Download Quantitative ultrastructural analysis in cardiac membrane physiology

Quantitative
ultrastructural
analysis
in cardiac membrane
physiology
heart muscle; plasma membrane;
lum; mitochondrial
cristae
T system; nexus; sarcoplasmic reticu-
MICROGRAPHS
of heart muscle can be analyzed to obtain quantitative
data about membrane
areas and membrane structure in heart muscle cells.
The application of this analysis to the plasma membrane, sarcoplasmic reticulum,
and mitochondrial
membranes provides quantitative
structural information needed by cell physiologists interested in cardiac
membrane physiology. This review will consider quantitative measurements made on ultrathin sections of
fixed heart muscle photographed by transmission electron microscopy (TEM) and on electron micrographs of
membrane replicas prepared from freeze-fractured tissue. The methods applied to thin sections include the
use of map readers and planimeters, point and intersection counting (“stereology”), and other morphometric
techniques (9, 15, 37, 39, 42, 74, 75). These methods
yield values for the areas of the various types of
membranes in myocardial cells and for the volumes of
cells and of membrane-limited
cellular subcompartments. The methods applied to freeze-fractured membrane replicas yield data on the size, spacing, and
distribution of particles and other structures dispersed
in the membrane (1, 29, 37, 39, 42).
ELECTRON
STRUCTURALASPECTSOFCARDIACCELLULARPHYSIOLOGY
Morphometry of thin sections obtained by TEM is
useful for problems in membrane physiology that de0363-6143/78/0000-0000$01.25
Copyright
0
1978 the American
Physiological
pend on a knowledge of membrane area. Examples of
the application of such morphometric data to the membrane physiology of heart muscle have included the
following: determination of the membrane area of highresistance plasmalemma for the purpose of expressing
solute fluxes (52), ionic conductances (22, 23), and
membrane capacitance (23, 43) in absolute units
(moles, mhos, and farads per unit area of membrane);
measurement of isotopic potassium fluxes across lowresistance (nexal) junctions between myocardial cells
with concomitant morphometric estimation of nexal
membrane area (73); determination of the membrane
area of sarcoplasmic reticulum (SR) (47, 49) and mitochondrial cristae (69, 70) relative to the volume of
myofibrils for which these two membrane systems
serve, respectively, as a critical control system for
contraction and relaxation (the SR) and the primary
source of energy (ATP) supply (the cristae); measurement of volume changes in membrane-limited subcellular compartments (SR and mitochondria) in intact
hearts after osmotic perturbations or alterations in
extracellular ionic composition (55); and compartmental analysis of heart muscle by the simultaneous use of
both extracellular
tracers and morphometry to estimate cellular and extracellular
volumes (61). These
selected examples of morphometry represent applications of morphometry to frequently encountered aspects
of classical cardiac membrane physiology. In addition
Society
Cl47
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
PAGE, ERNEST. Quantitative
ultrastructural
analysis in cardiac membrane physiology.
Am. J. Physiol. 235(5): C147-Cl%,
1978 or Am. J.
Physiol: Cell Physiol.
4(3): C147-Cl!%,
1978. -Quantitative
measurements on electron micrographs
of heart muscle can yield information
useful for cellular
physiologists
and at present not obtainable
in other
ways. These methods are subject to preparative
artifact, sampling problems, and problems inherent
in the mathematical
description
of ultrastructure.
Nevertheless
they provide the best available
data for membrane areas of the plasmalemma
and its components,
as well as for
membrane
areas of the sarcoplasmic
reticulum
and mitochondria.
Morphometric methods can be used to study growth of membranes.
Changes
in the volumes of intracellular
membrane-limited
subcompartments
can
also be measured.
Quantitative
analysis of freeze-fractured
membrane
replicas can be carried out either by a statistical
approach or by optical
diffraction.
In this way, physiological
perturbations
or developmental
events leading to changes in membrane
permeability
can be studied for
correlated changes in membrane structure.
Cl48
E. PAGE
METHODOLOGICAL
,
CONSIDERATIONS
Quantitative
analysis of cardiac membranes
in electron micrographs
is subject to uncertainties
similar to
those that beset qualitative
interpretations
based on
the same material:
sampling
problems
and artifacts
arising during preparation
of the tissue for electron
microscopy.
To these must be added problems peculiar
to the quantitative
analysis: consideration
of the adequacy of the assumed
structural
model and of the
approximations
used to derive the morphometric
equations.
Sampling of the tissue for electron microscopy
must
be appropriate
both in space and in time. Both thin
sections and freeze-fractured
membrane
replicas can
sample only a very small fraction
of the tissue. In
working
with mammalian
heart muscle,
the tissue
must be sampled at multiple sites. Multiple sampling
is required
to obtain a useful average value for the
membrane
content of the tissue as well as to rule out
systematic
nonuniformities
in cellular membrane
content. In practice the number of samples can be held to
a workable
minimum
of two or three tissue blocks only
by using small tissues like papillary
muscles or Purkinje fibers, or well-defined
and small regions of the
atria1 or ventricular
walls from small animals.
The
appropriate
timing
in obtaining
a sample becomes
important
in following
time-variant
changes in the
amount or structure
of membranes.
Distortions
(shrinking,
swelling,
fragmentation
of
membrane-limited
ultrastructures,
appearance,
disappearance, fusion or redistribution
of membrane
particles, etc.) may arise during any of the multiple steps by
which the final electron micrograph
is prepared from
the intact tissue. Indeed, all electron micrographs
of
heart muscle (whether
from TEM of tissue preserved
by chemical fixation or freeze substitution
or electron
micrographs
of membrane
replicas
prepared
from
freeze-fractured
material)
are experimental
artifacts
with properties
significantly
different from those of the
native
(in situ) state. The issue is not whether
the
electron micrograph
represents
a distortion
of biol .ogical reality,
but how to quantify
the distortion.
For
examined
glutaraldehyde-fixed
skeletal
muscle
bY
TEM, systematic measurements at each step in the
preparative procedure disclosed that the distortions,
though significant, were not large enough to vitiate the
quantitative
analysis (14). In mammalian heart muscle, an in vivo compartmental analysis with isotopitally measured extracellular tracers yielded the same
values for cell volume and extracellular volume as a
morphometric analysis performed after fixing the same
hearts and processing them as for TEM (61). For the
freeze-fracture method, the structural distortions introduced during preparation are just beginning to be
studied (8, 19, 25, 58). Thus, in order to validate
morphometric data, it is desirable to confirm them by
other methods (46, 49).
A discussion of morphometric theory, for which the
reader is referred to the appropriate monographs and
reviews (9, 15, 74, 75), is beyond the scope of this
article. Physiologists using data obtained by morphometry should, however, be aware that the equations used
to calculate membrane areas from morphometric data
are derived on the basis of simplifying assumptions.
Thus, assumptions are made about the extent, orientation, and shape of heart muscle cells, membranes, and
intracellular membrane-limited structures. If, as often
happens, the assumed model corresponds only partially
to the geometrical arrangement of membranes in the
tissue, the equations based on this model may lead to
of membrane
systematic over- or underestimation
areas.
MORPHOMETRY
OF CARDIAC
MUSCLE
EXAMINED
BY TEM
This section will tabulate and discuss selected morphometric data obtained by TEM of fixed heart muscle
that has been examined in thin, sections stained with
salts of uranium and lead. The following two sections
(also dealing with TEM) will consider, respectively, the
changes in the volumes of membrane-limited intracellular subcompartments in response to experimental
perturbations, and the growth of some of these membrane systems.
Components of the Plasma Membrane
TEM and freeze-fracture techniques show that the
plasma membranes of both myocardial cells and Purkinje fibers are each made up of several structurally
differentiable components (18, 41, 48, 63). Among the
identifiable components that have been measured in
myocardial cells are the external sarcolemmal envelope, the T system, the caveolae, the areas of diadic
junctional complexes of the plasmalemma with subjacent terminal cisterns of the SR, the gap junctions or
nexuses, the insertions of the myofibrils into the ends
of the cells (fasciae adhaerentes), and the desmosomes.
