Stem Cell Derived Nodal-Like Cardiomyocytes as a …pharmrev.aspetjournals.org/content/pharmrev/67/2/368...aimed at preferentially driving stem cells down a nodal lineage. Finally,

1521-0081/67/2/368–388$25.00 http://dx.doi.org/10.1124/pr.114.009597PHARMACOLOGICAL REVIEWS Pharmacol Rev 67:368–388, April 2015Copyright © 2015 by The American Society for Pharmacology and Experimental Therapeutics

ASSOCIATE EDITOR: YOSHIHIRO ISHIKAWA

Stem Cell–Derived Nodal-Like Cardiomyocytes asa Novel Pharmacologic Tool: Insights from Sinoatrial

Node Development and FunctionAndrea Barbuti and Richard B. Robinson

Department of Biosciences, Università degli Studi di Milano, Milano, Italy (A.B.); and Department of Pharmacology, Columbia UniversityMedical Center, New York, New York (R.B.R.)

Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368I. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368II. The Adult Sinoatrial Node. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370III. Developmental Origin of the Sinoatrial Node. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371

A. Elucidation of the Gene Program Leading to Sinoatrial Node Formation . . . . . . . . . . . . . . . . . . 374B. Sinoatrial Precursor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

IV. Mouse Stem Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375A. Evidence for Possibility of Generating Sinoatrial-Like Cells from Murine Embryonic

Stem Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376B. Improvements in the Generation of Sinoatrial-Like Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377

V. Human Stem Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381A. Human Embryonic Stem Cell–Derived Cardiomyocyte Selection and Sinoatrial-Like

Cell Enrichment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383B. Human Induced Pluripotent Stem Cells and Automaticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384

VI. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385

Abstract——Since the first reports on the isolationand differentiation of stem cells, and in particular sincethe early success in driving these cells down a cardiaclineage, there has been interest in the potential of suchpreparations in cardiac regenerative therapy. Much ofthe focus of such research has been on improvingmechanical function after myocardial infarction; how-ever, electrophysiologic studies of these preparationshave revealed a heterogeneous mix of action potentialcharacteristics, including some described as “pace-maker” or “nodal-like,” which in turn led to interest inthe therapeutic potential of these preparations in thetreatment of rhythm disorders; several proof-of-conceptstudies have used these cells to create a biologicalternative to electronic pacemakers. Further, thereare additional potential applications of a preparation of

pacemaker cells derived from stem cells, for example, inhigh-throughput screens of new chronotropic agents. Allsuch applications require reasonably efficient methodsfor selecting or enriching the “nodal-like” cells, however,which in turn depends on first defining what constitutesa nodal-like cell since not all pacemaking cells arenecessarily of nodal lineage. This review discusses thecurrent state of the field in terms of characterizingsinoatrial-like cardiomyocytes derived from embryonicand induced pluripotent stem cells, markers that mightbe appropriate based on the current knowledge of thegene program leading to sinoatrial node development,what functional characteristics might be expected anddesired based on studies of the sinoatrial node, andrecent efforts at enrichment and selection of nodal-likecells.

I. Introduction

Since the early 1990s, numerous studies have beenconducted in which stem cells of various types (embry-onic, adult mesenchymal, induced pluripotent) and frommultiple species (mouse, human) have been manipulated

in some way to produce a preparation of cardiac-likecells (Wobus et al., 1991; Kehat et al., 2001; Chow et al.,2013; Robertson et al., 2013). What is generally obtained,however, is a heterogeneous and immature population ofcells with characteristics reminiscent of (but not necessarily

Address correspondence to: Dr. Richard. B. Robinson, Department of Pharmacology, Columbia University Medical Center, 630 W.168th St, Room PS7-445, New York, NY 10032. E-mail: [email protected]

dx.doi.org/10.1124/pr.114.009597.

368

by guest on July 1, 2020D

ownloaded from

http://dx.doi.org/10.1124/pr.114.009597

mailto:[email protected]

http://dx.doi.org/10.1124/pr.114.009597

identical to) cells from distinct adult cardiac regions,including ventricle, atrium, and sinoatrial node (SAN)(Maltsev et al., 1993; Mummery et al., 2003). The vastmajority of research studies and review articles havefocused on producing ventricular-like cells that might besuitable for repairing the working myocardium aftermyocardial infarction or heart failure.More recently, investigators have begun addressing

