5-s intervals are shown. We illustrate representative examples ofthe motion of mitochondria in the cell cortex (polar section;
Fig. 4A) or in an equatorial section deep in the cell interior(Fig. 4B). ER motion in or an equatorial section is also illustrated(Fig. 4C). Interestingly, the slopes of the averaged plots of MSD
against time (Fig. 4D) indicated that there was a highermitochondria mobility in the cell interior (perinuclear) ascompared with in the cortical region (cortical), suggesting thatmotion is slower when mitochondria approach the plasma
Table 1. Morphometry of mitocondria and ER structures in cultured bovine chromaffin cells
Structures Localization Density (particles/mm2) P-value Area (mm2) P-value Number of structures
Mitochondria Cortical 0.4660.05 P,0.005 0.2260.01 P,0.0001 79Perinuclear 0.2160.03 0.5160.06 69
ER structures Cortical 0.9460.17 P.0.05 0.2560.02 P,0.0001 104Perinuclear 0.8760.26 0.5760.04 104
Results are mean6s.e.m. P-values were calculated with a non-parametric Mann–Whitney test.
Fig. 4. Motion of mitochondria and ER in differentregions of chromaffin cells. Chromaffin cellsexpressing RFP–LifeAct were stained with eitherMitotracker Green or ERtracker Green as described inthe previous figures, and imaged at 1-s intervals using aconfocal microscope at room temperature (21–22˚C).(A,B) Sequence of representative images ofmitochondria separated by 5 s taken either at the polarcortical plane (A) or at the equatorial plane (B). Images ofER taken at the equatorial plane are shown in(C). Motion was analyzed, after creation of particles byusing an appropriate image threshold, by tracking ofparticle center and measuring displacement. (D) Plots(mean6s.e.m.) of MSD against time for a large numberor mitochondria (MITO, n5202 particles) and ERparticles (n5148) in both the cortical and the perinuclearregions. The plots indicate that ER particles were almostimmobile, whereas mitochondria presented faster motilityin the perinuclear than in the cortical region. Scale bars:3 mm. (E,F) Cells were incubated with 2 mMjasplakinolide (JASP) or 1 mM taxol for 30 min to studythe role of cytoskeletal systems in mitochondrial mobility.The apparent mobility coefficient (mean6s.e.m.) wasestimated as the slope of the plot of MSD against time(Qian et al., 1991, slope54D). In the perinuclear region,mitochondria dynamics is sensitive to F-actin inhibitorsand microtubule stabilizers, whereas in the corticalregion mitochondria motion seems to only be controlledby F-actin. **P,0.01, ***P,0.001 (one-way ANOVAKruskal–Wallis test when compared to control values).
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5108
Jour
nal o
f Cel
l Sci
ence
membrane. When compared to mitochondria, ER motion throughthe cell cytoplasm was extremely slow, both in the cortical and in
the perinuclear region (Fig. 4D).The mitochondrial transport seems to involve cytoskeletal
proteins, as the mobility coefficient in the perinuclear region(calculated from the slope of the MSD against time plot) was
reduced by incubation of the cells with the F-actin inhibitorjasplakinolide or with the microtubule stabilizer taxol (Fig. 4E).In contrast, in the cortical region, the motion of mitochondria
seems to be controlled only by F-actin given that taxol did notsignificantly influence the mobility coefficient (Fig. 4F).
To summarize, perinuclear mitochondria are transported
through F-actin and microtubular structures and experienceincreasing mobility restrictions when approaching the corticalregion, which is formed by a dense meshwork of F-actin. This
retention could be the mechanism responsible for stabilizing thelocation of a mitochondrial subpopulation in the vicinity of thesecretory sites.
A further demonstration of the retentive role of cortical F-actin
in stabilizing a mitochondrial population was obtained when wefound that Latrunculin A, an stabilizer of monomeric actin(Spector et al., 1983), affects the distribution of chromaffin cell
mitochondria, decreasing the cortical population and favoring themitochondria in transition (supplementary material Fig. S2).
