Cardiovascular Research 72 (2006) 271 – 281

www.elsevier.com/locate/cardiores

Extended atrial conduction system characterised by the expression

of the HCN4 channel and connexin45

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

Mitsuru Yamamoto a , Halina Dobrzynski b , James Tellez b , Ryoko Niwa a , Rudi Billeter c ,
Haruo Honjo a , Itsuo Kodama a , Mark R. Boyett b,⁎
a
Research Institute of Environmental Medicine, Nagoya University, Nagoya 464-8601, Japan
b
Cardiovascular Research Group, School of Medicine, University of Manchester, Core Technology Facility, 46 Grafton Street, Manchester M13 9NT, UK
c
Institute of Membrane and Systems Biology, University of Leeds, Leeds, LS2 9JT, UK
Received 8 June 2006; received in revised form 20 July 2006; accepted 25 July 2006
Available online 2 August 2006
Time for primary review 18 days

Abstract

Objective: In the heart, there are multiple supraventricular pacemakers involved in normal pacemaking as well as arrhythmias and the
objective was to determine the distribution of HCN4 (major isoform underlying the pacemaker current, If) in the atria.
Methods: In the atria of the rat, the localisation of HCN4 and connexins was determined using immunohistochemistry, and electrical activity
was recorded using extracellular electrodes.
Results: As expected, HCN4 and Cx45 (but not Cx43) were expressed in the sinoatrial node extending from the superior vena cava down the
crista terminalis. The same pattern of expression of HCN4 and connexins was observed in a novel tract of nodal-like cells extending from the
superior vena cava down the interatrial groove. Although the sinoatrial node was usually the leading pacemaker site, the novel tract of HCN4-
expressing cells was capable of pacemaking and could act as the leading pacemaker site; there was evidence of a hierarchy of pacemakers.
The same pattern of expression of HCN4 and connexins was also observed in the atrioventricular ring bundle (including the atrioventricular
node) encircling the tricuspid valve, but not in the atrioventricular ring bundle encircling the mitral valve. HCN4 was not expressed in the
pulmonary veins.
Conclusions: The widespread distribution of HCN4 can explain the widespread location of the leading pacemaker site during sinus rhythm,
the extensive region of tissue that has to be ablated to stop sinus rhythm, and the widespread distribution of ectopic foci responsible for atrial
tachycardia.
© 2006 European Society of Cardiology. Published by Elsevier B.V.

1. Introduction pacemaker activity of many of the atrial pacemakers

[7,15,16]. Four cDNAs encoding If-type currents have
In the heart, there are multiple supraventricular tissues been cloned from vertebrates (HCN1-4); HCN4 has been
capable of showing pacemaker activity including the reported to be a major isoform in the cardiac conduction
sinoatrial node (SAN), atrioventricular node (AVN), system (SAN, AVN, Purkinje fibres) [15,17,18]. The aim of
superior vena cava [1–4], AV ring bundles [5–12] and the present study was to determine the distribution of
pulmonary veins [13]. These tissues constitute the normal HCN4 in the atria. A characteristic feature of the SAN and
pacemaker (SAN), back-up pacemakers (e.g. AVN) and AVN is that there is little or no expression of connexin43
ectopic foci (e.g. pulmonary veins) responsible for atrial (Cx43; medium conductance gap junction protein), which is
tachycardia and fibrillation. The hyperpolarization-activated abundantly expressed in the remainder of the heart, and
current, If, plays a major role in pacemaker activity in the instead there is expression of connexin45 (Cx45; low
heart [14] and If is known to play a significant role in the conductance gap junction protein). This results in poor
electrical coupling in the nodal tissues and this is
⁎ Corresponding author. Tel.: +44 161 275 1192; fax: +44 161 275 1233. considered essential to protect the pacemaking tissue from
E-mail address: [email protected] (M.R. Boyett). the inhibitory electrotonic influence of the surrounding non-
0008-6363/$ - see front matter © 2006 European Society of Cardiology. Published by Elsevier B.V.
doi:10.1016/j.cardiores.2006.07.026
272 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281

pacemaking atrial muscle [19]. The distribution of Cx43 was quickly removed. From the rat hearts, the atria (or AV
and Cx45 in the atria was also determined in the present junction) were isolated. Some of the rat preparations were
study. As well as in the SAN and AVN, HCN4 and Cx45 frozen and cryosectioned; the sections were subsequently
(but not Cx43) were expressed in tracts of nodal-like cells stained with Masson's trichrome or immunolabelled for
in the interatrial groove and around the tricuspid valve, but HCN4, Cx43 and Cx45 using standard techniques [20].
surprisingly not in the pulmonary veins. Alternatively, extracellular electrodes were used to measure the
activation sequence and spontaneous beating rate of the rat atria
2. Methods [21]. One rabbit heart was frozen and cryosectioned and the
sections stained with Masson's trichrome. From seven rabbit
Wistar rats (male; 250–300 g) and New Zealand white hearts, total RNA was isolated from the proximal and distal
rabbits (male; 1.9–2 kg; used for histology and quantitative pulmonary veins and the right atrium; cDNA was generated

