Franklin Sedarat, Liqun Xu, Edwin D.€W. Moore and Glen F ...

279:202-209, 2000. Am J Physiol Heart Circ PhysiolFranklin Sedarat, Liqun Xu, Edwin D. W. Moore and Glen F. Tibbits

You might find this additional information useful...

23 articles, 14 of which you can access free at: This article cites http://ajpheart.physiology.org/cgi/content/full/279/1/H202#BIBL

15 other HighWire hosted articles, the first 5 are: This article has been cited by

  [PDF]  [Full Text]  [Abstract]

, March 15, 2006; 90 (6): 1999-2014. Biophys. J.X. Koh, B. Srinivasan, H. S. Ching and A. Levchenko

A 3D Monte Carlo Analysis of the Role of Dyadic Space Geometry in Spark Generation 

[PDF]  [Full Text]  [Abstract], June 1, 2006; 290 (6): H2267-H2276. Am J Physiol Heart Circ Physiol

J. Huang, L. Xu, M. Thomas, K. Whitaker, L. Hove-Madsen and G. F. Tibbits L-type Ca2+ channel function and expression in neonatal rabbit ventricular myocytes

  [PDF]  [Full Text]  [Abstract]

, July 1, 2006; 291 (1): H344-H350. Am J Physiol Heart Circ PhysiolM. Roth H. H. Patel, B. P. Head, H. N. Petersen, I. R. Niesman, D. Huang, G. J. Gross, P. A. Insel and D.

microdomains and {delta}-opioid receptorsProtection of adult rat cardiac myocytes from ischemic cell death: role of caveolar 

[PDF]  [Full Text]  [Abstract], October 1, 2007; 93 (7): 2504-2518. Biophys. J.

P. Dan, E. Lin, J. Huang, P. Biln and G. F. Tibbits during Development

Three-Dimensional Distribution of Cardiac Na+-Ca2+ Exchanger and Ryanodine Receptor 

[PDF]  [Full Text]  [Abstract], December 1, 2007; 293 (6): H3506-H3516. Am J Physiol Heart Circ Physiol

A. Wasserstrom J. M. Cordeiro, J. E. Malone, J. M. Di Diego, F. S. Scornik, G. L. Aistrup, C. Antzelevitch and J.

Cellular and subcellular alternans in the canine left ventricle

on the following topics: http://highwire.stanford.edu/lists/artbytopic.dtlcan be found at Medline items on this article's topics

Physiology .. Lagomorpha Cell Biology .. Confocal Microscopy Oncology .. Immunofluorescence Physiology .. Sarcoplasmic Reticulum Biochemistry .. Ryanodine Receptor Calcium Release Channel

including high-resolution figures, can be found at: Updated information and services http://ajpheart.physiology.org/cgi/content/full/279/1/H202

can be found at: AJP - Heart and Circulatory Physiologyabout Additional material and information http://www.the-aps.org/publications/ajpheart

This information is current as of January 7, 2008 .  

http://www.the-aps.org/.ISSN: 0363-6135, ESSN: 1522-1539. Visit our website at Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the American Physiological Society. intact animal to the cellular, subcellular, and molecular levels. It is published 12 times a year (monthly) by the Americanlymphatics, including experimental and theoretical studies of cardiovascular function at all levels of organization ranging from the

publishes original investigations on the physiology of the heart, blood vessels, andAJP - Heart and Circulatory Physiology

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

Colocalization of dihydropyridine and ryanodine receptorsin neonate rabbit heart using confocal microscopy

FRANKLIN SEDARAT,1 LIQUN XU,1 EDWIN D. W. MOORE,2 AND GLEN F. TIBBITS1

1Cardiac Membrane Research Laboratory, Simon Fraser University, Burnaby, British ColumbiaV5A 1S6; and 2Department of Physiology, University of British Columbia, Vancouver,British Columbia V6T 1Z3, CanadaReceived 4 August 1999; accepted in final form 29 December 1999

Sedarat, Franklin, Liqun Xu, Edwin D. W. Moore,and Glen F. Tibbits. Colocalization of dihydropyridine andryanodine receptors in neonate rabbit heart using confocalmicroscopy. Am J Physiol Heart Circ Physiol 279:H202–H209, 2000.—Because of undeveloped T tubules andsparse sarcoplasmic reticulum, Ca21-induced Ca21 release(CICR) may not be the major mechanism providing contrac-tile Ca21 in the neonatal heart. Spatial association of dihy-dropyridine receptors (DHPRs) and ryanodine receptors(RyRs), a key factor for CICR, was examined in isolatedneonatal rabbit ventricular myocytes aged 3–20 days bydouble-labeling immunofluorescence and confocal micros-copy. We found a significant increase (P , 0.0005) in thedegree of colocalization of DHPR and RyR during develop-ment. The number of voxels containing DHPR that alsocontained RyR in the 3-day-old group (62 6 1.8%) was sig-nificantly lower than in the other age groups (76 6 1.3 in6-day old, 75 6 1.2 in 10-day old, and 79 6 0.9% in 20-dayold). The number of voxels containing RyR that also con-tained DHPR was significantly higher in the 20-day-oldgroup (17 6 0.5%) compared with the other age groups (10 60.7 in 3-day old, 11 6 0.6 in 6-day old, and 11 6 0.5% in10-day old). During this period, the pattern of colocalizationchanged from mostly peripheral to mostly internal couplings.Our results provide a structural basis for the diminishedprominence of CICR in neonatal heart.