The distribution of cortical mitochondria has a relationshipwith the location of the secretory machineryThe cortical mitochondria are best positioned to support the needsof energy supply and Ca2+ clearance in the proximity of thesecretory apparatus (Montero et al., 2000; Villalobos, 2002;Garcıa et al., 2006). In order to address this idea directly, we
expressed RFP–SNAP-25 in cultured chromaffin cells; thisprotein dimerizes with syntaxin-1 at the plasma membrane, andthe position of the dimers define the place of granule fusion
(Rickman et al., 2004; Lopez et al., 2009; Lopez-Font et al.,2010). After cell treatment with Mitotracker Green, the corticalregion was observed in polar images (see example in Fig. 4), and
the distances between the fusion machinery (RFP–SNAP-25microdomain labeling) and the nearest visualized mitochondria(the nearest pixel corresponding to mitochondria labeling) were
measured. Interestingly, 25–30% of the SNAP-25 microdomainscolocalized with mitochondria and therefore no distance could bemeasured under our optical resolution (Fig. 5A,B). Thedistribution of distances shown in Fig. 5B, suggested, in
addition, that there was a second subpopulation of mitochondriathat were located within 0.3 mm, the average diameter of achromaffin granule. Therefore, it seems that separate
subpopulations of mitochondria could be distinguished in thecortical region.
Fig. 5. Distances of corticalmitochondria and ER to thesecretory machinery. Cellsexpressing RFP–SNAP-25 wereincubated with Mitotracker Green orERtracker Green as described for theprevious figures. Then corticaldistributions where imaged using aconfocal microscope to study thedistances between the RFP–SNAP-25 microdomains and themitochondria and ER structures.Depicted are cortical polar confocalplanes of chromaffin cells showingmitochondria and SNAP-25microdomains (A) or ER and SNAP-25 microdomains (C). Scale bars:3 mm. These images where used tocalculate the distributions ofdistances (expressed asmean6s.e.m.) between SNAP-25clusters and the nearest mitochondria(B, 107 measurements from 10 cells)or ER structure (D, 118measurements from 17 cells) using a0.1-mm bin size.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5109
Jour
nal o
f Cel
l Sci
ence
The same approach was used to study the distribution of ER inrelation to the position of the secretory machinery (Fig. 5C).
Again, distances to the nearest RFP–SNAP-25 microdomain werecalculated. The distribution for a large number of ER particlesis shown in Fig. 5D. Approximately 20% of ER structurescolocalized with the secretory machinery patches, whereas a
second pool of cortical ER particles was found within the distanceof a chromaffin granule diameter (0.3 mm; Fig. 5D). Theexistence of these two pools is reminiscent of the mitochondrial
organization, suggesting that the same organizing influencesapply to mitochondria and ER.
Exocytotic sites are located in the immediate vicinity ofcortical mitochondriaA more direct procedure to evaluate the proximity of cortical
mitochondria and ER to secretory sites is to estimate thedistances between these organelles and the fusion sites, whichcan be labeled with anti-dopamine-b-hydroxylase (DBH)antibody because DBH incorporates into the plasma
membrane after exocytosis (Patzak et al., 1984; Lopez et al.,2007). In these experiments, chromaffin cells labeled witheither Mitotracker Red or ERtracker Red were stimulated by
cell depolarization for 1 min by perfusion with 59 mM KClsolution, Then DBH was immobilized in the membrane bysudden lowering of the temperature to 2 C prior to DBH
staining with anti-DBH antibody for 2 h in ice-cold buffer.After cell permeabilization and fixation, incubation with asecondary antibody coupled to Cy2 (which emits green
fluorescence) allowed simultaneous observation of the fusionsites (DBH) and organelle distribution (in red). The results areillustrated in Figs 6 and 7, which compare polar planes(Fig. 6A; Fig. 7A) and equatorial planes (Fig. 6B; Fig. 7B).
Polar sections of the cortical region were used to measure thedistances separating exocytotic sites and the nearest organellein the plane of the membrane (xy-distance), whereas equatorial
sections allowed measuring the distance from the cellmembrane to the nearest organelle (z-distance, perpendicularto the membrane plane). The distributions found are compared
in Fig. 6C and Fig. 7C. Interestingly, both distance distributionswere similar revealing that: (1) 30–35% of secretory sitescolocalized with mitochondria and, again, that (2) there is animportant population of mitochondria centered around the
300 nm distance. Taking into consideration that the corticalmitochondria amounted to 40% of total in the cell (Fig. 1), thismeans that ,12–14% of total mitochondria localize within a
extremely short distance of active sites, whereas a secondpopulation, corresponding to 25% of the total mitochondria, arelocated in close vicinity (a distance of one to two granule
diameters away) to the secretory sites.A similar analysis of the ER distributions (Fig. 7) indicated
that ,10% of the cortical population is in close association
with exocytotic sites (and again the same results are obtainedwhen calculating distances in both polar or equatorialdistributions). As for mitochondria (see Fig. 6C), there is asubstantial ER subpopulation that locates within the range of
one to two vesicle diameters away from the secretory sites(Fig. 7C). Taken together, our results indicate that two distinctsubpopulations of cortical mitochondria and ER exist that can
be stated as being closely associated with the secretory sites.This physical proximity could be very important for regulationof the exocytotic process in chromaffin cells, as discussed
below.