the question of whether the SAN subtype within thisheterogeneous cell population can be preferentiallygenerated or selected (Zhu et al., 2010; Hashem andClaycomb, 2013). If successful, such an approach wouldhave multiple applications. First, if a relatively purepreparation of cells characteristic of the adult SAN couldbe produced, it would provide an ideal model system forexploring the ionic basis of normal and abnormalautomaticity within the SAN. Although there has beenextensive study of this point, controversy remains (seesection II). Further, relatively little of the research isbased on human tissue, and thus an appropriate cellpreparation of human origin would be a valuable tool.Second, such a preparation might provide an idealsource with which to create a “biological pacemaker” toreplace current electronic pacemakers in treating rhythmdisorders. Proof-of-concept experiments with biologicpacemakers have demonstrated some success with theimplantation of fetal (Ruhparwar et al., 2002) or adult(Zhang et al., 2011) SAN cells and beating embryoidbodies (EBs) derived from embryonic stem cells (ESCs)(Kehat et al., 2004; Xue et al., 2005); cardiomyocytesfrom induced pluripotent stem cells (iPSCs) also havebeen produced and characterized (Novak et al., 2010;Mandel et al., 2012) but not yet implanted to form abiologic pacemaker. In addition, adult mesenchymal stemcells, not driven down a cardiac lineage but geneticallymodified to generate a pacemaker current (If), havealso proven successful (Potapova et al., 2004; Rosenet al., 2008).A third potential application for an improved nodal-

like population of stem cells would be to screen newchronotropic agents. The development of the bradycar-dic agent ivabradine (DiFrancesco, 2010) has demon-strated a therapeutic market for such drugs, and theexistence of a spontaneously active cell line with ap-propriate nodal characteristics would greatly facili-tate high-throughput screening of new molecules. Atpresent, one is limited to either screening such agentsagainst a single target channel that is a contributor to

automaticity or conducting labor-intensive single-cellpatch-clamp studies on SAN cells isolated from variousanimal preparations that may differ in unknown waysfrom human SAN cells. Studies have demonstrated thefeasibility of recording spontaneous activity from cardio-myocytes derived from stem cells either through the useof electrodes embedded in the culture substrate (Mandelet al., 2012) or by the use of calcium-sensitive fluorescentdyes (Zahanich et al., 2011), although additional work isneeded to adapt either method to a high-throughputscreen and to validate these approaches on an enrichedpopulation of nodal-like cells. Finally, a variation on thisis the use of patient-specific iPSCs to explore the cellularor molecular basis of those congenital cardiac disordersof complex origin (either because of a multigenic origin orbecause of mutation in a gene not directly related tomembrane excitability) and to screen for appropriatetherapeutic agents for these disorders. Such an approachhas been suggested and demonstrated with ventricular-like cells derived from iPSCs of patients with diverselong QT genetic mutations (Moretti et al., 2010; Itzhakiet al., 2011; Malan et al., 2011; Yazawa et al., 2011).Recently, we reported (Nawathe et al., 2013) on a patientwith long QT resulting from a mutation in a b-subunitthat not only interacts with a K channel associatedwith long QT syndrome but also with additional K chan-nels and with the hyperpolarization-activated cyclicnucleotide–gated (HCN) channels important to automa-ticity. The patient exhibited marked bradycardia, anda nodal-like cell preparation of patient-specific iPSCsmight be beneficial in elucidating the mechanism(s) bywhich this single point mutation in a b-subunit interactswith multiple ion channels that together result in sinusbradycardia.

All these applications would benefit from amethod thatallows for preferential induction of nodal-like cells sincecurrent standard methods for driving stem cells downa cardiac lineage typically result in only a few percent ofthe cells being nodal-like; however, even if purity remainslow, it may be sufficient if combined with an efficientmethod of selecting the nodal-like cells from within thetotal population. Although progress has been maderecently on these fronts, the extent of ultimate successremains to be determined. Further, it is not yet knownwhether one cell source (adult stem cells, ESCs, iPSCs)is more suitable than another for each of the potentialapplications. Finally, none of the current approachesproduces truly mature cardiac cells, at least with respect

