Formation and growth of gap junctions in mouse myocardium during ontogenesis: Quantitative data and...

J. Cell Sci. 30, 45-61 (1978) 4 5

Printed in Great Britain © Company of Biologists Limited 197S

FORMATION AND GROWTH OF GAP

JUNCTIONS IN MOUSE MYOCARDIUM

DURING ONTOGENESIS:

A FREEZE-CLEAVE STUDY

D. GROS,* J. P. MOCQUARD.f C. E. CHALLICEJAND J. SCHREVEL*

* Laboratoire de Zoologie et de Biologie Cellulaire et Service General deMicroscopie Electronique Applique'e a la Biologie. U.E.R. SciencesFondamentales et Applique'es. 40, Avenue du Recteur Pinean. Poitiers cedex86022, France,•\Laboratoire de Physiologie et Genetique des Crustace's. U.E.R. SciencesFondamentales et Applique'es. 40, Avenue du Recteur Pineau. Poitiers cedex86022, France, and XPhysics Department, University of Calgary, Calgary,Alberta TzN 1N4, Canada

SUMMARY

The freeze-cleave technique demonstrates the presence of gap junctions at early stages ofmouse cardiac muscle ontogenesis. The formation and growth of these junctions were studiedat 4 stages of development: 10, 14, 18 days post-coitum (dpc) and at the adult stage. Thediverse aspects of the gap junctions are interpreted as different steps in their formation. Thefirst indication of this formation seems to be the presence of linear arrays of g-nm particles onPF faces. At one end of these arrays a small aggregate of particles appears which acts as nuclea-tion site and grows by incorporating individual gap particles and/or linear arrays. Nexuses witharms and/or central particle-free zones would represent intermediate steps in the formation ofjunctions. The largest nexuses could be formed by fusion of smaller ones and/or by accretionof gap particles. Analysis of the size distribution of gap junctions shows their growth duringtheir development. At 10 dpc the surface area (S) of nexuses ranges from o-i to 3 x io~2 /<m2,at 14 dpc from o-i to 15 x io~2 /tm2, at 18 dpc from o-i to 26 x io~2 / 'm2, and at the adult stagefrom o-i to 54 x io~2/tm2. The percentage of large nexuses ( 5 > o - s x i o " ! / i m ! ) steadilyincreases from 10 dpc to the adult stage.

Fixation by glutaraldehyde before glycerol infiltration does not induce any modification inthe size distribution of adult heart gap junctions.

INTRODUCTION

Ultrathin sections, negative staining, diffraction analysis, and freeze-etch replicashave demonstrated the basic structure of gap junctions or nexuses (Revel & Karnov-sky, 1967; Chalcroft & Bullivant, 1970; McNutt & Weinstein, 1970, 1973; Steere &Sommer, 1972; Benedetti, Dunia & Bloemendal, 1974; Gilula, 1974a; Amsterdam,Joseph, Lieberman & Lindner, 1976). Biochemical studies on isolated gap junctionshave provided some data on their composition (Evans & Gurd, 1972; Goodenough &Stoeckenius, 1972; Goodenough, 1974; Gilula, 19746; Benedetti et al. 1976). Thesejunctions have been observed in both adult and embryonic tissues (McNutt, 1970;Friend & Gilula, 1972; Benedetti et al. 1974; McNutt & Weinstein, 1973; Revel,

4 C1£I. 30

46 D. Gros, J. P. Mocquard, C. E. Challice andj. Schrevel

Yip & Chang, 1973; Satir & Gilula, 1973) and also between both normal and trans-formed cultured cells (Revel et al. 1971; Johnson & Sheridan, 1971; Pinto da Silva &Gilula, 1972; Gilula, Reeves & Steinbach, 1972; Johnson, Hammer, Sheridan &Revel, 1974). Present between both non-excitable and excitable cells, they are prob-ably the most frequent junctions in animal tissues. Although it is now generallyaccepted that they establish ionic and metabolic coupling between cells (Furshpan &Potter, 1968; Pitts, 1971; Revel, Yee & Hudspeth, 1971; Dehaan & Sachs, 1972;Gilula et al. 1972; Bennett, 1973; Cox, Krauss, Balis & Dancis, 1974; Johnson et al.1974; Sheridan, 1974) definitive proof of this hypothesis is still lacking. However,low-resistance pathways and metabolic cooperation always correlate with the presenceof gap junctions.

The formation of nexuses has been studied in vitro and in situ using the freeze-cleave technique. Yee (1972) and Johnson et al. (1974) have reported their mor-phogenesis, respectively, between hepatocytes in the regenerating rat liver, andreaggregated Novikoff hepatoma cells, while Benedetti et al. (1974, 1976), Decker &Friend (1974), and Albertini, Fawcett & Old (1975) have described the process oftheir formation and growth, respectively, in the differentiating eye lens epithelium,the neural groove during amphibian neurulation and the ovarian follicle. Decker(1976) has demonstrated the thyroxine regulation of the differentiation of gap junc-tions between ependymoglial cells of Rana pipiens.

