Distribution of Atrial and Nodal Cells Within the
Rabbit Sinoatrial Node
Models of Sinoatrial Transition
E. Etienne Verheijck, PhD; Andy Wessels, PhD; Antoni C.G. van Ginneken, PhD; Jan Bourier;
Marry W.M. Markman; Jacqueline L.M. Vermeulen; Jacques M.T. de Bakker, PhD;
Wouter H. Lamers, PhD; Tobias Opthof, PhD; Lennart N. Bouman, PhD
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Background—In the sinoatrial node (SAN) the course of the action potential gradually changes from the primary
pacemaker region toward the atrium. It is not known whether this gradient results from different intrinsic characteristics
of the nodal cells, from an increasing electrotonic interaction with the atrium, or from both. Therefore we have
characterized the immunohistochemical, morphological, and electrophysiological correlates of this functional gradient.
Methods and Results—The distribution of rabbit nodal myocytes in the SAN has been studied by immunohistochemistry.
After cell isolation, the electrophysiological characteristics of different nodal cell types were measured. (1) The staining
pattern of a neurofilament protein coincides with the electrophysiologically mapped pacemaker region in the SAN. (2)
Enzymatic digestion of the SAN reveals three morphologically different nodal cell types and one atrial type. Of each
nodal cell type, neurofilament-positive as well as neurofilament-negative myocytes are found. Atrial cells are all
neurofilament-negative. (3) In contrast to previous findings, we observed atrial cells in the very center of the SAN. The
relative number of atrial cells gradually increases from the central pacemaker area toward the atrium. (4) Differences
in electrophysiological characteristics between individual nodal cells are not associated with differences in cell type.
Conclusions—(1) The expression of neurofilaments can be used to delineate the nodal area in the intact SAN but is not
sufficiently sensitive for characterizing all individual isolated nodal cells. (2) A fundamentally different organization of
the SAN is presented: The gradual increase in density of atrial cells from the dominant area toward the crista terminalis
in the SAN causes a gradual increase of atrial electrotonic influence that may be an important cause of the gradual
transition of the nodal to the atrial type of action potential. (Circulation. 1998;97:1623-1631.)
Key Words: pacemakers • cells • electrophysiology • immunohistochemistry
he mammalian sinoatrial node (SAN) displays inhoniogeneity in structure and function.1"4 The site of the
earliest activation (ie, the primary pacemaker area) is invariably localized in an area with low myofilament density in
rabbit,5 guinea pig,6 cat,7 pig,8 and monkey.9 The action
potentials in the primary pacemaker area are characterized by
low upstroke velocity in association with substantial diastolic
depolarization.5 In several species a more or less gradual
transition in action potential characteristics from a nodal type
(low upstroke velocity, steep diastolic depolarization) to an
atrial type (high upstroke velocity, no diastolic depolarization) has been demonstrated. The area of transition has been
extensively studied in rabbit both by morphological and
electrophysiological methods.5-10'12 From the earlier work the
conceptual model arose of a compact center, homogeneously
composed of typical nodal cells, surrounded by an area in
which structure and electrophysiology of the fibers gradually
T
changes from nodal to atrial. In accordance with this view,
gap junction density was found to increase from the center of
the rabbit node toward the surrounding atria.5'10 In a model
study it was established that proper pacemaking in the SAN
required a gradual increase in electrical coupling from the
center of the SAN to the atrial myocardium.13 However, more
recent data suggest a less gradual transition. Measurements of
electrotonic spread of subliminal pulses demonstrated an
irregular pattern consistent with an inhomogeneous distribution of nodal and atrial fibers.12 In guinea pig SAN, a
combined immunohistochemical and electrophysiological
study demonstrated that the center of the node is surrounded
by an area in which small strands of atrial cells interdigitate
with strands of nodal cells.14 On the other hand, this study
also showed that the small gap junctions that have been
observed with the electron microscope technique15 could not
be demonstrated immunohistochemically. Thus these studies
Received July 17, 1997; revision received October 24, 1997; accepted November 19, 1997.
From the Department of Physiology (E.E.V., A.C.G.v.G., J.B., L.N.B.), the Department of Anatomy and Embryology (A.W., M.W.M.M., J.L.M.V.,
W.H.L.), and the Department of Clinical and Experimental Cardiology (J.M.T.d.B., T.O.), Academic Medical Center, University of Amsterdam, The
Netherlands.
Correspondence to Dr E. Etienne Verheijck, Department of Physiology, Academic Medical Center, University of Amsterdam, PO Box 22700, 1100
DE Amsterdam, Netherlands.
E-mail [email protected]
© 1998 American Heart Association, Inc.
1623
1624
Atrial and Nodal Cells in the SA Node
brought conflicting data on the question of whether the
transition from the center of the node to the atrial myocardium is composed of gradually changing fiber types or is
mainly formed by interdigitating strands of nodal and atrial
cells.
The objective of the current study was to investigate the
immunohistochemical, morphological, and intrinsic electrophysiological substrate of the gradual transition in action
potential configuration when going from the primary pacemaker region toward the atrium. In contrast to previous
findings, we observed that atrial cells can even be observed in
the very center of the SAN. Furthermore, we have found that
the SAN comprises three morphologically distinct nodal cell
types that do not exhibit different electrophysiological
characteristics.
Methods
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Animals
For all experiments, New Zealand albino rabbits weighing 1.8 to 2.5
kg of either sex were used. Rabbits were anesthetized with Hypnorm
(1 mL/kg, 0.32 mg/mL fentanylcitrate, and 10 mg/mL fluanison IM,
Janssen Pharmaceutics). Under artificial ventilation, the thorax was
opened and 0.1 mL heparin sodium (5000 lU/mL) was injected into
the left ventricle, after which the heart was excised. Animal care was
in accordance with institutional guidelines.
Immunohistochemistry
Tissue and Cell Preparation
Hearts were dissected and fixed in a mixture of methanol, acetone,
acetic acid, and water (35:35:5:25 vol/vol) for 48 hours, dehydrated
in a graded series of ethanol, cleared in chloroform, embedded in
Paraplast Plus, and cut in a transverse plane into 5-p,m-thick serial
sections.