ABBREVIATIONS: AG1478, N-(3-chlorophenyl)-6,7-dimethoxyquinazolin-4-amine; ALCAM, activated leukocyte cell adhesion; ANP, atrialnatriuretic peptide; AV, atrioventricular; BQ123, cyclo(D-Trp-D-Asp-Pro-D-Val-Leu); BQ788, N-[(cis-2,6-dimethyl-1-piperidinyl)carbonyl]-4-methyl-L-leucyl-1-(methoxycarbonyl)-D-tryptophyl-D-norleucine; CACNA1C, CaV1.2 L-type calcium channel; Cx, connexin; E, embryonic day;EB, embryoid body; EBIO, 1-ethyl-2-benzimidazolinone; EGFP, enhanced green fluorescent protein; ESC, embryonic stem cell; ET-1,endothelin-1; FHF, first heart field; GFP, green fluorescent protein; HCN, hyperpolarization-activated cyclic nucleotide–gated; hESC, humanembryonic stem cell; ICaL, L-type calcium current; If, pacemaker current; IK1, inward rectifier K current; iPSC, induced pluripotent stem cell;mESC, mouse embryonic stem cell; MHC, myosin heavy chain; MLC, myosin light chain; MSC, mesenchymal stem cell; NF-M, neurofilament-M;SAN, sinoatrial node; SIRPA, signal-regulatory protein a; SR, sarcoplasmic reticulum; TMRM, tetramethylrhodamine methyl ester perchlorate;ZD 7288, 4-ethylphenylamino-1,2-dimethyl-6-methylaminopyrimidinium chloride.

Nodal-Like Stem Cell Applications 369

to ventricular function (Yang et al., 2014). The issue issomewhat less clear in the case of nodal-like cells since inthis case there is less (but nonzero) functional distinctionbetween cells of the young and adult SAN.In this review, we consider the current state of the art

and discuss what remains to be accomplished beforea preparation of human nodal-like cells derived fromstem cells can be used in a range of therapeutic and basicscience applications. A critical issue is how one definesand identifies cells of nodal lineage from the heteroge-neous population of spontaneous cardiomyocytes origi-nating within these stem cell preparations. In thatregard, we first briefly review the critical molecular andfunctional characteristics of the adult SAN cell. We thenreview what is known about normal SAN developmentsince this has obvious relevance to devise novel approachesaimed at preferentially driving stem cells down a nodallineage. Finally, we review the existing literature on thepreparation, selection, and characterization of nodal-like cells derived from various mouse and human stemcell sources and consider current limitations and poten-tial future directions.

II. The Adult Sinoatrial Node

A major challenge in developing a homogeneous cellpopulation of cells to replicate SAN functionality isthat the SAN itself is heterogeneous, and that veryheterogeneity—along with the nature of the connec-tions to surrounding atrial tissue—is probably criticalto the SAN’s syncytial behavior in vivo. There aredifferences across the nodal region in cell size, densityof ionic currents, connexin (Cx) expression, and othermolecular markers (Boyett et al., 2003), which mayimpact function. For example, superfusion of theisolated node with a b-adrenergic agonist results ina shift of the initiation site within the node, suggestinglocalized differences in adrenergic responsiveness(Boyett et al., 2000). Thus, at present, the best onemay be able to accomplish is to settle on a collection ofcharacteristics or cellular parameters that are mostrepresentative of the SAN as a whole and most relevantto its function and use these as screening criteria at themolecular or cellular level, recognizing that the result-ing homogeneous preparation will recapitulate only asubset of the complex functionality of the syncytial SAN.Eventually, tissue engineering techniques (Camellitiet al., 2005) may allow more complete reconstruction ofcomplex nodal tissue architecture; however, an accuraterepresentation would require not just replicating themulticellular pattern but also any heterogeneity in cellcoupling and ion channel expression, both of which varyas a function of location within the node and diseasestate (Kohl et al., 2005; Hao et al., 2011).The most obvious and relevant feature of the SAN

is spontaneous activity, and some clear qualitativecommonalities throughout the SAN contribute to this

behavior, even if there are regional quantitative differ-ences across the tissue. One such hallmark of SAN cellsthat is important for their automatic function is therelative low abundance of inward rectifier K current, IK1

(Cohen and Robinson, 2006). The paucity of backgroundoutward current during diastole allows a relativelysmall inward current to drive the SAN cell to threshold.In this regard, the second typical feature of SAN cells isthe presence of the If, which is generated by the HCNgene family, which is highly expressed in the SAN.Conceptually, this current is ideally suited for pace-maker function in that it activates on hyperpolarization(at the end of the action potential), deactivates on de-polarization (thus minimally contributing to actionpotential duration), and directly binds the b-adrenergicsecond-messenger cAMP (which shifts If activation toless negative potentials), thereby providing autonomicresponsiveness (Robinson and Siegelbaum, 2003). Whereasdebate on the importance of this current continues(Lakatta and DiFrancesco, 2009), there is no questionthat HCN subunits (specifically HCN4) are highlyexpressed throughout the SAN of different species(Shi et al., 1999; Yamamoto et al., 2006; Liu et al., 2007;Brioschi et al., 2009; Chandler et al., 2009), that HCN4mutations result in cardiac rhythm disorders (Baruscottiet al., 2010), and that an If-selective blocker slows (butdoes not stop) sinus rate (DiFrancesco, 2010). Thus,both functionally and as a molecular marker, the pres-ence of HCN4 can be considered a minimal criterion foridentifying a nodal-like cell. As detailed in section II,developmental data argue that HCN4 is a general markerof pacemaker cells rather than a specific marker of onlysinoatrial cells. That is, although all nodal cells may havepacemaking functionality, not all pacemaker cells are ofnodal lineage. Further, whereas If typically activates atless negative voltages in SAN than other cardiac regionsin the adult, the same is not necessarily true in theimmature heart (Robinson et al., 1997). Thus, HCN4expression and a positively activating If are necessarybut not sufficient selection criteria.