Although the ultrastructure and the physiology of the gap junctions in the adultheart (Dewey & Barr, 1964; Barr, Dewey & Berger, 1965; Dreifus, Girardier & Forss-mann, 1966; Revel & Karnovsky, 1967; Muir, 1967; McNutt, 1970; McNutt &Weinstein, 1970; Mazet & Cartaud, 1976) as well as their distribution and extent(Spira, 19710; Mobley & Page, 1972; Matter, 1973; Page & McAllister, 1973; Arluk& Rhodin, 1974; Hirakow & Gotoh, 1976) have provided the subject of numerousstudies, relatively little is known about gap junctions in the embryonic and developingheart. However, a few nexuses have been observed in the embryonic cardiac musclesof chick, mouse and rat (Muir, 1965; Pager & De Ceccaty, 1969; Polinger & McNutt,1971; Nanot, 1971; Spira, 19716; Challice & Viragh, 1973; Hirakow & Gotoh, 1976)and in human foetal heart (McNutt, 1970; Dehaan & Sachs, 1972). Gap junctionshave also been demonstrated between myocardial cells in culture (Dehaan & Hirakow,1972; Dehaan & Sachs, 1972; Masson-Pevet, Jongsma & Bruijne, 1976). Fromstudies of pulse-rate synchronization in pairs of isolated cardiac cells (Dehaan &Hirakow, 1972; Dehaan & Sachs, 1972), it has been found that synchrony is attainedin only a few minutes, suggesting that the electrical coupling and formation of gapjunctions are rapidly established.

The present report concerns a freeze-cleave study of the formation and growthof gap junctions in the mouse myocardium during ontogenesis. The importance ofthis structure during tissue differentiation has been emphasized (Loewenstein, 1973)and its particular role in cardiac muscle has been reviewed (Dehaan & Sachs,1972). Reports of part of this study have been published previously (Gros & Challice,19760,6).

Gap junctions in myocardium during ontogenesis 47

Fig. i . 10 dpc embryonic heart. The cleavage plane has broken through 2 myocardialcells, C] and C2) revealing the intracellular organelles and the plasma membranefracture faces Px and Z?, of the cell C\. The P faces are studded with numerous intra-membranous particles, whereas the E faces show only a few particles. Arrows indicateclosely packed arrays of particles which represent the aspect of the P fracture facesof gap junctions, es, extracellular space; g, Golgi apparatus; in, mitochondria; vif,myofibrils. x 22 400.

4-2


MATERIALS AND METHODS

Freeze-cleaving

Mice were mated overnight and the fertilized females were selected in the morning on thebasis of the presence of the vaginal plug. Mating occurs mostly during the night, generallyaround 2 a.m. (Rugh, 1968). The embryonic hearts were used between 9 and 10 a.m. The 6 to8 h that elapsed between the probable time of mating and the following morning were not in-cluded in the recorded age of the embryos. 10, 12, 14 and 18 days post-coitum (dpc) embryonichearts were obtained from pregnant mice; adult hearts from 3-month-old female mice. Theventricles were dissected in physiological salt solution (130 mM NaCl, 27 mM KC1, 2-2 mMCaCl3, 1 mM NaHjPO4, 11-3 mM NaHCO3, 075 mM MgCL, n i m M glucose) and smallpieces immersed for 20 min sequentially in 10, 20 and 30 % glycerol in physiological salt solu-tion, at 4 °C. Adult hearts were also fixed for 30 min, at 4 °C, in 5 % glutaraldehyde solution in01 M cacodylate buffer at pH 7-4, before being infiltrated with glycerol in physiological saltsolution. The specimens were mounted on gold disks and frozen in Freon 22 cooled with liquidnitrogen. Freeze-fracturing at — 100 °C, occasionally etching (from 1 to 2 mm), and platinumcarbon shadowing were carried out using a Balzers freeze-etch apparatus according to Moor &Miihlethaler (1963). Cleaned replicas were examined with a Hitachi HU 11 Cs or SiemensElmiskop 1 a electron microscope.

Surface measurements

Electron microscopes were calibrated with a germanium-shadowed carbon replica (54864lines in."1). Measurements of the surfaces of the gap junctions were all performed from micro-graphs of magnification 90000 x using graph paper. Surface measurements were carried out on3 hearts at each stage studied.

RESULTS

Formation of gap junctions

Freeze-cleaving splits the plasma membrane, producing 2 fracture faces (Branton,1966): P face directed toward the exterior of the cell and E face directed toward thecytoplasm (Branton et al. 1975). The fracture faces P and E of the plasma membranesof embryonic and adult myocardial cells are studded with intramembranous particlesdistributed at random, but in both cases the P faces show from 5 to 6 times moreparticles than E faces (Fig. 1).

At 10 dpc, the earliest stage which has been studied, gap junctions are rare and in

Fig. 2. 10 dpc embryonic heart. Arrows indicate 2 linear arrays of particles on the Pface of a cell 1 (CJ. Note the paucity of intramembranous particles around the lineararrays. E3, E fracture face of the cell 3 (C3); P2, P fracture face of the cell 2; mf,myofibrils. x 45 000.Fig. 3. 10 dpc embryonic heart. Formation plaque with exceptionally long lineararrays of 9-nm particles and aggregates (arrows). The extracellular space (es) is quitereduced in the formation plaque and widens on each side. Sometimes the particles inthe linear arrays seem to be coupled in pairs (black-headed arrows). White-headedarrows indicate io-11-nm particles, x 120000.Fig. 4.10 dpc embryonic heart. E fracture face with linear array of pits demonstratingthe gap-like nature of the linear arrays of particles of the P faces, x 99 000.Fig. 5. 10 dpc embryonic heart. P (/) and E (2, j) fracture faces of small gap junctions.Hexagonal arrays of pits (circled) clearly appear on the EF face, x 120000.