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

PCR) were sacrificed humanely according to the United from the total RNA. The relative abundance of HCN1 and
Kingdom Animals (Scientific Procedures) Act, 1986; in addi- HCN4 cDNA (a measure of the abundance of HCN1 and
tion, the investigation conformed with the Guide for the Care HCN4 mRNA in the original samples) was then measured with
and Use of Laboratory Animals published by the US National quantitative PCR. From three rabbit hearts, the right superior
Institutes of Health (NIH Publication No. 85-23, revised 1996). pulmonary vein was isolated and the membrane potential
For the recording of membrane potential from the pulmonary recorded with glass microelectrodes. Data are presented as
veins, Japanese white rabbits (of either sex; 1.5–2.0 kg) were means ± SEM. A one- or two-way ANOVA completed by a
anesthetized with pentobarbital (30–40 mg/kg, iv); animal Bonferroni t-test was used to test differences; a difference was
procedures were approved by the Animal Care and Use considered significant if P b 0.05. Further details of the
Committee, Research Institute of Environmental Medicine, techniques are available in the Electronic Supplementary
Nagoya University. Following death of the animals, the heart Material.

Fig. 1. Expression of HCN4 and connexins at base of superior vena cava. A, Masson's trichrome stained section of junction of superior vena cava with right
atrium showing nodal cells (lightly stained; demarcated by dashed lines) surrounded by atrial cells (darkly stained). Dashed lines in inset show location of
sections through right atrium in Figs. 1, 2 and 3; red bar highlights section in this figure. B, HCN4 labelling in region identified by box in A. Dashed lines
highlight area strongly expressing HCN4 (dull green labelling outside this region may be background labelling only). C, high-magnification image of HCN4
labelling in box in B. D, high-magnification image of Cx45 (red) and Cx43 (green) labelling in box in B. Scale bars, 200 μm (A, B) and 50 μm (C, D). In this and
all figures: AS, atrial septum; AVN, atrioventricular node; CS, coronary sinus; CT, crista terminalis; D, dorsal; IVC, inferior vena cava; L, left; LA, left atrium;
LAA, left atrial appendage; LSARB, left sinoatrial ring bundle; LV, left ventricle; MV, mitral valve; PV-P and PV-D, proximal and distal parts of pulmonary vein;
R, right; RA, right atrium; RAA, right atrial appendage; RSARB, right sinoatrial ring bundle; RS-PV, right superior pulmonary vein; RV, right ventricle; SAN,
sinoatrial node; SEP, interatrial septum; SVC, superior vena cava; TV, tricuspid valve; V, ventral; VS, ventricular septum.
 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281 273

3. Results HCN4-positive/Cx43-negative/Cx45-positive cells are SAN

cells.
3.1. Superior vena cava and rear wall of right atrium Fig. 2 shows sections at a more inferior level (Fig. 2A, inset).
At this level, there were two groups of nodal (or nodal-like) cells:
Isolated atrial preparations were serially cryosectioned in a a large group of nodal cells was located alongside the crista
plane perpendicular to the axis from the superior to the inferior terminalis (this group constitutes the SAN [20]) and a smaller
vena cava. Representative results from one heart are shown in group of nodal-like cells was located on the epicardial surface
Figs. 1–3; similar results were obtained from four hearts. Fig. 1 next to the interatrial groove. Fig. 2A shows that both groups
shows adjacent sections at the level of the junction of the (tissue within dashed lines in the main panel of Fig. 2A)
superior vena cava with the right atrium (see inset in Fig. 1A). appeared pale when stained with Masson's trichrome (unlike
The dorsal region of the junction showed strong red staining surrounding atrial muscle). Fig. 2F shows that both groups

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

with Masson's trichrome, but in the ventral region, there were (tissue within dashed lines in Fig. 2F) showed strong expression
cells with paler staining (Fig. 1A). The adjacent section was of HCN4. Details of the HCN4 labelling are shown in Fig. 2D,E.
labelled for HCN4 and the cells showing strong Masson's Both groups also showed expression of Cx45 and no expression
trichrome staining showed no labelling of HCN4, whereas the of Cx43. As an example, Fig. 2C shows labelling of Cx43 and
cells showing pale Masson's trichrome-staining showed strong Cx45 in the nodal-like cells next to the interatrial groove — in
HCN4 labelling (see tissue demarcated by dashed lines in Fig. these cells, note the close correspondence between the labelling
1A,B). At high magnification, labelling of HCN4 was observed of HCN4 and Cx45 (Fig. 2B,C; see cells highlighted by arrows).
at the cell membrane and, in addition, there was substantial Fig. 3 shows sections at a more inferior level — at the level of
intracellular labelling (perhaps protein just translated or being the right superior pulmonary vein (Fig. 3A, inset). Fig. 3A shows
trafficked) (Fig. 1C). Another adjacent section was double part of the rear wall of the right atrium and the neighbouring right
labelled for Cx43 and Cx45 — the cells showing strong superior pulmonary vein (RS-PV) — the crista terminalis is not
Masson's trichrome staining showed labelling of Cx43, but not included. The small group of HCN4-positive/Cx43-negative/
of Cx45 (data not shown), whereas the cells showing pale Cx45-positive nodal-like cells next to the interatrial groove was
Masson's trichrome staining showed labelling (punctate) of still present at this level (Fig. 3B,C). At this level, a group of
Cx45, but not of Cx43 (Fig. 1D). The strong Masson's SAN cells was also still present next to the crista terminalis (data
trichrome-stained/HCN4-negative/Cx43-positive/Cx45-nega- not shown). The results presented in Figs. 1–3 show that nodal/
tive cells are atrial cells and the pale Masson's trichrome-stained/ nodal-like tissue forms an inverted U shape centred on the

