© 1999 American Heart Association, Inc.
Original Contribution |
Examination of the Transverse Tubular System in Living Cardiac Rat Myocytes by 2-Photon Microscopy and Digital Image–Processing Techniques
From the Department of Physiology, University of Auckland, Auckland, NZ.
Correspondence to Dr Mark Cannell, Department of Physiology, University of Auckland, Private Bag 92019, Grafton St, Auckland, NZ. E-mail m.cannell{at}auckland.ac.nz
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abstract—The transverse tubular system (t-system) of cardiac muscle is a structure that allows rapid propagation of excitation into the cell interior. Using 2-photon molecular excitation microscopy and digital image–processing methods, we have obtained a comprehensive overview of the t-system of rat ventricular myocytes in living cells. We show that it is possible to quantify the morphology of the t-system in terms of average local tubule diameter, branching pattern, and local abundance of the t-system by immersing living myocytes in a dextran-linked fluorescein solution. Our data suggest that previous electron microscopic examinations of t-system structure have underestimated both the geometric complexity of the t-system morphology and the fraction of cell volume occupied by the t-system (3.6% in this species). About 40% of tubules occur between Z-lines, and the t-tubule diameter is 255±0.85 nm (mean±SEM). The t-tubules leave the outer surface of the cell in an approximately rectangular array; however, at some points junctions between the t-tubules and the surface membrane are missing. In view of the complexity of the t-system apparent from our images, we propose that the t-system be renamed the "sarcolemmal Z rete." The methods presented here are generally applicable to the quantification of the sarcolemmal Z rete and other structures within cells by fluorescence microscopy in a variety of cell types.
Key Words: heart structure • t-system • imaging • myocyte • rat
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mammalian ventricular heart muscle cells have a fine sarcotubular system that was first identified by electron microscopy (EM).1 Subsequent EM studies revealed that the tubular system was continuous with the sarcolemma and located at the level of the Z-lines (eg, Reference 22 ). Since most of the tubules were associated with the Z-lines, they were assumed to run transversely and were named the transverse tubular system or "t-system" (for review, see Reference 33 ). However, subsequent studies have shown that considerable numbers of tubules run in the axial direction,4 5 leading to the descriptive term "transverse-axial tubular system"6 or TATS.7
Most studies have used EM to provide sufficient resolution to identify
the tubular membranes. However, this method suffers from the
disadvantage that the specimen must be fixed (or frozen), dehydrated,
thin sectioned and heavy-metal stained. Obtaining structural
information through the thickness of a cell with EM is
problematic because of the poor penetration of the electron
beam. Even "thick EM sections" are limited to a thickness of
5 µm,8 and because of this problem only an
11x24x2–µm section of the TATS in a mouse myocardial cell has been
shown.6 EM of thin sections relies on absorption and so
has no intrinsic 3-dimensional (3D) resolution. This leads to images of
thicker sections becoming blurred and losing internal detail.
Three-dimensional reconstruction from stacks of thin EM sections is
complicated by the difficulty of obtaining the necessary registration
between separate thin sections that are subject to distortion. Indeed,
the TATS has only been reconstructed over a 5.5-µm-thick segment of a
heart cell with manual section registration.7
The t-system in skeletal muscle provides a rapid propagation of electrical excitation into the cell interior.9 10 11 Evidence for such a role for the t-system in heart12 13 has been less clear, since there is no correlation between contraction speed and abundance of t-system.3 However, optical measurements of latencies for calcium release across rat ventricular cells show that the t-system provides both a rapid inward spread of electrical excitation and the calcium influx that triggers calcium release from the sarcoplasmic reticulum.14 The microarchitecture of the t-system has been suggested to be important in determining the levels of ionic accumulation/depletion during excitation.7 15 16 The t-system may not be a fixed structure within the cell, since short-term cell culture has been reported to be associated with a loss of t-tubules,17 and t-tubule proliferation occurs with hypertrophy.13 These observations suggest that a method for investigating t-tubular morphology in living myocytes may be useful for studying pathological changes in myocyte structure.
The confocal microscope provides a means for optically sectioning living specimens, thereby overcoming problems associated with fixation and dehydration. As shown by Shacklock et al18 and Cheng et al,19 it is possible to detect t-tubules by immersing cells in a soluble fluorescent marker that penetrates the tubular system. We have extended this approach by using 2-photon molecular excitation (TPME) microscopy20 and staining the extracellular compartment with a fluorescent probe that does not cross the cell membrane. Although the t-tubules themselves are generally below the optical limit of resolution, this does not preclude quantitative measurement of the fractional cell volume occupied by these structures, since the imaging properties of the microscope are known from its point-spread function (PSF). We have used this approach to obtain estimates of fractional volumes and tubule diameters in rat ventricular myocytes. A preliminary account of some of these methods has been published elsewhere.21
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cells
Enzymatically isolated rat cardiac myocytes were prepared as described elsewhere.22 Briefly, 250-g Wistar rats were killed by cervical dislocation and the hearts quickly removed and mounted on a Langendorff perfusion system. The heart was initially perfused for 5 minutes at 37°C with a nominally calcium-free solution containing (in mmol/L) NaCl 120, KCl 5.4, MgCl2 1, HEPES 10, pyruvate 5, glucose 20, and taurine 20 (pH 7.05 with NaOH). This basic solution was then switched to one containing 4 U/mL protease (Sigma type XXIV) and 100 µmol/L CaCl2. After 3 minutes, the perfusate was replaced with one containing collagenase (Worthington type II, 1 mg/mL) in place of protease for a further 5 to 10 minutes. On completion of the enzyme treatment, the ventricles were cut free and chopped into small pieces, and serial triturations were performed using a wide-bore pipette. Following each trituration, the resultant cell suspension was filtered through gauze and centrifuged briefly, and the pellet was resuspended in isolation solution to which 0.2 mmol/L CaCl2 was added. Cell suspensions were stored at room temperature until being placed on the stage of the TPME microscope in a chamber that incorporated a glass coverslip as its bottom. Only cells that were quiescent during the observation period were used. All experiments were conducted at room temperature (22°C to 23°C).