Other studies have explored the role of calciumhomeostasis in basal and adrenergically-stimulated au-tomaticity (Ju and Allen, 1999; Lakatta et al., 2008).Although laboratory and species differences in themagnitude of the effect were found, there is generalagreement that disruption of calcium homeostasis,typically by use of ryanodine to deplete sarcoplasmicreticulum (SR) Ca2+ stores, slows sinus rate, and reducesthe extent to which rate increases in response tob-adrenergic agonists (Bogdanov et al., 2001; Bucchiet al., 2003). It has been argued that the basis for thiscontribution of calcium homeostasis to SAN automaticityis the electrogenic sodium (Na)/calcium (Ca) exchangecurrent, which generates a depolarizing current whenremoving Ca2+ released from the SR from the cell. Theprimary source of the Ca2+ released from the SR andremoved by the Na/Ca exchange current is the L-type

370 Barbuti and Robinson

Ca current, ICaL, which is largely responsible for theaction potential upstroke in SAN cells. In this respect,other potential identifying characteristics of the SANcell are a high abundance of the L-type calcium channelisoform CaV1.3 and T-type isoforms CaV3.1 and CaV3.2(Mangoni et al., 2003; Marionneau et al., 2005; Chandleret al., 2009; Scavone et al., 2013) and also the relativeabsence of fast inward Na (NaV1.5) current, although thelatter is not absolute; several studies have reported thepresence of Na current and Na channel isoforms in SAN,depending on species and developmental stage (Baruscottiet al., 1996, 1997; Maier et al., 2003).More recent data suggest that the If and calcium

homeostasis pathways may not be independent, andthat it is the adenylyl cyclase (AC) signaling cascadethat provides the interconnection. Unlike workingmyocardium, SAN expresses several AC isozymes thatare activated by Ca2+ in the ;100 nM range, specifi-cally, AC1 and AC8 (Mattick et al., 2007; Younes et al.,2008). In comparison, the ventricle predominantlyexpresses AC5/6, which are inhibited by Ca . 1 mM.Thus, greater Ca2+ influx might not only increase thedepolarizing Na/Ca exchange current but also activateAC1/8, thus increasing cAMP and, through it, If. Thisobservationmost likely explains why disruption of calciumhomeostasis also reduces the ability of b-adrenergic agoniststo increase If (Kryukova et al., 2012). Further, thepresence of Ca2+-activated AC isozymes in SAN alsomay contribute to the fact that basal cAMP is higher inSAN than elsewhere in the heart Vinogradova et al.,2006; Kryukova et al., 2012). This characteristic of SANcells is itself important since it provides a means bywhich cholinergic agonists reduce the spontaneous rate,even in the absence of prestimulation by adrenergicagonists (DiFrancesco et al., 1989).Much of what has been described about the molecular

and ionic basis of SAN automaticity derives from animalstudies, for which the rabbit is the prototypical and bestcharacterized preparation. Considerably less data havebeen derived at the single cell level from human SAN,but recent reports of SAN dysfunction associated withHCN4 mutations (Baruscotti et al., 2010) and thera-peutic efficacy of the If blocker ivabradine (DiFrancesco,2010) support at a qualitative level a role for If inhuman SAN automaticity. Several studies from theWilders Laboratory have provided some quantitationfor human If, and they report that If is present in humanSAN but has a markedly lower current density andmore negative activation threshold than in rabbit,compatible with the difference in heart rates betweenthe species (Verkerk et al., 2007a,b). In an interestinganalysis, they used a computer simulation to assessthe current level of understanding of the functionalimpact of the reported HCN4 and KCNE2 (an HCN4b-subunit) mutations (Verkerk and Wilders, 2014).Although the simulations indeed predicted an impacton SAN function, they also highlighted inconsistencies

with the clinical data and illustrate the limited state ofour knowledge concerning the cellular and molecularbasis of automaticity and its autonomic regulation inhuman SAN.