Fig. 2. Expression of HCN4 and connexins below base of superior vena cava. A, Masson's trichrome stained section. Dashed circles highlight two pale
Masson's-trichrome stained nodal/nodal-like regions. Inset shows location of section in right atrium (red bar). B, C, high-magnification images of labelling of
HCN4 (B) and Cx45 (red) and Cx43 (green) (C) from tract of nodal-like cells in interatrial groove (upper box in A). Note that myocytes highlighted by arrows
expressed both HCN4 (B) and Cx45 (C). D, E, high-magnification images of labelling of HCN4 from tract of nodal-like cells in interatrial groove (D) and SAN
(E) (from boxed regions in F). F, low-magnification image of HCN4 labelling in another preparation. Level of section was similar to that of A. Dashed circles
highlight two regions strongly expressing HCN4. White line marks cavity of superior vena cava. Scale bars, 200 μm (A, F) and 50 μm (B–E).
274 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

Fig. 3. Expression of HCN4 and connexins at level of right superior pulmonary vein. A, Masson's trichrome stained section including right superior pulmonary
vein. Inset shows location of section in right atrium (red bar). B, C, high-magnification images of labelling of HCN4 (B) and Cx45 (red) and Cx43 (green) (C)
from tract of nodal-like cells in interatrial groove (right box in A). Note that group of myocytes highlighted by arrows expressed both HCN4 (B) and Cx45 (C). D,
E, high-magnification images of labelling of HCN4 (D) and Cx45 (red) and Cx43 (green) (E) from proximal right superior pulmonary vein (left box in A). Scale
bars, 200 μm (A) and 50 μm (B–E).

superior vena cava: one arm of the U forms the SAN and the shifted from the SAN adjacent to the crista terminalis to the
other forms a tract of nodal-like cells extending down the junction between the superior vena cava and the interatrial
interatrial groove. groove (ISO site) — Fig. 4A,B shows examples. Of these four
preparations, in the presence of ISO, two exhibited a permanent
3.2. Functional importance of tract of HCN4-expressing pacemaker shift to the ISO site and in the other two the leading
nodal-like cells in interatrial groove pacemaker site shifted back and forth between the SAN and the
ISO site. In one of five preparations, the leading pacemaker site
Under control conditions, the location of the leading pace- shifted from the interatrial groove to the ISO site (Fig. 4C). This
maker site was investigated in 18 preparations. Fig. 4 shows distribution of leading pacemaker sites (under control conditions
representative activation maps (epicardial views shown). In 11 and in presence of ISO) corresponds to the distribution of HCN4
of 18 preparations, the leading pacemaker site was located in the described above.
SAN next to the crista terminalis and Fig. 4A,B shows two such We compared the pacemaker activity of the SAN and the
examples (see left hand activation maps obtained under control tract of nodal-like cells in the interatrial groove. Preparations
conditions) — the leading pacemaker site (asterisk) was located were divided into two as shown in Fig. 5 (inset). One part
on or next to the crista terminalis (grey shaded bundle) and the included the SAN and the other part included the interatrial
right sinoatrial ring bundle (RSARB). However, in the groove (Fig. 5, inset; endocardial view of tissue shown).
remaining seven preparations, the leading pacemaker site was Under control conditions, after the division of the tissue into
located in the interatrial groove and Fig. 4C (left) shows one two, the preparation containing the interatrial groove showed
example — the leading pacemaker site (asterisk) is next to the spontaneous beating for a brief period (0.5–5 min), whereas
left sinoatrial ring bundle (LSARB). Next, we investigated the the preparation containing the SAN showed continuous
effect of 0.05 μM isoproterenol (ISO) on the leading pacemaker beating. The cycle length of the preparation including the
site. In four of five preparations, the leading pacemaker site SAN was 362 ± 26 ms and it was not changed by the separation
 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281 275

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

Fig. 4. Activation sequence in absence and presence of isoprenaline. A–C, activation sequence from three representative cases under control conditions (left) and
in presence of 0.05 μM isoprenaline (ISO; right). Asterisks indicate position of leading pacemaker site and isochrones (dashed lines; at 2.5 ms intervals) show
spread of activation.

of the tissue into two. In the presence of 0.05 μM ISO (Fig. 5A, cycle length was prolonged by 23 ± 7% to 204 ± 18 ms
left), the preparation including the interatrial groove started (P b 0.05 versus control) (Fig. 5A,B). This suggests that
beating spontaneously and the cycle length was similar to that HCN4 plays an important role in pacemaking in both regions.
of the preparation including the SAN in the presence of ISO
(166 ± 8 and 174 ± 8 ms, respectively; not significantly 3.3. AV ring bundles and AVN
different). In the presence of 0.05 μM ISO, 2 mM Cs+ was
applied to block If, for which HCN4 is responsible (Fig. 5A, In the atria, there are rings of nodal-like cells around the
right). Cs+ caused both preparations to slow down: in the atrioventricular valves (AV ring bundles) [5–12]. Fig. 6A
preparation including the SAN, the cycle length was shows a Masson's trichrome stained sagittal section through
prolonged by 31 ± 5% to 228 ± 16 ms (P b 0.05 versus control) the right and left atria and ventricles. Small nodal-like cells
and, in the preparation including the interatrial groove, the were present alongside both the tricuspid and mitral valves —
276 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