Imaging
The t-system was visualized by immersing cells in a bathing
medium containing 1 mmol/L dextran-linked
fluorescein (molecular weight 70 000; Molecular Probes).
Images were recorded with an LSM 410 laser scanning confocal
microscope (Zeiss), which was modified for TPME
microscopy.23 Illumination for 2-photon
fluorescence excitation was supplied by a mode-locked
Ti:Sapphire laser (Mira 900F; Coherent), which generated a train of
ultrashort pulses at a repetition frequency of 76 MHz. For these
experiments, the laser was tuned to a center wavelength of 860 nm with
the intracavity prisms adjusted to produce a bandwidth of 19 nm. The
pulse width was 45 fs (measured by autocorrelation) and the average
illumination power at the sample was 4 mW. An external prism
compressor was adjusted to (re)establish minimum pulse width at
the specimen.23 The cell was imaged through a Zeiss x63
1.25-numerical aperture oil-immersion objective. Images from the
isolated cells were obtained at a nominal spacing of 200 nm along the
optical axis. The optical performance of the microscope was
determined by imaging 200-nm fluorescent latex beads
(Fluospheres, Molecular Probes). Since the PSF of the microscope
depends on the distribution of refractive index in the
sample,24 25 the PSF was measured by dispersing the
subresolution beads in the bath containing the myocytes in bathing
medium without fluorescein. Beads that spontaneously stuck
to the cell membrane were imaged at several heights above the
coverslip, and from these images a resolution of 400 nm (full width at
half maximum; FWHM) in x-y and 800 nm in z was
measured. Figure 1
illustrates the
TPME microscope PSF measured in this way. To exceed the Nyquist
criterion for image sampling, the voxel size was set to
0.1x0.1x0.2 µm (x, y, and z,
respectively). This fine voxel resolution also improved convergence of
the deconvolution procedure (see below), but it limited the cell area
that could be imaged to 102x102 µm. Since this encompassed
most of the chosen myocytes, this limitation was considered
acceptable.
|
Data Processing
To improve the resolution and signal-to-noise ratio of the
volume data, digital deconvolution26 was applied to
"stacks" of images. In a recent comparison of image-restoration
methods for confocal microscopy, the Richardson-Lucy
algorithm27 was judged superior to other
algorithms.28 For this study, the deconvolution algorithm
was implemented in the language IDL (Research Systems Inc) to
produce a maximum likelihood reconstruction for the specimen in the
presence of Poisson (photon shot) noise. A Poisson noise model was
used, because the number of photons collected was probably insufficient
to justify using a gaussian noise model. The update step of the
procedure can be written as follows27 28 :
 | (1) |
are the convolution and
cross-correlation operators, respectively. In general, about 15
iterations of Equation 1
were used to generate a deconvolved
image data set. All calculations were performed with an
Indigo2 workstation (Silicon Graphics).
Volume Visualization
To assist visualization of the t-system architecture, a
topological "skeleton" was constructed from the deconvolved stack
of images. In this 3D skeleton,29 all information about
tubule signal intensity was removed by replacing tubule segments with a
thin line of uniform thickness, which followed the direction of the
t-tubule in space. Since the 3D connectivity of the skeleton is
identical to that of the t-tubular system, this method of data
presentation clarifies the organization of the t-tubular
system. In any case, the calculated t-tubule diameters (see below) were
generally less than the wavelength of light so the original light image
contained no more spatial information than the topological skeleton.
The first step in skeleton construction was generation of a binary mask
from the enhanced image stack. Pixels with intensities greater than the
mean background+5 SD were assumed to represent t-tubular
segments and were set to 1.0; all other pixels were set to 0.0. From
this 3D binary mask, a topological skeleton was generated by deleting
all pixels that did not change the connectivity of the 3D
structure.30
The need to construct a topological skeleton forced us to adopt a
lower-limit threshold for t-tubular cross-section. For the data sets
presented here, tubules less than 50 nm in diameter would be
lost in the noise and therefore missed. Since the smallest t-tubule
diameter reported in left ventricular myocytes of rat is
70 nm,31 this limit was considered acceptable. However,
more extensive signal averaging (to reduce photon noise) or a
relaxation of the stringency of the binary mask would allow any desired
lower limit to be achieved.
Estimation of Fractional t-Tubule Volume
Since we were unable to detect any entry of dextran-linked
fluorescein into the intracellular compartment, and cell
autofluorescence was negligible (not shown), the calculation of
fractional t-tubular volume FVtt was
straightforward and required no assumptions about t-tubular
microarchitecture. The total intensity of fluorescence
recorded within the boundary of the sarcolemma is directly
proportional to the amount of dye contained within the t-system, as
shown in the following:
 | (2) |
Estimation of t-Tubule Diameters
The peak signal from a piece of t-tubule is a function of the
tubule geometry and PSF of the microscope (see Figure 2A) as follows:
 | (3) |
with
respect to the optical axis) being 1 at points inside the segment and 0
elsewhere.
|
) also determines the strength of the
fluorescence signal. The angular dependence is removed by
making the PSF spherical (C). D, Various tubule morphologies and
branching patterns sketched and labeled with the correction factor (w)
they cause. E, Relationship between normalized image intensity and
local tubule radius calculated from the spherical modified microscope
PSF (see text).