Consideration of the specific transmembrane chan-nels and other proteins contributing to automaticityand its autonomic regulation is obviously relevant tocreating functional criteria by which to assess the“nodal” characteristics of cardiomyocytes derived fromstem cells. In addition, these proteins can serve aspositive or negative molecular markers either to aid inthe selection or subsequent identification of cells ofnodal lineage. In addition to those channel isoformsand other proteins described herein that are eitherhighly expressed or relatively absent in the SAN, thereare additional molecular markers of nodal cells such asneurofilament-M (NF-M) and the connexin isoformsCx45 and Cx30.2 (Marionneau et al., 2005; Boyett et al.,2006; Scavone et al., 2013) as well as transcriptionfactors that may also serve as molecular markers. Theseare summarized in Table 1 and described in subsequentsections.

III. Developmental Origin of the Sinoatrial Node

Ancients already knew that the heart was “endowedwith life,” as Aristotle observed many centuries ago andas Claudius Galen from Pergam (some 2000 years ago)confirmed when he recognized that both ventriclespulsate even when their nerves are severed or the heartis removed from the thorax. Thus, the power of pulsationhas its origin in the heart itself (Aird, 2011). It was notuntil 1882, however, that Gaskell proposed his theory onthe myogenic origin of the heartbeat (Gaskell, 1882).Furthermore, he identified the sinus-auricle (the sinusnode) as the region from which the contractions startand propagate to the ventricle-auricle [the atrioventric-ular (AV) node] and then to the ventricle. Finally, in1907, 1 year after Tawara (1906) described the AV node,Keith and Flack (1907) identified the SAN in the heartof a mole and of several other animals: “The nature ofthis remnant (of the sinus venosus) is perhaps bestexemplified in the heart of the mole. Here it is seen thatat the sinoauricular junction there is a mass of remark-able tissue. It appears to the eye as a very intimate networkof palely stained undifferentiated fibers with a largenumber of well-stained nuclei. It is totally differentfrom the surrounding musculature.”

Although the anatomic location and the physiologicfunction of the sinus node were clear, at the middle ofthe 20th century, the embryologic origin of the pace-maker tissue was less clear. On the basis of thecontraction rate of the tubular heart, it seemed that ata very early stage of development the pacemaker waslocated in the bulboventricular part (the prospectiveventricle), and only later, when atrium and sinusvenosus formed, the pacemaker shifted to this location


(Patten, 1949; DeHaan, 1965; Van Mierop, 1967). In1967, Van Mierop clearly demonstrated that in chickembryos at 8- and 9-somite stages—although heartseither beat faintly or did not visibly beat at all—it waspossible to use an intracellular microelectrode to recordaction potentials in both the bulboventricular andsinoatrial regions, with sinoatrial beats always pre-ceding ventricular ones (Van Mierop, 1967). A few yearslater, using an optical mapping technique, it was alsodemonstrated that spontaneous electrical activity inchick embryo can be measured in the cardiac primordia,even before their fusion to form the linear heart tube(Kamino, 1991). These data clearly demonstrated thata functional pacemaker region is already present atearly stages of development. Cells constituting theprimary myocardium of the early embryonic heart tubepresent structural and functional properties similar tothose of mature nodes (sinoatrial and AV). In particular,these cells display spontaneous beating characterizedby weak contractions and morphologically appear pale,with a high glycogen content and a poorly organizedsarcomeric structure.The preceding features and the lack of specific

markers have delayed the identification and localizationof sinoatrial precursors. One of the first markersidentified to be specifically expressed in the SAN and in

the other portions of the conduction system is NF-M,a protein expressed mainly in neuronal cells. Using insitu hybridization, it was demonstrated that in theadult rabbit, NF-M transcript is localized in myocytes ofall the components of the cardiac conduction system, theSAN and AV nodes, and the AV bundle and bundlebranches, but not in working (atrial and ventricular)cardiomyocytes. During rabbit embryonic development,NF-MmRNA is already detectable in a subpopulation ofcardiac myocytes at embryonic day (E) E9.5. The ex-pression of NF-M in the cardiac conduction system initiallysuggested a neuroectodermic origin of the conductionsystem, later proven incorrect (Gorza et al., 1988; Vitadelloet al., 1996).