calcium; dyadic coupling; excitation-contraction coupling

EXCITATION-CONTRACTION COUPLING (E-C coupling) in adultcardiac myocytes requires Ca21 influx through sar-colemmal L-type Ca21 channel or dihydropyridine re-ceptor (DHPR), followed by Ca21-induced Ca21 release(CICR) from the sarcoplasmic reticulum Ca21-releasechannel or ryanodine receptor (RyR) (5, 6). A majorstructural specialization in adult cardiac myocytes isdyadic couplings formed between sarcoplasmic reticu-lum (SR) and either the external or T-tubular sarco-lemma (SL) (12, 21). The close apposition of the SR andSL defines a functionally restricted space that acts likean incomplete barrier to diffusion underneath the sar-colemmal membrane. In this space, which has beenreferred to as fuzzy space, RyRs are located very close(,20 nm) to the SL and T-tubule membranes and thus

are exposed to a high local intracellular calcium con-centration ([Ca21]i) whenever neighboring DHPRsopen (17). The macroscopic behavior of CICR dependscritically on the spatial relationship of the DHPR andRyR in dyadic couplings, as well as on SL and SRCa21-channel kinetics (22).

During mammalian heart development, the mor-phology undergoes significant changes as does themechanism of E-C coupling (7). The SR volume inneonates is less than in the adult, and both the amountof SR and SR Ca21 uptake per gram of muscle increasewith age (19). Most newborn mammalian cardiomyo-cytes do not develop T tubules until 8–10 days of age(11). With the lack of a developed T-tubular system inneonatal heart, the spatial relationship of DHPR toRyR may be different from that in the adult heart.Because the proximity of DHPR and RyR is a key factorfor CICR, neonatal myocardium may not rely on CICRand the triggered release of Ca21 from SR to providecontractile Ca21. It has been shown, for example, thatCa21-channel blockers have little effect on tension gen-eration in neonatal cardiac muscle (16). This suggeststhat the Ca21 current may not directly contribute Ca21

for contraction or for CICR in neonatal cells. Alterna-tively, reverse-mode Na1/Ca21 exchange has been sug-gested as a major source of Ca21 influx for contractionin neonatal myocytes (3). The cardiac SL Na1/Ca21

exchanger is abundant and functionally well developedin the late fetal/early newborn rabbit heart, and thedensity appears to decline postnatally (1). Given thatthe colocalization of DHPR and RyR is an essentialelement in E-C coupling in adult heart, it is of interestto determine the developmental changes in the geomet-ric arrangement of DHPRs and RyRs in neonatalheart. In the present study we report the distributionpattern and degree of colocalization of DHPR and RyRin rabbit myocardial cells during ontogeny.

MATERIALS AND METHODS

Animals. Male or female neonatal New Zealand Whiterabbits were used in four age groups: 1- to 3-day old, 6- to7-day old, 10- to 11-day old, and 20- to 21-day old. For each

Address for reprint requests and other correspondence: G. F.Tibbits, Cardiac Membrane Research Laboratory, Simon FraserUniv., Burnaby, BC V5A 1S6, Canada (E-mail [email protected]).

The costs of publication of this article were defrayed in part by thepayment of page charges. The article must therefore be herebymarked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734solely to indicate this fact.

Am J Physiol Heart Circ Physiol279: H202–H209, 2000.

0363-6135/00 $5.00 Copyright © 2000 the American Physiological Society http://www.ajpheart.orgH202

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

age group, 5 hearts and 10 cells/heart (50 cells/age group)were studied.

Antibody characterization. Mouse monoclonal anti-RyRantibodies (IgG1, clone C3–33, 1 mg/ml) were purchased fromAffinity Bioreagents. Specificity of this antibody in rabbitcardiac muscle was determined with the use of Western blotsas previously described (2). Rabbit polyclonal anti-DHPRantibody (CNC1, 0.469 mg/ml) was used; the specificity ofthis antibody has been previously described (10). CNC1 spe-cifically binds to the class C a1-subunit of the L-type Ca21

channel.Cell isolation procedure. The method of cell isolation was

adapted from Mitra and Morad (18). All of the solutions wereprepared with double-deionized water and filtered with a0.2-mm filter. The solutions were aerated with 100% O2before and during the isolation. Neonatal rabbits were anes-thetized and heparinized by an intraperitoneal injection ofpentobarbital sodium (60 mg/kg body wt) and heparin (2,700USP/kg body wt). The heart was excised and kept in cold(4°C) dissection solution (in mM: 126 NaCl, 4.4 KCl, 5 MgCl2,24 HEPES, 22 glucose, 20 taurine, 5 creatine, 5 sodiumpyruvate, and 1 NaH2PO4, pH 7.4 at 37°C) for ;1 min toarrest the heart. After aortic cannulation, the coronary ar-teries were washed out by 5–10 ml of cold modified Kraft-bruhe (KB) solution (in mM: 10 taurine, 70 glutamic acid, 25KCl, 10 KH2PO4, and 22 dextrose, pH 7.3 at 22°C), which was