Fig. 5. Comparison of intrinsic pacemaker activity of SAN and tract of nodal-like cells in interatrial groove. Preparation was divided into two parts as shown in
inset. One part included SAN and second part included interatrial groove. A, cycle length of two preparations in presence of 0.05 μM isoprenaline (ISO; left) and
0.05 μM isoprenaline with 2 mM Cs+ (right). Individual data points as well as means ± SEM (n = 6) shown. B, mean (+SEM; n = 6) percentage increase in cycle
length of two preparations on addition of Cs+. ⁎P b 0.05; NS, not significant.

in the boxed areas B and C. Fig. 6B (from boxed area B in pattern of HCN4, Cx45 and Cx43 in the AVN (seen in
Fig. 6A) shows that in the AV ring bundle around the tri- Fig. 6A) was similar to that of the AV ring bundle around the
cuspid valve, the expression pattern of HCN4, Cx45 and tricuspid valve (data not shown). This is not surprising,
Cx43 was similar to that of the SAN, although the amount of because the AVN can be thought of as part of the AV ring
HCN4 and Cx45 was lower than in the SAN. The expression bundle [22]. Fig. 6C (from boxed area C in Fig. 6A) shows

Fig. 6. Expression of HCN4 and connexins at AV junction. A, Masson's trichrome stained section (cut along sagittal plane) showing AV junction. B, C, high-
magnification images of labelling of HCN4 (left), Cx45 (red; right) and Cx43 (green; right) at tricuspid (B) and mitral (C) valves. Location of images shown by
boxes in A. Scale bars, 1 mm (A) and 50 μm (B, C).
 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281 277

that in the AV ring bundle around the mitral valve, the ex- quantitative PCR (it was not possible to immunolabel HCN4
pression of HCN4 was low or absent, whereas the expression protein as in previous experiments, because the antibody used
of Cx45 and absence of Cx43 were similar to those in the AV was raised in the rabbit). In the proximal and distal parts of the
ring bundle around the tricuspid valve. Similar results were rabbit pulmonary veins, the abundance of both HCN1 and
obtained from four hearts. HCN4 mRNA was comparable with that in right atrial muscle,
i.e. the abundance of HCN4 mRNA in the pulmonary veins
3.4. Pulmonary veins was low (Fig. 7B). We have previously shown that, in the
presence of ryanodine, rapid stimulation routinely induces
In four rat hearts, expression of HCN4 and Cx45 was not pacemaker activity in the pulmonary veins from the rabbit
detected in the myocardial sleeves of the pulmonary veins, [13]; similar behaviour has been observed in the dog [24]. Fig.
whereas Cx43 was expressed throughout the myocardial 7C shows action potentials recorded from the rabbit right

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

sleeves (Fig. 3D,E). This expression pattern was same as that superior pulmonary vein after exposure to 0.5 μM ryanodine.
of the right and left atrial free walls. The absence of the HCN4 Following a rest period of 1 min, the preparation was
in the pulmonary veins is surprising, because in the embryonic stimulated at 3.3 Hz for 20 pulses; this resulted in a substantial
mouse (ED 12) HCN4 mRNA has been observed in tissue that depolarization of the membrane (Fig. 7C). When 3.3 Hz
will give rise to the common pulmonary vein [23]. The stimulation was stopped, a train of spontaneous action
presence of HCN4 in the pulmonary veins was, therefore, potentials was recorded; a pacemaker potential was observed
checked in another species. The abundance of HCN4 (and also between action potentials (Fig. 7C). Block of If by 2 mM Cs+
HCN1) mRNA was measured in the proximal and distal parts had no effect on this behaviour: in the absence and presence of
of the pulmonary veins of the rabbit (Fig. 7A) using Cs+, the train of spontaneous action potentials continued for

Fig. 7. HCN4 and pulmonary veins of rabbit. A, Masson's trichrome stained section through right superior pulmonary vein. Proximal and distal parts of
pulmonary vein are shown. Scale bar, 1 mm. B, abundance of HCN1 and HCN4 mRNAs in right atrial muscle and proximal and distal parts of pulmonary veins
as detected by quantitative PCR. Means ± SEM (n = 7) shown. C, effect of rapid pacing (20 pulses at 3.3 Hz) on membrane potential of right superior pulmonary
vein treated with 0.5 μM ryanodine. Left, stimuli and slow time base recording of membrane potential before (top) and after (bottom) the application of 2 mM
Cs+. Preparation was stimulated at 2 Hz, rested for 1 min, and then briefly stimulated at 3.3 Hz (20 pulses). Burst of spontaneous action potentials was induced by
rapid pacing. Right, superimposed recordings of spontaneous action potentials before and after application of Cs+ (from times shown on left).
278 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281

