View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by PubMed Central THE MYOCARDIAL INTERSTITIUM : ITS STRUCTURE AND ITS ROLE IN IONIC EXCHANGE J . S . FRANK and G . A . LANGER From the Departments of Medicine and Physiology and the Los Angeles County Heart Association Cardiovascular Research Laboratory, University of California at Los Angeles, Center for the Health Sciences, Los Angeles, California 90024 ABSTRACT The structures present in the rabbit myocardial interstitium have been defined and quanti- fied . Stereological methods were used for the quantification . The extracellular space con- tains abundant ground substance (23%) distributed in a homogeneous mat throughout the space and within the T tubules . The remainder of the space contains 59% blood ves- sels, 6% "empty" space, 4 .0% collagen, and 7 .0% connective tissue cells . The arrangement of the interstitium in relation to the myocardial cells and the capillaries has been described . '4C In addition, the extracellular space was measured using extracellular markers : sucrose 140 (neutrally charged), 35S04 (negatively charged), and La (positively charged) . The La+++ space differed markedly from the other two (P « 0 .001), indicating extensive binding of La+++ to polyanionic extracellular structures . Cetylpyridinium chloride, a cationic de- tergent specific for polysaccharides, caused precipitation of the ground substance and marked decrease in the La+++ space . This study indicates the considerable structural complexity of the interstitium . The effects of an abundant negatively charged protein- polysaccharide within the interstitium has been discussed in terms of cation exchange in arterially perfused tissue . INTRODUCTION Many studies have described the cytology of the (ground substance) present on the surface and in mammalian heart (6, 13), but little information the vicinity of muscle cells, and he suggested that is available on the ultrastructural aspects of the this material with its fixed negative charges might extracellular space in the myocardium . The modify the ionic environment of the plasma mem- region between the myocardial cell surface and brane and the cell itself . A number of flux and the capillary wall has not been studied in detail perfusion studies have emphasized the fact that nor have the structures within this space been coupling calcium is derived from superficial sites quantified . on or near the myocardial cell membrane which Kinetic studies on the evaluation of cation are in very rapid equilibrium with the intestitial exchange in vascularly perfused mammalian space (24, 29) . Displacement of calcium from myocardium are not consistent with an inter- these superficial sites and from the interstitium stitium composed simply of a vascular transudate results in a proportional loss of tension, indicating (9, 29) . Indeed, 10 years ago, Bennett (1) pointed the importance of "extracellular" calcium sources out the abundance of protein polysaccharide to excitation-contraction coupling (24) . Ki- 5 86 THE JOURNAL OF CELL BIOLOGY . VOLUME 60, 1974 . pages 586-601 netically, it is not possible to separate, in the ness . Another group of muscles (5) were perfused with intact myocardium, superficial binding sites of glutaraldehyde which contained 0 .1% cetylpyri- calcium on or near the myocardial cell membrane dinium chloride (CPC) . CPC is a cationic detergent from sites within the interstitial space . However, which in a water solution forms micelles that act as polyvalent cations with a high charge . Such cationic Langer and Frank were able to investigate the micelles associate with polyanionic polysaccharides, role of the myocardial basement membrane as a causing the polysaccharides to precipitate (26, 27) . possible binding source of coupling calcium by After 20 min of perfusion, the septum was cut down using myocardial cells grown in tissue culture . and minced into small cubes of tissue ( < 1 mm per In the present study we investigate the interstitial side) . Pieces were sampled from the endocardial sur- space in intact tissue with respect to its detailed faces and from the middle portion of the septum and structure and ionic binding characteristics . placed in the glutaraldehyde fixative for an additional The interventricular septum was used, since it is 1% h. The pieces of septum were rinsed in buffer and a physiologically perfused preparation and since then postfixed for 1 h in 1 % Os04-0 .2 M Na-caco- dylate buffer solution . After a further rinse the tissue many functional studies involving the exchange of was dehydrated in alcohol and propylene oxide, em- cationic substances have been studied using this bedded in Epon 812, and sectioned on Porter-Blum procedure (10) . Our findings indicate that, indeed, Mt-2 ultramicrotome . the interstitium is of considerable structural com- Thin unoriented sections giving a silver interference plexity. Using the morphometric techniques of pattern were stained with lead citrate and with