nominally free of Ca21 and Na1. The heart was thenmounted on a Langendorff apparatus. The heart was retro-gradely perfused with prewarmed (37°C) and oxygenated KBsolution for 4 min. The heart was then digested with acollagenase (0.5 mg/ml, collagenase type II; Worthington)solution for 5–8 min at a flow rate of 1–4.7 ml/min at34–35°C (enzyme concentration, flow rate, and enzyme di-gestion time depend on animal age; Table 1). This was fol-lowed by perfusion with an EGTA (0.5 mM) containing KBsolution for 5 min to wash out the digestive enzymes. Thedigested heart was transferred to a petri dish containing 5 mlEGTA-KB solution, and the ventricles were dissected fromthe heart. The ventricles were gently teased with forceps todisperse individual myocytes. The cell suspension was fil-tered through a coarse nylon mesh (200 mm) to remove tissuechunks.

Indirect immunofluorescence labeling. Double labeling wasperformed on isolated rabbit myocytes with anti-RyR andanti-DHPR primary antibodies. Isolated myocytes were ini-tially fixed with 2% buffered paraformaldehyde for 10 min.The fixed cells were quenched for aldehyde groups in 0.75%glycine buffer for 10 min. The cells were then permeabilizedwith Triton X-100 (0.1%) for 10 min. After being washed withPBS for 10 min, the cells were incubated with nonconjugatedgoat anti-rabbit IgG (Molecular Probes, Eugene, OR; 2 mg/ml) to block possible cross-reactions between the secondaryanti-rabbit antibody and rabbit myocardial cells. The anti-body dilution was 1:100 (2-h incubation time in 1.5 ml mi-crocentrifuge tubes with gentle agitation). Excess antibodywas washed off with the use of antibody wash solution [0.05%Triton X-100 in SSC (150.7 mM sodium chloride and 17.5 mMsodium citrate)] for 10 min and then PBS for another 10 min.Subsequently, the cells were incubated overnight with pri-mary antibodies in antibody buffer solution (2% goat serum,1% BSA, 0.05% Triton X-100, and 3 mM NaN3 in SSC). Theantibody buffer solution contained goat serum and BSA toblock nonspecific binding sites. Antibody dilution for primaryantibodies was 1:80 to 1:100 for the polyclonal anti-DHPRand 1:100 to 1:200 for the monoclonal anti-RyR. Excess

Fig. 1. A: 3-dimensional confocal image ofdistribution of dihydropyridine receptor(DHPR) in a 20-day-old rabbit myocyte.Data were introduced by superposition(max-z projection) of a series of optical sec-tions from front to back surfaces of cell. B:control experiment related to myocytes fromsame rabbit as in A. Cell was prepared withuse of double-labeling procedure outlined inMATERIALS AND METHODS except that primaryantibodies were omitted. Confocal micro-scope settings for images in A and B wereidentical. Absence of any specific signal oncontrol image indicates that blocking of pos-sible cross-reactions between secondaryanti-rabbit antibody and rabbit myocardialcells was complete. C: same image as in B,but intensity and contrast of pixels in imagewere adjusted to make it possible to seepixels related to nonspecific binding of sec-ondary anti-rabbit antibody. In calculationof colocalization degree, a threshold that ex-cluded ;99% of signal found in image in Bwas applied to image in A to exclude anyfluorescent signal related to nonspecificbinding. Same procedure was applied to im-age of distribution of ryanodine receptor(RyR; not shown).

Table 1. Protocols for cell isolationin different age groups

Age, days

3 6 10 20

KB, ml 4 5 6 16KB1collagenase, ml 5 8 10 20KB1EGTA, ml 4 5 6 16Perfusion speed, ml/min 1 1.5 1.6 4.7

KB, Kraftbruhe solution.

H203DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

primary antibodies were washed off with the use of antibodywash solution (2 3 10 min) and then PBS (10 min). The cellswere then incubated with Alexa-conjugated secondary anti-bodies (Molecular Probes; 2 mg/ml), diluted 1:100, for 2 h.Highly cross-adsorbed Alexa488 goat anti-rabbit IgG (H1L)and Alexa594 goat anti-mouse IgG (H1L) were used. Afterapplication of labeled secondary antibodies, the test tubeswere wrapped in aluminum foil to prevent light exposure.The cells were then washed twice with the antibody washsolution (10 min) and once with PBS (10 min). In each step,solutions were added to disposable centrifuge tubes contain-ing a suspension of isolated myocardial cells, and the tubeswere gently agitated. At the end of each step, the tubes werecentrifuged for 5 min at 5,000–10,000 rpm (depending on thesize of the cells), and then the pelleted cells were used for thenext step of the experiment. The cells were rinsed in a 90%glycerol-PBS mixture containing 2.5% 1,4-diazabicyclo-[2.2.2] octane (DABCO) and then mounted on a slide. Incontrol experiments, single staining (to ensure the function-ality of each primary antibody) and single staining withreverse secondary antibody were performed. To determinenonspecific binding, staining control experiments with sec-ondary antibody without primary antibody were also per-formed.

Wheat germ agglutinin. Wheat germ agglutinin (WGA)was used to visualize SL and T-tubular patterns. Nonperme-abilized isolated myocytes were incubated with WGA (100mg/ml) coupled to tetramethylrhodamine isothiocyanate(TRITC; Sigma) for 30 min. The cells were rinsed three timeswith PBS and then mounted on a slide.