D. Gros, J. P. Mocquard, C. E. Challice and J. Schrevel


certain replicas no intercellular junctions or even any indication of their formationhave been found. This does not prove that nexuses are not present in such hearts,but it does indicate that if present they are rare. In other replicas small gap junctionsand what we believe to be images of their formation can be observed. The first indica-tion of the formation of nexuses seems to be the presence of linear arrays of 9-nmparticles (usually from 6 to 20 particles, sometimes more) on the P faces, in certainareas of the plasma membrane which are generally characterized by the relative lackof intramembranous particles of smaller and more commonly observed sizes (Figs. 2,3). These particular regions were called 'formation plaques' by Johnson et al. (1974).Aggregates of 2 or 3 rows of linear arrays and loosely organized clusters of particlesare also present in such regions (Fig. 3) where the intercellular space between theapposed membranes is greatly reduced (Fig. 3). On the E faces, alinements of smalldepressions which would correspond to the linear arrays of particles of the P facesare observed (Fig. 4). In the neighbourhood of the linear arrays some io-11-nmparticles can be seen (Fig. 3). Less frequently we can observe small hexagonal arraysof particles with a 9-nm centre-to-centre spacing, with the corresponding fractureface E showing similar arrays of 3'5-5-nm depressions (Fig. 5). These structures arecharacteristic of gap junctions. Sometimes, these early nexuses have straight orcurved 'arms' formed by one or two rows of particles (Figs. 6, 8). These arms arepart of the gap junctions as verified by the complementary EF face pits (Fig. 7).Some larger nexuses possess, within their central region, one or two circular particle-free zones, or zones containing only a few particles (Figs. 9, 10). At this stage (10 dpc)we observed 93 junctions, of which 10 were in the form of linear arrays and 83 weremacular gap junctions. 8-4% of these macular nexuses displayed arms or centralparticle-free zones, or both.

The above description applies to both 12 and 14 dpc hearts with only slight modi-fications. At 12 dpc (Figs. 11, 12) both linear and hexagonal arrays of particles are

Fig. 6. 10 dpc embryonic heart. Gap junction formed mainly by closely packedparticles but also by loosely packed particles (arrow). Arrowhead indicates a lineararray forming an arm. Note the small aggregates of 9-nm particles in the left lowerpart and in the neighbourhood of the main nexus the presence of 10-1 i-nm granules(small arrowheads), x 90 000.

Fig. 7.10 dpc embryonic heart. E fracture face of a gap junction with an arm (arrow).E face replicas reveal the complementary image of the arms demonstrating theirgap-like structure, x 120000.

Fig. 8. 10 dpc embryonic heart. In the right part of the micrograph a stretched gapjunction with 3 arms (arrowheads). The rows of particles indicated by the arrow areperhaps joining the closely packed particles, x 90000.

Fig. 9. 10 dpc embryonic heart. P fracture face of a hexagonally arrayed gap junctionwith 2 central particle-free zones, x 130000.

Fig. 10. 10 dpc embryonic heart. E fracture face of a gap junction with 2 pit-freezones (arrows) demonstrating that the presence of particle-free zones on the PF faceis not due to the pulling out of gap particles during the fracture process, x 132 000.

Fig. 11. 12 dpc embryonic heart. Linear arrays of particles (arrowheads) and a gap junc-tion with a particle-free zone and a long branched arm (arrow), x 48000.

D. Gros, J. P. Mocquard, C. E. Challice and J. Schrevel

wmm


observed and both are more common than at 10 dpc. 25-2% of the macular gapjunctions had arms and/or central particle-free zones. In 163 junctions observed wecounted 55 single linear arrays of particles. By 14 dpc (Fig. 13) the linear arrays ofparticles have become rarer, whereas the hexagonal arrays are now relatively numer-ous; 9-3 % of the macular gap junctions (215) displayed arms or central particle-freezones or both.

At 18 dpc (Fig. 14) and at the adult stage (Fig. 15) the rows of particles are nolonger seen; only macular gap junctions are present, a significant number havingattained a large size. Some micrographs suggest that the large nexuses, for these 2latter stages, may have been formed by the fusion of smaller junctions (Fig. 14) and/orby the accretion of io-u-nm individual particles which span the membrane (gapparticles) (Fig. 15). As the formation of nexuses proceeds the formation plaques areencroached upon by the common 7-5-nm intramembranous particles and only aroundthe relatively large gap junctions does a particle-free halo persist. The above observa-tions are summarized in Table 1.