28.9 ± 5.7 s (n = 3) and 26.6 ± 3.6 s (n = 3), respectively (Fig. Cx43) looped around the ventral surface at the junction of the
7C). This demonstrates that If plays no role in pacemaker superior vena cava with the right atrium and extended from
activity in the pulmonary veins. here down the intercaval region, both next to the crista termi-
nalis (to form the SAN) and the interatrial groove. This roughly
3.5. Summary corresponds to the distribution of leading pacemaker sites
observed in 18 preparations under control conditions (asterisks
Fig. 8A,B shows ventral and dorsal views of the atria and in Fig. 8C). Fig. 8A,B also shows the expression of HCN4 and
summarises the distribution of HCN4 and Cx45 (and Cx43) Cx45 in the AV ring bundles as well as the AVN; as indicated
observed in the present study. Nodal or nodal-like cells (as by the grey scale, expression of HCN4 in the SAN N expression
defined by the expression of HCN4 and Cx45 and lack of in the AV ring bundle encircling the tricuspid valve N

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

Fig. 8. Summary of HCN4 and Cx45 distribution in atria and position of leading pacemaker site. A, B, expression patterns of HCN4 (A) and Cx45 (B). Ventral
view of atria is shown on left and dorsal view is shown on right. Dark area has abundant expression and dotted area has less abundant expression. Tissues
expressing Cx45 did not express Cx43. C, dorsal view of atria showing location of leading pacemaker site (asterisks) under control conditions in 18 preparations.
 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281 279

expression in the AV ring bundle encircling the mitral valve, for different heart rates. Our data are consistent with these
whereas the expression of Cx45 was the same in the three observations.
tissues. The tract of HCN4-expressing nodal-like cells in the
interatrial groove may be important pathologically: recently,
4. Discussion Yamada et al. [27] showed that atrial tachycardia in a group of
patients could be attributed to a focus in the interatrial groove
In the present study, in the atrial muscle of the rat, Cx43 was on the rear wall of the right atrium. Perhaps the focus was a
expressed, whereas HCN4 and Cx45 were not expressed. tract of HCN4-expressing nodal-like cells in the interatrial
However, HCN4 and Cx45 were expressed, whereas Cx43 groove (as observed in the present study — Fig. 8). In another
was not expressed, in tracts of nodal-like cells extending down study, Kistler et al. [28] showed that tachycardia, in one group
the interatrial groove and surrounding the tricuspid valve, as of patients, could be attributed to a focus in the crista ter-

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

well as in the SAN and AVN (Fig. 8). Evidence showed that minalis and, in a second group, to a focus in the interatrial
the tract of HCN4-expressing cells in the interatrial groove has septum. Perhaps in the first group of patients the focus was
probable If-dependent pacemaker activity and in some cir- the SAN itself, whereas in the second group the focus was
cumstances can be the leading pacemaker site (taking over once again a tract of HCN4-expressing nodal-like cells in the
from SAN) ((Figs. 4, 5 and 8C)). interatrial groove.

4.1. Tract of nodal-like cells in interatrial groove 4.2. AV ring bundles

This study showed the presence of a tract of HCN4- It has been reported that in several mammalian species,
positive/Cx43-negative/Cx45-positive small nodal-like cells including the human, myocytes in the atrioventricular annuli
extending downwards from the superior vena cava in the have automatic or triggered activities [5–12]. The automa-
interatrial groove (Figs. 2 and 3). The tract has not been ticity in rabbit tricuspid valve myocytes is sensitive to Cs+
described previously and is distinct from the SAN, which block of If (Cs+ results in a slowing of the beating rate [6,7])
consists of a tract of HCN4-positive/Cx43-negative/Cx45- and, furthermore, If has been recorded from the tricuspid
positive small nodal cells extending downwards from the valve myocytes [7]. In the rabbit, pacemaker activity is faster
superior vena cava next to the crista terminalis (Fig. 2). By in the tricuspid annulus than in the mitral annulus [5]. Con-
separating the interatrial groove from the SAN (Fig. 5), the sistent with these observations, in the present study, we
nodal-like cells in the interatrial groove were demonstrated demonstrated expression of HCN4 in the AV ring bundle of
to be capable of independent pacemaker activity. However, the tricuspid valve (Fig. 6). In the AV ring bundle of the
the pacemaker activity of the nodal-like cells in the mitral valve there was little or no expression of HCN4 (Fig.
interatrial groove was less robust than that of the SAN, 6). This finding may account for the difference in pacemaker
because, in isolated preparations, they did not show activity between the two annuli. In the AV ring bundles of
pacemaker activity under control conditions (ISO was both the tricuspid and mitral valves, we observed expression
required), whereas the SAN did (Fig. 5). This explains of Cx45 (and no expression of Cx43; Fig. 6) (cf. [8,22]). The
why in intact preparations under control conditions, the tract of HCN4-expressing nodal-like cells around the
SAN was normally the leading pacemaker, although in some tricuspid valve may again be important pathologically:
preparations the tract of nodal-like cells in the interatrial Kistler et al. [28] showed that atrial tachycardia in another
groove was the leading pacemaker site (Figs. 4C and 8C). group of patients (see above) could be attributed to a focus in
HCN4 is responsible for the Cs+ -sensitive pacemaker the right atrium around the tricuspid valve — this could
current, If, and it was demonstrated that the pacemaker correspond to a tract of HCN4-expressing nodal-like cells
activity of both the SAN and the nodal-like cells in the surrounding the tricuspid valve (as observed in the present
interatrial groove is Cs+-sensitive (Fig. 5). Although under study — Fig. 8).
control conditions the pacemaker activity of the nodal-like
cells in the interatrial groove was less robust than that of the 4.3. Pulmonary veins
SAN, in the presence of ISO the pacemaker activity of the
nodal-like cells in the interatrial groove was as fast as that of The pulmonary veins are capable of pacemaking and
the SAN (Fig. 5) and this perhaps explains why in intact pulmonary vein cells can have a pacemaker potential [13]. In
preparations the leading pacemaker shifted from the SAN to the present study, no significant expression of HCN4 protein or
the interatrial groove (Fig. 4). In the dog, Boineau et al. [25] mRNA was observed in the pulmonary veins (Figs. 3D
(see also [26]) reported a consistent relationship between the and 7B) and 2mM Cs+ had no effect on spontaneous activity of
leading pacemaker site and the heart rate: vagal stimulation the pulmonary veins (Fig. 7C). Consistent with this, Ehrlich et
decreased the heart rate and resulted in a caudal shift of the al. [29] reported that If is not present in dog pulmonary vein
leading pacemaker site, whereas β-adrenergic stimulation cells. These observations suggest that If plays no role in the
increased the heart rate and resulted in a cranial shift. They pacemaker activity of the pulmonary veins and in this case
suggested that different regions of the SAN are specialised pacemaking must be the result of other ionic currents.
280 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281