DISCUSSIONIn chromaffin cells, mitochondria distribute within twodistinct poolsTwo major conclusions can be derived from our study, first
that mitochondria distribute within two major populations, aperinuclear and a cortical group, whereas the ER seems todistribute in a relatively homogeneous manner throughout the
cytosol of chromaffin cells. Second, that in relation to the positionof exocytotic sites, both, cortical mitochondria and ER particlesform two distinctly distributed subpopulations, a first one that isclosely associated with the secretory sites and a second one that is
located ,0.3 mm away from the exocytotic sites. These two localcortical subpopulations could be directly involved it theregulation of Ca2+ signals and ATP supply in the immediate
vicinity of exocytotic sites.It is accepted that different subpopulations of mitochondria can
exist in a variety of cellular types, such as synaptic and non-
synaptic mitochondria in neurons (Hollenbeck, 1996), andsarcolemmal and miofibrillar mitochondria in myocytes(Chikando et al., 2011). In secretory tissues, the best-documented case is the existence of three different
Fig. 6. Distances of cortical mitocondria to the secretory sites. Culturedchromaffin cells incubated with Mitotracker Red were stimulated bydepolarization with a 59 mM KCl solution for 1 min at room temperature (21–22˚C), and then the secretion was stopped by lowering the temperature bymixing with ice-cold buffer. Then secretory sites were labeled using an anti-DBH antibody followed by a secondary antibody coupled to Cy2 (greenfluorescence) in ice-cold buffer to prevent endocytosis. The distribution ofcortical mitochondria and secretory sites was studied by acquiring confocalimages of the apical cortical area in polar sections (A) to calculate distancesin the membrane plane (XY, 122 measurements from 8 cells), or equatorialsections (B) to calculate the radial distances (Z, 225 measurements from 12cells). These distances were used to build a distribution (mean6s.e.m.)using 0.1-mm bin size (C). Scale bars: 3 mm.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5110
Jour
nal o
f Cel
l Sci
ence
mitochondrial subpopulations, namely, perinuclear, perigranular
and sub-plasmalemmal mitochondria, in pancreatic acinar cells(Park et al., 2001). In these cells, the distinct mitochondriasubpopulations ensure the independent control of cytosolic Ca2+
levels in the different regions of the cell. For example, theentry of Ca2+ though voltage-dependent Ca2+ channels atthe plasmalemma is rapidly buffered by subplasmalemmal
mitochondria, whereas the nuclear Ca2+signal is spatiallylimited by the perinuclear population. By contrast, the [Ca2+]c
signal that controls exocytosis of secretory granules is confined to
the secretory pole by the perigranular mitochondrial pool(Petersen and Tepikin, 2008; Petersen, 2012). Therefore, evenif we envision mitochondrial structures as forming part of a widerreticular network, it also seems evident that mitochondria could
be modulating distinct physiological processes in differentregions of the cytosol.
In neuroendocrine chromaffin cells, a number of contributions
have provided evidence for the functional coupling ofmitochondria and ER to VOCCs, resulting in subtle regulationof [Ca2+]c signals in the subplasmalemma (Montero et al., 2000;
Garcıa-Sancho et al., 2012). This coupling is so tight and efficientthat it even affects the characteristics of individual fusion eventsas measured by amperometry (Miranda-Ferreira et al., 2009). Ourpresent study precisely locates a substantial subpopulation of
mitochondria and ER structures as being in the immediatevicinity of the secretory apparatus, thus meaning it is feasible that
they strictly control of fusion events in the local microdomains,shape [Ca2+]c signals and also regulate the supply of energy forthe needs associated with exocytosis. Could these conclusions beextended to chromaffin cells in other species? At least in guinea
pig chromaffin cells, a cortical population of mitochondriaconstituting around 20% of the total has been described usingelectron microscopy (Inoue et al., 1996). This percentage is of the
same order as that described here (between 28% and 38%), andindicates that chromaffin cells from different species could havecortical mitochondria.
