Quantitative ultrastructural analysis in cardiac membrane physiology.

Quantitative ultrastructural analysis in cardiac membrane physiology 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. 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 www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (185.223.164.034) on November 7, 2018. Copyright © 1978 American Physiological Society. All rights reserved.

Cl48

E. PAGE

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. 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,


INVITED

Cl49

REVIEW

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 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 myofibrillar 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


Cl50

E. PAGE

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). 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


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

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

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


Cl52

E. PAGE

240 l 0

20 -

0

16 -

0 l

0 ID 8

0 0 0

O

IY

o*

m

0015

;o :&o

0

0

,

I

-025

00%

a m

J28 IL 0 I~24 8

a

a

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


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

N ‘g

.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.

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-

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

1 300

BODY WEIGHT (g) Downloaded from www.physiology.org/journal/ajpcell by ${individualUser.givenNames} ${individualUser.surname} (185.223.164.034) on November 7, 2018. Copyright © 1978 American Physiological Society. All rights reserved.

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) .34.32.30-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


INVITED

Cl55

REVIEW

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. 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


Cl56 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

_.,

..,

. .... . _

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-


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

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. 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-


Cl58

E. PAGE

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.

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,


EHD3-dependent endosome pathway regulates cardiac membrane excitability and physiology.

Quantitative analysis of ultrastructural changes in developing rat cardiac muscle during normal growth and during acute volume load.

A role for membrane shape and information processing in cardiac physiology.

Quantitative analysis of membrane trafficking in regulation of Cdc42 polarity.

Cardiac fibroblast physiology and pathology.

Does single ventricle physiology affect survival of children requiring extracorporeal membrane oxygenation support following cardiac surgery?

SIRT3 in Cardiac Physiology and Disease.

Quantitative physiology and immunohistochemistry of oral lesions.

Quantitative ultrastructural analysis of basket and axo-axonic cell terminals in the mouse hippocampus.

Quantitative ultrastructural study of cat abducens interneurons.

Quantitative analysis of annexin V-membrane interaction by flow cytometry.

Quantitative analysis of left ventricular strain using cardiac computed tomography.

Anatomy and physiology of the peritoneal membrane.

Cardiac optogenetics: using light to monitor cardiac physiology.

Ultrastructural localization and semiquantitative analysis of glycoconjugates in the tectorial membrane.

Cardiac Physiology of Aging: Extracellular Considerations.

An ultrastructural study of mammalian cardiac ganglia.

Ultrastructural analysis of platelets during three phases of pregnancy: a qualitative and quantitative investigation.

Quantitative ultrastructural analysis of hepatoprotective effects of (+)-cyanidanol-3 on alcoholic liver damage.

Synapses in newborn rat cerebral cortex: a quantitative ultrastructural study.

Cardiac extracellular proteome profiling and membrane topology analysis using glycoproteomics.

Development of neocortical circuitry: quantitative ultrastructural analysis of putative monoaminergic synapses.

Quantitative ultrastructural analysis of sperm tails reveals flagellar defects associated with persistent asthenozoospermia.

Cognitive Collaboration Found in Cardiac Physiology: Study in Classroom Environment.