As shown in Figure 2B, even for a (typical) PSF that is
symmetrical about the z-axis, Equation 3 remains a
function of both radius R and angle . However, if the PSF
is completely symmetrical, then the signal does not depend on (see
Figure 2C) and becomes only a function of the PSF and
R, as follows:
 | (4) |
),
if the tubules are adequately described by cylinders and the PSF is
symmetrical. Although the microscope does not have a symmetrical PSF,
the effective PSF applied to the data (cell structure) can be made
symmetrical by convolving each of the images of the 3D data set in the
x-y plane with a gaussian kernel so that the in-plane FWHM
equals the axial FWHM (0.8 µm). Blurring the images to create
identical resolutions in x, y, and z
results in the image data losing the directional dependence of the
local fluorescence signal. For segments of t-tubules of which
the architecture is more complicated than simple nonbranching
cylindrical segments, the binary topological skeleton contains the
information needed to correct the relationship between
IR and R for any branching that
may occur in the t-system. The correction factor is obtained by
convolving the binary skeleton with the PSF used to process the data to
yield a weighting (w) that is applied to the data to allow
for elements such as blind-ended tubules (w<1.0) and
branches (w>1.0) (see Figure 2D). Figure 2E
shows the relationship between
IR/IR=
and wxR.
We also used the calculated distribution of tubule diameters to
estimate the total surface area of t-system within the sampled cell
volume. In the calculation, pieces of t-tubule are approximated by
cylinders of varying diameter so that a piece of tubule of length
si and radius
Ri contributes a surface area
Si=si2
Ri.
The total t-system area STT in a given
volume can then be obtained by summing all the contributions of pieces
of t-tubule within the sample volume, as follows:
 | (5) |
provides the
smallest boundary for the enclosed volume. To visualize the distribution of tubule diameters, the skeleton was intensity coded according to the calculated t-tubule diameter and displayed by isosurface rendering using IrisExplorer software (Silicon Graphics).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Figure 3A
shows a typical
fluorescence micrograph from a stack of 60 images that were
acquired from a quiescent rat ventricular myocyte. In this
image, t-tubules and the exterior of the cell are bright because of the
presence of dextran-linked fluorescein in the bathing
medium. An enlarged part of the image is shown in Figure 3B, and some fine structure is visible, although there is photon noise in the
image (which could have been reduced by further signal averaging). The
lower panel (Figure 3C) shows the results of applying digital
deconvolution to the stack of images (see Materials and Methods). It is
notable that (1) there is a considerable reduction in photon noise in
the image, while (2) the image contrast is increased, and (3) the image
"sharpness" is also improved. These effects arise from the
increased data present in the volume data being convolved with the
PSF (compared with a single-image plane), as well as some implicit
smoothing due to the deconvolution algorithm. The resulting improvement
in signal-to-noise ratio and resolution allows the t-tubules to be seen
more easily. The fractional volume of the t-system estimated by
Equation 2 was 3.6±0.4% (mean±SD, n=11). This value is
significantly higher than previously reported for rat (1%; eg,
References 12 and 3212 32 ).
|
To simplify and clarify the general macroarchitecture of the cell
t-tubular system, the volume data were reduced to a 3D skeleton (see
Materials and Methods). The 3D skeleton of a rat
ventricular myocyte t-system is shown in Figure 4. Visual inspection of such skeletons
suggested that the overall macroarchitecture of the t-tubular system
resembles a complicated mesh with some limited regularity at the level
of the Z-line (at a regular axial spacing of 1.8 µm). There
are also a few regions relatively devoid of t-tubules that probably
coincide with the locations of cell nuclei. To quantify the
preponderance of t-tubules at the Z-lines, Figure 5A shows a graph of tubule density as a
function of the longitudinal position inside the cell. Although 60% of
the tubule volume occurs at the Z-line (as defined by their presence in
a region 0.55 µm around Z-lines), the remaining 40% of the
tubules are found at other locations along the myofibrils. It is
apparent from Figures 4 and 5A that a considerable amount
of the t-tubular system forms longitudinal connections between
t-tubules, although most of the t-tubules are organized in transverse
planes. However, the t-tubules do not run across the cell at the level
of the Z-lines for great distances but instead run at the level of 1
Z-line for a short distance and then change direction to join other
tubules at other Z-lines. In some places, no t-tubules ran in the
transverse direction (across the cell) but instead passed through the
cell at an angle to its longitudinal axis (see also stereo views in
Figure 6). Having reconstructed the
skeleton of t-tubular architecture, it is straightforward to determine
the distribution of the length of t-tubular segments between branching
points. Figure 5B shows a graph of t-tubule lengths between
branching points obtained from the skeleton data in Figure 4.
The distribution of interbranch lengths is approximately exponential,
with short segments being most prominent and a mean branch length of
6.87 µm. To help clarify the complex macroarchitecture of the
tubular system, Figure 6A shows an enlarged stereo view of the
tubular system from a randomly selected part of the cell. In the upper
right portion of this view, it can be seen that t-tubules form
ring-shaped connections, possibly wrapping around myofibrils. A region
almost devoid of tubules is apparent in the center of this section of
tubular network. The predominance of t-tubules in planes located at the
Z-line can be identified most clearly in the region at the left of this
view.
|
|
|
Electron micrographs suggest that tubules have nearly circular cross
sections, so it is possible to calculate the apparent width of the
tubule (see Materials and Methods). The calculated tubule diameters
have been imposed on the skeletons by gray coding in Figure 6B.
