Research ArticlePhysiology

A coupled-clock system drives the automaticity of human sinoatrial nodal pacemaker cells

See allHide authors and affiliations

Science Signaling  12 Jun 2018:
Vol. 11, Issue 534, eaap7608
DOI: 10.1126/scisignal.aap7608
  • Fig. 1 Single SANC isolation from a human heart.

    (A) A posterior right atrial (RA) tissue including SAN region was removed from the rest of the heart. The SAN region was identified anatomically (white dashed lines), from which small strips were excised for subsequent enzymatic digestion. IAS, intra-atrial septum; IVC, inferior vena cava; CT, crista terminalis. Scale bar, 2 cm. (B) Examples of isolated single human SANCs under low power (left scale bar, 200 μm) and high power (right scale bar, 10 μm). An example is also shown in movie S1. (C) A cross section of the SAN region marked with a solid line and an asterisk in (A) shows SAN tissue surrounding the SAN artery. Scale bar, 5 mm. (D) Immunohistochemical (right and middle) and immunofluorescence staining (left) for the SANC marker HCN4. Left scale bar, 30 μm. Middle and right scale bars, 20 μm. (E) HCN4 expression normalized to 18S ribosomal RNA as determined by qPCR was ~2.2-fold higher in SAN than in right atria. Data were from three independent experiments.

  • Fig. 2 2D Ca2+ signal and membrane potential measurements in single isolated human SANCs that fired spontaneous action potentials at baseline.

    Local Ca2+ releases (LCRs) were observed in all human SANCs studied (22 cells from 4 hearts; 16 from female, 6 from male). (A) 2D high-speed Ca2+ recording of a typical human SANC. Left and right: 2D original data and local Ca2+ release detection through our custom program (white), respectively (28). Scale bar, 20 μm. (B) Local Ca2+ release period is defined as the time that elapses between the peak of the prior action potential–induced global cytosolic Ca2+ transient and the onset of each local Ca2+ releases. a.u., arbitrary units. (C) Action potential–induced Ca2+ transients (magenta), individual local Ca2+ releases (plotted in gray-black), and local Ca2+ release ensemble area (blue). The ensemble area for local Ca2+ releases sums the total area for all local Ca2+ releases. A dynamic example of events is presented in movie S2. (D) Example of membrane potential measured from 1 of 4 of the 12 human SANCs in which Ca2+ signals were measured (12). MDP, maximum diastolic potential; TIP, time to ignition phase. (E) Effect of β-AR stimulation on action potential, local Ca2+ releases, and global cytosolic Ca2+ transient. Membrane potential (black), global cytosolic Ca2+ transient (magenta), individual local Ca2+ releases (plotted in colors), and local Ca2+ release ensemble (blue) in a human SANC that generated spontaneous action potentials at baseline (left) and during β-AR stimulation (right) are illustrated. See Tables 1 and 2 for quantitative data.

  • Fig. 3 The effect of cAMP in skinned isolated human SANCs.

    (A) Representative examples of confocal line-scan images of permeabilized human SANCs bathed in 50 nM free [Ca2+] under baseline conditions (upper panel) and in response to cAMP (10 μM) in the same cell (lower panel). (B to D) Histogram distributions of local Ca2+ release characteristics in permeabilized human SANCs under baseline conditions (white bars, 61 local Ca2+ releases from three cells) and in response to 10 μM cAMP (red bars, 131 local Ca2+ releases from three cells): (B) size, (C) duration, and (D) Ca2+ signals of individual local Ca2+ releases. Data are from three independent experiments. Fisher’s exact test was used to determine the statistical significance of a shift in the distribution of local Ca2+ release in permeabilized SANCs in response to β-AR stimulation. The asterisk indicates a significant shift in the distribution of local Ca2+ release parameters toward higher values (>50% percentile) in response to cAMP.

  • Fig. 4 Local Ca2+ releases in the absence of action potential–induced Ca2+ transients in arrested isolated human SANCs.