The caveolae and diadic junctional complexes are distributed over the surfaces of both external sarcolemma
and T system. In adult ventricular heart muscle the
nexuses, fasciae adhaerentes, and desmosomes are localized predominantly to the ends of the cells. In the
plasmalemma of atria1 muscle and Purkinje fibers,
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
the technique
is useful for quantifying
membrane
growth
in response
to developmental,
hormonal,
or
other stimuli.
In heart muscle, areas of particular
membranes
as
determined
by morphometry
of thin sections are often
constant or change only slowly compared to the rates of
many physiologically
important
changes in membrane
properties.
By contrast, the freeze-fracture
technique is
potentially
applicable
to the detection
of relatively
rapid changes in membrane structure.
The detection of
such rapid changes requires
that the membrane
be
preserved
quickly
enough to prevent
an artifactual
redistribution
of membrane
components.
If a redistribution can be prevented,
the freeze-fracture
technique
should be useful for examining the structural
effects of
various experimental
perturbations
commonly studied
in cardiac cellular
electrophysiology
and membrane
transport.
INVITED
Cl49
REVIEW
TABLE
1. Components of plasma membrane area in
mammalian ventricular myocardial cells
pm2 membrane
area/pm:’
volume
Component
Rabbit Right
Ventricular
Papillary
Muscle
External
sarcolemma
Total
“Unmodified”
plasmalemma
Caveolae
Junctional
(diadic)
complexes
terminal
cisterns
of SR
T System
Total
Unmodified
Caveolae
Junctional
terminal
Nexus
0.33*
0.28
0.04*
with
plasmalemma
(diadic)
cisterns
(gap junction)
complexes
of SR
Rat Left Ventricle
0.31
0.29
not known
0.010
0.023
0.23*
0.15
0.08
not known
0.16
with
cell
0.03*
0.042
0.017-t
0.069
0.0047
Table compiled
from Refs. 32, 33, 54, 72, and unpublished
studies
of E. Page and M. F. Surdyk.
Data are for -4.O-kg
rabbits
and 220to 300-g rats. Because
no data for caveolae
are available
for rat
ventricles,
the values for rat heart are uncorrected
for caveolae.
The
contribution
of the transverse
cell boundary
(“intercalated
disk”)
to
surface area/unit
cell volume
has not been measured
directly.
For
the rat ventricle,
assuming
a cell length
of 102 pm, a cell diameter
of 13.3 pm, and a folding
factor of 3.0 for the transverse
boundary
(50), each transverse
boundary
can be calculated
to make a contri* Assuming
two caveolae
bution of -0.029
~rn2/~rn:~ cell volume.
t Preliminary
estimates
of nexal
area for
per caveolar
neck.
myocardial
cells of rabbit
left ventricular
free wall,
made with a
more accurate
method
of Dr. K. Nakata
in the author’s
laboratory,
gave the somewhat
lower estimate
of 0.0042 + 0.0003 ~rn”/~rn%
TABLE
2. Membrane
in rat left ventricular
areas of intracellular
myocardial
cells
Mitochondria
Area
of inner
per unit
cell volume
20
membrane
per unit
mitochondrial volume
57
Sarcoplasmic
+ cristae
per unit
myofibrillar
volume
42
Area
of cisternal
SR
membranes
Reticulum
Area
of noncisternal
SR
per unit
cell volume
per unit
myofibrillar
volume
per unit
cell volume
per unit
myofibrillar
volume
0.19
0.40
1.03
2.2
All relationships
are expressed
as ,zrnz/prn:‘.
Left ventricular
free
walls of 200- to 260-g female
Sprague-Dawley
rats. References
for
original
data are: mitochondria
(69) and sarcoplasmic
reticulum
(47,
54). All values
for mitochondrial
and SR membrane
areas
are
minimal
estimates
because the morphometric
method
used systematically
underestimates
membrane
area. Additional
data on cardiac
mitochondrial
membranes
may be found in Refs. 24, 36, 60, 64, 65.
involved in diadic junctions with underlying terminal
cisterns. By contrast, such junctions are relatively rare
(7.7% and 3.2% of total membrane area, respectively)
at the external sarcolemmas of these two tissues.
Nexal contacts between cells occupy only a small
fraction of the total cell surface. It seems probable that
the morphometric technique underestimates the membrane area of the gap junction because nexuses oriented
obliquely (tilted) or tangent to the plane of section may
be missed by the analysis. Matter (38) attempted to
determine the effect of section tilting on the gap junction area in the rat by the use of a goniometer. The
estimate of mean nexal area per unit cell volume
calculable from Matter’s data is somewhat smaller
than that in Table 1, perhaps because of differences in
the assumptions underlying the two morphometric determinations.
The external sarcolemmal envelope of mammalian
ventricular heart muscle, like that of amphibian skeletal muscle (12), is plicated at periodic intervals along
the length of the myocardial cell (33). The resulting
sarcolemmal folds are always oriented perpendicularly
to the long axis of the cells. The model underlying the
equations used to derive the data in Table 1 takes these
folds into account (72). Although sarcolemmal folds can
be caused to unfold by passive stretch (12, 33), the data
in Table 1 are, to a first approximation, independent of
the degree of sarcolemmal folding, at least over the
range of sarcomere lengths that occur under physiological conditions. Sarcolemmal folds are also a prominent
feature of cardiac Purkinje fibers (22, 43).
Intracellular Membranes: Mitochondrial
Cristae and SR
Table 2 gives membrane areas for mitochondrial
respiratory membrane (mitochondrial cristae plus inner membrane), as well as for the cisternal and noncisternal parts of the SR. The data are for myocardial cells
of rat left ventricular
free wall, for which additional
data are given in Tables 1 and 3. In Table 2 the units of
reference for the areas of mitochondrial and SR membranes are myocardial cell volume and also myofibrillar volume. By referring these membrane areas to the
volume of the myofibrils, one can express the amount
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
these three specializations
may occur extensively
in
areas where cells are laterally apposed.
The most extensive
morphometric
data on plasma
membrane
components
in heart muscle are available
for rabbit right ventricular
papillary
muscle (72) and
rat left ventricle
(47, 51, 69, 72) (Table 1); useful but
somewhat
more limited data are also available
for
sheep Purkinje
fibers
(22, 23, 43). For ventricular
myocardial
cells of rats and rabbits
over the body
weight ranges included in Table 1, from 27-36% of total
plasma membrane area is in the T system. Preliminary
observations
on rabbit right ventricular
papillary muscle (32,33) suggest that the caveolar plasma membrane
increases total plasmalemmal
area by 14-21%; the final
surface amplification
depends on the measured
or assumed number of caveolae per caveolar neck, a ratio as
yet unknown.
(The contributions
of caveolae to plasma
membrane
area are similar
for the T system
and
external
sarcolemma).
In rat ventricle
the area of
plasma membrane
involved
in junctional
complexes
with the terminal cisterns of the SR is about 20% of the
total plasmalemmal
area in the external sarcolemmal
envelope and T system
(exclusive
of caveolae and
intercalated
disks) (Table l), and this figure is surprisingly large. Moreover,
about 75% of the plasmalemmal
area so involved is localized in the T system. Indeed,
about half of the area of the plasmalemma
in the T
system of rat left ventricle and 21% of the corresponding area in rabbit
papillary
muscle (Table 1) are
Cl50
E. PAGE
TABLE
3. Components of myocardial cell volume in
mammal ian ventricle and atrium
pm’{ component/pm.’
Component
Myofibrils
Mitochondria
Matrix
Cristae
+ inner membrane
“Intercristal
space”*
Sarcoplasmic
reticulum
Terminal
cisterns
Noncisternal
SR
Nucleus
Other
(e.g., sarcoplasm)
cell
Rat Left Ventricular Free Wall
Guinea Pig Left
Atrium
0.467
0.36
0.17
0.15
0.04
0.035
0.0035
0.0315
0.02
0.118
0.414
0.144
0.022
0.005
0.017
0.041
0.379
* Recent
studies
by Sjostrand
(68) on rat heart
mitochondria
suggest that the “intercristal
space” is a preparative
artifact.
References for data are rat left ventricle
(47, 69), guinea pig atrium
(20).