4.4. SAN and superior vena cava extension is the slow pathway into the AVN, and the
posterior nodal extension is the part of the AV ring bundle
In the present study, HCN4 and Cx45, but not Cx43, were proximal to the AVN. Both the AVN and the posterior
expressed in the SAN as has been observed previously nodal extension had a similar expression of HCN4 and
[17,30]. In the present study, the SAN extended downwards connexins as the rest of the AV ring bundle of the tricuspid
from the superior vena cava next to the crista terminalis. In valve (data not shown). Recently, we have shown that, after
many experimental animals (rabbit, guinea-pig, monkey, the destruction of the SAN, the junctional rhythm, which is
mouse, cat, pig), the SAN has been reported to be located in Cs+-sensitive (i.e. If-dependent), originates in the posterior
this position [16,31–35]; for example, see our three- nodal extension [15].
dimensional anatomical model of the SAN of the rabbit
[20]. In contrast, in the human, early studies by James [36] 4.6. Conclusion

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

and Truex et al. [37] described the SAN as a discrete tissue at
the junction of the superior vena cava with the right atrium In the atria, HCN4 is expressed in four structures: (i) a
(see well known diagrams of Netter). However, subsequently, tract of nodal-like cells in the interatrial groove, (ii) a tract of
in the human, it has been shown that the SAN can have a tail nodal-like cells encircling the tricuspid valve, (iii) the SAN
projecting down the crista terminalis [38]. Furthermore, the and (iv) the AVN. This will be of importance for normal and
leading pacemaker site in the human (as well as dog and pig) abnormal pacemaking in the atria.
can be located at many sites: as well as at the junction
between the superior vena cava and the right atrium as Appendix A. Supplementary Data
expected from the work of James [36] and Truex et al. [37], it
can be located at any point between the superior and inferior Supplementary data associated with this article can be found,
vena cava along the line of the sulcus terminalis (a groove in the online version, at doi:10.1016/j.cardiores.2006.07.026.
marking the border of the crista terminalis) in the rear wall of
the right atrium [25,26,39–42]. This widespread distribution References
of leading pacemaker sites is consistent with the distribution
of SAN tissue observed in the present study (Fig. 8C). The [1] Tsai CF, Tai CT, Hsieh MH, Lin WS, Yu WC, Ueng KC, et al. Initiation
dog SAN is considered to be similar to the human SAN and, of atrial fibrillation by ectopic beats originating from the superior vena
cava: electrophysiological characteristics and results of radiofrequency
in the dog, total ablation of SAN pacemaker function requires
ablation. Circulation 2000;102:67–74.
the destruction of a block of tissue ∼ 4 cm long and 1.5 cm [2] Liu TY, Tai CT, Lee PC, Hsieh MH, Higa S, Ding YA, et al. Novel
wide extending from the lateral junction of the superior vena concept of atrial tachyarrhythmias originating from the superior vena
cava with the right atrial appendage, along the crista cava: insight from noncontact mapping. J Cardiovasc Electrophysiol
terminalis to the inferior vena cava [26,43]. Once again, 2003;14:533–9.
[3] Shah DC, Haissaguerre M, Jais P, Clementy J. High-resolution
this corresponds to the distribution of SAN tissue as
mapping of tachycardia originating from the superior vena cava:
described in the present study (Fig. 8). The superior vena evidence of electrical heterogeneity, slow conduction, and possible
cava has been reported to have ectopic pacemaker activity circus movement reentry. J Cardiovasc Electrophysiol 2002;13:
[1–4]; it is possible that this is the result of abnormal 388–92.
pacemaker activity in the nodal cells that are wrapped around [4] Goya M, Ouyang F, Ernst S, Volkmer M, Antz M, Kuck KH.
Electroanatomic mapping and catheter ablation of breakthroughs from