    Local Ca2+ releases were present in human SANCs that were devoid of spontaneous action potential–induced global cytosolic Ca2+ transients at baseline (Table 1). Local Ca2+ release characteristics in arrested responder cells (n = 4) and nonresponder cells (n = 4) to β-AR stimulation are listed in Table 1. (A) Example of an arrested responder (left) that began to fire action potential–induced Ca2+ transients in response to β-AR stimulation (right). Scale bar, 20 μm. (B) Example of nonresponder or a cell that did not fire action potential–induced Ca2+ transients in the presence of β-AR stimulation. Local Ca2+ releases are present in the arrested state and did not became augmented by β-AR stimulation.

  • Fig. 5 The evolution of automaticity in arrested isolated human SANCs in response to β-AR stimulation.

    (A) Membrane potential from a typical arrested responder human SANCs before and during early β-AR stimulation. (B) Simultaneous recordings of membrane potential (black), global cytosolic Ca2+ signal (magenta), and local Ca2+ release ensemble (LCR ensemble, blue area) in an arrested SANC before and during action potential firing in response to β-AR stimulation at times from sections indicated by (a) and (e) in (A). (C) Phase plane diagrams of membrane potential and Ca2+ from sections indicated as “Cycle 1” and “Cycle 2” in (B). Roman numerals i to ix in (C) correspond to time points i to ix indicated in (B). (D) Membrane potential at a later stage of β-AR stimulation. (E) Simultaneous recordings of membrane potential (black), global cytosolic Ca2+ transient (magenta), and local Ca2+ release ensemble (blue) from a section marked as (a) in (D). (F) Phase plane plot diagrams of membrane potential and Ca2+ from sections indicated as “Cycle 2” in (B) and “Cycle 3” in (C). Roman numerals x to xiv correspond to time points x to xiv in (C), and Roman numerals vi to ix correspond to time points vi to ix in (B). (G) Simultaneous recordings of membrane potential (black), global cytosolic Ca2+ signal (magenta), and local Ca2+ release ensemble (LCR ensemble, blue area) in an arrested SANC after washout of β-AR stimulation.

  • Fig. 6 A unique continuum of clock coupling in human SANCs, which fire spontaneous action potentials over a wide range of rates.

    (A) Evolution of membrane potential and local Ca2+ release ensemble during the transition from arrested state to action potential firing. Detail of high-resolution action potential recordings during transition phase from arrested state to action potential firing (upper panel) and local Ca2+ release ensemble measured simultaneously during early and late stages of the β-AR stimulation response in critical stages in Fig. 5 (lower panel). Corresponding action potential tracings from upper panel are overlaid as dashed lines. Color-coded numbers 1 to 4 in each panel corresponds to numbers 1 to 3 in Fig. 5A and number 4 in Fig. 5D. (B to D) Correlations between cycle length, local Ca2+ release period, and time to ignition phase observed across SANCs over a wide range of action potential firing rates. Color codes of points are as follows: gray, steady-state firing at baseline (Fig. 2E, left, and Table 1; n = 4 cells); orange, those during β-AR stimulation (Fig. 2E, right, and Table 1; n = 4 cells); white, steady state (Fig. 4A, right; n = 4); magenta, transition state (Fig. 5 and table S2; n = 5 cells) of initially arrested responder SANCs that generated spontaneous action potential in response to β-AR stimulation. Correlations between cycle length and time to ignition phase (B), cycle length and local Ca2+ release period (C), and local Ca2+ release period and time to ignition phase (D) in each group of cells tested separately did not statistically differ from each other. Therefore, all points in each panel conform to common line: cycle length compared to time to ignition phase, Y = 1.073*X + 241.1, R2 = 0.96 (B); cycle length compared to local Ca2+ release period, Y = 0.718*X − 136, R2 = 0.85 (C); local Ca2+ release period compared to time to ignition phase, 1.264*X + 196, R2 = 0.82 (D).