Although the majority of tubules appear to have diameters of 200 to
300 nm, there are also local increases and decreases in tubule diameter
at (apparently) random points in the cell. The data shown in Figure 6B are summarized in a histogram of tubule diameters shown in
Figure 7. It is apparent that the
distribution is approximately gaussian, and almost no tubules larger
than 450 nm were found. The histogram has a maximum at a diameter of
240 nm, and most t-tubules (51%) are between 180 and 280 nm wide.
Using the calculated distribution of tubule diameters to estimate the
surface area of t-tubular membrane per unit of cell volume (see
Materials and Methods), we obtained a value of
STT/VCell=0.44
µm2/µm3.
|
The positions at which the t-tubules leave the cell surface can be
visualized from the same data set. In Figure 8, the topology of the surface membrane
has been extracted from the data set and isosurface rendered. It is
apparent that the t-tubules form extensive regions of regular patterns
on an almost rectangular mesh. The tubule mouths often run for some
tens of micrometers in linear arrays with a transverse
spacing between mouths of 1.8 µm. However, there are points
within the mesh that are missing a tubule contact. Although
freeze-fracture images also support this view,33 the more
extensive data set presented here allows a greater appreciation
of the extent of order and disorder of the tubule contacts with the
surface membrane.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have shown that it is possible to examine and quantify the macroarchitecture of the t-system by 3D reconstruction from a stack of 2-dimensional TPME images. Although a conventional confocal microscope could be used for this type of analysis, the TPME microscope has some advantages because of the minimization of out-of-focus light as well as reduced internal scattering (as a result of the longer excitation wavelength). With this microscope, we have been able to show the complex reticular nature of the t-tubular system in living cells, as well as to quantify some of its morphometry. It is important to note that our optical methods overcome the registration problems associated with serial reconstruction from the thin sections (needed for EM). Indeed, the only comparable EM study that we are aware of required manual identification of t-tubules before reconstruction could be performed.7 It is inevitable that manual identification will lead to some tubular elements being missed, while distortion of thin sections without fiducial marks will lead to geometric errors.
Our data suggest that the fractional volume
(VFtt) of the t-tubular system is
3.6%. Earlier measurements based on EM studies suggested a much
lower VFtt in the rat (typically 1% for
adult rat; eg, References 12 and 3212 32 ). This large difference is not
easily explained by an error in our method of measurement, since the
total fluorescence from the extracellular marker should depend
only on the volume of the t-tubules and the concentration of marker
(see Materials and Methods). Although a large contribution of
autofluorescence to the recorded signal would lead to an
overestimate of VFtt, inspection of the
images shows that the autofluorescence background was barely
detectable compared with the t-tubule signal. Similar arguments rule
out the possibility that marker dye penetration into the cytoplasm
could explain the larger VFtt (since any
entry of the dye into the cell would appear as generalized cytoplasmic
staining). Although low levels of dextran binding to phospholipids have
been reported34 and such an effect would cause an
overestimate of VFtt, 4 pieces of evidence
suggest that this explanation is unlikely. (1) The reported densities
of dextran binding by phospholipids are too low (0.4 mg D-40/g
1,2-dipalmitoyl-3-sn-phosphatidyl-choline) to
materially affect our results. (2) Our estimates of t-tubule diameters
and EM data are in reasonable agreement (see below), which suggests
that our method accurately measures t-tubules. (3) The level of
membrane binding needed to explain the difference between our results
and those of Stewart and Page32 would have led to an
elevated level of fluorescence near the cell edge, but this was
not observed. (4) Further support for the idea that EM studies have
underestimated the extent of the t-tubular system is provided by
electrophysiological studies. We estimate a
surface-to-volume ratio
STT/VCell=0.44
µm2/µm3 that is almost
3 times larger than previously reported stereological measurements from
EM data (0.15
µm2/µm3).35
A recent study of capacity/volume ratios in cardiac myocytes from
different species suggests a specific membrane capacitance
CM/VCell of
6.76 pF/pL for myocytes from young (3-month-old) and 8.88 pF/pL from
older (6-month-old) rats.36 Since the specific
membrane capacitance is 1 µF/cm2 (see, eg,
Reference 37 ), these values can be converted into a ratio
of total sarcolemmal membrane area (outer sarcolemma+ t-tubular
system) that is
STotal/VCell0.68
µm2/µm3 for the younger
and 0.89 µm2/µm3
for older rats. For a typical myocyte 100 µm long with a
diameter of 20 µm (ie, a volume of
3.1x104 µm3, we
can estimate from our data that
STotal/VCell
would be 0.68
µm2/µm3, which is in
good agreement with the
electrophysiological data.
The above agreement between the estimate of cell capacitance and our data suggests that the lower VFtt and STT/VCell derived from previous EM studies must be in error. Since there can be problems in identifying t-tubules in electron micrographs (eg, see Reference 3838 ), as well as specimen shrinkage during section preparation, such an underestimate of VFtt and STT/VCell might easily occur. It is also possible that the method of measurement used by Stewart and Page32 will have underestimated VFtt, since, as they noted, (1) average t-tubule diameter is much less than the EM section thickness, and (2) t-tubules were assumed to run at right angles to the long axis of the cell only, whereas we have detected large amounts of t-tubules running in other directions.
"t-System" Morphology
The volume data show that the term "transverse-tubular system"
is something of a misnomer, since t-tubules run in all directions. The
general impression of the t-tubular organization is that of a rete,
although such a term does not fully describe the preponderance of
elements near the Z-lines. We have been able to confirm the presence of
large numbers of tubules running in axial directions,4 the
presence of which led to the suggestions that the t-system be referred
to as the TATS, or T-Ax.5 6 38 However, there are
also large numbers of tubules that run in neither axial nor transverse
directions, so a more accurate descriptive term might be the
"sarcolemmal Z rete" (SZR). For convenience, this nomenclature will
be adopted in the following discussion, although the possible general
acceptance of such a term will depend on the findings of future
studies on other species.