As shown by VanMierop, themost important functionalcharacteristic of a pacemaker cell is its ability to re-peatedly generate the so-called prepotential or diastolicdepolarization. A prominent role in the generation ofthe diastolic depolarization is played by the If currentgenerated by HCN channels. Of the four known isoformsof HCN channels, HCN4 is the most expressed one in theadult sinus node of different species (Yamamoto et al.,2006; Liu et al., 2007; Brioschi et al., 2009; Chandleret al., 2009). On this basis, it has been shown thatexpression of HCN4 mRNA can be detected as early asE7.5 in the precardiac mesoderm (cardiac crescent) and

TABLE 1Positive and negative SAN and stem cell markers

EmbryonicSAN

PostnatalSAN Species References CD166

mESCaSHOX2-NEO

mESCbCD166hESCc

GATA6-GFPhESCd

Tbx18+ h, m Wiese et al., 2009; Mommersteeg et al., 2010; +Tbx3+ Tbx3+ m, r, h Hoogaars et al., 2004; Hoogaars et al., 2007,

Chandler et al., 2009+ + +

Shox2+ m Espinoza-Lewis et al., 2009 + +Isl-1+ Isl-1+ m Mommersteeg et al., 2007 + 2 +mef2C m Vedantham et al., 2013 + + +Nkx2.52 m Hoogaars et al., 2004; Liu et al., 2012 Low Low + +Pitx2c2 m Mommersteeg et al., 2007;HCN4+ HCN4+ h, m, r,

rbGarcia-Frigola et al., 2003; Yamamoto et al.,

2006; Liu et al., 2007; Brioschi et al., 2009;Chandler et al., 2009; Christoffels et al., 2010

+ + +

HCN1+ HCN1+ h, m, r,rb

Stieber et al., 2003 + + +

CaV1.3+ CaV1.3

+ h, m Mangoni et al., 2003; Marionneau et al., 2005;Chandler et al., 2009; Scavone et al., 2013

+

CaV3.2+ m Marionneau et al., 2005 +

CaV3.1+ CaV3.1

+ h Chandler et al., 2009 + +Cx30.2+ Cx30.2+ m Hoogaars et al., 2007; Christoffels et al., 2010 + +Cx402 Cx402 m, Hoogaars et al., 2007; Christoffels et al., 2010CX432 CX432 m, h Hoogaars et al., 2007; Chandler et al., 2009;

Christoffels et al., 20102 2

HCN22 HCN22 h, m, r,rb,

Stieber et al., 2003; Chandler et al., 2009 2

Kir2.12 r, rb, h Marionneau et al., 2005; Chandler et al., 2009

ssTnI+ M Marionneau et al., 2005 +NF-M+ NF-M+ rb Gorza et al., 1988; Vitadello et., al 1996

AC1 g Mattick et al., 2007AC8 g Mattick et al., 2007

CD166+ m, h Lin et al., 2012; Scavone etal., 2013 6Nppa2 Nppa2 m, h Hoogaars et al., 2004; Chandler et al., 2009;

Christoffels et al., 20102 +

g, goat; h, human; m, mouse; r, rat; rb, rabbit.aScavone et al., 2013.bHashem and Calycomb, 2013.cRust et al., 2009.dZhu et al., 2010.


then in the sinus venosus (E8). During the subsequentphases of development, HCN4 expression delineatesthe formation of the SAN and the conduction system,even though after E10, it also can be detected in thedeveloping nervous system (Garcia-Frigola et al., 2003;Liang et al., 2013). Thus, rather than being a specificmarker of SAN precursor cells, HCN4 is most likelya specific (functional) marker of pacemaker cells. Twoarticles have indeed shown that HCN4 is expressedearly during cardiogenesis in progenitors of the firstheart field (FHF) (Liang et al., 2013; Spater et al., 2013)well before formation of the sinus venosus and SAN.Lineage-tracing experiments in mice have clearly shownthat cells expressing HCN4 at gastrulation (E6.0) andat the crescent stage (E7.0) give rise, at later stages(E19.5), to compact and trabecular layers of the leftventricle and parts of both atria. HCN4 expression isthen downregulated in these progenitors and starts tobe upregulated in the Tbx18+ progenitors that will giverise to sinus venosus (Spater et al., 2013). Even if FHFprogenitors expressing HCN4 at very premature stages(,E7.5) will not be part of the mature cardiac conductionsystem, they represent the spontaneously rhythmic cellsthat start peristaltic contraction as soon as the hearttube is formed (Christoffels et al., 2010) (Fig. 1); however,HCN4 starts to play a fundamental functional roleonly when its expression shifts to the prospective SAN

region, as evidenced by the observation that bothglobal and cardiac-specific HCN4 knockout mice diein utero between E9.5 and E11.5, a time matching thedevelopment of the SAN (Stieber et al., 2003). Thesepieces of evidence clearly show that at the cardiaclevel, HCN4 is a marker that specifically character-izes pacemaker cells at stages of development whenthe sinus node is not yet formed (,E9.5).