  • Table 1 Action potential–induced global cytosolic Ca2+ transient cycle lengths and local Ca2+ release characteristics in human SANCs.

    Data were from 20 single SANCs isolated from four hearts (three male, one female). *P < 0.05, compared to baseline by two-tailed paired t test. CaT CL, Ca2+ transient cycle length; NA, not applicable.

    nCaT CL (ms)LCR ensemble (μm2/10 ms)LCR period (ms)
    Baselineβ-AR
    stimulation
    Baselineβ-AR
    stimulation
    Baselineβ-AR
    stimulation
    Firing SANCs121676 ± 241971 ± 234*15.9 ± 4.228.1 ± 3.7*870 ± 126530 ± 129*
    Arrested SANCs
    (responder)
    4NA3328 ± 111022 ± 339 ± 10*NA1432 ± 528
    Arrested SANCs
    (nonresponder)
    4NANA11 ± 53 ± 3NANA
  • Table 2 Action potential characteristics of single isolated firing human SANCs.

    Data were from 11 single SANCs isolated from two hearts (one female, one male). *P < 0.05, compared to baseline by two-tailed paired t test. AP CL, action potential cycle length; MDP, maximum diastolic potential; TIP, time to ignition phase.

    nArrested
    potential
    AP CL (ms)MDP (mV)TIP (ms)
    BaselineBaselineβ-AR
    stimulation
    Baselineβ-AR
    stimulation
    Baselineβ-AR
    stimulation
    Firing SANCs4NA1698 ± 4821211 ± 374*−50.1 ± 2.3−55.3 ± 5.41582 ± 483958 ± 357*
    Arrested SANCs
    (responder)
    4−39.7 ± 3.3NA2936 ± 608NA−56.2 ± 4.2NA2302 ± 616
    Arrested SANCs
    (nonresponder)
    3−31.3 ± 8.8NANANANANANA

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/11/534/eaap7608/DC1

    Fig. S1. Action potential repolarization failure in the transition from arrest to spontaneous action potential firing in response to β-AR stimulation.

    Fig. S2. A schematic of the clock coupling in SANCs.

    Table S1. Characteristics of the donor hearts used in the present study.

    Table S2. Time-dependent evolution of action potential parameters in the transition state of an initially arrested human SANC.

    Movie S1. A freshly isolated single spontaneously beating human SANC from heart 3.

    Movie S2. LCR detection software on a spontaneously beating human SANC from heart 4.

    Movie S3. Simultaneous measurements of membrane potential and a 2D Ca2+ signal.

  • Supplementary Materials for:

    A coupled-clock system drives the automaticity of human sinoatrial nodal pacemaker cells

    Kenta Tsutsui, Oliver J. Monfredi, Syevda G. Sirenko-Tagirova, Larissa A. Maltseva, Rostislav Bychkov, Mary S. Kim, Bruce D. Ziman, Kirill V. Tarasov, Yelena S. Tarasova, Jing Zhang, Mingyi Wang, Alexander V. Maltsev, Jaclyn A. Brennan, Igor R. Efimov, Michael D. Stern, Victor A. Maltsev, Edward G. Lakatta*

    *Corresponding author. Email: lakattae{at}grc.nia.nih.gov

    This PDF file includes:

    • Fig. S1. Action potential repolarization failure in the transition from arrest to spontaneous action potential firing in response to β-AR stimulation.
    • Fig. S2. A schematic of the clock coupling in SANCs.
    • Table S1. Characteristics of the donor hearts used in the present study.
    • Table S2. Time-dependent evolution of action potential parameters in the transition state of an initially arrested human SANC.
    • Legends for movies S1 to S3

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). A freshly isolated single spontaneously beating human SANC from heart 3.
    • Movie S2 (.mp4 format). LCR detection software on a spontaneously beating human SANC from heart 4.
    • Movie S3 (.mp4 format). Simultaneous measurements of membrane potential and a 2D Ca2+ signal.

    © 2018 American Association for the Advancement of Science

Stay Connected to Science Signaling

Navigate This Article