A model to understand the impact of mitochondria and ERpopulations in shaping Ca2+ signals during stimulationThe data reported here could be used to complete and redefine therole of mitochondria and the ER in shaping Ca2+ signals duringstimulation in the neuroendocrine model of chromaffin cells andits impact on secretion. Our findings are summarized in the model
shown in Fig. 8. Mitochondria are distributed in three differentgroups: (1) cortical (subplasmalemmal), (2) transitional, and (3)perinuclear (Fig. 8). In addition, the cortical subpopulation can be
subdivided into two further subpopulations: one that is closelyassociated with the secretory machinery, and a second that islocated at a distance of one to two granule diameters away from
the plasma membrane, which would be embedded into thecortical cytoskeleton. The cortical mitochondria can bedistinguished from the perinuclear not only by the distribution
within the cell, but also by their average size and density(Table 1); cortical mitochondria are smaller and more denselypacked, perhaps because of the reduced space left by cytoskeletalstructures to accommodate organelles in this region (Giner et al.,
2007).The ER distribution in the cytoplasm is different and increases
in density from the cell periphery to the perinuclear region
(Fig. 3D; Table 1). This distribution is consistent with thatexpected for a reticular network and seems to be relatively static,with very slow motion that seems does not seem to be sensitive to
perturbation of the transport systems based in cytoskeletalstructures. In any case, ER appears as a dense network thatbecomes somewhat more tightly packed on approaching theperinuclear area in our confocal images. Particle analysis
indicates that this increase is mainly due to the increase in sizeof the average ER structure rather than a change in density, whichseems relatively homogeneous through the cytosolic space.
The non-homogeneous distribution of mitochondria could bedue to the transport mediated by F-actin and microtubularstructures, and their retention in the F-actin-dense cortex; a
similar mechanism has been proposed for other organelles such asdense core granules (Neco et al., 2003; Rudolf et al., 2003;Gutierrez, 2012). How do the discussed ER and mitochondria
distributions affect the management of exocytotic Ca2+ signals inchromaffin cells? Existence of two mitochondrial populations atsubplasmalemmal and cell core locations had been proposedpreviously on the basis of the extremely bimodal mitochondrial
Ca2+ uptake generated by Ca2+ entry though VOCCs during celldepolarization. This was attributed to the building up of twodifferent cytosolic microdomains with [Ca2+]c values near to
50 mM (subplasmalemmal) and 1 mM (cell core), respectively(Montero et al., 2000; Garcıa et al., 2006). The subplasmalemmalmitochondrial pool would include ,50% of mitochondria, which
would be packed within the first 1 mm beneath the plasma
Fig. 7. Distance of the cortical ER structures to the secretory sites.Cultured chromaffin cells incubated with ERtracker Red (100 nM) werestimulated by depolarization with a 59 Mm KCl solution for 1 min at roomtemperature (21–22˚C). Secretory sites were labeled in green using anti-DBH as indicated in Fig. 6. The distribution of cortical ER and secretory siteswas studied acquiring confocal images of the cortical area in polar sections(A) to calculate distances in the membrane plane (XY, 122 measurementsfrom 15 cells), or equatorial sections (B) to calculate the radial distances(Z, 222 measurements from 23 cells). These distances were used to build adistribution (mean6s.e.m.) using a 0.1 mm-bin size (C). Scale bars: 3 mm.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5111
Jour
nal o
f Cel
l Sci
ence
membrane (Villalobos et al., 2002). The maximal capacity of themitochondrial Ca2+ uptake is so high that 10% of the totalmitochondria, occurring at only 0.25 mm beneath the plasmamembrane, would be enough to sink essentially all the Ca2+
entering though maximally activated VOCCs (Villalobos et al.,2002). These data fit with the pools and sizes proposed here. Theexistence of the two subpopulations of cortical mitochondria and
a dense but relatively homogeneous ER ensures the presence ofboth (1) a general system for amplification of Ca2+ signals usingCa2+-induced Ca2+ release (CICR), and (2) an specialized
mitochondrial-buffering system controlling the progression ofthe [Ca2+]c wave at two levels: the immediate space surroundingCa2+ channels and the fast-releasable vesicle pool of the secretory
machinery (Heinemann et al., 1994; Montero et al., 2000;Villalobos et al., 2002; Garcıa et al., 2006; Garcıa-Sancho et al.,2012), and a second level controlling the lower Ca2+ levelsneeded for recruiting new vesicles embedded in the cortical
network (Montero et al., 2000; Villalobos et al., 2002; Garcıaet al., 2006; Garcıa-Sancho et al., 2012). In addition, thesecortical subpopulations of mitochondria will be supplying the
ATP needed for active transport, as well as for vesicular primingand maturation (Gunter et al., 1994; Kumakura et al., 1994;Rizzuto et al., 2000; Villalobos et al., 2002).