Size distribution of gap junctions during the development

Single linear arrays of particles (or 'monolinear' gap junctions) were not takeninto account in the calculation of size distribution because of the difficulty in measur-ing their surface areas. When present they represent a small percentage of the totalarea of gap junctions. The central particle-free zones present in certain nexuses wereexcluded from the surface measurements. The smallest macular junctions measuredwere composed of from 8 to 10 closely packed particles.

Fig. 16 shows the distribution of the area 5 of the gap-junctions through 4 stagesexamined. At 10 dpc the area of the gap junctions observed (n = 83) ranges from o-ito 3 x io~2/tm2; at 14 dpc (n = 215) from o-i to 15 x io~2/fm2; at 18 dpc (« = 158)from o-i to 26-3 x io~2/tm2 and at the adult stage (n — 92) from o-i to 54 x io~2/tm2.

Fig. 12. 12 dpc embryonic heart. Zone rich in linear arrays of particles (arrowheads).Note also 2 small gap junctions with arms (arrows), x 48 000.

Fig. 13. 14 dpc embryonic heart. Fracture through a gap junction-containing zone.Large gap junctions are indicated by arrows and a small aggregate of particles by anarrowhead. Note in the right upper part a junction with a central particle-free region.

Fig. 14. 18 dpc embryonic heart. Convergence in a small region of numerous gap junc-tions of different sizes. Note in the neighbourhood of nexuses 10-11-nm particles(arrowheads). In such a region, the largest junctions could be formed by the fusion ofsmaller ones or by addition of gap particles, x 120000.

Fig. 15. Adult heart. Part of a large gap junction from glutaraldehyde-fixed prepara-tion. Dashed line delimits the P fracture face from the E fracture face. A small partof the EF face of the junction, with a hexagonal array of pits, can be seen (large blackarrow). Note in close proximity to the nexus, the presence of 10-11-nm particles on thePF face (small black arrows) and pits on the EF face (small white arrow). The peri-pheral particles seem to leave deeper pits on the EF face than the usual gap junctionparticles: compare the pits indicated by the white arrows and these of the pitted lattice(large black arrow). Such a micrograph could suggest the incorporation of the 10-11-nm particles in the neighbouring gap junction, x 210000.

54 D. Gros, J. P. Mocquard, C. E. Challice and J. Schrevel

An arbitrary division was made at the size of 0-5 x io~2/tm2 to define 2 classes(class I and class II) of gap junctions, this number representing the median area of theoverall number (N = 548) of nexuses observed. At 10 dpc the percentage of gapjunctions of class I (S ^ 0-5 x io~2/<.m2) is 67%; at 14 dpc, 57%; at 18 dpc, 44%and at the adult stage, 31 %. The histogram in Fig. 17 depicts these measurements.

All the above studies were performed on unfixed specimens treated with glycerolbefore freezing. To detect a possible action of glycerol on the size of the gap junctions,

Table 1. Summary of the observations about the diversity of the gapjunctions during the heart development

See explanations in the text.

10 dpc 12 dpc 14 dpc 18 dpc Adult

Linear arrays ofparticles observed

Macular gap junctionsobserved

Macular gap junctionswith ' arms ' or centralparticle-free zone/orboth

Small gap junctionsLarge gap junctions

83

55

108

(8-4%) (25-2%)

2 0

215 158 92

(9-3 %)

to

c

:tio

c

a.CO

30--

20-

10-

A

L^L

2?

CDCM rf COO^- csi ' j

S. TO"2/ im2

CD CO

in

S, 10-2 ,

CDO5

S, I O - 2 ,

40-

30-

20-

10-

a> CM coocMincM "i-^3-, _ ,_CM CMCMCO m in

S, 10-2/im2

Fig. 16. Histograms showing size distribution of gap junctions observed in io, 14, 18dpc and adult mouse hearts (A, B, C, D, respectively). Abscissa, area, S, of gap junc-tions. Ordinate, percentage of ratios no. of gap junctions in class intervals/total no. (n)of gap junctions observed at each stage: for A, B, C, D, n = 83, 215, 158, and 92, res-pectively. Three hearts used for each stage, unfixed specimens.


3 adult hearts were fixed before being infiltrated with the glycerol solutions (seeMaterials and methods). In the specimens treated in this way (Fig. 15) the ultra-structure of the gap junctions in the replicas is identical with that of unfixed speci-mens. Their size distribution after glutaraldehyde fixation is given in Fig. 18. Theareas of the nexuses observed (« = 125) range from o-i x io~2 to 32-4 x io~2/<m2

and 34% of these junctions have an area S ^ 0-5 x io~2/mi2.

80-

60-

40-

20-

I

10dDC 14dpc 18dpc Adu'tFig. 17. Histogram illustrating the percentages of class I (HO), S<o-s x io~2/tm2) andclass II (0 , S>o-5 x io-2/tm2) gap junctions in relation to their total number (w)observed at each stage.

COin

CO

oCM

COinCM CM

COCMCO

S. 10-2/;m2

Fig. 18. Histogram showing size distribution of gap junctions in 3 adult mouse hearts.Specimens fixed with glutaraldehyde before being infiltrated with glycerol, n = 125.Axes as in Fig. 16.