Microscopy. Samples were examined with the use of aZeiss LSM 410 laser scanning confocal microscope. An Ar-Kr488/568 laser provided the excitation light beam, and a neu-tral density filter (T 0.01) was used to uniformly attenuatethe intensity of laser light. The excitation light (488 and 568nm) passed through a dual-band dichroic beam splitter (FT488/568) that allowed the capture of images in both greenand red channels simultaneously. The green and red emis-sions were separated by a dichroic splitter (FT 560) andfiltered (515- to 540-nm band-pass filter for green and.610-nm long-pass filter for red emission). The Z intervalwas adjusted to 0.25 mm, and the number of sections wasadjusted to 50–70 planes depending on the diameter of thecells. Two three-dimensional (3-D) images were acquiredfrom the two different emission wavelengths representingDHPR and RyR. To determine the amount of colocalization ofDHPR and RyR, images were imported into Optimas 5.2image processing software. With the use of Analytical Lan-guage for Images, macros were written to analyze the 3-Ddata sets. A threshold was applied to the images to exclude;99% of the signal found in the control images. The two 3-Dimages were compared voxel by voxel to determine the degreeof colocalization. To get the best possible resolution, weoptimized the performance of the optical system. A highlycorrected objective (Zeiss Plan-Apochromat 363, numericalaperture 1.40 oil), standard Zeiss immersion oil, and cover-slips of 0.17-mm thickness were used. To reduce sphericalaberration due to the refractive index (h) mismatch, mount-ing medium was prepared with glycerol (h 5 1.47) to make itsimilar to that of immersion oil with h 5 1.518 (96% similar-ity). Because there is lower resolution when recording opticalsections in focal planes away from the coverslip, the slideswere left upside down after slide preparation; this allowedthe myocardial cells to attach to the coverslip. Despite theuse of a plan objective, just a small part of the field of viewnearest the optical axis was imaged to reduce the effect ofoff-axis aberrations. A plan-apochromat objective was used to

eliminate chromatic aberrations. It should be noted that evenin confocal microscopy, to some extent images may sufferfrom degradation because of out-of-focus light contributing toin-focus areas and also from anisotropy in the imaging prop-erties (i.e., inferior axial resolution compared with the lateralresolution). In this study, the best compromise between sig-nal level and spatial resolution was found by setting theconfocal pinhole to the diameter of the Airy disk. By appli-cation of the Nyquist theorem, it was determined that a pixeldiameter of ;100 nm (in reference to the specimen) wasneeded to properly sample the data in the xy-plane. Byadjustment of the zoom setting in the confocal microscope, apixel size equal to 100 nm was obtained.

Materials. All chemicals used were purchased from SigmaChemicals unless specified otherwise.

RESULTS

Immunofluorescence. Immunoblots of crude mem-brane extracts from ventricular tissue of neonatal rab-bits indicated that the MA3–916 anti-RyR antibodybinds specifically to an antigen of the predicted relativemolecular mass (;565 kDa). Control experiments dem-onstrated that there is no cross-reactivity between thetwo sets of primary and secondary antibodies, asshown in Fig. 1.

Fig. 2. Wheat germ agglutinin (WGA) staining pattern in myocar-dial cells before and after T-tubular formation. A: 6-day-old rabbitmyocyte labeled with TRITC-WGA. Note fluorescent labeling as aboundary around cell (no T tubules). Arrows, staining related tointernal organelles (refer to RESULTS); n, nuclear shadow. B: nearlyfully developed T tubules in a myocardial cell isolated from 20-day-old rabbit heart.

H204 DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

WGA. WGA staining patterns before and after T-tubular formation are illustrated in Fig. 2, A and B. Ttubules were not observed in 3- and 6-day-old rabbitmyocytes. In these age groups, WGA stained the SL,and a boundary around the periphery of the cell wasobserved. Although the cells were not permeabilized, afew internal spots were observed. This suggests that

fixation of myocardial cells may induce some pores inthe SL. T tubules were first observed in myocardialcells from 10-day-old rabbits. At this age, the T tubuleswere observed as small invaginations, whereas in moremature myocytes, WGA labeling was distributedthroughout the entire cell, indicating a more developedT-tubular system. Different degrees of T-tubular devel-

Fig. 3. RyR staining pattern (left, pseudocolored red), DHPR staining pattern (middle, pseudocolored green), andcolocalization staining pattern (right, colocalized pixels are pseudocolored yellow) in myocardial cells isolated from3- (A), 6- (B), 10- (C), and 20-day-old (D) rabbit hearts. Nos. of voxels containing DHPR were 8,778, 11,563, 25,192,and 75,799 in A, B, C, and D, respectively. Nos. of voxels containing RyR were 91,612, 86,177, 185,275, and 339,733in A, B, C, and D, respectively. Nos. of voxels containing both DHPR and RyR were 5,677, 8,609, 18,949, and 59,623in A, B, C, and D, respectively. Note that above nos. were calculated from whole image stack, but just 1 focal planefrom each image stack is shown in each panel. Refer to RESULTS for description of staining patterns.