It is of interest to convert the relative membrane
areas and volume fractions in Tables l-3 into absolute
values (membrane area and organelle volume in a
single myocardial cell or in a known volume or weight
of tissue). Average values for single cells could be
derived by multiplying the tabulated quantities by the
mean myocardiai cellular volume. This quantity depends on the stage of cardiac growth. Korecky and
Rakusan (30) and Rakusan et al. (62) have published
data on cell volume of myocardial cells from the left
ventricular free walls of male rats. If these figures can
be extrapolated to the left ventricles of female rats
(Tables i--3), the membrane and organelle contents of
single rat myocardial cells could be obtained by multiplying the tabulated values by 24 x lo3 J-Lrn:’(the
approximate volume of a single myocardial cell for rats
with body weights corresponding to these data). Alternatively the absolute membrane and organelle contents
in a given dry weight of rat left ventricle can be derived
if the total volume of myocardial cells per unit dry
weight and the density of the tissue are known. Polimeni (61), using a combination of morphometric and
tracer techniques in the author’s laboratory, has measured with unusual precision the cell water content per
unit dry weight of rat left ventricular free wall. After
subtracting the volume of the T system (reckoned as
“cellular”
in Tables l-3)) Polimeni’s morphometric
value for the cell volume and his tracer measurements
for cell water content were, respectively, 2.46 cmVg dry
wt or -2.46 g cell water/g dry wt (0.57 cm3/g wet wt or
-0.57 g cell water/g wet wt). The volume and weight of
cell water are not exact indices of cell volume, but can
be used as a rough approximation (in converting tissue
weight and volume for use in Tables 1-3, 1 cm:] = 101’
,um3). For the female Sprague-Dawley rats from whose
hearts the data in Tables l-3 were derived, the weight
of the left ventricular
free wall + interventricular
septum (W& is a linear function of body weight (W,)
over a range of body weights from 40 to 300 g; the
regression equations, which have a correlation coefficient of + 0.99, are WL\.,,, = 52.45 + 1.98 W, and WLi-Cir,
= 9.32 + 0.49 W, when ventricular weights (WJ are
expressed on a basis of milligrams wet weight and dry
weight, respectively, and W, is given in grams (72). It
should thus be possible to make approximate predictions of the membrane and organelle contents of rat left
ventricles from body weight. For example, given that
there are about 2.46 x 1012,urn:’ myocardial cells per
gram dry left ventricle and that for a 250-g rat the left
ventricular
dry weight is 131.8 mg, then the total
volume of cells is (0.1318)(2.46 x 1012) = 324.2 x 10”
pm:’ and the total external
sarcolemmal area =
(0.31)(324.2 x log) = 100.5 X 10” pm% Similarly, one
could derive the left ventricular content of each of the
membrane types and organelles from the rat data in
Tables l-3. If myocardial cell volume is in fact 24 x 10”
,urn” (30, 62), then there are 324 x log/24 x lo;{ = 13.5
x 10” myocardial cells in left ventricles of this size, and
the average contents of each membrane type and organelle in a single myocardial cell can readily be computed.
These calculations are uncorrected for the volume of
nonmyocardial cells; this correction is small because
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
of each membrane
with respect to the volume of the
organelle to which it is functionally
most closely related: as a supplier of ATP to the contractile
system
(the cristae) and as a control system for contraction
and
relaxation
(the SR).
In rat left ventricular
myocardial
cells, the area of
mitochondrial
respiratory
membrane
per unit cell volume exceeds that of SR by 16-fold (Table 2) and of the
total plasmalemma
(exclusive of caveolae) by about 41fold (Tables 1 and 2). This fact helps to explain why,
when membrane vesicles of fragmented
SR and plasma
membrane
are isolated by centrifugation
from homogenates of heart muscle, they are invariably
contaminated with
fragmented
mitochondrial
membranes,
with sedimentation
characteristics
similar to those of
fragmented
SR and plasma membrane.
As indicated in
the table, all values of membrane area for both SR and
cristae are minimal estimates,
again due to systematic
errors in the morphometric
method. In this case the
errors result from the fact that the spatial extent of the
structures
analyzed may be only slightly greater than
the thickness
of the plane of section; that some structures (particularly
the cisternal
and noncisternal
SR
components
oriented transversely
to the long axis of
the cell) may be sectioned tangentially;
and-that
the
equations used to calculate area are relatively
insensitive to transversely
oriented components of the SR.
Table 3 gives the components
of muscle cell volume
for rat left ventricular
free wall and guinea pig left
atrium. The units of the table are pm3 of each component per pm3 of cell volume. For rat left ventricular
myocardial
cells, the volume fractions
of Table 3 may
therefore
be combined with the membrane
areas of
Table 2 to compute the surface area-volume
ratios for
the mitochondrial
matrix and the two portions of the
SR. If the volumes of the matrix
and of the two SR
compartments
vary in different functional
states, such
ratios might be useful physiological
tools. The measurement
of volume changes in the SR by point and
intersection
counting methods like those used to obtain
the data in Tables 2 and 3 is, however,
too timeconsuming
to be practically
useful; a more suitable
method is the use of a device for sizing the crosssectional areas of SR tubules (55).
INVITED
Cl51
REVIEW
such cells, though n umerous,
myocardia .l volume.
Use of Morphometry
across Intrkcellular
to Monitor
Membranes
occupy a small fraction
of
Transport
of Membranes
of Myocardial
in Growth
Cells
and
In utero and during the first few days or weeks after
birth, mamma .lian ventricu .lar heart muscle grows by a
simultaneous
increase
in both size and number
of
myocardial
cells. Thereafter
these cells enlarge in diameter and length, but do not, except under abnormal
conditions, increase in number (78, 79). It is now widely
type both the
appreciated
that for each membrane
amount of membrane
surface and the membrane properties change during embryonic
development
of heart
muscle. It is much less well understood
that the relative amounts of a particular
membrane type in myocardial cells may vary, not only during embryogenesis,
or pathological
postnatal
but also during
normal
growth.
For cardiac cellular physiologists
it is often
important
to know
whether
an observed change in
electrophysiological
or transport
properties
of heart
muscle results
from a change in the properties
of a
particular
membrane,
from a change in the amount of
membrane
surface, or from changes in both of these
variables. This differentiation
is difficult in heart muscle. Biochemical
characterization
of membrane vesicles
prepared by fractionation
of mechanically
or enzymatitally disrupted
heart muscle gives inaccurate
values
for the membrane
content of the undisrupted
tissue.
The inaccuracy
results
from the loss of membrane
during fractionation
and from the cross-contamination
of one membrane
type with another. Loss and crosscontamination
introduce particularly
serious errors in
the estimation
of the SR and the various subdivisions
of the plasma membrane.
Physiologists
who need to decide to what extent
changes in solute fluxes, ion conductances,
membrane
capacity, and other area-intensive
membrane
properties are due to growth-related
changes in membrane
area must thus depend on a morphometric
analysis of
membrane
growth.
Some applications
of morphometry
to the participation
of plasma membrane,
SR, and
mitochondrial
membranes
in the growth of myocardial
cells are illustrated
below ‘. In these applications
it
turned out to be most useful to ask the question: During
growth of the heart, how does the area of the membrane
increase relative
to the increase
in myocardial
cell
volume? And sometimes: How does the membrane area
increase relative to the volume of the organelle (e.g.,
the mitochondrion)
of which the membrane
forms a
part, or relative to the volume of the organelle which
the membrane
controls or supplies with energy (e.g.,
the myofibrils)?
Growth
of external sarcolemma
and T system.
A
morphometric
analysis of normal postnatal
growth
of
rat left ventricular
myocardial
cell s (47) suggested that
the plasmalemma
in this tissue grows in accordance
with a relatively
simple rule: The sum (referred to unit
cell volume) of the membrane
areas in the external
plasmalemmal
envelope and in the T system remains
approximately
constant.