DISCUSSION

Embryonic myocardium and gap junctions

Muir (1965) observed a few nexuses in ultrathin sections of the 14 dpc rat embryonicmyocardium, as did Pager (1968) and Pager & de Ceccaty (1969). Spira (1971 b) foundclose oppositions (4-nm gap) between chicken myocytes, whereas the freeze-cleavestudies of Polinger & McNutt (1971) demonstrated the presence of small nexuses.Gap junctions have also been observed in human foetal myocardium (McNutt, 1970;Dehaan & Sachs, 1972). In the literature the earliest stage at which gap junctionshave thus far been reported in mouse heart is 13 dpc (Nanot, 1971; Challice & Viragh,1973). The present freeze-cleave study unquestionably demonstrates the existence ofgap junctions as early as 10 dpc in the developing cardiac muscle. The small size ofthese junctions and the limited membrane area that they occupy explain why it hasbeen so difficult to detect them in sectioned material. Since the first myocardial con-tractions appear in the tubular heart of 8-dpc embryos (Rugh, 1968) gap junctionsmight be expected to exist at this time if they represent the sites of electric couplingbetween cardioblasts.

Formation of gap junctions

Pinto da Silva & Gilula (1972) observed different gap particle arrays, especiallysmall macular and linear rows and they speculated that such different arrays mayrepresent aspects of the gap junction differentiation. Yee (1972), Decker & Friend(1974), Benedetti et al. (1974), Johnson et al. (1974), and Decker (1976) from studiesof freeze-cleave replicas of regenerating or differentiating tissues, all reached the sameconclusion. Fujisawa, Morioka, Nakamura & Watanabe (1976) suggested that sucha structural diversity may reflect the existence of functionally different gap junctions.

The ontogenetic sequence found in the present study suggests that the arrays ofparticles and depressions, respectively, observed on the P and E fracture faces duringcardiac muscle development represent stages of gap junction formation and growth.The following observations are pertinent: the presence of isolated io-11-nm particlesin the formation plaques; the presence of linear arrangements of gap particles at theearliest stages (10, 12 and 14 dpc) and then, later their absence (18 dpc and adult)(Table 1); the general increase in size of the gap junctions during heart development(Figs. 16, 17).

The mode of development of nexuses suggested by the present study is similar tothat proposed by other authors (Benedetti et al. 1974; Decker & Friend, 1974; John-son et al. 1974; Decker, 1976). In the formation plaques linear arrays of gap particlesof various lengths appear. At one end of these rows, small aggregates develop eitherby a coiling of the linear arrays, or by the accretion of other gap particles. Suchaggregates of particles act as nucleation sites which grow by incorporation of individualparticles, on other linear arrays, or both. The possibility that small aggregates developalone, without going through the linear-array stage cannot be eliminated, but thelinear arrangements appear to be characteristic of differentiating junctions. Theyhave been observed in the neural groove of amphibian neurulae (Decker & Friend,


1974), between differentiating lens fibres (Benedetti et al. 1974), between differentiat-ing ependymoglial cells (Decker, 1976), in the neural retina of chicken (Fujisawa et al.1976), and in the cardiac muscle of amphibian larvae (Mazet, 1977). The greatestnumber of linear arrays were found at the 12 dpc stage, which suggests that thetwelfth and thirteenth days are a particularly active period for gap junction differen-tiation. Such linear arrays have not been observed in the 18 dpc and adult hearts.

Nexuses with arms and/or central particle-free zones could represent intermediatesteps in the formation of the junctions. The free end of a junction arm could benduntil it touches the 'body' of the junction, forming a ring enclosing a particle-freezone. The particle-free zones formed would then become filled with gap particles byaccumulation. Larger nexuses would then be formed by the fusion of smaller onesand/or by accretion of io-11-nm particles. In the later stages (18 dpc and adult)these io-11-nm particles have been demonstrated to be gap particles. They have alsobeen observed by most authors studying the formation of gap junctions and wereinterpreted as precursors of the 9-nm subunits of the gap junctions (Johnson et al.1974; Decker & Friend, 1974; Decker, 1976). With this hypothesis it is interesting tonote the decrease in size of the gap particles. Could this change be attributed to theinteraction of particles during their packing? The reasons for this change in size arestill unclear. In the earlier stages studied io-11-nm particles are also present in thePF faces, but they are never very numerous and we have not been able to detect theircorresponding depressions in the EF faces, which is probably due to the difficulty indiscovering and observing such small isolated pits.

Such a process of formation of the gap junctions presupposes, at least, 2 conditions:translational mobility of the junctional particles, and of their linear arrays in theplasma membrane; and a mutual affinity between the gap particles. These factorshave been discussed by Decker & Friend (1974) on the basis of the fluid mosaicmembrane model proposed by Singer & Nicolson (1972).

Size distribution of the gap junctions

Figs. 16 and 17 show a general increase in size of the gap junctions with increasingage, although a significant number (31 %) of junctions of class I (S ^ 0-5 x io~2/<m2)remain in the adult heart. Such a dynamic process is a feature characteristic of apopulation growing in size and it supports the present hypothesis of gap junctionformation and growth in cardiac muscle. In the 18 dpc and adult cardiac musclesanalysed, roughly 10% of the largest nexuses (S > 0-5 x io~2/tm2) were incompletein the replicas, which suggest a slight underestimate of their size distribution. Theexistence of a virtually continuous spread of size in the adult myocardium suggeststhat even there nexuses continue to develop. However, at this stage no linear arraysof particles (supposed to be the first manifestations of junction differentiation) wereseen. Observations of linear arrays at the earlier stages are not rare, but they are notthat frequent: roughly 2 linear arrays for 10 macular nexuses observed (see Table 1).Thus, in the adult heart, the differentiation of new gap junctions appears to havereached a minimum, making the linear arrays rare, or absent.