H205DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

opment were observed in different cells obtained fromthe same animal. A nonuniform appearance of T-tubu-lar development in the cells acquired from the sameanimal, particularly in myocytes from 10-day-old rab-bits, indicated that in a given heart myocardial cellscan be in different stages of development.

RyR, DHPR, and colocalization staining patterns.Figure 3 is composed of representative confocal imagesacquired from isolated myocytes in each of the four agegroups. Each panel shows RyR (red), DHPR (green),and colocalization (yellow) staining patterns in onefocal plane close to the center of the cell.

Staining pattern of RyR was similar in all age groupsand included striations, spaced at regular intervals of;2 mm, that clearly indicated the existence of cytoplas-mic arrays of RyR (red pixels in Fig. 3, A–D). Thispattern is consistent with that of the Z lines in adultmyocardial cells but was observed in even the youngestanimals at a time when the T tubules had not yetformed.

The pattern of distribution of DHPR was different inthe different age groups. In young animals, before thedevelopment of T tubules, fluorescence associated with

the DHPRs was seen as discrete spots only in theperiphery of the cell, most likely in the SL (green pixelsin Fig. 3, A and B). After the formation of T tubules,however, fluorescence associated with DHPRs could beseen both at the periphery of the cell and in the cellinterior (green pixels in Fig. 3, C and D). There was aclose correlation between the age at which the T tu-bules began to form and the time at which DHPRsbegan to be seen in the cell interior. In 10-day-oldrabbits, fluorescence associated with DHPRs wascloser to the periphery than the center of the cell, andthis parallels the pattern of T tubules at this age,which are detected as only small invaginations fromthe surface. In 20-day-old rabbits, however, DHPRsappeared to be distributed in a pattern similar to thatof RyR: regularly spaced transverse bands (and at thisage the T tubules have nearly fully developed). In pilotstudies, live myocardial cells were labeled, by suspend-ing them in a buffer that resembled the intracellularenvironment, and skinned with saponin. Because thisapproach did not use Triton X-100 and there were nosignificant differences in the staining pattern of DHPR,we believe there were either no or a negligible number

Fig. 4. A: cross sections of a 6-day-old rabbit myocardial cell. Cell has been rotated about x- and y-axes. Top: DHPR(green), RyR (red), and colocalization (yellow) staining patterns. Bottom: only voxels containing DHPR. It isapparent from cross sections that DHPR and also coincident voxels are located almost exclusively on cell surfaceat this age. Note elongation of distribution patterns along the z-axis because of lower resolution in z (axialresolution) compared with lateral resolution. B: cross (yz) sections of a 20-day-old rabbit myocyte. Colocalizedvoxels (yellow, top) and DHPR voxels (green, bottom) are distributed both on cell surface (arrowheads) and ininterior.

H206 DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

of epitopes removed by the mild Triton X-100 treat-ment used in the present study.

In superimposed images, a voxel containing bothDHPR and RyR is pseudocolored yellow. The pattern ofdistribution of colocalized DHPR and RyR was strik-ingly similar to the pattern of distribution of DHPR inevery age group. In young animals, there were discretepoints of colocalization in the periphery before T-tubu-lar formation (yellow pixels in Fig. 3, A and B). In olderanimals, there was an increase in the number of colo-calized voxels in the interior of the cell, and, finally, thecolocalized voxels appeared as regularly spaced arraysin animals in which the T tubules had nearly fullydeveloped (yellow pixels in Fig. 3, C and D). In 3- and6-day-old rabbits, the sarcolemmal region was demar-cated by discrete yellow spots. Many of the colocalizedvoxels were in register with the regularly spaced trans-verse bands of RyR in the cytoplasm. This suggeststhat DHPRs may be clustered in the SL in close asso-ciation with subsarcolemmal junctional SR to makeperipheral couplings. At the same time, a substantialamount of RyR-specific staining was observed in theinterior of the cell that was not in close associationwith DHPRs and probably represents corbular SR. In10-day-old rabbits, colocalization was detected in boththe periphery and in discrete spots within the cell. Thisindicates both peripheral and internal couplings, sug-gesting that DHPRs are appearing inside the develop-ing T tubules and are making internal couplings withSR elements. In the 20-day-old group, yellow voxelswere most likely to appear in transversely orientedbands along the entire length of the cell. Figure 4shows cross-sectional images of myocardial cells beforeand after T-tubular formation.

Degree of colocalization. With the use of Optimassoftware, the degree of colocalization was calculated inthe different age groups. The results are shown inTable 2.

For both DHPR and RyR, the degree of colocalizationincreased with age. Even in the youngest group, .60%of DHPRs were colocalized, but the colocalized voxelswere restricted to the periphery. At the same time, 90%of RyRs were not colocalized. A lower degree of colocal-ization for RyR compared with DHPR was observed inall age groups because of a higher number of voxelscontaining signal specific for RyR compared with the

number of voxels containing signal specific for DHPR.The number of voxels containing RyR was almost five-to seven-fold greater than the number of voxels con-taining DHPR. In the older age groups, the degree ofcolocalization increased. In 20-day-old animals, almost80% of DHPRs and 17% of RyRs were colocalized. Atthis age, although .80% of RyRs were not colocalized,the pattern of colocalization was totally changed com-pared with younger animals. In these more maturecells, colocalized voxels were detected along the entirewidth of the myocardial cell. To quantify peripheral vs.internal couplings, the ratios of pixels containing pe-ripheral couplings to the total number of coincidentpixels were calculated in different age groups. In eachage group, 20 cells were randomly selected. In eachcell, images from one plane in the middle of the stackwere chosen. In these image planes, numbers of coin-cident pixels located at the border of the cells weremeasured as peripheral couplings. As shown in Fig. 5,the percentages of peripheral couplings in the myo-cytes from 3-, 6-, 10-, and 20-day-old rabbits were96.8 6 1.0, 88.1 6 1.8, 40.5 6 3.5, and 24.8 6 2.5,respectively.