That is, additional
plasma
membrane area accumulates
in the T system in such a
way as to maintain
constant the composite surface-tovolume ratio, namely
(membrane
area of external
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
Net movements
of water into and out of membranelimited compartments
in response to tran .smembrane
gradients in the activity of water or to net movements
of solute across the membrane
are associated
with
changes in compartmental
volume. The measurement
of changes in cell volume based on this principle is a
very old technique for studying the permeability
of the
plasma membrane
to solutes and water. The classical
studies of Boyle and Conway
(5) and of Hodgkin
and
Horowitz
(27), who applied this principle to frog skeletal muscle cells, were models for its subsequent
application to mammalian
heart muscle cells (45, 53). In
heart muscle
other, more direct,
techniques
have
of cell volume for the
superseded
th .e measurement
study of plasma membrane
permebility
to ions. However, even today. there are no comparably
direct methods for studying the permeability
of intracellular
membranes in intact or %kinned”
heart muscle. Although it
has proved possible to perfuse the sarcoplasmic
face of
the SR directly
by skinning
myocardial
cells (i.e.,
removing
the sarcolemma)
(16), inferences
about the
permeability
of the SR membrane to ions are still made
very indirectly
by measuring
the contractile
response
of the cells (17).
These considerations
suggest that it might be worthwhile to use morphometric
techniques
for studying
more directly
the volume changes that intracellular
compartments
undergo
in response
to experimental
perturbations
of interest
for cardiac cellular physiology. To be useful for studying
net movements
of ions
and water, the technique would have to fix the tissue
rapidly enough to avoid or minimize volume changes
during fixation. At the present stage of technical development, this requirement
would seem to be most nearly
met in intact muscle fixed uniformly
by perfusion
through its blood supply, or in single cells (either intact
or skinned),
fixed by immersion
in fixative. So far, the
feasibility
of this approach has been explored only by
measurements
of SR volume in intact, perfused
rat
hearts (55). For this purpose it proved expedient
to
estimate the distribution
of cross-sectional
areas for
tubules of noncisternal
SR with an instrument
that
permits a much larger number of measurements
than
can be conveniently
obtained by conventional
methods
based on point and intersection
counts or planimetry.
Figure 1 shows the frequency distributions
of SR crosssectional areas obtained in -intact rat hearts perfused
through the coronary circulation
with solutions of two
different osmolalities
and then fixed. Figure 2 is a plot
of mean cross-sectional
area (an index of volume)
against the reciprocal of extracellular
osmolality.
The
osmolality
was varied by adding NaCl to an isotonic
rat Ringer
solution.
Figures
1 and 2 indicate that
morphometry
applied to electron micrographs
of rapidly fixed heart muscle can in fact detect osmotically
induced volume changes in the SR.
Participation
Development
Cl52
E. PAGE
240
l
0
20 -
0
16 -
0
l
0
ID
8
0
0
0
o*
m
0015
;o
:&o
0
0
,
I
-025
00%
a
m
I
,035
X
I
I
.045
,040
1
,050
In
,055
op
,060
04,
r/
q.
.065y.070
I
'.b5
A
-
a
f20
t
a
A
X
A
Ls
.015
.020
.025
,030
,035
EQUIVALENT
,040
.045
DIAMETER
.050
.055
( p METERS)
.060
,065
1. Frequency
distributions
of equivalent
cross-sectional
diof longitudinally
oriented
sarcoplasmic
reticulum
in mycells of rat left ventricles
perfused
through
the coronary
on the Langendorff
cannula.
Measurements
were made on
micrographs
with Zeiss model TGZ3 particle
size analyzer.
A: after perfusion
for 5 min with control
electrolyte
solution
approximately
isosmolal
with blood plasma. B: after perfusion
with similar
solution
made 1.88 x isosmolal
by addition
of NaCl.
Symbols
refer
to experiments
from different
hearts.
From Ref. 55, reproduced
by
permission.
plasmalemma + T system)/(unit cell volume). For rat
left ventricle this relationship was also found to hold
approximately for growth stimulated by thyroxin or by
aortic constriction (49). Over the somewhat narrower
range of body weights examined (2.5-4.0 kg), the relationship also predicted the growth of plasma membrane
FIG.
ameters
ocardial
vessels
electron
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
O
IY
J28
IL
0
I~24
8
a
a
INVITED
Cl53
REVIEW
in rabbit right ventricular
papillary
muscles
(72). A
recent, more rigorous, analysis and more extensive
measurements based on the model of a folded sarcolemma developed by Stewart (72) (Fig. 3) have confirmed these conclusions for rat ventricles of 75 to 300g rats. For unknown reasons, the conclusion does not
hold for ventricles from very small rats (72).
Growth of gap junctional membrane. Figure 4 shows
that the area of gap junctional membrane that electrically couples rat ventricular myocardial cells is signifi.00160
.00100
?itOOO80
\
b
.00060
.00040
/
r
F //
,’
0
.00020
w
02
0
04
06
08
OShOLAL=ITY
(R;LATI;E
IO
I-’
l
2. Dependence
of volume
of longitudinally
oriented
sarcoplasmic reticulum
(LSR) on relative
osmolality
of perfusing
solution
(protocol
as for Fig.
1). Relative
osmolality
(w) was defined
as
osmolality
of perfusing
solution/322
mosmol/kg
HzO, the denominator being
the osmolality
of the isosmolal
control
solution.
The
equivalent
volume
of a unit length of LSR lumen was approximated
for each heart by a right circular
cylinder
with diameter
equal to d’
(diam in pm of circle equivalent
in area to mean cross-sectional
area
of LSR profiles)
and is given by (7~/4)(d’)~.
The data have been fitted
by the method
of least squares to the line (~/4)(d’)~
= 0.00100 ~r)-l +
0.000271,
with
r = +0.94
and standard
error
of the estimate
=
0.000094.
From Ref. 55, reproduced
by permission.
w
FIG.
0
0
0.0
’
I
I
0
40
80
I
I
120
160
FIG. 3. Composite
surface-to-volume
ratio
(membrane
area of external
plasmalemmal
envelope + area of plasma membrane
in T system)
per unit myocardial
cell volume
remains
approximately
constant
during
postnatal
growth
of rats
from 44 to 300 g. During
this time left ventricular
wet weight
increases
linearly
by K&fold
predominantly
through
enlargement
of myocardial
cells.
Replotted
from Ref. 72.
a
:
I
I
I
200
240
280
BODY WEIGHT (g)
1
300
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
N
‘g
cantly greater for the small myocardial cells of very
young animals than for the larger cells of older animals; the figure shows further that nexal area per unit
cell volume approaches a constant value relatively
early during postnatal growth of the heart (72). A
larger gap junctional area per unit myocardial cell
volume in myocardial cells from small animals was
also found in rabbit right ventricular papillary muscles
(72). In the author’s laboratory, Stewart (73) has measured the longitudinal self-diffusion of 42K across nexal
couplings in the same papillary muscles subsequently
used for morphometric determination of nexal membrane area. He showed that longitudinal self-diffusion
occurred at the same rate in papillary muscles from
large and small rabbits, even though in smaller animals nexal impediments to sarcoplasmic diffusion are
encountered at shorter intervals along the diffusion
path.
Growth of SR . In rat left ventricle an interesting
relationship also exists between the total membrane
area of the SR and the respective volumes of the
myocardial cell and myofibrils. Both ratios (area SR/
cell volume, area SR/myofibrillar
volume) remain constant during the entire interval of postnatal growth
examined (a range of body weights of 36-227 g). As
already discussed, the more fundamental relationship
from a physiological viewpoint is that between SR
membrane area and myofibrillar volume (47).
Growth of mitochondrial respiratory membrane. In
the third curve in Fig. 5, the area of respiratory
membrane per unit of mitochondrial volume for rabbit
left ventricle is plotted against age in days before or
after birth. The measurements were made during a
developmental period when two stimuli of particular
physiological interest are acting to promote the accumulation of respiratory membrane (70). The interval
chosen is that from 3 days before birth to 4 days after
birth. It includes the neonatal transition from partially
anaerobic metabolism to aerobic metabolism. This
transition is brought about at birth by the functional
closure of the ductus arteriosus and foramen ovale, the
fall in pulmonary vascular resistance, and the initiation of air breathing by the lungs. At the same time
the occlusion of the placental shunt raises the resistance to flow in the systemic arterial circuit and thereby
increases the work done by the left ventricle in ejecting
blood. Figure 5 shows that this transition is accompanied by an increase in the cellular concentration of
mitochondria and an even larger increase in the cellu-
Cl54
E. PAGE
160
140
e
1
w
f 120
-I
p 100
FIG. 4. Gap junctional
area per unit myocardial cell volume
during
growth
of rat left ventricle. Data for same hearts as Fig. 3. Note that gap
junctional
area decreases
initially,
then remains
approximately
constant.