The SZR often formed complete loops around the (assumed) location of
myofibrils. However, there are also regions within the cell that are
relatively devoid of tubules. We suspect that some of these regions may
be associated with parts of the cell that are not involved in
excitation-contraction coupling (such as the nucleus), but further work
will be required to confirm this view. Visual inspection of the SZR
skeletons suggested that there was little or no correlation of tubule
width with position along the sarcomere (see Figure 6B).
However, quantitative analysis suggested that the average
tubule diameter was 15% larger between the Z-lines than in
longitudinal elements (not shown). Since the cells were relaxed, this
increase in average diameter could be the result of the constant volume
behavior of the cell. We suspect that if the cell were stretched to
working lengths, this difference would largely disappear, and there
would be almost no correlation between tubule diameter and longitudinal
position.
Although the optical methods used here cannot resolve the fine microarchitecture of the SZR, they give good insight into the gross morphology of the SZR. For surveying the general organization of the SZR, our optical approach produces a comprehensive overview of it, which would not be possible (for any reasonable amount of effort) with EM. This is because with one image stack we are able to resolve the entire SZR within a selected myocyte. In addition, we have shown that it is straightforward to extract quantitative information about the fractional cell volume occupied by the SZR and equivalent tubule diameters. In contrast to the utility of our optical method, stereological analysis of electron micrographs is extremely time consuming. In addition, our fluorescent method does not require extensive geometrical correction factors to take account of the variations in the plane of section of t-tubules.32
Surface Membrane Topology
Using the methods introduced here, we have been able to produce
images that portray the organization of junction between the SZR and
the outer sarcolemmal surface. These images are very similar to
freeze-fracture images of myocyte surfaces.33 However, we
have been able to examine a much larger area of the cell surface than
has been achieved by freeze-fracture techniques. In addition, we can be
certain that the organization of SZR "mouths" is not confused with
caveolae. The SZR mouths are mostly arranged in linear arrays with a
longitudinal spacing of 1.8 µm. However, these linear arrays
are interrupted by missing mouths or by a misregistration with another
region of linear arrays (Figure 8). It is possible that such an
organization may reflect the nonuniform development of the SZR during
maturation, since it is known that in some species neonatal myocytes
have no SZR (eg, cats40 ). It is likely that during
hypertrophy the SZR will grow and form new connections with
the surface membrane. Whether this occurs as a result of new
invaginations of the surface membrane or by extension of the SZR with
subsequent fusion with the surface of the cell is unknown. It may be
possible to distinguish between these possibilities by applying our
methods to myocytes that are actively undergoing
hypertrophy.
The fluid contained within the SZR can only directly exchange with the extracellular fluid at the junction of the SZR with the outer cell surface. Since accumulation/depletion phenomena will be limited by the efficiency of this exchange, it is possible that a part of the propensity of hypertrophied hearts to develop arrhythmias (eg, Reference 4141 ) may reflect a mismatch between SZR volume and the number of SZR mouths. Again, future work should be able to directly examine this possibility.
In conclusion, we have shown that it is possible to examine the gross morphology of the SZR using optical methods combined with digital image processing. The advantage of this approach is that most of the analysis and data acquisition can be automated. In addition, we can examine the cell microarchitecture in identified living myocytes that have not suffered from the extensive preprocessing required for EM.
|
Acknowledgments |
|---|
We thank the Wellcome Trust (UK) for financial support.
Received August 10, 1998; accepted October 6, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Lindner E. Die submikroskopische Morphologie des Herzmuskels. Zeitschrift Zellforsch. 1956;45:702–746.
2. Girardier L, Pollet M. Demonstration de la continuité entre l'espace interstitiel et la lumière de canaux intercellulaires dans le myocarde de rat. Helv Physiol Pharmacol Acta. 1964;22:C72–C73.[Medline] [Order article via Infotrieve]
3. Sommer JR, Johnson EA. Ultrastructure of cardiac muscle. In: Berne RM, Sperelakis N, Geiger SR, eds. Handbook of Physiology: The Cardiovascular System. Vol. 1, Section 2. Bethesda, Md: American Physiological Society; 1979:113–186.
4.
Forssmann WG, Girardier L. A study of the T-system in
rat heart. J Cell Biol. 1970;42:1–19.
5. Sperelakis N, Rubio R. An orderly lattice of axial tubules which interconnect adjacent transverse tubules in guinea-pig ventricular myocardium. J Mol Cell Cardiol. 1971;2:212–220.
6. Forbes MS, Hawkey LM, Sperelakis N. The transverse-axial tubular system (TATS) of mouse myocardium: its morphology in the developing and adult animal. Am J Anat. 1984;170:143–162.[Medline] [Order article via Infotrieve]
7. Amsellem J, Delorme R, Souchier C, Ojeda C. Transverse-axial tubular system in guinea pig ventricular cardiomyocyte: 3D reconstruction, quantification and its possible role in K+ accumulation-depletion phenomenon in single cells. Biol Cell. 1995;85:43–54.[Medline] [Order article via Infotrieve]
8. Peachey, LD. The extraction of three-dimensional information from stereo micrographs of thick sections using computer graphics methods. Ann N Y Acad Sci. 1986;483:161–170.[Medline] [Order article via Infotrieve]
9.
Costantin LL. The role of the sodium current in the
radial spread of contraction in frog muscle. J Gen
Physiol. 1970;55:703–715.
10.
González-Serratos H. Inward spread of activation
in vertebrate muscle fibres. J Physiol (Lond). 1971;212:777–799.