Isolated cardiomyocytes from the SAN region (see below) were
collected on slides and fixed in a fixative containing methanol,
acetone, acetic acid, water (35:35:5:25 vol/vol) for 20 minutes and
washed with phosphate-buffered saline (PBS).
Immunohistochemical Staining
To detect the binding of the specific monoclonal antibodies with
the respective antigens on the paraplast sections and on the
isolated cells, the indirect unconjugated peroxidase-antiperoxidase technique (PAP technique) was applied essentially as described before.16 After deparaffination, sections were treated with
hydrogen peroxide (3% vol/vol in pH 7.4, PBS) for 30 minutes to
reduce endogenous peroxidase activity. This procedure proved
not to be necessary for isolated cells. To reduce nonspecific
binding, sections and cells were preincubated with TENG-T
(10 mmol/L Tris, 5 mmol/L EDTA, 150 mmol/L NaCl, 0.25%
gelatin, 0.05% Tween-20, pH 8.0) for 15 (cells) or 30 minutes
(sections). After incubation with the monoclonal antibodies
(overnight at room temperature), antibody binding was detected
by subsequent incubation with goat anti-mouse, donkey anti-goat,
and goat peroxidase-antiperoxidase complex diluted in PBS. This
immunocomplex was visualized by incubating the sections with
0.5 mg/mL 3,3'diaminobenzidine, 0.02% hydrogen peroxidase in
30 mmol/L imidazole, 1 mmol/L EDTA (pH 7.0). Tissue and
cell-containing slides were mounted in Entellan (Merck).
Antibodies
The monoclonal antibody reacting with the low-molecular-weight
fraction of mammalian neurofilament (68 kD), was purchased from
DAKOPATTS. Anti-human desmin was purchased from Sanbio.
SAN Preparation
Electrophysiological Mapping of the SAN
Hearts were excised and immersed in a solution containing (mmol/
L): 130.6 NaCl; 5.6 KC1; 2.2 CaCl2; 0.6 MgCl2; 24.2 NaHCO3; 11.1
glucose; 13.2 sucrose, saturated with 95% O2 and 5% CO2 at a
temperature of 37±0.3°C and a pH of 7.4. The isolated right atrium
preparation including the SAN and the crista terminalis was mounted
on a perforated silicon rubber block, endocardial side up, and
perfused with the same solution. Transmembrane potentials were
recorded with glass microelectrodes filled with 2.7 mol/L KC1 and
2 mmol/L potassium citrate. Generally, impalements were made
0.4 mm apart and in the primary pacemaker area up to 0.2 mm apart.
A unipolar surface electrogram, derived from the crista terminalis,
provided a time reference. The activation moment of a cell was
defined as the moment that the voltage was halfway between the
maximum diastolic potential (MDP) and the top of the action
potential and was timed with respect to the time reference. Beat-tobeat interval and temperature were recorded continuously. Signals
were stored on magnetic tape for off-line data analysis. The mapping
procedure lasted 1.5 to 2 hours.
In one set of experiments the distribution of upstroke velocities of
the action potentials was correlated with neurofilament staining
pattern. To align the electrical and immunohistological map, the
tissue was marked with 10- to 50-jam-diameter Alcian blue dots.
Therefore the electrode was backfilled with 1% Alcian blue in
0.5 mmol/L sodium acetate (pH 4.0) at the end of the mapping
procedure, the tip of the electrode was broken to lower resistance,
and dots were obtained iontophoretically by applying rectangular
0.5-mA pulses (500 Hz; duration 30 to 300 ju,s) for 10 to 15
seconds.17 Dots were placed at an interval of 400 /xm along lines
parallel and perpendicular to the crista terminalis. In total, about 12
dots were placed. After localization of the dots in the tissue sections,
a two-dimensional reconstruction of the neurofilament labeling
patterns in the SAN region was projected on the maximum upstroke
velocity map.
In another set of experiments we constructed activation maps
before cell isolation to estimate the distribution of morphologically
different nodal cell types in functionally different areas of the SAN.1
Therefore, the leading pacemaker cell group, that is, the group of
earliest-discharging cells (area 1 to 1.5 mm2) and an adjacent piece
of tissue (2 to 3 mm2) located superior to the leading pacemaker cell
group containing latent pacemaker cells, were dissected and separately dissociated (see below). After isolation, different cell types
were counted and their ratio calculated.
Cell Distribution Without Electrophysiological Mapping
To determine local differences in cell distribution we dissected three
2.5 mm2 adjacent pieces of nodal tissue, along a line perpendicular to
the crista terminalis: (1) the central portion of the putative node,
expected to contain nodal tissue only, (2) the crista terminalis part,
containing nodal and some atrial tissue from the site of the crista
terminalis, and (3) the septal part, existing of nodal and some atrial
wall tissue. These pieces were dissociated separately, after which
cells were counted and rated.
Single Sinoatrial Nodal Cells
Isolation Procedure
Single SAN myocytes were isolated according to the method of
DiFrancesco et al,19 with some modifications as described previously.20 Single cells for immunohistochemical staining as well for
experiments in which the electrical activity of the nodal cells was
studied were obtained following cell isolation procedures 1 and 2
(see below). For the experiments in which cells were isolated after
regional dissection of the SAN and in experiments in which the SAN
was first electrophysiologically mapped, only cell isolation procedure 2 was followed (see below).