the ventral surface of the superior vena cava (Fig. 8). The tract
the right atrium to the superior vena cava in patients with atrial
of nodal-like cells in the interatrial groove is a continuation of fibrillation. Circulation 2002;106:1317–20.
the tract of nodal cells wrapped around the ventral surface of [5] Rozanski GJ, Jalife J. Automaticity in atrioventricular valve leaflets of
the superior vena cava (Fig. 8). rabbit heart. Am J Physiol 1986;250:H397–406.
Specialised internodal pathways (between the SAN and [6] Rozanski GJ. Electrophysiological properties of automatic fibers in
rabbit atrioventricular valves. Am J Physiol 1987;253:H720–7.
AVN) have been proposed. Could nodal or nodal-like cells
[7] Anumonwo JM, Delmar M, Jalife J. Electrophysiology of single heart
form the pathways? There was not a continuous tract of cells from the rabbit tricuspid valve. J Physiol 1990;425:145–67.
HCN4-expressing cells between the SAN and AVN (Fig. 8) [8] McGuire MA, de Bakker JM, Vermeulen JT, Moorman AF, Loh P,
and this is consistent with the finding that an increase in Thibault B, et al. Atrioventricular junctional tissue. Discrepancy
extracellular K+ (which abolishes the action potential in between histological and electrophysiological characteristics. Circula-
tion 1996;94:571–7.
atrial muscle but not in nodal tissue) blocks action potential
[9] Bassett AL, Fenoglio Jr JJ, Wit AL, Myerburg RJ, Gelband H.
transmission from the SAN to the AVN [44]. Eelctrophysiologic and ultrastructural characteristics of the canine
tricuspid valve. Am J Physiol 1976;230:1366–73.
4.5. AVN [10] Wit AL, Fenoglio Jr JJ, Hordof AJ, Reemtsma K. Ultrastructure and
transmembrane potentials of cardiac muscle in the human anterior
mitral valve leaflet. Circulation 1979;59:1284–92.
In the present study, HCN4 and Cx45, but not Cx43,
[11] Wit AL, Fenoglio Jr JJ, Wagner BM, Bassett AL. Electrophysiological
was expressed in the AVN (c.f. [15]). The AVN can be properties of cardiac muscle in the anterior mitral valve leaflet and the
considered to be part of the AV ring bundle of the tricuspid adjacent atrium in the dog. Possible implications for the genesis of
valve (Fig. 8A,B). The posterior (or inferior) nodal atrial dysrhythmias. Circ Res 1973;32:731–45.
 M. Yamamoto et al. / Cardiovascular Research 72 (2006) 271–281 281

[12] Makarychev VA, Kosharskaia IL, Ul'ianinskii LS. Automatic activity of term follow-up in a large consecutive single center series. Heart
pacemaker cells in the atrioventricular valves of the rabbit heart. Biulleten Rhythm, vol. 1; 2004. p. S236.
Eksperimentalnoi Biologii I Meditsiny, vol. 81; 1976. p. 517–20. [29] Ehrlich JR, Cha TJ, Zhang L, Chartier D, Melnyk P, Hohnloser SH, et al.
[13] Honjo H, Boyett MR, Niwa R, Inada S, Yamamoto M, Mitsui K, et al. Cellular electrophysiology of canine pulmonary vein cardiomyocytes:
Pacing-induced spontaneous activity in myocardial sleeves of pulmonary action potential and ionic current properties. J Physiol 2003;551:801–13.
veins after treatment with ryanodine. Circulation 2003;107:1937–43. [30] Coppen SR, Kodama I, Boyett MR, Dobrzynski H, Takagishi Y, Honjo
[14] DiFrancesco D. Pacemaker mechanisms in cardiac tissue. Annu Rev H, et al. Connexin45, a major connexin of the rabbit sinoatrial node, is
Physiol 1993;55:455–72. co-expressed with connexin43 in a restricted zone at the nodal-crista
[15] Dobrzynski H, Nikolski VP, Sambelashvili AT, Greener ID, Boyett MR, terminalis border. J Histochem Cytochem 1999;47:907–18.
Efimov IR. Site of origin and molecular substrate of atrioventricular [31] Opthof T, de Jonge B, Mackaay AJC, Bleeker WK, Masson-Pevet M,
junctional rhythm in the rabbit heart. Circ Res 2003;93:1102–10. Jongsma HJ, et al. Functional and morphological organization of the
[16] Boyett MR, Honjo H, Kodama I. The sinoatrial node, a heterogeneous guinea-pig sinoatrial node compared with the rabbit sinoatrial node.
pacemaker structure. Cardiovasc Res 2000;47:658–87. J Mol Cell Cardiol 1985;17:549–64.