11.
Nakajima S, Gilai A. Radial propagation of muscle
action potential along the tubular system examined by potential
sensitive dyes. J Gen Physiol. 1980;76:751–762.
12.
Page E. Stereological measurements of cardiac
ultrastructures implicated in excitation-contraction coupling.
Proc Natl Acad Sci U S A. 1971;68:1465–1466.
13. Page E, McCallister, LP. Quantitative electron microscopic description of heart muscle cells. Am J Cardiol. 1973;31:172–181.[Medline] [Order article via Infotrieve]
14. Cheng H, Cannell MB, Lederer WJ. Propagation of excitation-contraction coupling into ventricular myocytes. Pflügers Arch. 1994;428:415–417.
15. Bers DM, Peskoff A. Diffusion around a cardiac calcium channel and the role of surface-bound calcium. Biophys J. 1991;59:703–721.[Medline] [Order article via Infotrieve]
16. Yasui K, Anno T, Kamiya K, Boyett MR, Kodama I, Toyama J. Contribution of potassium accumulation in narrow extracellular spaces to the genesis of nicorandil-induced large inward tail current in guinea-pig ventricular cells. Pflügers Arch. 1993;422:371–379.
17.
Lipp P, Hüser J, Pott L, Niggli E. Spatially
non-uniform Ca2+ signals induced by the reduction
of transverse tubules in citrate-loaded guinea-pig
ventricular myocytes in culture. J Physiol
(Lond). 1996;497:589–597.
18.
Shacklock PS, Wier WG, Balke CW. Local
Ca2+ transients (Ca2+
sparks) originate at transverse tubules in rat heart cells.
J Physiol. 1995;487:601–608.
19.
Cheng H, Lederer MR, Lederer WJ, Cannell MB. Calcium
sparks and [Ca2+]i waves
in cardiac myocytes. Am J Physiol. 1996;270:C148–C159.
20.
Denk W, Strickler JH, Webb WW. Two-photon laser
scanning fluorescence microscopy. Science. 1990;248:73–76.
21. Soeller C, Cannell MB. Investigation into the morphology of the t-tubular system in living cardiac myocytes by two photon excitation microscopy. Biophys J. 1997;72:A370. Abstract.
22.
Evans AM, Cannell, MB. The role of L-type
Ca2+ current and Na+
current-stimulated Na/Ca exchange in triggering SR calcium release in
guinea-pig cardiac ventricular myocytes. Cardiovasc
Res. 1997;35:294–302.
23. Soeller C, Cannell MB. Construction of a two-photon microscope and optimisation of illumination pulse duration. Pflügers Arch. 1996;432:555–561.
24. Carlsson K. The influence of specimen refractive index, detector signal integration, and non-uniform scan speed on the imaging properties in confocal microscopy. J Microsc. 1991;163:167–178.
25. Hell S, Reiner G, Cremer C, Stelzer EHK. Aberrations in confocal fluorescence microscopy induced by mismatches in refractive index. J Microsc. 1993;169:391–405.
26. Van der Voort HTM, Strasters KC. Restoration of confocal images for quantitative analysis. J Microsc. 1995;178:165–181.
27. Richardson HW. Bayesian-based iterative method of image restoration. J Opt Soc Am. 1972;62:55–59.
28. Van Kempen GMP, Van Vliet LJ, Verveer PJ, Van der Voort HTM. A quantitative comparison of image restoration methods for confocal microscopy. J Microsc. 1997;185:354–365.
29. Lee T, Kashyap RL. Building skeleton models via 3-D medial surface/axis thinning algorithms. CVGIP: Graphical Models and Image Processing. 1994;56:462–478.
30. Bertrand G, Malandain G. A new characterisation of three-dimensional simple points. Pattern Recognition Lett. 1994;15:169–175.
31. Ogata T, Yamasaki Y. High-resolution scanning electron microscopic studies on the three-dimensional structure of the transverse-axial tubular system, sarcoplasmic reticulum and intercalated disc of the rat myocardium. Anat Rec. 1990;228:277–287.[Medline] [Order article via Infotrieve]
32. Stewart JM, Page E. Improved stereological techniques for studying myocardial cell growth: application to external sarcolemma, t system, and intercalated disks of rabbit and rat hearts. J Ultrastruct Res. 1978;65:119–134.[Medline] [Order article via Infotrieve]
33.
Rayns DG, Simpson FO. Transverse tubule apertures in
mammalian myocardial cells: surface array. Science. 1967;156:656–657.
34. Minetti M, Aducci P, Viti V. Interaction of neutral polysaccharides with phosphatidylcholine multilamellar liposomes: phase transitions studied by the binding of fluorescein-conjugated dextrans. Biochemistry. 1979;18:2541–2548.[Medline] [Order article via Infotrieve]
35.
Page E, Surdyk-Droske M. Distribution, surface density,
and membrane area of diadic junctional contacts between plasma membrane
and terminal cisterns in mammalian ventricle. Circ Res. 1979;45:260–267.
36. Satoh H, Delbridge LMD, Blatter LA, Bers DM. Surface:volume relationship in cardiac myocytes studied with confocal microscopy and membrane capacitance measurements: species dependence and developmental effects. Biophys J. 1996;70:1494–1504.[Medline] [Order article via Infotrieve]
37. Hille B. Ionic Channels of Excitable Membranes. Sunderland, Mass: Sinauer Associates; 1992.
38. Tomita Y, Ferrans VJ. Morphological study of the transverse-axial tubular system (TaxTS) in rat heart using ferrocyanide-osmium method and thick sectioning. J Submicrosc Cytol. 1987;19:523–535.[Medline] [Order article via Infotrieve]
39. Forbes MS, van Neil EE. Membrane systems of guinea pig myocardium: ultrastructure and morphometric studies. Anat Rec. 1988;222:362–379.[Medline] [Order article via Infotrieve]
40. Maylie JG. Excitation-contraction coupling in neonatal and adult myocardium of cat. Am J Physiol. 1982;11:H834–H843.
41.