Statistical analysis. With the use of SPSS statisticalanalysis software, one-way ANOVA was performed toexamine the equality of the means for the amount ofcolocalization in the different age groups.

A multiple-comparisons test indicated that theDHPR colocalization differs significantly between thefirst age group (1- to 3-day old) and all other groups(P , 0.0005). For RyR, the amount of colocalization inthe oldest group (20-day old) differs significantly fromall other groups (P , 0.0005).

Fig. 5. Percentages of peripheral couplings in different age groupsrepresenting ratio of colocalized pixels located in the periphery of cellto total no. of colocalized pixels. Values are means 6 SE; n 5 20 foreach age group.

Table 2. Degree of colocalizationin different age groups

Age,days

No. of Voxels Colocalization Degree

DHPR RyR Colocalized %DHPR %RyR

3 14,780 91,458 9,163 6261.8 1060.76 21,129 145,789 16,058 7661.3 1160.6

10 23,264 158,198 17,448 7561.2 1160.520 47,154 216,910 37,251 7960.9 1760.5

For no. of voxels, values are means. For colocalization degree,values are means 6 SE. DHPR, dihydropyridine receptor; RyR,ryanodine receptor; %DHPR, no. of colocalized voxels divided by totalno. of DHPR voxels; %RyR, no. of colocalized voxels divided by totalno. of RyR voxels.

H207DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

DISCUSSION

Immunofluorescent localization of DHPR and RyR.This study examined the subcellular distribution of SLL-type Ca21 channels and SR Ca21-release channels inneonate rabbit ventricular myocytes aged 3–20 days.

RyR staining pattern. We observed that the stainingpattern of RyR was in transverse bands at regularlyspaced intervals of ;2 mm. This is consistent with theRyR pattern in adult rabbit myocardial cells, as re-ported by Carl et al. (2). Jorgensen et al. (14) detecteda similar RyR-specific pattern of transversely orientedrows of fluorescent foci in rat papillary myofibers. Wefound well-organized arrays of RyRs in the cytoplasmof myocytes even in 3-day-old rabbits at a time whenthe T tubules had not yet formed. This parallels thereport of Carl et al. (2) of RyR staining pattern in adultrabbit atrial cardiac myocytes, which also lack a T-tubular system. The RyR pattern was similar in all agegroups. Parameswaran et al. (20) studied the distribu-tion of the RyR in postnatal developing porcine cardiacmuscle (at 3, 5, 10, and 20 days) and reported nodifference in immunofluorescence labeling of RyR inthese age groups compared with adult porcine cardiacmyocytes.

RyRs were partly colocalized and partly noncolocal-ized with SL or T-tubular DHPRs. Because the prom-inent structural difference between junctional and cor-bular SR is that junctional SR is connected to either Ttubules or to SL via “feet” structures, whereas corbularSR is not (13), we conclude that RyRs in rabbit myo-cardial cells are located in both junctional and corbularSR. This is in agreement with the immunoelectronmicroscopy studies by Jorgensen et al. (14), whichreported that RyRs were localized to junctional andcorbular SR. Another consideration is whether or notthe RyR staining pattern is different in junctional andcorbular SR. In our images, the RyR staining patternwas identical in junctional and corbular SR. This par-allels the study of the structure of corbular SR in rabbitcardiac muscle by freeze fracture. Dolber and Sommer(4) reported that the processes on the surface of corbu-lar SR had all the anatomical features of junctionalprocesses of junctional SR.

DHPR staining pattern. Before T-tubular formation,we observed the pattern of DHPR staining as discretespots limited to the periphery of the cell. This patternis consistent with DHPR staining pattern in rabbitatrial cells that lack a T-tubular system, as reported byCarl et al. (2). In 10-day-old rabbits, DHPRs wereobserved as discrete spots in the interior of the cell aswell as in the periphery. In 20-day-old rabbits, DHPRswere distributed mostly in transverse bands similar toRyR staining pattern. The DHPR staining pattern inthe 20-day-old rabbit is similar to that in adult rabbit,as reported by other investigators (2).

An interesting observation in our study was thepresence of some intensive green fluorescent signalsaround the nuclear area in some of the 3- and 6-day-oldrabbit myocytes. This was somewhat unexpected inimmature cells that are essentially devoid of T tubules,

and, as a result, the fluorescence associated withDHPR should be restricted to cell periphery. Thesespots were never observed in control experiments,which indicates that they are not related to nonspecificbinding. This fluorescent staining is most likely relatedto a1-subunits located in cytoplasmic organelles, eitherin the process of posttranslational assembly and pack-aging for the delivery to the SL or in the process ofdegradation.