Figures
for gap junctional
area are minimal
estimates.
Replotted
from Ref. 72.
3 80
$ 60
“E
a
40
20
0
BODY WEIGHT (g)
.32.30-l
d .282
.26-
$24.22-
FIG. 5. Developmental
changes
in area of mitochondrial
respiratory
membrane
(cristae
+ inner
membrane)
during
perinatal
transition
to aerobic
metabolism.
Rabbit
left ventricular
myocardial
cells. A: myofibrillar
volume
per unit myocardial
cell volume.
B: mitochondrial
volume
per unit myocardial
cell volume.
C: area of respiratory
membrane
per unit mitochondrial
volume.
D: area of
respiratory
membrane
per unit myofibrillar
volume.
Replotted
from Ref. 70.
.20.1864.
y
-J
w
Z
a
0
=
0
6056.
52.
48
“E 44’
=t 40.
36
I
I
-3
-2
I
-I
DAYS BEFORE
I
I
0
+I
(0) OR AFTER
I
I
I
+.2
43
+‘4
(+) BIRTH
lar concentration of myofibrils. The mitochondria become more densely packed with respiratory membrane,
as measured by the area of respiratory membrane per
unit mitochondrial volume. At the same time, the area
of respiratory membrane per unit myofibrillar volume
increases, i.e., the amount of ATP-producing
membrane increases more rapidly than the amount of the
organelle that consumes most of the ATP.
It is noteworthy that in ventricular heart muscle the
respective areas of respiratory membrane per unit cell
volume, per unit mitochondrial volume, and per unit
myofibrillar volume also change in several other functional states (69). Moreover, these ratios vary directly
among various mammalian species in approximately
the same sequence as that for the heart rate characteristic of the species (46). A similar interspecies variation
with heart rate prevails for the membrane area of
diadic junctional coupling between plasma membrane
and terminal cistern of SR, again referred to myofibrillar volume (E. Page and M. F. Surdyk, unpublished).
Functional subunits in myocardial cells. These selected examples of membrane participati .on i n the
growth of myocardial cells have raised two new issues
about the division of mammalian heart muscle into
functionally significant subunits. It was recognized by
classical light microscopists (31) that mammalian heart
muscle, unlike fast skeletal muscle, falls into the
category of “Felderstruktur.”
This terminology recognized that the contractile material is not organized into
discrete myofibrils; instead, the interior of the heart
muscle cell is occupied by a matrix of contractile
material within which other organelles appear to be
embedded. However, the observation that the area of
plasma membrane (external envelope + T system) per
unit cell volume is constant during growth suggests
that one role of the T system may be to divide the
myofibrillar
matrix of ventricular
heart muscle into
functional units; the similar constancy of the membrane area of SR per unit myofibrillar
volume is
consistent with further subdivision yielding a second
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
.34-
INVITED
Cl55
REVIEW
Quantitative
Analysis
of
Freeze-Fractured
Membrane
Replicas
The quantitative
analysis of the membrane replicas
obtained by freeze fracture
i.s a technique for correlating physiologically
significant
changes of membrane
structure
with membrane
function.
Despite the fact
that an extensive analysis of this sort has not yet been
applied to heart muscle, several directions
in which
research
on heart muscle can proceed are already
evident from preliminary
studies as well as from experiments on other tissues.
Replicas of freeze-fractured
membranes
consist of particles or more complex structures (e.g., caveolar necks) dispersed
in the plane of
the membrane
to various
depths and with various
degrees and kinds of order. A logical goal for physiologists is to identify each type of particle (and each more
complex structure)
with its function or functions.
This
goal may be approached
by at least two experimental
designs: an analysis of developmental
changes in membrane structure
accompanying
developmental
changes
in membrane physiology,
and an analysis of structural
changes accompanying
physiologically
interesting
exThe results
of these two
perimental
perturbations.
experimental
designs may, in turn, be subjected to
whichever
of two methods of analysis is more appropriate: a statistical
analysis of particle distributions
in the
replicas of the membrane fracture faces or an analysis
of-the Fourier
transforms
of the replica images produced by optical diffraction
of the electron micrograph.
presupposes
a reasonable
Each of these approaches
attempt to define and mini .mize artifact in the preparationof the membrane replicas.
Particles in freeze-fractured
membrane replicas have
been reliably
identified
with transport
functions
in
erythrocyte
plasma membrane
(21) and in SR from
skeletal muscle (66). Although the physiological
role of
gap junctional
particles
(nexuses)
in heart muscle is
well established,
functional
identification
of other
membrane
particles in the cardiac plasma membrane,
SR, and mitochondrial
membranes
has not been
achieved. A developmental
approach to functional identification of plasmalemmal
particles would be based on
the electrophysiological
observation
that more or less
ion-specific
channels
for Na, K, Ca, and perhaps Cl
become detectable at rather sharply defined stages in
the embryonic
development
of the heart (3, 40, 71). It
seems useful to determine at each such stage whether
the appearance of a change in membrane permeability
is associated with the appearance of a new population
of membrane particles or with a change in the existing
population.
A positive result would suggest identification of particles with transport
mechanisms,
though it
would not by itself be conclusive.
Negative results are
more difficult
to interpret:
a given transport
mechanism may not manifest itself as a particle, or a preexmay change from an inactive to an
isting “particle”
active transport
mechanism
without
undergoing
a
structural
change detectable
by the freeze-fracture
method.
Physiological
perturbations
applied to mature heart
muscle might be directed first at the gap junctions.
In
particular,
it seems worthwhile
to seek structural
correlations for the apparent changes in nexal resistance
of heart muscle recently described after experimental
increases
in the sarcoplasmic
concentration
of ionized
calcium (lo), as well as after other manipulations
(11,
76). These calcium-induced
changes in nexal resistance
of heart muscle are special examples of the control of
nexal permeability
by the cytoplasmic
ionized calcium
concentration
recently reviewed
by Loewenstein
(34).
The expectation
that functionally
significant
changes
in nexal structure
may be demonstrable
in heart muscle is rendered plausible by the finding of such changes
in nexuses of crayfish
(59), rat stomach and liver (58),
and calf heart (57) by Peracchia,
and in nexuses of
mouse liver by Caspar et al. (6) and Makowski
et al.
(35). These experiments
suggest that increases in nexal
resistance
produced by a rise in cytoplasmic
ionized
calcium concentration
should be correlated
(in freezefractured
replicas of the nexus) with a transition
from
a flexible structure
showing
considerable
short-range
disorder to a more highly ordered and closely spaced
hexagonal structure.
At the same time, the intercellular space at the nexus as seen in TEM should become
narrower.
These experimental
approaches
require
sensitive
techniques for detecting structural
changes in replicas
of membrane
fracture
faces. This review will consider
only two such techniques
for obtaining
quantitative
information
from electron
micrographs.
The choice
between these two techniques depends on whether
the
particles (or other structures)
are present in the membrane at relatively
low density per unit area or whether
they are closely spaced, preferably
in some relatively
orderly array. For example, most of the particles present in the sarcolemmal
fracture
faces are dispersed
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
and different functional
subunit structure.
Thus, for a
given rate of Ca uptake by the SR, there may be two
critical distances for diffusion of Ca. One such distance
might be from the most distal Ca-binding
sites on
trononin C in the thin filaments
to that point of the
noncisternal
SR which takes up the Ca2+- from those
sites during
relaxation,
and a second such critical
distance might be the length of the path for diffusion of
Ca released by excitation from the terminal cisterns to
the most distant troponin C-binding
sites controlled by
those terminal
cisterns.