The perinuclear mitochondria could fulfill a different role,protecting the nucleus from the [Ca2+]c signals generated in theperiphery by limiting progression to the nucleoplasm (Chameroet al., 2002; Gerasimenko and Gerasimenko, 2004; Alonso et al.,
2006). The relatively small increase in the Ca2+ concentrationinside the matrix of these mitochondria might nevertheless beenough to stimulate respiration, thus also increasing the ATP
supply at the cell core (Villalobos et al., 2002). This energy supplymight be required for mobilizing new secretory vesicles from thereserve pool as well as supporting the activity of other perinuclear
organelles, such as the Golgi, nucleoplasm or even the nuclearstructures (Montero et al., 2000; Villalobos et al., 2002; Garcıaet al., 2006; Garcıa-Sancho et al., 2012). The size of the
mitochondrial Ca2+ load provides a ‘memory’ that forces anincrease of energy production that lasts until clearance of the Ca2+
debt by the mitochondrial Na+/Ca2+ exchanger is completed. This
takes longer for the subplasmalemmal mitochondria than for thecell core mitochondria (Villalobos et al., 2002; Garcıa et al., 2006).
In summary, our study supports the hypothesis that there arespecialized mitochondrial populations serving different roles in
the cortical and perinuclear areas of neuroendocrine cells, andgives structural support for idea of fine tuning of Ca2+ signals bythe plasma-membrane–mitochondria–ER triads (Montero et al.,
2000) in the proximity of secretory sites. Future studies modelingthe influence of the organelle distributions proposed here and theinsertion of organelles in the cytoskeletal cages surrounding
secretory sites (Torregrosa-Hetland et al., 2011) will be needed tounderstand the impact of such distributions in the fine tuning ofCa2+ signals and secretion.
MATERIALS AND METHODSChromaffin cell preparation and cultureChromaffin cells were isolated from bovine adrenal glands by
collagenase digestion, and they were further separated from the debris
and erythrocytes by centrifugation on Percoll gradients as described
previously (Gutierrez et al., 1988; Gil et al., 2002). The cells were
maintained as monolayer cultures in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% fetal calf serum, 10 mM
cytosine arabinoside, 10 mM 5-fluoro-29-deoxyuridine, 50 U/ml
penicillin and 50 mg/ml streptomycin. The cells were harvested at a
density of 150,000 cells/cm2 in 22-mm diameter coverslips coated with
poly-lysine. Cells were used between the third and sixth day after plating.
For experiments, the cell culture medium was replaced with Krebs/
HEPES (K/H) basal solution containing (in mM): NaCl, 134; KCl, 4.7;
KH2PO4, 1.2; MgCl2, 1.5; CaCl2, 2.5; glucose, 11; ascorbic acid, 0.56
and Na-HEPES, 15, pH 7.4. Cells were stimulated for 1 min using a
depolarizing solution with high levels of K+ (59 mM; obtained by
isosmotically replacing NaCl by KCl) in the K/H basal solution.
Dynamic confocal images of the F-actin cytoskeleton,mitochondria and the ERF-actin structures were labeled by expression of enhanced GFP- or RFP-
tagged LifeAct, a 17-amino-acid peptide that binds to F-actin without
altering its dynamics in vivo (Riedl et al., 2008) or in vitro studies. In these
experiments, chromaffin cells were transfected using the Amaxa basic
nucleofector kit for primary mammalian neuronal cells according to the
manufacturer’s instructions (Program 0-005, Amaxa GmbH. Koehl,
Fig. 8. A model of the distribution ofmitochondria and ER structures in chromaffincells. Subpopulations of mitochondria located atthe cortical (i), intermediate (ii) and perinuclear(iii) regions are represented, taking into accounttheir calculated mean size and density (Table 1).Subplasmalemmal mitochondria and ERstructures show a high degree of colocalizationwith the secretory machinery at the cortical area.A second pool of both organelles was embeddedin the F-actin cortex, represented as a greymeshwork. The ER density is relatively uniformthroughout the entire cytosol. The increase influorescence in the perinuclear area (Figs 1 and2) is due to the increased size of the ERstructures. The distances of the differentmitochondria and ER pools to the secretory sitesare shown schematically in the right panel.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5112
Jour
nal o
f Cel
l Sci
ence
Germany) as described previously (Villanueva et al., 2012). At 2 days after
cell transfection, cells were incubated either with a 1 mM Mitotracker
Green or Red or with the same concentration of ERtracker Green
(BODIPY–FL-glybenclamide) or ERtracker Red (BODIPY–TR-g-
glibenclamide) for a 15-min period at room temperature.