Cell Isolation Procedure 1
The heart was excised and mounted on a Langendorff perfusion
system and cleared from blood for 5 minutes with a solution
Verheijck et al
April 28, 1998
1625
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
containing (in mmol/L) NaCl 140, KC1 5.4, CaCl2 1.8, MgCl2 1.0,
HEPES 5.0, and glucose 5.5 (normal Tyrode solution). The solution
was kept at 37°C; pH was adjusted to 7.4 with NaOH. The SAN
region was excised and cut into small strips (width, 0.5 to 1 mm;
length «*2 mm) perpendicular to the crista terminalis. Strips were
allowed to equilibrate for 15 minutes in normal Tyrode solution at
room temperature.
was photographed, and the cell profile was traced on a transparent
sheet. For the reconstruction the successive profiles of the cell were
fed, using a Summa-Sketch Plus MM 1201 (Summagraphics) digitizing tablet, into a Macintosh computer using MacStereology,
written by V.A. Moss.24
Cell Isolation Procedure 2
Strips of nodal tissue were placed in a test tube with an oxygenated
calcium-free Tyrode solution at room temperature containing
(in mmol/L) NaCl 140, KC1 5.4, MgCl2 0.5, KH2PO4 1.2, HEPES
5.0, and glucose 5.5. The pH was adjusted to 6.9 with NaOH. Next,
the strips were transferred to a calcium-free Tyrode solution to which
collagenase B (0.28 U/mL, Boehringer Mannheim), pronase E (0.92
U/mL, Serva), elastase (12.4 U/mL, Serva), and bovine serum
albumin were added (dissociation solution) and incubated at 37°C
for «12 minutes and gently triturated through a pipette with a tip
diameter of 2.0 mm. Dissociation was stopped by transferring the
strips into a modified Kraft-Briihe (KB) solution21 containing
(in mmol/L) KC1 85, K2HPO4 30, MgSO4 5.0, glucose 20, pyruvic
acid 5.0, creatine 5.0, taurine 30, EGTA 0.5, j3-hydroxybutyric acid
5.0, succinic acid 5.0, Na2ATP 2.0, and polyvinylpyrolidone 50 g/L
(pH adjusted with KOH to 6.9). Thereafter, strips were triturated in
KB solution through a pipette (tip diameter 0.8 to 1.2 mm) for 2
minutes. Cells were placed in a recording chamber on the stage of an
inverted microscope (Nikon Diaphot) and superfused with normal
solution (0.6 mL/min). Experiments were performed at 35±0.5°C,
maintained by a translucent heating plate underneath the bottom of
the recording chamber22 and continuously monitored.
Results are presented as mean±SD. For statistical analysis of the
action potential parameters we used the mean values of 15 subsequent action potentials. Statistical significance was determined by
one-way or two-way ANOVA combined with a Student's t test for
paired observations with Bonferroni correction. A probability value
of P<.05 was considered significant.
Membrane Potential Recording and Analysis
Membrane potentials were recorded using the whole-cell technique.23 Electrodes were pulled from borosilicate glass (outer diameter, 1 mm) by use of a vertical one-stage patch-electrode puller and
thereafter heat-polished and backfilled with a micropore-filtered
electrode solution containing (mmol/L): K-gluconate 120; KC1 20;
HEPES 5; MgCl2 5; CaCl2 0.6; Na2ATP 5; cAMP 0.1; and EGTA 5
(pH 7.2). Electrode resistances were between 3 and 5 MO. Series
resistance was compensated for by :=25%. Apart from zeroing the
potential before touching the surface of cell by pipette tip, no
attempts were made to correct for junction potential.
Recordings were made with a custom-build current-voltage clamp
amplifier. Data were sampled directly into a Macintosh Quadra 650
microcomputer (Apple Computer, Inc) with the use of customwritten data acquisition software (J. Zegers, University of Amsterdam, The Netherlands) and stored on digital audio tape using a
digital tape recorder (DTR-1204, Bio-Logic Co), for off-line processing using custom-written data analysis software (A.C.G. van
Ginneken, University of Amsterdam, The Netherlands).
The membrane capacitance, Cm, was defined as the amplitude of a
hyperpolarizing current pulse (20 to 40 pA, 200 ms) divided by the
initial slope of the transmembrane voltage in response to this current
pulse. The current pulse was switched on shortly after the action
potential had reached its maximum diastolic potential. The current
pulse was adjusted to produce a membrane hyperpolarization of ^ 10
mV. Mean membrane capacitance of the nodal cells was 46± 11 pF
(n=21). Atrial cells were paced at 3 Hz with current pulses of 2-ms
duration, 10% to 15% suprathreshold. For nodal cells the following
action potential parameters were measured: cycle length, action
potential overshoot, MDP, diastolic depolarization rate measured
over the first 100 ms starting at the MDP (DDR), maximum upstroke
velocity (dV/dtmax) and duration between 50% depolarization and
50% repolarization (APD50), and duration between 50% depolarization and maximum diastolic potential (APD,00). For atrial cells we
determined cycle length, action potential overshoot, membrane
resting potential (Vrest), dV/dtmax, APD,0, and duration between 50%
depolarization and 90% repolarization (APD90).
Electron Microscopic Reconstruction of a Nodal
Cell In Situ
Tissue sections were prepared as described by De Maziere et al.' 5 A
series of 580 tissue sections of 50 nm was used. Every 15th section
Statistical Analysis
Results
Neurofilament Staining Pattern in the SAN
In the SAN, a gradual change in electrophysiological properties exists when going from the center of the node toward
the atrium. Two conceptual models exist to explain this
gradual transition. In one, the center of the node is surrounded
by an area in which the structure and electrophysiology of the
fibers gradually change from nodal to atrial. In the other
model, the center of the node is surrounded by an area in
which the atrial cell density gradually increases toward the
atrium. Nodal cells can specifically be stained with a neurofilament antibody,25 which enables us to discriminate between
both models.
Therefore, three hearts were serially sectioned in the
transverse plane to study the neurofilament distribution in the
SAN and the surrounding atrium. In these sections we aimed
to discriminate the border between myocardial cells and
nonmyocardial cells. Myocardial cells were identified from
nonmyocardial cells by an antidesmin antibody that reacted
with a phosphorylated isoform of intermediate filament
desmin that is expressed in the working myocardium as well
in the conduction system.26'27 Fig 1, a and b, shows that
antidesmin homogeneously reacts with myocytes from the
atrium and SAN region.