uranyl Weibel (31, 32), we have quantified the majority acetate and examined on a Siemens Elmiskop IA of the structural components of the extracellular electron microscope. In recording the micrographs, space and have demonstrated that the ground the part of the section in the upper left-hand corner of substance occupies, exclusive of the blood vessels, the squares of the supporting copper grid was photo- graphed in an attempt to obtain an unbiased sample . over 58% of the extracellular space . The arrange- For calibration, a photograph of a carbon replica ment of the interstitium in relation to the myo- cross grating (21,600 lines/cm) was taken with each cardial cells and the capillaries has been described . series of photographs . Measurements were made on In addition, the anionic nature of the abundant 20 X 23 cm prints having a final magnification of ground substance was documented by isotopic 9500. extracellular marker studies . Finally, the effect of TERMINOLOGY : We determined the volume per- an interstitium, which contains fixed negative cent of the following constituents of the extracellular charges, on the exchange of charged and un- space : charged solutes is discussed . Xg = extracellular space filled with ground sub- stance . METHODS XX = extracellular space apparently empty of any structural components . Fixation and Microscopy X,, = collagen fraction . Xb v = the space occupied by blood vessels. Arterially cannulated rabbit interventricular septa Xf = cells of connective tissue origin, i.e., fibro- (five muscles) were perfused with oxygenated per- blast and pericytes . fusate of the following millimolar composition : NaCl, The data are reported in two ways : (a) All the com- 133 ; KCI, 3 .6 ; CaC1 2, 1.0 ; MgC12, 0.3 ; glucose, 16 .0 ; ponents of the extracellular space Xg , Xe, X,,, Xf , and and buffered with 3 mM HEPES (N-2-hydroxyethyl- Xb, are given as a volume percent of the extracellular piperazine-N'2-ethane sulfonic acid) (pH 7 .4) for 15 space (extracellular space = 100%) ; (b) the extra- min-thus allowing for a complete flushing of vascu- cellular components have been presented as a volume lar space . The perfusion rate was 0 .75-0 .90 ml /min at percent of the whole myocardial tissue (myocardial hydrostatic pressure between 55 and 66 mm Hg . This cells and all components of extracellular space flow rate was chosen since it is in the normal physio- 100%). logical range for rabbit myocardium, and the hydro- All volumes are presented as mean t standard static pressures needed to deliver this flow rate were error. Student's T test was applied to controls and were well under the normal rabbit diastolic blood CPC data . Figs . 1 a and I b give examples of what pressure (i.e., ti 80 mm Hg) . The perfusion medium was switched to 2 .0% glutaraldehyde in 0 .1 M Na- were morphologically characterized as Xg and Xe. cacodylate buffer (pH 7 .4, osmolarity 424 mosmol) Fig. 2 demonstrates examples of components X,,, X1, and the muscle was perfused with the fixative for 20 and Xb,, . min, at which time the septum was quite firm and ap- MEASUREMENTS : The relative volumes of Xg ,X,, peared uniformly dark-yellow throughout its thick- X,,, Xf and Xb„ were determined using the follow- J . S . FRANK AND G . A . LANGER Myocardial Interstitium and Ionic Exchange 587 ing formula derived from stereologic principles (2, of its activity as compared to its value at initiation 32) : V„ = Pi/PT, the volume fraction V„ of a com- of washout . [14C]Sucrose and 35SO4 declined to less ponent contained in unit volume of a material equals than 0 .1 % of initial activity after 30 min of washout. the number of points (Pi) falling on profiles of that After initial washout, the septum was perfused for component divided by the total number of test 5 min with standard perfusate to which 0 .025% points (PT) . Point counting was performed on the CPC was added . It was then relabeled for 10 min photographic prints of unoriented sections from the with the appropriate isotopically labeled marker outer and inner segments of the septum . The prints (no CPC in the labeling solution) and again washed were covered with a clear plastic sheet on which a out for 30 min, and the venous effluent was collected . square grid was imprinted (See Fig . 3) . The points The control and post-CPC washout curves were for point counting were formed by the intersections plotted (from 0 to 30 min) on linear paper . The of lines on the grid. Since there was considerable vari- areas under each curve were cut out and weighed ation in size of the volume of the different structures in order to integrate each of the washouts . Com- measured (30-1 %), the optimal point density (num- parison of the post-CPC integration with the control ber of points counted) was quite different for the var- integration gave a measure of relative change in the ious volumes (5) . The squares of the grid used for magnitude of the spaces induced by 5 min of per- analysis of large volumes, i .e.