Fluorescence confocal images were obtained using an Olympus
Fluoview FV300 confocal laser system mounted on an IX-71 inverted
microscope incorporating a 1006 PLAN-Apo oil-immersion objective
with 1.45 NA. Excitation was achieved using argon and helium-neon
visible light lasers. Images were processed using the ImageJ program
with different plugins [particle centroid tracking, region of interest (ROI)
measurements, image average, multiple channel image comparison, and
colocalization analysis]. For the analysis of the motion of the
mitochondria and the ER, particles were generated by image
thresholding, obtaining ellipsoid forms, and the centroid position was
tracked at 1-s intervals. Further analysis such as the mean square
displacement (MSD) determinations according to Qian et al., 1991, was
performed using in-house macros for Igor Pro (WaveMetrics Inc., Lake
Oswego, OR). The plot of MSD against time was used to analyze the area
explored by a particle in motion from an initial position, and it is essential
to calculate the diffusion coefficient and to define the motion as confined
or not confined (Qian et al., 1991).
Electron microscopy imagesChromaffin cell pellets were fixed with 2.5% glutaraldehyde in 0.2 M
cacodylate buffer at pH 7.0 for 2 h at 4 C. Then pellets were washed with
a solution of 2.5% (w/v) sucrose in 0.2 M cacodylate buffer overnight and
then post-fixed using 1% osmium tetroxide in 0.2 M cacodylate buffer
over 2 h. After extensive washing, the pellets were stained with 2%
aqueous uranyl acetate for 2 h. Pellets were washed again, and
dehydrated through an ethanol series (30%, 50%, 70%, 80%, 96% and
99%, 15 min each) and propylene oxide for 30 min at room temperature.
Finally, cell pellets were embedded in epoxy resin. Ultrathin sections
(70 nm) were obtained using a Leica UC6 ultramicrotome and transferred
onto copper grids (200 mesh). After staining with uranyl acetate for
5 min and lead citrate for 1 min, the ultrathin sections were analyzed
with a JEOL 1011 a 80-kV transmission electron microscope using a
Gatan BioScam mod 792 digital camera to capture the images.
Graphics were obtained with IgorPro, Graphpad Prism (GraphPad
software, San Diego, CA) and Adobe Photoshop 7.0. A non-parametric
Mann–Whitney test for paired samples or a one-way ANOVA Kruskal–
Wallis test was used to establish statistical significance (differences
were considered significant when P,0.05). Data are expressed as
mean6s.e.m. obtained from experiments performed in a number (n) of
individual cells from at least three different cultures.
Competing interestsThe authors declare no competing interests.
Author contributionsJ.V. performed most of the experiments and analysed data. S.V. performed andanalysed the experiments concerning the motion of organelles. Y.G.-M. performedand analysed electron microscopy experiments. V.G.-M. and M.d.M.F. assisted inthe over-expression experiments and chromaffin cell cultures. G.E.-R. assisted inthe confocal fluorescence experiments. J.G.-S. helped in the design and writing ofthe paper. L.M.G. designed the experiments and wrote the paper.
FundingThis study was supported by grants from the Spanish Ministerio de Economia yCompetitividad [grant numbers BFU2011-25095 to L.M.G., and BFU2010-17379BFI to J.G.S.].
Supplementary materialSupplementary material available online athttp://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.160242/-/DC1
ReferencesAlonso, M. T., Villalobos, C., Chamero, P., Alvarez, J. and Garcıa-Sancho, J.(2006). Calcium microdomains in mitochondria and nucleus. Cell Calcium 40,513-525.
Burgoyne, R. D., Morgan, A., Robinson, I., Pender, N. and Cheek, T. R. (1993).Exocytosis in adrenal chromaffin cells. J. Anat. 183, 309-314.
Camacho, M., Machado, J. D., Alvarez, J. and Borges, R. (2008). Intravesicularcalcium release mediates the motion and exocytosis of secretory organelles: astudy with adrenal chromaffin cells. J. Biol. Chem. 283, 22383-22389.
Chamero, P., Villalobos, C., Alonso, M. T. and Garcıa-Sancho, J. (2002).Dampening of cytosolic Ca2+ oscillations on propagation to nucleus. J. Biol.Chem. 277, 50226-50229.
Chikando, A. C., Kettlewell, S., Williams, G. S., Smith, G. and Lederer, W. J.(2011). Ca2+ dynamics in the mitochondria – state of the art. J. Mol. Cell.Cardiol. 51, 627-631.