Fig Ic shows the neurofilament expression pattern in a
section adjacent to the one used for the desmin staining.
Neurofilament is not expressed in the atrial myocardium. To
our surprise, the neurofilament expression in the SAN region
was heterogeneous. Fig Id shows an enlargement from the
SAN area and clearly demonstrates that in the SAN, not all of
the desmin-positive cells react with the antineurofilament
antibody (Fig Ib and Id). These cells are either nodal cells
that do not stain with the neurofilament antibody, or, alternatively the neurofilament negative cells are atrial cells.
Neurofilament Distribution Pattern in the
Electrophysiologically Mapped SAN
To correlate the immunohistochemical and electrophysiological characteristics, we studied the neurofilament distribution
pattern of two right atrial preparations of which an electrophysiologically determined maximum upstroke velocity map
was made. From sections cut perpendicular to the crista
terminalis and spaced by 200 jam, the area of neurofilament
expression was determined. Fig 2 shows a composite of the
neurofilament expression (striated area) and the electrophysiologically determined maximum upstroke velocity map
within the SAN. The area in which cells show an upstroke
1626
Atrial and Nodal Cells in the SA Node
SAN
./
AO
Figure 3. Cell types in the sinoatrial node. Photomicrograph of
morphologically different cell types observed in cell suspension
after enzymatic dispersion of the sinoatrial node. Cells are
stained with hematoxylin-eosin. a, Elongated spindle cell; b,
spindle cell; c, spider cell; and d, atrium cell. Bar, 30 /urn.
SAN
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
CT
LA
the SAN region, similar to what was described previously in
guinea pig,14 or nodal cells in which neurofilament is not
expressed. Therefore, we questioned whether enzymatic digestion of the SAN would (1) provide us with atrial cells and
(2) whether single nodal cells would all express the
neurofilament.
Morphology of Single Nodal Cells
• AO
Figure 1. Photomicrographs of the sinoatrial node, a, Overview
of the staining pattern of desmin. The box in a is shown in detail
in b. c, Overview of the staining pattern of neurofilament in an
adjacent section. The box in c is shown in detail in d and shows
that not all cells within the nodal area react with antineurofilament antibody. The bar in a and c is 1 mm and in b and d is
100 jum. LA indicates left atrium; RA, right atrium; SAN, sinoatrial node; CT, crista terminalis; and AO, aorta.
velocity of <15 V/s coincides with the area in which cells
stain with neurofilament antibody. These data clearly demonstrate that the neurofilament expression can be used to
delineate the SAN from the surrounding atrium. The desminpositive (Fig Ib) but neurofilament-negative spots in the SAN
(Fig Id) represent either atrial cells that might invaginate into
Figure 2. Correlation of immunohistochemistry with electrophysiology in a sinoatrial node preparation. The spatial distribution of neurofilament is represented by the dashed area in the
schematic representation of the sinoatrial node. In addition, the
pacemaker area in which cells have an upstroke velocity of <15
V/s is enclosed by an iso-15 V/s line. It is clear that both areas
coincide well. C.T. indicates crista terminalis; I.V.C., inferior vena
cava; and S.V.C., superior vena cava.
Cell morphology was studied on freshly isolated myocytes in
storage (KB) solution and hematoxylin-eosin (H-E)-stained
cells. Under both conditions, most of the cells kept their
elongated appearance, although a few rounded up. From the
nodal area, four types of isolated myocytes could be isolated.
"Elongated spindle cells" have a slender, faintly striated cell
body and contain one or two nuclei. The length of these cells
ranges between 50 and 80 |u,m (Fig 3a). Spindle cells have a
similar shape as elongated spindle cells but are considerably
shorter (30 to 40 /Ltm) with blunted ends and contain 1 nucleus
only (Fig 3b). Spider cells have a varying number of
irregularly shaped branches. After H-E staining, we observed
that the majority of the spider cells were single cells containing one or sometimes two nuclei (Fig 3c). Only a few
appeared to be composed of more than one cell. Rod cells
have clear cross-striations and contain one or two nuclei (Fig
3d). In contrast to the first three cell types, these rod cells
were quiescent but excitable in normal Tyrode solution. For
this reason and because of their shape, which is similar to
myocytes from the working atrial myocardium, these cells are
considered to be atrial cells. Because elongated spindle cells,
spindle cells, and spider cells all exhibit spontaneous activity
in normal solution, they are all considered to be nodal
pacemaking cells.
Denyer and Brown,18 who described the same types of
nodal cells, assumed that spider cells are bundles of two or
more closely packed spindle cells. In contrast, we found that
the vast majority of spider cells have only one nucleus,
indicating that they are single cells. To demonstrate that
spider cells are naturally present in the SAN and not caused
by the isolation procedure, we made a three-dimensional
reconstruction of a cell from the central portion of the SAN
on basis of electron microscopic photographs (Fig 4). This
Verheijck et al
I
April 28, 1998
162!
I I
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Figure 4. Electron microscopic reconstruction of a spider cell.
Reconstruction of a part of a spider cell from the central portion
of the sinoatrial node. Three cell branches develop from the cell
body. The nucleus is shaded in the middle of the cell.
reconstruction shows a part of a spider cell. From the central
part of the cell, which contains the nucleus (striated area),
three branches develop (indicated by arrows). On the basis of
this reconstruction, it can be concluded that cells with a
number of branches exist in the native rabbit SAN as a
separate entity. In conclusion, after enzymatic dissociation of
the rabbit SAN, three morphologically distinct nodal cell
types as well as atrial cells can be distinguished.
Responsiveness of Nodal Cells to
Neurofilament Staining
It was questioned whether the neurofilament negative spots in
the SAN (Fig Id) can be explained fully by the presence of
atrial cells, which are all neurofilament negative. Therefore,
the individual responsiveness of the three nodal cell types to
neurofilament staining was tested.