Finally, for any particular
set
of rates of mitochondrial
ATP production
and myofibrillar ATP consumption,
there may be a critical perimitochondrial
radius for ATP supply of the contractile
mass surrounding
the mitochondrion.
The second issue raised by these functional aspects of
membrane
structure
has to do with growth.
Is the
apparent association
of membranes
like the SR and T
system with “units”
of myofibrillar
volume related to
the way contractile
material
is added when mammalian ventricular
muscle grows or hypertrophies?
In
other words, is contractile
material added as a unit of
volume defined by the membranes
of the T system, SR,
or both? Are these membranes
involved in the assembly
or degradation
of the myofibrillar
matrix,
and if so,
how? These questions are at present unanswered.
Cl56
_.,
..,
. .... . _
4
FIG. 6. Diffractometry
of cardiac gap junction
from bovine cardiac
Purkinje
fiber fixed in glutaraldehyde.
In the diffractometer
a beam
from a helium-neon
gas laser was passed through
the photographic
negative
of the gap junctional
membrane
replica
shown
in the
electron
micrograph,
yielding
an approximately
hexagonal
Fourier
transform
(insert).
After calibration
of the diffractometer,
the average interparticle
spacing
can be determined
by measuring
the
distances
between
the central
(zero order)
spot and the first order
diffraction
spots at the points of the hexagonal
pattern.
Calibration
line on electron
micrograph
is 0.1 pm. Freeze-fractured
replica
and
diffraction
pattern
were prepared
by J. Upshaw-Earley
in author’s
laboratory.
tissue fractionation
resulting in isolation of gap junction membrane fragments satisfactory for diffractometry has not yet been reported for heart muscle; when
such isolation is achieved, it should become possible to
use data from optical diffraction of negatively stained
gap junctions for a reconstruction
of the gap junction
(6, 35).
Optical diffraction of freeze-cleaved gap junctions in
rat Graafian follicles has been reported by Amsterdam
et al. (1). Figure 6 presents a similar analysis from the
author’s laboratory for the gap junction of the bovine
cardiac Purkinje fiber. The imperfections
in the quasihexagonal first order reflection suggest that, in common with permeable gap junctions from other tissues,
the cardiac gap junctions manifest considerable shortrange disorder (6), that is, the packing in the fracture
face of the connexon units containing
the permeable
channels (6, 35) is imperfectly hexagonal. Preliminary
diffractometric
studies on freeze-fractured
replicas of
gap junctions from rat and rabbit ventricular
myocardial cells (E. Page and J. Upshaw-Earley,
unpublished
observations)
indicate that the range of inter-particle
spacing for junctions from electrically
coupled cells is
B-10 nm. Accordingly,
both the range of interparticle
spacing and the presence of short-range disorder in the
hexagonal array from cardiac nexuses resemble the
result obtained in the much more extensive measurements reported for other tissues (6, 35, 57, 58). Thus,
the diffractometric
evidence suggests that permeable
cardiac gap junctions,
like permeable
gap junctions
from other tissues, have a flexible structure. Transitions from this flexible structure to the structure char-
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
rather widely with little apparent order, as are the
caveolar necks; the particles and pits on the nexal
fracture faces are, on the other hand, closely spaced
and usually arrayed in a more or less regular polygonal
pattern.
For widely dispersed particles and structures, the
surface density and distribution
of each type of particle
recognizable in the fracture faces may be determined
by some convenient technique. One such technique is
to place the print on a digitizing
tablet and to store
information
about the spatial distribution
of particles
in a computer by means of a digitizing
probe (39, 42).
The questions to be answered are whether there are
changes in the surface density (number of particles per
unit area) and the type of particle distribution
(random
or other). The hypothesis to be tested is that a permeability change at a particular developmental
stage or in
response to a particular
experimental
perturbation
is
associated with a change in surface particle density or
with a transition
from one statistical
distribution
of
particles to some other distribution.
The expectation
that such changes may be found in freeze-fractured
heart muscle membranes derives from a growing number of observations on other tissues (4, 7, 26, 42, 77). In
all these systems, the precondition underlying
changes
in the distribution
of particles and structures observed
with freeze fracture is the lateral mobility
of these
membrane components in the plane of the membrane,
as predicted by the fluid-mosaic
model (13, 67). Some
examples of statistical methods suitable for this problem are given in papers by Mehlhorn and Packer (42),
Maul et al. (39), and Markovics et al. (37).
For a quantitative
description of the gap junction, as
it appears in the membrane fracture face, it has proved
expedient to apply optical diffraction
analysis to the
replica of the gap junction
(1, 28). This technique
transforms the image of the object (a negative transparency of the electron micrograph
of either of the two
fracture faces of the gap junction)
into its Fourier
components. The transformation
makes it possible to
do spatial filtering
in the plane of the transform,
thereby modifying
the image. Optical analysis and
spatial filtering
facilitate
recognition
and measurement of the periodic structure and its orientation.
At
the same time, periodic or random background
patterns, which often interfere with visual interpretations, can be removed; the residual structures, which
are presumably
the features of interest, can thus be
more closely defined.
With the use of an optical diffractometer,
it is possible to determine to what extent the particles in the gap
junction are organized as a hexagonal lattice and to
estimate the particle-to-particle
spacing for both fracture faces. The analysis can also be applied to en face
sections of the gap junction obtained by TEM of fixed
tissues (i.e., to sections cut parallel to the plane of the
membrane or to similarly
oriented, negatively stained
nexuses); such en face sections also show the approximately hexagonal subunit structure of cardiac nexuses.
They can therefore be used to check on the effect of the
preparative
procedure on the diffraction
pattern of
freeze-fractured
membrane
replicas. At this writing,
E. PAGE
INVITED
Cl57
REVIEW
acteristic
of junctions
rendered less permeable by high
Ca2+ concentrations
should therefore
be detectable in
heart muscle as in other tissues. Indications
of such a
transition
have recently been found in ischemic heart
muscle, though with techniques
somewhat
less sensitive than the diffractometric
method (2).
CONCLUSIONS
I thank
Mrs. Judy Upshaw-Earley
for technical
assistance.
Portions
of the research
reported
in this review
were supported
by Public Health
Service
Grants
HL-10503
and HL-20592.
REFERENCES
1. AMSTERDAM,
A., R. JOSEPHS, M. E. LIEBERMAN,
AND H. R.
LINDNER.
Organization
of intramembrane
particles
in freezecleaved gap junctions
of rat Graafian
follicles:
optical diffraction
analysis.
J. CeZZ Sci. 21: 93-105,
1976.
2. ASHRAF, M., AND C. HALVERSON.
Ultrastructural
modifications
of nexuses
(gap junctions)
during
early myocardial
ischemia.
J.
MOL. Cellular
CardioZ.
10: 263-270,
1978.
3 BERNARD,
C. Establishment
of ionic permeabilities
of the myocardial
membrane
during
embryonic
development
of the rat.
In: DeveLopmental
and PhysioLogical
CorreZates
of Cardiac
MuscZe, edited by M. Lieberman
and T. Sano. New York:
Raven,
1975, p. 169-184.
4 BOURGUET,
J., J. CHEVALIER,
AND J. S. HUGON.
Alterations
in
membrane-associated
particle
distribution
during
antidiuretic
challenge
in frog urinary
bladder
epithelium.
Biophys.
J. 16:
627-639,
1976.
5 BOYLE, P. J., AND E. J. CONWAY.
Potassium
accumulation
in
muscle
and associated
changes.
J. Physiol.
London
100: l-63,
1941.
6 CASPAR, D. L. D., D. A. GOODENOUGH,
L. MAKOWSKI,
AND W. C.
PHILLIPS. Gap junction
structures.
I. Correlated
electron
microscopy and X-ray
diffraction.
J. CeZZ BioZ. 74: 605-628,
1977.
7 COPPS, T. P., W. S. CHELACK,
AND A. PETKAU.
Variation
in
distribution
of membrane
particles
in AchoZepZasma
ZaidZawii
B
with pH. J. UZtrastruct.
Res. 55: l-3, 1976.