Swynghedauw B, Chevalier B, Charlemagne D, Mansier P,
Carre F. Cardiac hypertrophy, arrhythmogenicity and the new
myocardial phenotype, II: the cellular adaptational process.
Cardiovasc Res. 1997;35:6–12.
This article has been cited by other articles:
 |
|
 |
|
  C. Soeller, I. D. Jayasinghe, P. Li, A. V. Holden, and M. B. Cannell Three-dimensional high-resolution imaging of cardiac proteins to construct models of intracellular Ca2+ signalling in rat ventricular myocytes Exp Physiol, May 1, 2009; 94(5): 496 - 508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Orchard, M. Pasek, and F. Brette The role of mammalian cardiac t-tubules in excitation-contraction coupling: experimental and computational approaches Exp Physiol, May 1, 2009; 94(5): 509 - 519. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hayashi, M. E. Martone, Z. Yu, A. Thor, M. Doi, M. J. Holst, M. H. Ellisman, and M. Hoshijima Three-dimensional electron microscopy reveals new details of membrane systems for Ca2+ signaling in the heart J. Cell Sci., April 1, 2009; 122(7): 1005 - 1013. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Fazakerley, S. P. Lawrence, V. A. Lizunov, S. W. Cushman, and G. D. Holman A common trafficking route for GLUT4 in cardiomyocytes in response to insulin, contraction and energy-status signalling J. Cell Sci., March 1, 2009; 122(5): 727 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. N. Edwards and B. S. Launikonis The accessibility and interconnectivity of the tubular system network in toad skeletal muscle J. Physiol., November 1, 2008; 586(21): 5077 - 5089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tao, S. C. O'Neill, M. E. Diaz, Y. T. Li, D. A. Eisner, and H. Zhang Alternans of cardiac calcium cycling in a cluster of ryanodine receptors: a simulation study Am J Physiol Heart Circ Physiol, August 1, 2008; 295(2): H598 - H609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Orchard and F. Brette t-tubules and sarcoplasmic reticulum function in cardiac ventricular myocytes Cardiovasc Res, January 15, 2008; 77(2): 237 - 244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Soeller, D. Crossman, R. Gilbert, and M. B. Cannell Analysis of ryanodine receptor clusters in rat and human cardiac myocytes PNAS, September 18, 2007; 104(38): 14958 - 14963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Launikonis and E. Rios Store-operated Ca2+ entry during intracellular Ca2+ release in mammalian skeletal muscle J. Physiol., August 15, 2007; 583(1): 81 - 97. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Lukyanenko, A. Ziman, A. Lukyanenko, V. Salnikov, and W. J. Lederer Functional groups of ryanodine receptors in rat ventricular cells J. Physiol., August 15, 2007; 583(1): 251 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Brette and C. Orchard Resurgence of Cardiac T-Tubule Research Physiology, June 1, 2007; 22(3): 167 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Swift, T. A. Stromme, B. Amundsen, O. M. Sejersted, and I. Sjaastad Slow diffusion of K+ in the T tubules of rat cardiomyocytes J Appl Physiol, October 1, 2006; 101(4): 1170 - 1176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pasek, J. Simurda, and G. Christe The functional role of cardiac T-tubules explored in a model of rat ventricular myocytes Phil Trans R Soc A, May 15, 2006; 364(1842): 1187 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-S. Song, E. A. Sobie, S. McCulle, W. J. Lederer, C. W. Balke, and H. Cheng Orphaned ryanodine receptors in the failing heart. PNAS, March 14, 2006; 103(11): 4305 - 4310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Hunter and P. Nielsen A Strategy for Integrative Computational Physiology Physiology, October 1, 2005; 20(5): 316 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. P. Head, H. H. Patel, D. M. Roth, N. C. Lai, I. R. Niesman, M. G. Farquhar, and P. A. Insel G-protein-coupled Receptor Signaling Components Localize in Both Sarcolemmal and Intracellular Caveolin-3-associated Microdomains in Adult Cardiac Myocytes J. Biol. Chem., September 2, 2005; 280(35): 31036 - 31044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Swietach and R. D Vaughan-Jones Relationship between intracellular pH and proton mobility in rat and guinea-pig ventricular myocytes J. Physiol., August 1, 2005; 566(3): 793 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Payne, Z. Zheng, E. Gonzalez, Z.-M. Wang, M. L. Messi, and O. Delbono External Ca2+-dependent excitation-contraction coupling in a population of ageing mouse skeletal muscle fibres J. Physiol., October 1, 2004; 560(1): 137 - 155. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Vadakkadath Meethal, K. T. Potter, D. Redon, D. M. Heisey, and R. A. Haworth Ca transients from Ca channel activity in rat cardiac myocytes reveal dynamics of dyad cleft and troponin C Ca binding Am J Physiol Cell Physiol, February 1, 2004; 286(2): C302 - C316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. O. Gray, H.