All cells of the four morphologically different cell types
proved positive for desmin, which indicates that the cytoskeleton is not affected by the isolation procedure. Fig 5 demonstrates the variation of neurofilament staining of the three
nodal cell types. In three experiments, 58% of all tested cells
responded positively to the neurofilament antibody. Although
there was a tendency to a higher responsiveness in the spider
cells (70%) the differences between the three cell types were
not significant. All isolated atrial cells showed a negative
response (not shown).
a
53±11%
b 59±11%
c
70±5%
d
47±8%
e
f
30±6%
41 ±8%
Figure 5. Neurofilament staining of morphologically different
nodal cell types. Neurofilament staining of morphologically different nodal cell types, isolated from one sinoatrial node preparation. Cells can stain positive (a, b, and c) or negative (d, e,
and f) for antineurofilament antibody, a and d, Elongated spindle
cells; b and e, spindle cells; c and f, spider cells. Bar, 30 /xm.
In conclusion, all nodal cell types show an inhomogeneous
response to neurofilament staining. Therefore, the neurofilament-negative spots in the SAN (Fig Id) can be caused by
either atrial cells or nodal cells that do not stain with
neurofilament antibody. The large percentage of neurofilament-positive cells in all three nodal cell types still renders
this technique suitable for marking the borders of the SAN in
the light microscopic preparation but is not suitable to
discriminate between atrial and nodal myocardium at the
cellular level.
Distribution of Nodal Cell Types Within the SAN
In the experiments described thus far, single cells were
isolated from the whole SAN region. However, in the intact
SAN, a gradual transition in action potential configuration
exists when going from the central area toward the atrium. It
was questioned whether this gradual transition can be explained by a gradient of a specific cell type. Therefore, we
established the distribution of the four morphologically different cell types within the more central area of the SAN.
In six experiments, the nodal area was divided into three
zones of 2.5 mm2 perpendicular to the crista terminalis, that
1628
Atrial and Nodal Cells in the SA Node
S.V.C.
I
Figure 6. Relative contribution of the four
different cell types in the sinoatrial node.
Schematic representation of the sinoatrial
node from which three adjacent areas
were excised and dissociated separately.
The relative contribution of the four different cell types in these areas is shown. C.T.
indicates crista terminalis; I.V.C., inferior
vena cava; and S.V.C., superior vena cava.
| elongated spindle
$y$\ spindle
EZ3 spider
& atrial
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
is, crista terminalis area, central area, and septal area (Fig 6).
Zones were excised and dissociated separately. After plating,
the number of each cell type per zone was counted and
expressed as a fraction of the total. In every experiment,
=500 cells were counted. In the three zones, all cell types
were present (Fig 6). Cell types were equally distributed in
the central area. In the crista terminalis area, the majority of
cells (63 ±18%) were atrial cells, whereas the elongated
spindle cells were the predominant type of nodal cells (21%).
In the septal area of the SAN, 88 ±19% of the cells were of
atrial origin. In this area, nodal cells types were distributed in
about equal quantities. In the central area, 41 ±10% were
atrial cells. The density of atrial cells was not significantly
between the central an crista terminalis area, whereas the
difference of both areas with the septal area was significant
(P<.0l after Bonferroni correction).
In summary, none of the cell types was exclusively present
in any specific area of the SAN. In the area bordering the
crista terminalis, the elongated spindle cells were the most
prominent of the nodal cells. Within the other areas no
difference in distribution of nodal cells was found.
To relate the distribution of cell types to the site of the
dominant pacemaker, the activation pattern of the SAN was
mapped electrophysiologically before cell dissociation in five
experiments. Cells were isolated from a piece of tissue of
< 1 mm2 around the site of the earliest discharge and from a
piece of 2 to 3 mm2 located directly superior to this area. Even
in the very restricted area around the dominant pacemaker
area, 22±15% of the cells were atrial cells. Between the
dominant and latent pacemaker area no significant difference
in distribution of any cell type was found (Table 1).
Electrical Activity of Morphologically Different
Cell Types
In the next series of experiments we investigated whether
morphologically different cell types could be distinguished
from each other electrophysiologically. In these experiments
the whole SAN was used for dissociation without prior
mapping.
Fig 7 shows that a large variability exists with respect to
pacemaker potency between isolated cells. Table 2 summarizes action potential parameters of the different cell types.
Spindle cells have the smallest Cm compared with elongated
spindle cells (P<.01) and spider cells (P<.01). The variability of the action potential and diastolic depolarization within
one morphological cell group is much larger than the variability between the groups. In other words, cell morphology
is not a predictor of pacemaker characteristics. All atrial cells
were quiescent but excitable. Fig 7D shows a typical electrophysiological recording of a paced (3 Hz) atrial cell. Table 2
also summarizes electrophysiological parameters of six atrial
cells. It is clear that the action potential configuration differs
drastically from that of all morphologically different nodal
cell types.
Discussion
Immunohistochemical Markers in the SAN
With the aid of immunohistochemistry it has been demonstrated that specific proteins are expressed in the conduction
system of the heart of various species.28"30 In bovine heart it
has been demonstrated that the SAN could be distinguished
from the atrial myocardium by a monoclonal antibody (4456E10) raised against the bovine conduction system.31 In the
SAN of the guinea pig we found that the impulse originates
in an a-smooth muscle actin-positive, virtually connexinnegative region. In the SAN of the adult rabbit,25 a specific
expression of an antineurofilament antibody was described.