Besides HCN4 expression, which confers intrinsicspontaneous activity to pacemaker cells, another fun-damental feature that makes pacemaker cells able topace and to drive neighboring cells in a functionalsyncytium is their junctional resistance. In fact, thesmall spontaneously beating area (source) and the muchlarger excitable surrounding tissue (sink) need to havea poor electrical coupling to allow the generation andpropagation of the excitatory stimulus (Joyner and vanCapelle, 1986). Among the Cxs expressed in the heart,Cx45 and Cx30.2 form low-conductance gap junctions,whereas Cx40 and Cx43 form high-conductance gapjunctions. Notably, gap junctions are scarce in the em-bryonic heart, expression of Cx40 and Cx43 is negligible(van Kempen et al., 1991; Delorme et al., 1997), andexpression of Cx45 is low and widespread (Alcolea et al.,1999). This is probably due to the fact that expression ofhigh-conductance Cxs is modulated by the same factorscontrolling the overall differentiation of the working

Fig. 1. (A) Schematic representation of the genes involved in the generation of spontaneous electrical activity and in the development of sinoatrialnode cells from mesodermal precursors in the mouse. Pacemaker activity is first detected at E7.5 in mesodermal cells of the FHF that express Nkx2-5and HCN4. At E8, posterior heart field (PHF) precursors characterized by expression of Tbx18 proliferate and at E8.5 they start to form the sinusvenosus (SV). At this stage, SV precursors expressing Tbx18, Shox2, HCN4, and CD166 become the leading pacemaker. From E9.5, a subgroup of SVprecursors start to express the second heart field (SHF) marker Isl-1 and transcription repressor Tbx3 generating the SAN precursors. With theprogression of development, these cells increase the level of expression of Tbx3 and HCN4 while decreasing the level of CD166 expression. Postnatal(PN) SAN cells are characterized by the expression of the transcription factors Isl-1, Tbx18, Shox2, Tbx3, and a high level of HCN4. (B) Schematicrepresentation of relation between factors inducing (↓) or inhibiting (’) SAN specification; the shaded area within the dotted line represents thesinoatrial lineage. Dotted arrows represent indirect associations with unknown mediators.


myocardium. It is known, for example, that expressionof Cx40 and Cx43 and other genes of the workingmyocardium, such as Nppa, are promoted by Tbx5 andNkx2.5, two transcription factors responsible for thespecification of the working myocardium but fullycomplementary to the expression of Tbx3, a transcriptionfactor specifically controlling the development of theconduction system (see later) (Hoogaars et al., 2004).Using transgenic mice, it was shown that Tbx3 de-ficiency causes the expansion of working myocardialgene expression (Cx40, Cx43, Nppa) into the SAN,whereas deficiency of Nkx2.5 causes ectopic expressionof Hcn4 and Tbx3 and decrease of Cx40 in the atrium(Hoogaars et al., 2007; Christoffels et al., 2010).Similarly, atrial ectopic expression of Tbx3 in mice led torepression of working myocardium genes and activation ofHcn4 and Cx30.2 (Hoogaars et al., 2007; Christoffels et al.,2010). Therefore, the expression of low-conductance Cxsand the low abundance of assembled gap junctions ensurethat the pacemaker cells are able to beat spontaneouslyand to drive the surrounding tissue during cardiacmorphogenesis.

A. Elucidation of the Gene Program Leading toSinoatrial Node Formation

With the advancement of lineage tracing techniquesand the help of transgenic mice, it was possible toidentify the progenitors of the sinus node cells wellbefore its formation. Mommersteeg and colleagues(2010) have shown that the sinus venosus develops fromTbx18+/Nkx2.52/Isl-12 progenitors that separate quiteearly (around E8) from the rest of the cardiac mesoderm.At E8.5, some cells among these progenitors start toexpress also Isl-1 (Tbx18+/Isl-1+/Nkx2.52), and preciselythis subgroup of cells will turn on the genetic pathwaythat leads to SAN development (Fig. 1).With progression of embryonic development, cardiac