Colocalization of DHPR and RyR. There are twomajor findings in this study related to colocalization ofDHPR and RyR during the period of ontogeny: 1) thedegree of colocalization of DHPR and RyR increaseswith age during development, and 2) there is a transi-tion from couplings restricted to the periphery tomostly internal couplings along the entire length of thecell. It is conceivable that by increasing the number ofcouplings and the distribution of these couplings overthe entire cell, the role of inward Ca21 current (ICa),CICR, and SR Ca21 release increases in the contrac-tion of developing myocardial cells.

Before T-tubular formation, we found that codistri-bution of DHPR and RyR was restricted to the periph-eral SL. Other studies in chick myocardium (23) withno T tubules and also in rabbit atrial cells (2) revealeddyadic couplings between DHPR and RyR restricted tothe peripheral SL. We observed more internal dyadiccouplings in the older animals, which paralleled thedevelopment of the T-tubular network. In the 20-day-old animals, couplings were most likely to appear alongthe entire width of the cell. This is consistent with thedistribution of DHPR and RyR couplings in adult rab-bit ventricular cells, as reported by other researchers(2). In a recent study, localized SR Ca21-release events(Ca21 sparks) were reported to occur predominantly atthe cell periphery of newborn (1–14 days) rabbit myo-cytes in contrast to adults, in which sparks occurredacross the entire width of the cell (9). In all age groups,we observed many of the RyRs as non-colocalized withDHPRs. Even in 20-day-old rabbits, only 17% of RyRswere colocalized. In one study by Moore (EDW Moore,unpublished data) in adult rat myocardial cells, ;30%of RyRs were reported as colocalized with DHPR. Itwould be worthwhile to determine the role of these“uncoupled” RyRs in corbular SR and whether they areactivated in the absence of a close association with theDHPR.

In this study, even in the youngest group, fluores-cence related to DHPR was detected in clusters mostlyin the regions of the SL that are in register withcytoplasmic arrays of RyR, which we refer to as mor-phological cross talk. This is in contrast to Na1/Ca21

exchanger (NCX) immunolocalization studies, whichhave shown that antibodies against NCX label SLalmost uniformly (3, 15). Although this morphologicalcross talk may suggest that the two proteins are inclose association with each other, it does not necessar-ily mean that they are close enough to have functionalcross talk. It should be noted that measurement ofcolocalization degree is always at the limits of theoptical system. It has been postulated that the L-type

H208 DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from

Ca21 channel should be within ;20 nm of the RyR totrigger Ca21 release from the SR (16). Even under idealconditions, the resolution of confocal microscopy is sev-eral times this distance. Changes in E-C coupling as-sociated with cardiac hypertrophy investigated by Go-mez et al. (8) suggest that functional cross talk is verysensitive to the geometric arrangement of DHPRs andRyRs in the dyad (8). With the use of a perforatedpatch-clamp technique, we studied inactivation kinet-ics of the L-type Ca21 channel before and afterapplication of 5 mM caffeine from myocytes of rabbitsof the same age groups as used in the present study(24). These data, which show only the inactivationkinetics of the 20-day-old group to be affected by caf-feine, are consistent with the notion that DHPR andRyR functional coupling is only observed in the oldestgroup.

Our data along with data from other investigatorssuggest that dyadic couplings in rabbit ventricularmyocytes undergo both structural and functionalchanges during the first 20 days of life. In immaturemyocytes, elements of dyadic couplings (DHPR andRyR) are less coupled, either morphologically or func-tionally. Because cross signaling of DHPR and RyR isthe cornerstone of CICR, and CICR is widely acceptedas the fundamental component of E-C coupling in ma-ture heart, there should be another alternative mech-anism in the immature myocardium to provide theCa21 required for contraction. Because the NCX isabundantly localized in the SL and demonstrates highactivity in fetal and newborn hearts, reverse-modeNa1/Ca21 exchange may be a major pathway for trans-membrane influx of Ca21 (3). We speculate that inthe rabbit myocytes at some time during ontogeny,most likely in the first month after birth, the transitionto adult E-C coupling mechanism occurs, and the de-pendence on the NCX to provide contractile Ca21 de-creases. The characterization of the control systemsthat integrate these changes remains to be achieved.

We thank the generous support of the Heart and Stroke Founda-tion of British Columbia and Yukon (to G. F. Tibbits) that made thesestudies possible. Rabbit polyclonal anti-DHPR antibody was gener-ously provided by Dr. William A. Catterall at the Department ofPharmacology, University of Washington, Seattle, WA.

REFERENCES

1. Artman M, Ichikawa H, Avkiran M, and Coetzee WA. Na1/Ca21 exchange current density in cardiac myocytes from rabbitsand guinea pigs during postnatal development. Am J PhysiolHeart Circ Physiol 268: H1714–H1722, 1995.

2. Carl SL, Felix K, Caswell AH, Brandt NR, Ball WJ Jr,Vaghy PL, Meissner G, and Ferguson DG. Immunolocaliza-tion of sarcolemmal dihydropyridine receptor and sarcoplasmicreticular triadin and ryanodine receptor in rabbit ventricle andatrium. J Cell Biol 129: 672–82, 1995.