Several papers allow us to compare the size of the gap junctions in the mouse adult


heart (o-i x io~2 /tm2-o-54 /tm2) to that of gap junctions in other kinds of tissues andto ascertain that they are similar (Revel et al. 1971; Friend & Gilula, 1972; Albertini& Anderson, 1974a, b; Fry, Devine & Burnstock, 1977).

Fixation by glutaraldehyde before glycerol infiltration does not induce any modi-fication in the size distribution of adult heart gap junctions. Replicas of nexuses offixed and unfixed specimens show identical particle arrays, as reported by Breathnach,Gross, Martin & Stolinski (1976). McNutt & Weinstein (1970, 1973) had noticed afterinfiltration of unfixed preparations of adult mouse heart by 40 % glycerol solutions,that the particle arrays tended to become looser. In both cases in the present study,both closely and loosely packed arrays were seen.

The present study shows that gap junctions form and increase in size duringdevelopment of the myocardium. With the hypothesis that gap junctions are the sitesof ionic and metabolic coupling between cells, such a development necessarily in-fluences cellular physiology and in particular the degree of intercellular communica-tion. Attempts to quantify the junctional membrane surface area involved in gapjunctions, during development and at the adult stage, are under investigation.

One of us (D.G.) is very grateful to Dr W. Costerton and Mr D. Cooper for their adviceand the use of the freeze-etch device in Calgary. The authors are indebted to Mrs C. Besse andJ. Riviere for their skilful technical assistance, to Miss D. Decourt for her help in the prepara-tion of the manuscript.

This work was supported by D.G.R.S.T. grant no. 76-7-1408, the Alberta Heart Founda-tion and the National Research Council of Canada. Part of this study was performed at theUniversity of Calgary while D. G. was a participant in the CNRS (France) and NRC (Canada)exchange programme.

REFERENCESALBERTINI, D. F. & ANDERSON, E. (1974 a). The appearance and structure of intercellular

connections during the ontogeny of the rabbit ovarian follicle with particular reference togap junctions. J. CellBiol. 63, 234-250.

ALBERTINI, D. F. & ANDERSON, E. (19746). Structural modifications of lutein cell gap junctionsduring pregnancy in the rat and the mouse. Anat. Rec. 181, 171-194.

ALBERTINI, D. F., FAWCETT, D. W. & OLDS, P. J. (1975). Morphological variations in gapjunctions of ovarian granulosa cells. Tissue & Cell 7, 389-398.

AMSTERDAM, A., JOSEPH, R., LIEBERMAN, M. E. & LINDNER, H. R. (1976). Organization ofintramembrane particles in freeze-cleavage gap junctions of rat graafian follicles: optical-diffraction analysis..?. CellBiol. 21, 93-106.

ARLUK, D. J. & "RHODIN, J. A. G. (1974). The ultrastructure of calf heart conducting fibreswith special reference to nexuses and their distribution. J. Ultrastruct. Res. 49, 11-23.

BARR, L., DEWEY, M. M. & BERGER, W. (1965). Propagation of action potentials and the struc-ture of the nexus in cardiac muscle. J. gen. Physiol. 48, 797-824.

BENEDETTI, E. L., DUNIA, I. & BLOEMENDAL, H. (1974). Development of junctions duringdifferentiation of lens fibres. Proc. natn. Acad. Sci. U.S.A. 71, 5073-5077.

BENEDETTI, E. L., DUNIA, I., BENTZEL, C. J., VERMORKEN, A. J. M., KIBBELAAR, M. & BLOE-MENDAL, H. (1976). A portrait of plasma membrane specializations in eye lens epitheliumand fibres. Biochim. biophys. Acta 457, 353-384.

BENNETT, M. V. L. (1973). Function of electronic junctions in embryonic and adult tissues.Fedn Proc. Fedn Am. Socs exp. Biol. 32, 65-75.

BRANTON, D. (1966). Fracture faces of frozen membranes. Proc. natn. Acad. Sci. U.S.A. 55,1048-1056.


BRANTON, D., BULLIVANT, S., GILULA, N. B., KARNOVSKY, M. J., MOOR, H., MOHLETHALER,K., NORTHCOTE, D. H., PACKER, L., SATIR, B., SATIR, P., SPETH, V., STAEHLIN, L. A.,STEER, R. L. & WEINSTEIN, R. S. (197s). Freeze-etching nomenclature. Science, N.Y. 190,54-56.

BREATHNACH, A. S., GROSS, M., MARTIN, B. & STOLINSKI, 0.(1976). A comparison of membranefracture faces of fixed and unfixed glycerinated tissue, jf. Cell Sci. 21, 437-448.