We demonstrated that the antineurofilament monoclonal anTABLE 1. Cell Distribution After Electrophysiological Mapping
in the Dominant and the Latent Pacemaker Areas (n=5)
Cell Type
Dominant, %
Mean±SD
Latent, %
Mean±SD
25±8
Elongated spindle
27±11
Spindle
21±10
18±9
Spider
30 ±23
22±15
28±15
Atrial
29±10
Verheijck et al
Elongated spindle cells
April 28, 1998
1629
Spindle cells
mV
0.2
0.3
0.4
0.2
time (s)
0.3
0.4
time (s)
Atrial cell
40
mV
mV
20-
A
\
0-
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
-20-40-60-
0.2
0.3
0.4
|
V
v
0
0.1
0.2
time (s)
0.3
0.4
0.5
0.6
time(s)
Figure 7. Action potentials of morphologically different nodal cells. A, Action potential recordings of three different elongated spindle
cells. B, Action potential recordings of three different spindle cells. C, Action potential recordings of three different spider cells. D,
Action potential recording of an atrial cell paced at 3 Hz.
tibody reacts with isolated nodal myocytes but not with atrial
cells. However, this marker did not identify all isolated nodal
myocytes. It is unlikely that the neurofilament-negative nodal
cells were negative because of an isolation artifact because all
nodal cells were desmin positive. These results demonstrate
that the expression of the antineurofilament antibody cannot
be used to distinguish individual nodal cells from atrial
myocytes. In the intact SAN, the neurofilament-positive area
coincides with the electrophysiologically defined pacemaker
area, where cells had an upstroke velocity of ^ 15 V/s (Fig 2).
This, together with the large percentage of neurofilament-
positive cells (58%), renders this technique very suitable for
marking the borders of the SAN.
Morphologically Different Nodal Cell Types
We described three morphologically different nodal cell
types, that is, elongated spindle, spindle, and spider cells as
they appear after enzymatic dissociation. According to Denyer and Brown,18 spider cells are actually small bundles of
two or more closely packed spindle cells. However, Rossi et
al32 demonstrated in human SAN the presence of cells with
polypoid branches. After H-E and immunohistochemical
TABLE 2. Action Potential Parameters of Morphologically Different Nodal Cells and Atrial Cells
Elongated Spindle
Spindle
Spider
(n=7)
Mean±SD
(n=8)
Mean±SD
(n=6)
Mean±SD
Cycle length, ms
441 ±206
308±85
288 ±80
Cycle length, ms
APDso, ms
111±22
102±16
111+36
APDso, ms
32±8
APD10o, ms
173±29
161 ±22
176±54
APDcx), ms
54±9
DDR, mV/s
106±54
134±42
178±63
(JV/dU, V/s
6.4±2.9
7.3±5.7
4.8±3.5
MDP, mV
-53±5
-49±6
-46±6
Atrial
(n=6)
Mean±SD
dV/dW, V/s
Vres,, mV
333 (paced)
160±20
-74±2
Overshoot, mV
33±7
32±12
27±9
Overshoot, mV
22±4
Q..PF
53±3f
32±5
54 ±4*
Cm, PF
95±11
APD indicates action potential duration; DDR, depolarization rate measured over the first 100 ms starting at the MDP; dV/dtmax, maximum upstroke velocity; MDP,
maximum diastolic potential; Cm, membrane capacitance; and Vrest, membrane resting potential.
'Statistical difference between spindle and spider cells (P<.02, with Bonferroni correction).
fStatistical difference between elongated spindle and spindle cells (f<.02, with Bonferroni correction).
1630
Atrial and Nodal Cells in the SA Node
CRISTA TERMINALIS
CRISTA TERMINALIS
CENTRAL PART OF SA NODE
CENTRAL PART OF SA NODE
Figure 8. Two conceptual models for the sinoatrial transition.
Left, Gradual transition from typical nodal to atrial cells separated by an area of transitional cells. Right, Presence of two different cell types: nodal (white) and atrial (black). The transition is
formed by a zone in which the density of atrial cells gradually
increases from the typical nodal area toward the crista
terminalis.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
staining, we observed that only few apparent spider cells were
composed of more than one cell (<1%). The majority were
single cells that contained one or two nuclei. A threedimensional reconstruction on the basis of electron microscopic photographs of a cell from the central portion of the
intact SAN (Fig 4) demonstrates that the node contains cells
with numerous branches that may run over a considerable
distance between other cells. Apparently spider cells exist in
the intact rabbit SAN and are not to be considered as an
isolation artifact.
The Transitional Zone
A gradual transition from nodal to atrial tissue has been
described in the rabbit SAN.5 In the present study we found
that elongated spindle cells are more abundant in the area
adjacent to the crista terminalis than spindle and spider cells,
which suggests that they are the previously described transitional fibers. Suggestions for an intermediate class of cells in
the rabbit were put forward, both on morphological10 and
functional33'34 grounds. Fig 8 (left) shows a scheme of a
central node composed of typical nodal cells surrounded by a
rim of transitional cells, as was originally suggested for the
rabbit SAN.5 This scheme did not prove applicable to the
feline, canine, or human SAN, in which typical nodal cells
were observed intermingled with and adjacent to atrial cells
(Fig 8, right).4'35 However, in the rabbit we also found
evidence for a less smooth transition.15 Previous studies have
suggested that typical (central) nodal cells are sensitive to
caseum but not to TTX.33 However, our group previously
challenged this concept: It was found that any nodal cell, if
sufficiently hyperpolarized, will open Na + channels.36 Preliminary results indicated larger If current density in larger single
nodal cells.37 It was hypothesized that large nodal cells are
located more to the border of the intact SAN. These cells
would be more likely to overcome the hyperpolarizing load of
adjacent atrium more effectively. We found that the area
adjacent to the crista terminalis has indeed more elongated
spindle cells than the central and septal area (Fig 6). On the
other hand, elongated spindle cells are not the exclusive cell
type of the crista terminalis area. In contrast with previous
findings,37 we recently found the same current density for If,
IK, and ICa. L in cells with different dimensions.38 In addition,
in the current we demonstrated that the action potential
configuration of small and large isolated nodal cells is similar
(Fig 7, Table 2). Therefore, cell size alone cannot explain
regional differences in action potential configuration, as
found in the intact SAN. Unfortunately, whole-cell current
clamp experiments on cells obtained after electrophysiological mapping of the SAN failed, because no calcium resistant
cells could be obtained after the long time span of the whole
procedure. Consequently, we were not able to relate regional
differences in action potential configuration as found in the
intact SAN to differences in action potential configuration
found in single cells of known location within the SAN.