3. Chen F, Mottino G, Klitzner TS, Philipson KD, and FrankJS. Distribution of the Na1/Ca21 exchange protein in developingrabbit myocytes. Am J Physiol Cell Physiol 268: C1126–C1132,1995.

4. Dolber PC and Sommer JR. Corbular sarcoplasmic reticulumof rabbit cardiac muscle. J Ultrastruct Res 87: 190–196, 1984.

5. Fabiato A. Appraisal of the physiological relevance of two hy-potheses for the mechanism of calcium release from the mam-

malian cardiac sarcoplasmic reticulum: calcium-induced releaseversus charge-coupled release. Mol Cell Biochem 89: 135–140,1989.

6. Fabiato A. Calcium-induced release of calcium from the cardiacsarcoplasmic reticulum. Am J Physiol Cell Physiol 245: C1–C14,1983.

7. Fabiato A and Fabiato F. Calcium-induced release of calciumfrom the sarcoplasmic reticulum of skinned cells from adulthuman, dog, cat, rabbit, rat, and frog hearts and from fetal andnew-born rat ventricles. Ann NY Acad Sci 307: 491–522, 1978.

8. Gomez AM, Valdivia HH, Cheng H, Lederer MR, SantanaLF, Cannell MB, McCune SA, Altschuld RA, and LedererWJ. Defective excitation-contraction coupling in experimentalcardiac hypertrophy and heart failure. Science 276: 800–806,1997.

9. Haddock PS, Coetzee WA, Cho E, Porter L, Katoh H, BersDM, Jafri MS, and Artman M. Subcellular [Ca21]i gradientsduring excitation-contraction coupling in newborn rabbit ven-tricular myocytes. Circ Res 85: 415–427, 1999.

10. Hell JW, Westenbroek RE, Warner C, Ahlijanian MK, Pry-stay W, Gilbert MM, Snutch TP, and Catterall WA. Identi-fication and differential subcellular localization of the neuronalclass C and class D L-type calcium channel alpha 1 subunits.J Cell Biol 123: 949–962, 1993.

11. Hoerter J, Mazet F, and Vassort G. Perinatal growth of therabbit cardiac cell: possible implications for the mechanism ofrelaxation. J Mol Cell Cardiol 13: 725–740, 1981.

12. Inui M, Saito A, and Fleischer S. Isolation of the ryanodinereceptor from cardiac sarcoplasmic reticulum and identity withthe feet structures. J Biol Chem 262: 15637–15642, 1987.

13. Jewett PH, Sommer JR, and Johnson EA. Cardiac muscle.Its ultrastructure in the finch and hummingbird with specialreference to the sarcoplasmic reticulum. J Cell Biol 49: 50–65,1971.

14. Jorgensen AO, Shen AC, Arnold W, McPherson PS, andCampbell KP. The Ca21-release channel/ryanodine receptor islocalized in junctional and corbular sarcoplasmic reticulum incardiac muscle. J Cell Biol 120: 969–980, 1993.

15. Kieval RS, Bloch RJ, Lindenmayer GE, Ambesi A, andLederer WJ. Immunofluorescence localization of the Na-Caexchanger in heart cells. Am J Physiol Cell Physiol 263: C545–C550, 1992.

16. Klitzner TS, Chen FH, Raven RR, Wetzel GT, and Fried-man WF. Calcium current and tension generation in immaturemammalian myocardium: effects of diltiazem. J Mol Cell Cardiol23: 807–15, 1991.

17. Lederer WJ, Niggli E, and Hadley RW. Sodium-calcium ex-change in excitable cells: fuzzy space. Science 248: 283, 1990.

18. Mitra R and Morad M. A uniform enzymatic method for dis-sociation of myocytes from hearts and stomachs of vertebrates.Am J Physiol Heart Circ Physiol 249: H1056–H1060, 1985.

19. Nakanishi T and Jarmakani JM. Developmental changes inmyocardial mechanical function and subcellular organelles.Am J Physiol Heart Circ Physiol 246: H615–H625, 1984.

20. Parameswaran S, Jones L, Kobayashi Y, and JorgensenAO. Subcellular distribution of triadin and the ryanodine recep-tor (RyR) in developing cardiac sarcoplasmic reticulum (SR)(Abstract). Biophys J 76: A469, 1999.

21. Rardon DP, Cefali DC, and Mitchell RD. High molecularweight proteins purified from cardiac junctional sarcoplasmicreticulum vesicles are ryanodine-sensitive calcium channels.Circ Res 64: 779–789, 1989.

22. Stern MD. Theory of excitation-contraction coupling in cardiacmuscle. Biophys J 63: 497–517, 1992.

23. Sun XH, Protasi F, Takahashi M, Takeshima H, FergusonDG, and Franzini-Armstrong C. Molecular architecture ofmembranes involved in excitation-contraction coupling of car-diac muscle. J Cell Biol 129: 659–671, 1995.

24. Xu L, Hove-Madsen L, and Tibbits GF. Ontogeny of myocar-dial excitation-contraction coupling (Abstract). Biophys J 76:A306, 1999.

H209DHPR AND RYR COLOCALIZATION IN DEVELOPING RABBIT HEART

on January 7, 2008 ajpheart.physiology.org

Dow

nloaded from