CHALCROFT, J. P. & BULLIVANT, S. (1970). An interpretation of liver cell membrane and junc-tion structure based on observation of freeze-fracture replicas of both sides of the fracture.J. Cell Biol. 47, 49-61.

CHALLICE, C. E. & VIRAGH, S. (1973). The embryonic development of the mammalian heart.In Ultrastructure of the Mammalian Heart (ed. C. E. Challice & S. Viragh), pp. 91-126.New York: Academic Press.

Cox, R. P., KRAUSS, M. R., BALIS, M. F. & DANCIS, J. (1974). Metabolic cooperation in cellculture. A model for cell-to-cell communication. In Cell Communication (ed. R. P. Cox),pp. 67-95. New York: Wiley.

DECKER, R. S. (1976). Hormonal regulation of gap junction differentiation. J. Cell Biol. 69,669-686.

DECKER, R. S. & FRIEND, D. S. (1974). Assembly of gap junctions during amphibian neurula-tion. J. Cell Biol. 62, 32-47.

DEHAAN, R. L. & HIRAKOW, R. (1972). Synchronization of pulsation rates in isolated cardiacmyocytes. Expl Cell Res. 70, 214-220.

DEHAAN, R. L. & SACHS, H. G. (1972). Cell coupling in developing systems: the heart-cellparadigm. Curr. Top. dev. Biol. 7, 193-228.

DEWEY, M. M. & BARR, L. (1964). Study of the structure and distribution of the nexus. J. CellBiol. 23, 553-585-

DREIFUSS, J. J., GIRARDIER, L. & FORSSMANN, W. G. (1966). Etude de la propagation de l'exci-tation dans le ventricule de rat au moyen de solutions hypertoniques. Pfliigers Arch. ges.Physiol. 292, 13-33.

EVANS, W. M. & GURD, J. W. (1972). Preparation and properties of nexus and lipid enrichedvesicles from mouse liver plasma membranes. Biochem. J. 128, 691-700.

FRIEND, D. S. & GILULA, N. B. (1972). Variations in tight and gap junctions in mammaliantissues. J. Cell Biol. 53, 758-776.

FRY, G. N., DEVINE, C. E. & BURNSTOCK, G. (1977). Freeze-fracture studies of nexuses be-tween smooth muscle cells. Close relationship to sarcoplasmic reticulum. J. Cell Biol. 72,26-35-

FUJISAWA, H., MORIOKA, H., NAKAMURA, H. & WATANABE, K. (1976). Gap junctions in thedifferentiated neural retinae of newly hatched chickens. J. Cell Sci. 22, 597-606.

FURSHPAN, E. J. & POTTER, D. D. (1968). Low-resistance junctions between cells in embryosand tissue culture. Curr. Top. dev. Biol. 3, 95-127.

GILULA, N. B. (1974 a). Junctions between cells. In Cell Communication (ed. R. P. Cox), pp.1-29. New York: Wiley.

GILULA, N. B. (19746). Isolation of rat liver gap junctions and characterization of the poly-peptides. J. Cell Biol. 63, 111 a.

GILULA, N. B., REEVES, R. & STEINBACH, A. (1972). Metabolic coupling, ionic coupling andcell contacts. Nature, Lond. 235, 262-265.

GOODENOUGH, D. A. (1974). Bulk isolation of mouse hepatocyte gap junction. Characterizationof the principal protein, connexin. J. Cell Biol. 61, 557-563.

GOODENOUGH, D. A. & STOECKENIUS, W. (1972). The isolation of mouse hepatocyte gap junc-tions. Preliminary chemical characterization and X-ray diffraction.^. Cell Biol. 54, 646-656.

GROS, D. & CHALLICE, C. E. (1976 a). Early development of gap junctions between the mouseembryonic myocardial cells. A freeze-etching study. Experientia 32, 996—997.

GROS, D. & CHALLICE, C. E. (19766). The growth of the gap junctions in mouse myocardiumduring development: a freeze-etch study. JRCS Med. Sci. 4, 416.

HIRAKOW, R. & GOTOH, T. (1976). A quantitative ultrastructural study on the developing ratheart. In Developmental and Physiological Correlates of Cardiac Muscle (ed. M. Lieberman &T. Sano), pp. 37-50. New York: Raven Press.

JOHNSON, R. G., HAMMER, M., SHERIDAN, J. & REVEL, J. P. (1974). Gap junctions betweenreaggregated Novikoff hepatoma cells. Proc. natn. Acad. Sci. U.S.A. 71, 4536-4540.


JOHNSON, R. G. & SHERIDAN, J. (1971). Junctions between cancer cells in culture: ultrastructureand permeability. Science, N.Y. 174, 717-719.

LOEVVENSTEIN, W. R. (1973). Membrane junctions in growth and differentiation. Fedn Proc.Fedn Am. Socs exp. Biol. 32, 60-64.

MASSON-PEVET, M., JONGSMA, H. J. & DE BRUIJNE, J. (1976). Collagenase - and trypsin-dissociated heart cells. A comparative ultrastructural study. J . molec. cell. Cardiol. 8, 747-759.

MATTER, A. (1973). A morphometric study on the nexus of rat cardiac muscle, Jf. Cell Biol. 56,690-696.