8 COSTELLO, M. J., AND T. GULIK-KRZYWICKI.
Correlated
X-ray
diffraction
and freeze-fracture
studies
on membrane
model systems: perturbations
induced
by freeze-fracture
preparative
procedures. Biochim.
Biophys.
Acta 455: 412-432,
1976.
9 DEHOFF,
R. T., AND F. N. RHINES. Quantitative
Microscopy.
New
York: McGraw,
1968.
10. DEMELLO,
W. C. Effect of intracellular
injection
of calcium
and
strontium
on cell communication
in heart.
J. Physiol.
London
250: 231-245,
1975.
11. DEMELLO,
W. C. Influence
of the sodium
pump on intercellular
communication
in heart fibres: effect of intracellular
injection
of
sodium
ion on electrical
coupling.
J. Physiol.
London
263: 171197, 1976.
12. DULHUNTY,
A. F., AND C. FRANZINI-ARMSTRONG.
The relative
contributions
of the folds and caveolae
to the surface membrane
of frog skeletal
muscle
fibres at different
sarcomere
lengths.
J.
PhysioZ. London
250: 513-539,
1975.
13. EDIDIN,
M. Rotational
and translational
diffusion
in membranes. Ann. Rev. Biophys.
Bioeng.
3: 179-201,
1974.
14. EISENBERG,
B. R., AND B. A. MOBLEY.
Size changes
in single
muscle fibers during
fixation
and embedding.
Tissue CeZZ 7: 383387, 1975.
15. ELIAS, H., A. HENNIG,
AND D. E. SCHWARTZ.
Stereology:
applications
to biomedical
research.
PhysioZ.
Rev. 51: 158-200,
1971.
16. FABIATO,
A., AND F. FABIATO.
Activation
of skinned
cardiac
cells: subcellular
effects
of cardioactive
drugs.
European
J.
CardioZ.
l-2: 143-155,
1973.
17. FABIATO, A., AND F. FABIATO. Contraction
induced
by a calciumtriggered
release of calcium
from the sarcoplasmic
reticulum
of
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
single
skinned
cardiac
cells. J. PhysioZ.
London
249: 469-495,
1975.
FAWCETT, D. W., AND N. S. MCNUTT.
The ultrastructure
of cat
myocardium.
I. Ventricular
papillary
muscle. J. CeZZ BioZ. 42: l45, 1969.
FINEGOLD,
L. Cell membrane
fluidity:
molecular
modeling
of
particle
aggregations
seen in electron
microscopy.
Biochim.
Biophys.
Acta 448: 393-398,
1976.
FRANK, M., I. ALBRECHT,
W. W. SLEATOR, AND R. B. ROBINSON.
Stereological
measurements
of atria1
ultrastructures
in the
guinea pig. Experientia
31: 578-579,
1975.
GUNN, R. B., AND R. G. KIRK. Anion
transport
and membrane
morphology.
J. Membrane
BioZ. 27: 265-282,
1976.
HELLAM,
D. C., AND J. W. STUDT. A core conductor
model of the
cardiac Purkinje
fibre based on structural
analysis.
J. PhysioZ.
London
243: 637-660,
1974.
HELLAM,
D. C., AND J. W. STUDT. Linear
analysis
of membrane
conductance
and capacitance
in cardiac
Purkinje
fibres.
J.
PhysioZ.
London
243: 661-694,
1974.
HERBENER,
G. H., R. H. SWIGART,
AND C. A. LANG. Morphometric
comparison
of the mitochondrial
populations
of normal
and hypertrophied
hearts. Lab. Invest.
28: 96-103, 1973.
HEUSER, J. E., T. S. REESE, AND D. M. D. LANDIS. Preservation
of synaptic
structure
by rapid
freezing.
CoZd Spring
Harbor
Symp. Quant.
BioZ. 40: 17-24, 1976.
H~CHLI,
M., AND C. R. HACKENBROCK.
Thermotropic
lateral
translational
motion
of intramembrane
particles
in the inner
mitochondrial
membrane
and its inhibition
by artificial
peripheral proteins.
J. CeZZ BioZ. 72: 278-291,
1977.
HODGKIN,
A. L., AND P. HOROWICZ.
The influence
of potassium
and chloride
ions on the membrane
potential
of single
muscle
fibres. J. PhysioZ. London
148: 127-160,
1959.
JOHANSEN,
B. V. Optical
diffractometry.
In: PrincipZes
and
Techniques
of ELectron
Microscopy:
BioLogicaL
AppZications,
edited by M. A. Hayat.
New York:
Van Nostrand,
1975, vol. 5, p.
114-173.
KAHN, C. R. Membrane
receptors
for hormones
and neurotransmitters.
J. CeZZ BioZ. 70: 261-286,
1976.
KORECKY,
B., AND K. RAKUSAN.
Dimensions
of cardiac
muscle
cells during
the life span of the rat (Abstract).
Physiologist
16:
366, 1973.
KRUGER, P., F. DUSPIVA, AND F. FORLINGER.
Tetanus
und Tonus
der Skelet-muskeln
des Frosches,
eine histologische,
reizphysiologische
und chemische
Untersuchung.
Pfluegers
Arch.
231:
750-786,
1933.
LEVIN,
K. R. Quantitative
studies
on caveolae
of rabbit
right
ventricular
myocardial
cells (PhD
thesis).
Chicago:
Univ.
of
Chicago,
1978.
LEVIN, K. R., AND E. PAGE. Sarcolemmal
caveolae
and folds in
rabbit
papillary
muscle (Abstract).
J. CeZZ BioZ. 75: 317, 1977.
LOEWENSTEIN,
W. R. Permeable
junctions.
CoZd Spring
Harbor
Symp. Quant.
BioZ. 40: 49-63, 1976.
MAKOWSKI,
L., D. L. D. CASPAR, W. C. PHILLIPS,
AND D. A.
GOODENOUGH.
Gap junction
structures.
II. An analysis
of the X-
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
Quantitative
measurements
on electron micrographs
of heart muscle can yield information
useful for cellular
physiologists
and at present not obtainable
in other
ways. These methods are subject to preparative
artifact, sampling problems, and problems inherent in the
mathematical
description
of ultrastructure.
Nevertheless they provide the best available data for membrane
areas of the plasmalemma
and its components,
as well
as for membrane
areas of the sarcoplasmic
reticulum
and mitochondria.
Morphometric
methods can be used
to study growth of membranes.
Changes in the volumes
of intracellular
membrane-limited
subcompartments
can also be measured.
Quantitative
analysis of freezefractured
membrane
replicas can be carried out either
by a statistical
approach or by optical diffraction.
In
this way, physiological
perturbations
or developmental
events leading to changes in membrane
permeability
can be studied for correlated
changes in membrane
structure.
Cl58
E. PAGE
57. PERACCHIA,
C. Calcium
effects on gap junction
structure
and cell
coupling.
Nature
271: 669-671,
1978.
58. PERACCHIA,
C. Gap junctions:
structural
changes
after uncoupling procedures.
J. CeZZ BioZ. 72: 628-641,
1977.
C., AND A. F. DULHUNTY.
Low resistance
junctions
59. PERACCHIA,
in crayfish:
structural
changes
with functional
uncoupling.
J.
CeZZ BioZ. 70: 419-439,
1976.
W., W. M. FISCHER,
AND E. LEITNER.
Quantitative
60. PFALLER,
changes
in mitochondria
of guinea
pig myocardium
following
anesthesia
with halothane.
Virchows
Arch. ZeZZpathoZ.
15: 229235, 1974.
P. I. Extracellular
space and ionic distribution
in rat
61. POLIMENI,
ventricle.
Am. J. PhysioZ.
227: 676-683,
1974.
K., S. RAMAN,
R. LAYBERRY,
AND B. KORECKY.
The
62. RAKUSAN,
influence
of aging and growth
on the postnatal
development
of
cardiac muscle in rats. CircuZation
Res. 42: 212-218,
1978.
AND W. S. BERTAUD.
Surface
63. RAYNS, D. G., F. 0. SIMPSON,
features
of striated
muscle.
I. Guinea-pig
cardiac muscle. J. CeZZ
Sci. 3: 467-474,
1968.