Duchen, M. R. (2012). Mitochondria, calcium-dependent neuronal death andneurodegenerative disease. Pflugers Arch. 464, 111-121.
Garcıa, A. G., Garcıa-De-Diego, A. M., Gandıa, L., Borges, R. and Garcıa-Sancho, J. (2006). Calcium signaling and exocytosis in adrenal chromaffincells. Physiol. Rev. 86, 1093-1131.
Garcıa-Sancho, J., de Diego, A. M. and Garcıa, A. G. (2012). Mitochondria andchromaffin cell function. Pflugers Arch. 464, 33-41.
Gerasimenko, O. and Gerasimenko, J. (2004). New aspects of nuclear calciumsignalling. J. Cell Sci. 117, 3087-3094.
Gil, A., Gutierrez, L. M., Carrasco-Serrano, C., Alonso, M. T., Viniegra, S. andCriado, M. (2002). Modifications in the C terminus of the synaptosome-associated protein of 25 kDa (SNAP-25) and in the complementary region ofsynaptobrevin affect the final steps of exocytosis. J. Biol. Chem. 277, 9904-9910.
Giner, D., Lopez, I., Villanueva, J., Torres, V., Viniegra, S. and Gutierrez, L. M.(2007). Vesicle movements are governed by the size and dynamics of F-actincytoskeletal structures in bovine chromaffin cells. Neuroscience 146, 659-669.
Gunter, T. E., Gunter, K. K., Sheu, S. S. and Gavin, C. E. (1994). Mitochondrialcalcium transport: physiological and pathological relevance. Am. J. Physiol.267, C313-C339.
Gutierrez, L. M. (2012). New insights into the role of the cortical cytoskeleton inexocytosis from neuroendocrine cells. Int. Rev. Cell Mol. Biol. 295, 109-137.
Gutierrez, L. M., Ballesta, J. J., Hidalgo, M. J., Gandia, L., Garcıa, A. G. andReig, J. A. (1988). A two-dimensional electrophoresis study of phosphorylationand dephosphorylation of chromaffin cell proteins in response to a secretorystimulus. J. Neurochem. 51, 1023-1030.
Heinemann, C., Chow, R. H., Neher, E. and Zucker, R. S. (1994). Kinetics of thesecretory response in bovine chromaffin cells following flash photolysis of cagedCa2+. Biophys. J. 67, 2546-2557.
Hollenbeck, P. J. (1996). The pattern and mechanism of mitochondrial transport inaxons. Front. Biosci. 1, d91-d102.
Inoue, M., Ogawa, K., Fujishiro, N., Yano, A. and Imanaga, I. (1996). Role andsource of ATP for activation of nonselective cation channels by AlF complex inguinea pig chromaffin cells. J. Membr. Biol. 154, 183-195.
Kumakura, K., Sasaki, K., Sakurai, T., Ohara-Imaizumi, M., Misonou, H.,Nakamura, S., Matsuda, Y. and Nonomura, Y. (1994). Essential role of myosinlight chain kinase in the mechanism for MgATP-dependent priming of exocytosisin adrenal chromaffin cells. J. Neurosci. 14, 7695-7703.
Lopez, I., Giner, D., Ruiz-Nuno, A., Fuentealba, J., Viniegra, S., Garcia, A. G.,Davletov, B. and Gutierrez, L. M. (2007). Tight coupling of the t-SNARE andcalcium channel microdomains in adrenomedullary slices and not in culturedchromaffin cells. Cell Calcium 41, 547-558.
Lopez, I., Ortiz, J. A., Villanueva, J., Torres, V., Torregrosa-Hetland, C. J., delMar Frances, M., Viniegra, S. and Gutierrez, L. M. (2009). Vesicle motion andfusion are altered in chromaffin cells with increased SNARE cluster dynamics.Traffic 10, 172-185.
Lopez-Font, I., Torregrosa-Hetland, C. J., Villanueva, J. and Gutierrez, L. M.(2010). t-SNARE cluster organization and dynamics in chromaffin cells.J. Neurochem. 114, 1550-1556.
Miranda-Ferreira, R., de Pascual, R., Caricati-Neto, A., Gandıa, L., Jurkiewicz,A. and Garcıa, A. G. (2009). Role of the endoplasmic reticulum andmitochondria on quantal catecholamine release from chromaffin cells ofcontrol and hypertensive rats. J. Pharmacol. Exp. Ther. 329, 231-240.