MAZET, F. (1977). Mise en Evidence et Etude des Variations Morphologiques des Gap-jonctionsdans les Tissus Electriquement Couples. Application aux Tissus Cardiaques. These de Doctoratd'Etat. Universite de Paris XI. Centre d'Orsay. Orsay.

MAZET, F. & CARTAUD, J. (1976). Freeze-fracture studies of frog atrial fibres. Jf. Cell Sci. 22,427-434-

MCNUTT, N. S. (1970). Ultrastructure of intercellular junctions in adult and developing cardiacmuscle. Am. J. Cardiol. 25, 169-182.

M C N U T T , N. S. & WEINSTEIN, R. S. (1970). Ultrastructure of the nexus: A correlated thin-section and freeze-cleave study. Jf. Cell Biol. 47, 666-689.

MCNUTT, N. S. & WEINSTEIN, R. S. (1973). Membrane ultrastructure at mammalian inter-cellular junctions. Prog. Biophys. molec. Biol. 26, 45-101.

MOBLEY, B. A. & PAGE, E. (1972). The surface area of sheep cardiac Purkinje fibres. Jf. Physiol.,Lond. 220, 547-564.

MOORE, H. & MOHLETHALER, K. (1963). Fine structure of frozen-etched yeast cells. Jf. CellBiol. 17, 609-628.

MUIR, A. R. (1965). Further observations on the cellular structure of cardiac muscle. J. Anat.99, 27-46.

MUIR, A. R. (1967). The effect of divalent cations on the ultrastructure of the perfused ratheart. J . Anat. 101, 239-261.

NANOT, J. (197I). Aspects Ultrastructuraux du Myocarde de VEmbryon de Souris. DeveloppementNormal, Differentiation in vitro et Action de Substances Pharmacologiques. These de 3emecycle. Nantes.

PAGE, F. & MCALLISTER, L. P. (1973). Studies on the intercalated disc of rat left ventricularmyocardial cells. Jf. Ultrastruct. Res. 43, 388-411.

PAGER, J. (1968). Evolution Structurale et Ultrastructuraie du Tissu Cardiaque en Developpe-vient cliez le Foetus de Rat. These de 3eme cycle. Poitiers.

PAGER, J. & PAVANS DE CECCATY, M. J. (1969). Evolution des rapports intercellulaires au coursde l'embryogenese du myocarde de Rat. Jf. Microscopie 8, 75 a.

PINTO DA SILVA, P. & GILULA, N. B. (1972). Gap junctions in normal and transformed fibro-blasts in culture. Expl Cell Res. 71, 393-401.

PITTS, J. D. (1971). Molecular exchange and growth control in tissue culture. In Growth Con-trol in Cell Cultures (ed. C. E. W. Wolstenholme & J. Knight), Ciba Fdn Syvip., pp. 89-105.Edinburgh and London: Churchill-Livingstone.

POLINGER, I. S. & MCNUTT, N. S. (1971). Intercellular junctions and freeze-cleaved andetched chicken heart. Commun. Am. Soc. Cell Biol., New Orleans.

REVEL, J. P. & KARNOVSKY, M. J. (1967). Hexagonal array of subunits in intercellular junctionsof the mouse heart and liver. J. Cell Biol. 33, C7-C12.

REVEL, J. P., YEE, A. G. & HUDSPETH, A. J. (1971). Gap junctions between electrotonicallycoupled cells in tissue culture and in brown fat. Proc. natn. Acad. Sci. U.S.A. 68, 2924-2927.

REVEL, J. P., YIP , P. & CHANG, L. L. (1973). Cell junctions in the early chick embryo. A freezeetch study. Devi Biol. 35, 302-317.

RUGH, R. (1968). The Mouse. Its Reproduction and Development. Minneapolis: Burgess.SATIR, P. & GILULA, N. B. (1973). The fine structure of membranes and intercellular communi-

cation in insects. A. Rev. Ent. 18, 143-166.SHERIDAN, J. D. (1974). Electrical coupling of cells and cell communication. In Cell Com-

munication (ed. R. P. Cox), pp. 31-42. New York: Wiley.SINGER, S. J. & NICOLSON, G. L. (1972). The fluid mosaic model of the structure of cell mem-

brane. Science, N.Y. 175, 720-731.


SPIRA, A. W. (1971 a). The nexuses in the intercalated disc of the canine heart: quantitativedata for an estimation of its resistance. J. Ultrastruct. Res. 34, 409-425.

SPIRA, A. W. (1971 b). Cell junctions and their role in transmural diffusion in the embryonicchick heart. Z. Zellforsch. mikrosk. Anat. 120, 463-487.

STEERE, R. L. & SOMMER, J. R. (1972). Stereo ultrastructure of nexus faces exposed by freeze-fracturing. J. Microscopie 15, 205-218.

YEE. A. G. (1972). Gap junctions between hepatocytes in regenerating rat liver. J. Cell Biol. 55,204. n.

(Received 14 June 1977)

CEL 30

Formation and growth of gap junctions in mouse myocardium during ontogenesis: Quantitative data and...

Documents

Transcript of Formation and growth of gap junctions in mouse myocardium during ontogenesis: Quantitative data and...