J. NOLTE, AND H. W. STAUDTE. The
64. REITH, A., D. BRDICZKA,
inner membrane
of mitochondria
under influence
of triiodothyronine
and riboflavin
deficiency
in rat heart muscle
and liver.
ExptZ. CeZZ Res. 77: 1-14, 1973.
muscle
of the rat under
65. REITH, A., AND S. FUCHS. The heart
influence
of triiodothyronine
and riboflavin
deficiency
with
special reference
to mitochondria.
Lab. Inuest. 29: 229-235,
1973.
of ATPase
protein
in
66. SCALES, D. J., AND G. INESI. Localization
sarcoplasmic
reticulum
membrane.
Arch.
Biochem.
Biophys.
176: 392-394,
1976.
The fluid mosaic
model of
67. SINGER, S. J., AND G. L. NICOLSON.
the structure
of cell membranes.
Science 175: 720-731,
1972.
of mitochondrial
membranes
68. SJ~STRAND, F. S. The arrangement
and a new structural
feature
of the inner mitochondrial
membranes.
J. UZtrastruct.
Res. 59: 292-319,
1977.
of rat heart mitochon69. SMITH, H. E., AND E. PAGE. Morphometry
drial subcompartments
and membranes:
application
to myocardial cell atrophy
after hypophysectomy.
J. Ultrastruct.
Res. 55:
31-41, 1976.
70. SMITH, H. E., AND E. PAGE. Ultrastructural
changes
in rabbit
heart mitochondria
during
the perinatal
period:
neonatal
transition to aerobic metabolism.
DeveZop. BioZ. 57: 109-117,
1977.
71. SPERELAKIS,
N., AND K. SHIGENOBU.
Changes
in membrane
properties
of chick
embryonic
hearts
during
development.
J.
Gen. PhysioZ.
60: 430-453,
1972.
72. STEWART, J., AND E. PAGE. Improved
stereological
techniques
for studying
myocardial
cell growth.
Application
to external
sarcolemma,
T-system,
and intercalated
disks of rabbit
and rat
hearts. J. UZtrastruct.
Res. In press.
73. STEWART, J. M. Correlated
morphometric
and flux measurements on rabbit
right ventricular
papillary
muscle (PhD thesis).
Chicago:
Univ.
of Chicago,
1975.
74. UNDERWOOD,
E. E. Quantitative
Stereology.
Reading,
PA: Addison-Wesley,
1970.
75. WEIBEL, E. R. Stereological
techniques
for electron
microscopy.
of ELectron Microscopy,
edited by
In: Principles
and Techniques
M. A. Hayat.
New York: Van Nostrand,
1973, vol. 3, p. 237-296.
76. WEINGART,
R. The actions of ouabain
on intercellular
coupling
and conduction
velocity
in mammalian
ventricular
muscle.
J.
Physiol. London
264: 341-365,
1977.
77 WEINSTEIN,
R. S. Changes
in plasma membrane
structure
associated with malignant
transformation
in human
urinary
bladder
epithelium.
Cancer Res. 36: 2518-2524,
1976.
78 ZAK, R. Cell proliferation
during
cardiac
growth.
Am. J. CardioZ. 31: 211-219,
1973.
79. ZAK, R. Development
and proliferative
capacity
of cardiac
muscle cells. CircuZation
Res. 34-35, Suppl. 2: 17-26, 1974.
Ernest Page
Departments
of Medicine and of Pharmacological
and Physiological
Pritzker
School of Medicine, The University
of Chicago,
Chicago, Illinois 60637
Sciences,
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on May 3, 2017
ray diffraction
data. J. CeZZ Biol. 74: 629-645,
1977.
36. MALL, G., K. KAYSER, AND J. A. ROSSNER. The loss of membrane
images from oblique
sectioning
of biological
membranes
and the
availability
of morphometric
principles-demonstrated
by the
examination
of heart muscle mitochondria.
Mikroskopie
33: 246254,1977.
37. MARKOVICS,
J., L. GLASS, AND G. G. MAUL.
Pore patterns
of
nuclear
membranes.
Exptl.
Cell Res. 85: 443-451,
1974.
38. MATTER, A. A morphometric
study on the nexus of rat cardiac
muscle. J. CeZZ BioZ. 56: 690-696,
1973.
39. MAUL,
G. G., J. W. PRICE, AND M. W. LIEBERMAN.
Formation
and distribution
of nuclear
pore complexes
in interphase.
J. CeZZ
BioZ. 51: 405-418,
1971.
40. MCDONALD,
T. F., H. G. SACHS, AND R. L. DEHAAN.
Development
of sensitivity
to tetrodotoxin
in beating
chick
embryo
hearts,
single cells and aggregates.
Science 176: 1248-1250,
1972.
41. MCNUTT,
N. S. Ultrastructure
of the myocardial
sarcolemma.
CircuZation
Res. 37: 1-13, 1975.
42. MEHLHORN,
R. J., AND L. PACKER. Analysis
of freeze-fracture
electron
micrographs
by a computer-based
technique.
Biophys.
J. 16:613-625,
1976.
43. MOBLEY, B. A., AND E. PAGE. The surface area of sheep cardiac
Purkinje
fibres. J. Physiol.
London
220: 547-563,
1972.
44. OJAKIAN,
G. K., AND P. SATIR. Particle
movements
in chloroplast
membranes:
quantitative
measurements
of membrane
fluidity
by freeze-fracture
technique.
Proc. NatZ. Acad. Sci. US
71:2052-2056,
1974.
45. PAGE, E. Cat heart muscle in vitro. II. The steady state resting
potential
in quiescent
papillary
muscles.
J. Gen. Physiol.
46:
189-199,
1962.
46. PAGE, E., J. EARLEY, L. P. MCCALLISTER,
AND C. BOYD. Copper
content
and exchange
in mammalian
hearts.
CircuZation
Res.
35:67-76,
1974.
47. PAGE, E., J. EARLEY, AND B. POWER. Normal
growth
of ultrastructures
in rat left ventricular
myocardial
cells. CircuZation
Res. 34-35, Suppl. 2: 12-16, 1974.
48. PAGE, E., AND H. A. FOZZARD. Capacitive,
resistive,
and syncytial properties
of heart muscleultrastructural
and physiologiof MuscZe,
cal considerations.
In: The Structure
and Function
edited by G. H. Bourne.
New York:
Academic,
1973, vol. 2, p.
91-158.
49. PAGE, E . , AND L. P. MCCALLISTER.
Quantitative
electron
microscopic description
of heart
muscle
cells: application
to normal,
hypertrophied
and thyroxin-stimulated
hearts.
Am. J. CardioZ.
31: 172-181,
1973.
50. PAGE, E., AND L. P. MCCALLISTER.
Studies
on the intercalated
disk of rat left ventricular
myocardial
cells. J. UZtrastruct.
Res.
43:388-411,
1973.
51. PAGE, E., L. P. MCCALLISTER,
AND B. POWER. Stereological
measurements
of cardiac
ultrastructures
implicated
in excitation-contraction
coupling.
Proc. NatZ. Acad. Sci. US 68: 14651466, 1971.
52. PAGE, E., AND P. I. POLIMENI.
Magnesium
exchange
in rat
ventricle.
J. Physiol.
London
224: 121-139,
1972.
53. PAGE, E., AND A. K. SOLOMON.
Cat heart muscle in vitro. I. Cell
volumes
and intracellular
concentrations
in papillary
muscle. J.
Gen. PhysioZ.
44: 327-344,
1960.
54. PAGE, E., AND M. F. SURDYK. Quantification
of plasma
membrane
specializations
in ventricular
muscle
cells (Abstract).
Biophys.
J. 21: 183, 1978.
55. PAGE, E., AND J. UPSHAW-EARLEY.
Volume
changes
in sarcoplasmic
reticulum
of rat hearts
perfused
with hypertonic
solutions. CircuZation
Res. 40: 355-366,
1977.
56. PAGER, J. Etude
morphometrique
du systeme
tubulaire
transverse du myocarde
ventriculaire
de rat. J. CeZZ BioZ. 50: 233-237,
1971.