Downloaded from https://academic.oup.com/cardiovascres/article/72/2/271/277438 by guest on 04 February 2022

[17] Biel M, Schneider A, Wahl C. Cardiac HCN channels: structure, [32] Alings AMW, Abbas RF, de Jonge B, Bouman LN. Structure and
function, and modulation. Trends Cardiovasc Med 2002;12:206–12. function of the simian sinoatrial node (Macaca fascicularis). J Mol
[18] Han W, Bao W, Wang Z, Nattel S. Comparison of ion-channel subunit Cell Cardiol 1990;22:1453–66.
expression in canine cardiac Purkinje fibers and ventricular muscle. [33] Opthof T, de Jonge B, Masson-Pevet M, Jongsma HJ, Bouman LN.
Circ Res 2002;91:790–7. Functional and morphological organization of the cat sinoatrial node.
[19] Joyner RW, van Capelle FJL. Propagation through electrically coupled J Mol Cell Cardiol 1986;18:1015–31.
cells: how a small SA node drives a large atrium. Biophys J 1986;50: [34] Opthof T, de Jonge B, Jongsma HJ, Bouman LN. Functional morphology
1157–64. of the pig sinoatrial node. J Mol Cell Cardiol 1987;19:1221–36.
[20] Dobrzynski H, Li J, Tellez J, Greener ID, Nikolski VP, Wright SE, et al. [35] Lei M, Jones SA, Liu J, Lancaster MK, Fung SS, Dobrzynski H, et al.
Computer three-dimensional reconstruction of the sinoatrial node. Requirement of neuronal- and cardiac-type sodium channels for
Circulation 2005;111:846–54. murine sinoatrial node pacemaking. J Physiol 2004;559:835–48.
[21] Yamamoto M, Honjo H, Niwa R, Kodama I. Low frequency [36] James TN. Anatomy of the human sinus node. Anat Rec 1961;141:
extracellular potentials recorded from the sinoatrial node. Cardiovasc 109–39.
Res 1998;39:360–72. [37] Truex RC, Smythe MQ, Taylor MJ. Reconstruction of the human
[22] Coppen SR, Severs NJ, Gourdie RG. Connexin45 (α6) expression sinoatrial node. Anat Rec 1967;159:371–8.
delineates an extended conduction system in the embryonic and mature [38] Sanchez-Quintana D, Anderson RH, Cabrera JA, Climent V, Martin R,
rodent heart. Dev Genet 1999;24:82–90. Farre J, et al. The terminal crest: morphological features relevant to
[23] Garcia-Frigola C, Shi Y, Evans SM. Expression of the hyperpolariza- electrophysiology. Heart 2002;88:406–11.
tion-activated cyclic nucleotide-gated cation channel HCN4 during [39] Boineau JP, Canavan TE, Schuessler RB, Cain ME, Corr PB, Cox JL.
mouse heart development. Gene Expr Patterns 2003;3:777–83. Demonstration of a widely distributed atrial pacemaker complex in the
[24] Chou CC, Nihei M, Zhou S, Tan A, Kawase A, Macias ES, et al. human heart. Circulation 1988;77:1221–37.
Intracellular calcium dynamics and anisotropic reentry in isolated canine [40] Boineau JP, Schuessler RB, Hackel DB, Miller CB, Brockus CW,
pulmonary veins and left atrium. Circulation 2005;111: 2889–97. Wylds AC. Widespread distribution and rate differentiation of the atrial
[25] Boineau JP, Schuessler RB, Roeske WR, Autry LJ, Miller CB, Wylds pacemaker complex. Am J Physiol 1980;239:H406–15.
AC. Quantitative relation between sites of atrial muscle origin and [41] Betts TR, Ho SY, Sanchez-Quintana D, Roberts PR, Anderson RH,
cycle length. Am J Physiol 1983;245:H781–9. Morgan JM. Three-dimensional mapping of right atrial activation
[26] Kalman JM, Lee RJ, Fisher WG, Chin MC, Ursell P, Stillson CA, et al. during sinus rhythm and its relationship to endocardial architecture.
Radiofrequency catheter modification of sinus pacemaker function J Cardiovasc Electrophysiol 2002;13:1152–9.
guided by intracardiac echocardiography. Circulation 1995;92: [42] Betts TR, Roberts PR, Ho SY, Morgan JM. High density endocardial
3070–81. mapping of shifts in the site of earliest depolarization during sinus rhythm
[27] Yamada T, Murakami Y, Muto M, Okada T, Okamoto M, Toyama J, et al. and sinus tachycardia. Pacing Clin Electrophysiol 2003;26:874–82.
Electrophysiologic characteristics of atrial tachycardia originating from [43] Euler DE, Jones SB, Gunnar WP, Loeb JM, Murdock DK, Randall
the right pulmonary veins or posterior right atrium: double potentials WC. Cardiac arrhythmias in the conscious dog after excision of the
obtained from the posterior wall of the right atrium can be useful to sinoatrial node and crista terminalis. Circulation 1979;59:468–75.
predict foci of atrial tachycardia in right pulmonary veins or posterior [44] Sanchis J, Opthof T, Wilde AAM, Janse MJ. Mutual electrotonic
right atrium. J Cardiovasc Electrophysiol 2004;15:745–51. interaction between the rabbit sinoatrial and atrioventricular node.
[28] Kistler P, Haqqani H, Fynn S, Stevenson I, Vohra J, Sparks P, et al. RF J Cardiovasc Electrophysiol 1990;1:318–26.
ablation of focal atrial tachycardia — anatomic distribution and long