-Z. Zhou, I. Schafhalter-Zoppoth, P. Zhu, D. Mochly-Rosen, and R. O. Messing Preservation of Base-line Hemodynamic Function and Loss of Inducible Cardioprotection in Adult Mice Lacking Protein Kinase C{epsilon} J. Biol. Chem., January 30, 2004; 279(5): 3596 - 3604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Jacobs, C. Soeller, A. M. G. Sisley, M. B. Cannell, and P. J. Donaldson Gap Junction Processing and Redistribution Revealed by Quantitative Optical Measurements of Connexin46 Epitopes in the Lens Invest. Ophthalmol. Vis. Sci., January 1, 2004; 45(1): 191 - 199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F R Quinn, S Currie, A M Duncan, S Miller, R Sayeed, S M Cobbe, and G L Smith Myocardial infarction causes increased expression but decreased activity of the myocardial Na+--Ca2+ exchanger in the rabbit J. Physiol., November 15, 2003; 553(1): 229 - 242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Balijepalli, A. J. Lokuta, N. A. Maertz, J. M. Buck, R. A. Haworth, H. H. Valdivia, and T. J. Kamp Depletion of T-tubules and specific subcellular changes in sarcolemmal proteins in tachycardia-induced heart failure Cardiovasc Res, July 1, 2003; 59(1): 67 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Brette and C. Orchard T-Tubule Function in Mammalian Cardiac Myocytes Circ. Res., June 13, 2003; 92(11): 1182 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rubart, E. Wang, K. W. Dunn, and L. J. Field Two-photon molecular excitation imaging of Ca2+ transients in Langendorff-perfused mouse hearts Am J Physiol Cell Physiol, June 1, 2003; 284(6): C1654 - C1668. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Inoue and J. H.B. Bridge Ca2+ Sparks in Rabbit Ventricular Myocytes Evoked by Action Potentials: Involvement of Clusters of L-Type Ca2+ Channels Circ. Res., March 21, 2003; 92(5): 532 - 538. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A Blatter, J. Kockskamper, K. A Sheehan, A. V Zima, J. Huser, and S. L Lipsius Local calcium gradients during excitation-contraction coupling and alternans in atrial myocytes J. Physiol., January 1, 2003; 546(1): 19 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. R. Heinzel, V. Bito, P. G.A. Volders, G. Antoons, K. Mubagwa, and K. R. Sipido Spatial and Temporal Inhomogeneities During Ca2+ Release From the Sarcoplasmic Reticulum in Pig Ventricular Myocytes Circ. Res., November 29, 2002; 91(11): 1023 - 1030. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Yang, C. Pascarel, D.S. Steele, K. Komukai, F. Brette, and C.H. Orchard Na+-Ca2+ Exchange Activity Is Localized in the T-Tubules of Rat Ventricular Myocytes Circ. Res., August 23, 2002; 91(4): 315 - 322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ozcelik, B. Erdmann, B. Pilz, N. Wettschureck, S. Britsch, N. Hubner, K. R. Chien, C. Birchmeier, and A. N. Garratt Conditional mutation of the ErbB2 (HER2) receptor in cardiomyocytes leads to dilated cardiomyopathy PNAS, June 25, 2002; 99(13): 8880 - 8885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S Launikonis and D G. Stephenson Tubular system volume changes in twitch fibres from toad and rat skeletal muscle assessed by confocal microscopy J. Physiol., January 15, 2002; 538(2): 607 - 618. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zou, A. Yao, W. Zhu, S. Kudoh, Y. Hiroi, M. Shimoyama, H. Uozumi, O. Kohmoto, T. Takahashi, F. Shibasaki, et al. Isoproterenol Activates Extracellular Signal-Regulated Protein Kinases in Cardiomyocytes Through Calcineurin Circulation, July 3, 2001; 104(1): 102 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J M Cordeiro, K W Spitzer, W R Giles, P E Ershler, M B Cannell, and J H B Bridge Location of the initiation site of calcium transients and sparks in rabbit heart Purkinje cells J. Physiol., March 1, 2001; 531(2): 301 - 314. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R Houser Reduced abundance of transverse tubules and L-type calcium channels: another cause of defective contractility in failing ventricular myocytes Cardiovasc Res, February 1, 2001; 49(2): 253 - 256. [Full Text] [PDF]
|
|
|
|
|
|
J.-Q. He, M. W Conklin, J. D Foell, M. R Wolff, R. A Haworth, R. Coronado, and T. J Kamp Reduction in density of transverse tubules and L-type Ca2+ channels in canine tachycardia-induced heart failure Cardiovasc Res, February 1, 2001; 49(2): 298 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. C. Britton, W. J. Hatton, C. F. Rossow, D. Duan, J. R. Hume, and B. Horowitz Molecular distribution of volume-regulated chloride channels (ClC-2 and ClC-3) in cardiac tissues Am J Physiol Heart Circ Physiol, November 1, 2000; 279(5): H2225 - H2233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. M. Sejersted and G. Sjogaard Dynamics and Consequences of Potassium Shifts in Skeletal Muscle and Heart During Exercise Physiol Rev, October 1, 2000; 80(4): 1411 - 1481. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lorente, A. Lacampagne, Y. Pouzeratte, S. Richards, B. Malitschek, R. Kuhn, B. Bettler, and G. Vassort gamma -Aminobutyric acid type B receptors are expressed and functional in mammalian cardiomyocytes PNAS, July 18, 2000; 97(15): 8664 - 8669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. J. Haynes, K.-Y. Ling, R. R. Preston, Y. Saimi, and C. Kung The Cloning and Molecular Analysis of pawn-B in Paramecium tetraurelia Genetics, July 1, 2000; 155(3): 1105 - 1117. [Abstract] [Full Text]
|
|
|
|
|
|
R. R. Kaprielian, S. Stevenson, S. M. Rothery, M. J. Cullen, and N. J. Severs Distinct Patterns of Dystrophin Organization in Myocyte Sarcolemma and Transverse Tubules of Normal and Diseased Human Myocardium Circulation, June 6, 2000; 101(22): 2586 - 2594. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||