Geometric Considerations
A completely different explanation for the functional differentiation within the intact SAN lies in its geometric relation
to the atrium. We observed in a restricted area (1 mm2)
around the electrophysiologically mapped dominant pacemaker area that 22% of the isolated cells were atrial cells. In
the larger central portion of the SAN (2 to 3 mm2), 41% of the
cells were from atrial origin, and when a piece of nodal tissue
bordering the crista terminalis was isolated, even a percentage
of atrial cells as high as 63% was obtained. These observations show that the number of atrial cells increases progressively from the dominant pacemaker site toward the crista
terminalis. These atrial cells probably originate from invaginating atrial myocardium, similar to what is suggested in
bovine31 and guinea pig heart.14 These data agree with the
close opposition of nodal and atrial cells in the center of the
rabbit SAN, as we demonstrated previously in an electron
microscopic study.15 Electrophysiological evidence for such
inhomogeneity was found by measurements of space constants in different parts of the rabbit SAN. These constants
appeared to vary considerably over very short distances.12
Watanabe et al39 performed experiments in which they modulated the SAN pacemaker activity by coupling SAN cells,
through an electronic artificial coupling resistance, to a
resistance-capacitance circuit resembling an atrial cell. Increasing the number of "atrial" cells coupled to the SAN cell
caused a gradual transition of the nodal to the atrial type of
action potential.
Data on specific connexins are lacking in the rabbit at the
level of contacts between atrial cells and nodal cells. Only in
guinea pig labeling with connexin43 antibodies between
atrial and nodal cells was found.14 Data on connexins between
rabbit nodal cells, however, have been shown to consist of at
least of connexin43.40 In other species, connexin40,41 connexin43,42 and connexin4541 have been described, but there is
no certainty whether these connexins are located between
individual nodal cells or that they may play a role in the
connection between nodal and atrial cells.
We hypothesize that in the intact SAN the gradual increase
in density of atrial cells from the dominant area toward the
crista terminalis (Fig 8, right) causes a gradual increase of
atrial electrotonic influence that may be an important cause of
the gradual transition of the nodal to the atrial type of action
potential. On the other hand, the variety in action potentials
recorded in isolated nodal cells (Fig 7) does not rule out the
possibility that differences in cellular properties of various
Verheijck et al
nodal cell types also contribute to the gradual transition of
action potential configuration from the center of the node to
the surrounding atrial myocardium.
Acknowledgments
This work was supported by a grant from the Netherlands Organization for Scientific Research (grant 900-516-093). The authors
thank M.J.A. van Kempen (Department of Medical Physiology and
Sport Medicine,
University
of Utrecht) for
her
stimulating discussions.
References
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
1. Bouman LN, Jongsma HJ. Structure and function of the sino-atrial node:
a review. Eur Heart J. 1986;7:94-104.
2. Tranum-Jensen J. The fine structure of the sinus node: a survey. In:
Bonke FIM, ed. The Sinus Node. The Hague, Netherlands: Martinus
Nijhoff; 1978:149-165.
3. Brown HF. Electrophysiology of the sinoatrial node. Physiol Rev. 1982;
62:505-530.
4. Opthof T. The mammalian sinoatrial node. Cardiovasc Drugs Ther.
1988; 1:573-597.
5. Bleeker WK, Mackaay AJC, Masson-Pevet MA, Bouman LN, Becker
AE. Functional and morphological organization of the rabbit sinus node.
Circ Res. 1980;46:11-22.
6. Opthof T, de Jonge B, Mackaay AJC, Bleeker WK, Masson-Pevet MA,
Jongsma HJ, Bouman LN. Functional and morphological organization of
the guinea-pig sinoatrial node compared with the rabbit sinoatrial node. J
Hoi Cell Cardiol. 1985;17:549-564.
7. Opthof T, de Jonge B, Masson-Pevet MA, Jongsma HJ, Bouman LN.
Functional and morphological organization of the cat sinoatrial node. J
Mol Cell Cardiol. 1986;18:1015-1031.
8. Opthof T, de Jonge B, Jongsma HJ, Bouman LN. Functional morphology
of the pig sinoatrial node. J Mol Cell Cardiol. 1987;19:1221-1236.
9. Alings AMW, Abbas RF, de Jonge B, Bouman LN. Structure and
function of the simian sinoatrial node (Macaca fascicularis). J Mol Cell
Cardiol. 1990;22:1453-1466.
10. Masson-Pevet MA, Bleeker WK, Besselsen E, Treytel BW, Jongsma HJ,
Bouman LN. Pacemaker cell types in the rabbit sinus node: a correlative
ultrastructural and electrophysiological study. J Mol Cell Cardiol. 1984;
16:53-63.
11. Masson-PeVet MA, Bleeker WK, Gros D. The plasma membrane of
leading pacemaker cells in the rabbit sinus node. Circ Res. 1979;45:
621-629.
12. Bouman LN, Duivenvoorden JJ, Bukauskas FF, Jongsma HJ. Anisotropy
of electrotonus in the sinoatrial node of the rabbit heart. J Mol Cell
Cardiol. 1989;21:407-418.
13. Joyner RW, Capelle FJL. Propagation through electrically coupled cells:
how a small SA node drives a large atrium. Biophys J. 1986;50:
1157-1164.
14. ten Velde I, de Jonge B, Verheijck EE, van Kempen MJ, Analbers L, Gros
D, Jongsma HJ. Spatial distribution of connexin 43, the major cardiac gap
junction protein, visualizes the cellular network for impulse propagation
from sinoatrial node to atrium. Circ Res. 1995;76:802-811.
15. De Maziere AMGL, van Ginneken ACG, Wilders R, Jongsma HJ,
Bouman LN. Spatial and functional relationship between myocytes and
fibroblasts in the rabbit sinoatrial node. / Mol Cell Cardiol. 1992;24:
567-578.
16. Sternberger LA. Immunohistochemistry. New York, NY: John Wiley &
Sons; 1986:90-209.
17. Lee BB, Mandl G, Stean JPB. Microelectrode tip marking in nervous
tissue: a new dye method. Electroencephalogr Clin Neurobiol. 1969;27:
610-613.