Montero, M., Alonso, M. T., Carnicero, E., Cuchillo-Ibanez, I., Albillos, A.,Garcıa, A. G., Garcıa-Sancho, J. and Alvarez, J. (2000). Chromaffin-cellstimulation triggers fast millimolar mitochondrial Ca2+ transients that modulatesecretion. Nat. Cell Biol. 2, 57-61.
Neco, P., Giner, D., del Mar Frances, M., Viniegra, S. and Gutierrez, L. M. (2003).Differential participation of actin- and tubulin-based vesicle transport systemsduring secretion in bovine chromaffin cells. Eur. J. Neurosci. 18, 733-742.
Nunez, L., Senovilla, L., Sanz-Blasco, S., Chamero, P., Alonso, M. T.,Villalobos, C. and Garcıa-Sancho, J. (2007). Bioluminescence imaging ofmitochondrial Ca2+ dynamics in soma and neurites of individual adult mousesympathetic neurons. J. Physiol. 580, 385-395.
Park, M. K., Ashby, M. C., Erdemli, G., Petersen, O. H. and Tepikin, A. V.(2001). Perinuclear, perigranular and sub-plasmalemmal mitochondria havedistinct functions in the regulation of cellular calcium transport. EMBO J. 20,1863-1874.
Patzak, A., Bock, G., Fischer-Colbrie, R., Schauenstein, K., Schmidt, W.,Lingg, G. and Winkler, H. (1984). Exocytotic exposure and retrieval ofmembrane antigens of chromaffin granules: quantitative evaluation ofimmunofluorescence on the surface of chromaffin cells. J. Cell Biol. 98, 1817-1824.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5113
Jour
nal o
f Cel
l Sci
ence
Petersen, O. H. (2012). Specific mitochondrial functions in separate sub-cellulardomains of pancreatic acinar cells. Pflugers Arch. 464, 77-87.
Petersen, O. H. and Tepikin, A. V. (2008). Polarized calcium signaling in exocrinegland cells. Annu. Rev. Physiol. 70, 273-299.
Qian, H., Sheetz, M. P. and Elson, E. L. (1991). Single particle tracking.Analysis of diffusion and flow in two-dimensional systems. Biophys. J. 60, 910-921.
Rickman, C., Meunier, F. A., Binz, T. and Davletov, B. (2004). High affinityinteraction of syntaxin and SNAP-25 on the plasma membrane is abolished bybotulinum toxin E. J. Biol. Chem. 279, 644-651.
Riedl, J., Crevenna, A. H., Kessenbrock, K., Yu, J. H., Neukirchen, D., Bista,M., Bradke, F., Jenne, D., Holak, T. A., Werb, Z. et al. (2008). Lifeact: aversatile marker to visualize F-actin. Nat. Methods 5, 605-607.
Rizzuto, R., Bernardi, P. and Pozzan, T. (2000). Mitochondria as all-roundplayers of the calcium game. J. Physiol. 529, 37-47.
Rudolf, R., Kogel, T., Kuznetsov, S. A., Salm, T., Schlicker, O., Hellwig, A.,Hammer, J. A., III and Gerdes, H. H. (2003). Myosin Va facilitates the
distribution of secretory granules in the F-actin rich cortex of PC12 cells. J. CellSci. 116, 1339-1348.
Spector, I., Shochet, N. R., Kashman, Y. and Groweiss, A. (1983). Latrunculins:novel marine toxins that disrupt microfilament organization in cultured cells.Science 219, 493-495.
Torregrosa-Hetland, C. J., Villanueva, J., Giner, D., Lopez-Font, I., Nadal, A.,Quesada, I., Viniegra, S., Exposito-Romero, G., Gil, A., Gonzalez-Velez, V. et al.(2011). The F-actin cortical network is a major factor influencing the organization ofthe secretory machinery in chromaffin cells. J. Cell Sci. 124, 727-734.
Villalobos, C., Nunez, L., Montero, M., Garcıa, A. G., Alonso, M. T., Chamero,P., Alvarez, J. and Garcıa-Sancho, J. (2002). Redistribution of Ca2+ amongcytosol and organella during stimulation of bovine chromaffin cells. FASEB J.16, 343-353.
Villanueva, J., Torres, V., Torregrosa-Hetland, C. J., Garcia-Martinez, V.,Lopez-Font, I., Viniegra, S. and Gutierrez, L. M. (2012). F-actin-myosin IIinhibitors affect chromaffin granule plasma membrane distance and fusionkinetics by retraction of the cytoskeletal cortex. J. Mol. Neurosci. 48, 328-338.
RESEARCH ARTICLE Journal of Cell Science (2014) 127, 5105–5114 doi:10.1242/jcs.160242
5114