18. Denyer JC, Brown HF. Rabbit sino-atrial node cells: isolation and electrophysiological properties. J Physiol (Land). 1990;428:405-424.
19. DiFrancesco D, Ferroni A, Mazzanti M, Tromba C. Properties of the
hyperpolarizing-activated current (if) in cells isolated from the rabbit
sino-atrial node. J Physiol (Lond). 1986;377:61-88.
April 28,1998
1631
20. Verheijck EE, van Ginneken ACG, Bourier J, Bouman LN. Effects of
delayed rectifier current blockade by E-4031 on impulse generation in
single sinoatrial nodal myocytes of the rabbit. Circ Res. 1995;76:
607-617.
21. Isenberg G, Klockner U. Calcium tolerant ventricular myocytes prepared
by preincubation in a 'KB medium.' Pflugers Arch. 1982;395:6-18.
22. van Ginneken ACG, Giles W. Voltage clamp measurements of the hyperpolarization activated inward current (If) in single cells from rabbit
sino-atrial node. J Physiol (Lond). 1991;434:57-83.
23. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved
patch-clamp techniques for high-resolution current recording from cells
and cell-free membrane patches. Pflugers Arch. 1981;391:85-100.
24. Moss VA. The computation of 3-dimensional morphology from serial
sections. Eur J Cell Biol. 1989;48(suppl 25):61-64.
25. Gorza L, Vitadello M. Distribution of conduction system fibers in the
developing and adult rabbit heart revealed by an antineurofilament
antibody. Circ Res. 1989;65:360-369.
26. Kjorell U, Thornell L-E, Lehto V-P, Virtanen I, Whalen RG. A comparative analysis of intermediate filament proteins in bovine heart
Purkinje fibres and gastric smooth muscle. Eur J Cell Biol. 1987;44:
68-78.
27. Thornell L-E, Price MG. The cytoskeleton in muscle cells in relation to
function. Biochem Soc Trans. 1991;19:1116-1121.
28. Skepper JN. An immunocytochemical study of the sinuatrial node and
atrioventricular conduction system of the rat for atrial natriuretic peptide
distribution. Histochem J. 1989;21:72-78.
29. Zahler R, Brines M, Kashgarian M, Benz EJ, Gilmore-Hebert M. The
cardiac conduction system in the rat expresses the alfa2 and alfa3
isoforms of the Na + , K+-ATPase. Proc Natl Acad Sci USA. 1992;89:
88-103.
30. Thornell L-E, Erikson A, Johansson B, Kjorell U. Intermediate filaments
and associated proteins in heart Purkinje fibers: a membrane-myofibril
anchored cytoskeletal system. Ann NY Acad Sci. 1985;245:273-279.
31. Oosthoek PW, Viragh S, Mayen AEM, van Kempen MJ, Lamers WH,
Moorman AFM. Immunohistochemical delineation of the conduction
system, I: the sinoatrial node. Circ Res. 1993;73:473-481.
32. Rossi L, Matturri L, Lotto A. Cardiac Conduction Blocks and Pacemaking: An Anatomical Collar Atlas. Brescia, Italy: Class International;
1988. ISBN 88-85091-00-8.
33. Kreitner D. Electrophysiological study of the two main pacemaker mechanisms in the rabbit sinus node. Cardiovasc Res. 1985;19:304-318.
34. Nathan RD. Two electrophysiologically distinct types of cultured
pacemaker cells from rabbit sino-atrial node. Am J Physiol. 1986;250:
H325-H329.
35. Opthof T, de Jonge B, Jongsma HJ, Bouman LN. Functional morphology
of the mammalian sinuatrial node. Eur Heart J. 1987;8:1249-1259.
36. Oei HI, van Ginneken ACG, Jongsma HJ, Bouman LN. Mechanisms of
impulse generation in isolated cells from the rabbit sinoatrial node. J Mol
Cell Cardiol. 1989;21:1137-1149.
37. Honjo H, Boyett MR. Correlation between action potential parameters
and the size of single sino-atrial node cells isolated from the rabbit heart.
J Physiol (Lond). 1992;452:128P. Abstract.
38. Wilders R, Verheijck EE, Kumar R, Goolsby WN, van Ginneken ACG,
Joyner RW. Model clamp and its application to synchronization of rabbit
sinoatrial node cells. Am J Physiol. 1996;271:H2168-H2182.
39. Watanabe E, Honjo H, Anno T, Boyett MR, Kodama I, Toyama J.
Modulation of pacemaker activity of sinoatrial node cells by electrical
load imposed by an atrial cell model. Am J Physiol. 1995;269:
H1735-H1742.
40. Anumonwo JMB, Wang H-Z, Trabka-Janik E, Dunham B, Veenstra RD.
Gap junctional channels in adult mammalian sinus nodal cells: immunolocalization and electrophysiology. Circ Res. 1992;71:229-239.
41. Davis LM, Kanter HL, Beyer EC, Saffitz JE. Distinct gap junction protein
phenotypes in cardiac tissues with disparate conduction properties. J Am
Coll Cardiol. 1994;24:1124-1132.
42. Trabka-Janik E, Coombs W, Lemanski LF, Delmar M, Jalife J. Immunohistochemical localization of gap junction protein channels in hamster
sinoatrial node in correlation with electrophysiologic mapping of the
pacemaker region. / Cardiovasc Electrophysiol. 1993;5:125-137.
Distribution of Atrial and Nodal Cells Within the Rabbit Sinoatrial Node: Models of
Sinoatrial Transition
E. Etienne Verheijck, Andy Wessels, Antoni C. G. van Ginneken, Jan Bourier, Marry W. M.
Markman, Jacqueline L. M. Vermeulen, Jacques M. T. de Bakker, Wouter H. Lamers,
Tobias Opthof and Lennart N. Bouman
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Circulation. 1998;97:1623-1631
doi: 10.1161/01.CIR.97.16.1623
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1998 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/97/16/1623
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answer
document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/