chambers (atria and ventricles) start to develop byproliferation and the addition of new progenitor cells(Moorman and Christoffels, 2003). Two conditions mustbe met in the region originating the sinus node: 1) thegenetic program for chamber specification needs to beinhibited; and 2) the genetic program for maintenance ofpacemaker properties needs to be promoted. Twotranscription factors, Tbx3 and Shox-2, have been foundto be fundamental for these processes.At E9.5, the subset of Tbx18+/Isl-1+/Nkx2.52 cells

within the sinus venosus starts to express Tbx3, atranscription factor that represses chamber developmentby specifically inhibiting expression of the atrial genesNppa and Cx40. Expression of Tbx3 is continuous duringcardiac development in those regions that will form themature conduction system, from the SAN to the bundlebranches of the ventricular conduction system (Hoogaarset al., 2004). The importance of Tbx3 in maintaining thepacemaker features of future SAN cells is evidenced bythe fact that its lack does not prevent the morphologic

development of the SAN but causes ectopic expression ofatrial genes in the SAN region.

Shox2 is a transcriptional repressor that, when knockeddown, causes embryonic lethality from the developmentof a hypoplastic SAN causing an abnormally low heartrate. Interestingly, Shox2 knockdown causes lack ofTbx3 and HCN4 expression in the SAN and a parallelectopic expression of the atrial genes nppa and Cx43.During normal cardiac development, Shox2 is expressedstarting from E8.5 in the region at the junction betweenthe common cardinal vein and primitive atrium (Espinoza-Lewis et al., 2009). Its expression becomes restricted tothe inflow tract and in particular to the sinus venosus.With the progression of development, Shox2 expressionis restricted to the right side of the sinus venosus in theregion coinciding with the forming SAN. This expres-sion pattern also has been demonstrated during humanembryonic development (Liu et al., 2012).

Shox2, like Tbx18, has an expression pathway com-plementary to that of Nkx2.5, the best marker of theFHF that during development marks the whole heart,with the exception of the conduction system (Liu et al.,2012). Nkx2.5 expression turns on the genetic pathway,leading to cardiac chamber formation; it can bind toCx40 and Nppa promoters and activate their expression(Hoogaars et al., 2004). Overexpression of Nkx2.5 in theatrium causes alteration of SAN function by inducingthe expression of specific atrial genes, and for thisreason, it is important to repress Nkx2.5 expression inthe conduction system. Shox2 is responsible for therepression of the Nkx2.5 gene, thus protecting theSAN from activating the genetic pathway of the chambermyocardium and allowing expression of Tbx3 and HCN4,which instead are required for setting the functionalproperties of pacemaker cells (Espinoza-Lewis et al.,2009).

Tbx18, Isl-1, Shox2, and Tbx3 are important forspecifying the region where the sinus venosus and thenthe SAN develop. However, whereas the sinus venosus isa symmetric structure, the sinus node develops only inthe right horn of the sinus venosus. This asymmetry isensured by the homeobox factor Pitx2c (Fig. 1). Pitx2c isnot expressed in the forming sinus node; indeed, inPitx2c-deficient mice, two SAN primordia develop, one onthe right and another on the left side. This left-sidedSAN primordium expresses HCN4 but lacks Cx40. Atlater stages (E15.5), when the four-chambered heart isformed, the left-sided SAN is morphologically similarto the right-sided SAN and expresses the same set ofspecific genes. Pitx2c thus specifically suppresses theSAN gene program, allowing the correct asymmetricdevelopment of the conduction system (Mommersteeget al., 2007).

These data clearly demonstrate that pacemaker cellsdo not represent unspecialized primitive myocardialcells but rather derive from the activation of a specificgene pathway in progenitors that separate early during


cardiogenesis. This distinction is critical to efforts toproduce an enriched preparation of nodal-like cells fromstem cells since functionally nodal cells and immatureventricular cells are superficially similar in that bothcan exhibit spontaneous activity arising from a some-what depolarized maximum diastolic potential followedby a diastolic depolarization.

B. Sinoatrial Precursor

Is it possible to identify and separate SAN precursorsat early stages of development? The acquired knowledgeon the genetic pathway leading to the specification ofsinoatrial pacemaker cells has laid the foundations tospecifically select SAN precursors. The development ofspecific transgenic mice has been particularly relevant tothis aim. For example, transgenic mice expressing thereporter genes Lac-Z or green fluorescent protein (GFP)in the Tbx18 locus have allowed dissecting from E9.5embryos the Tbx18+ mesenchyme destined to form thesinus venosus and to evaluate protein expression patternin acuto or after in vitro maturation (Wiese et al., 2009;Mommersteeg et al., 2010). Interestingly, Tbx18+ mesen-chyme stained negative for Nkx2.5, HCN4, and thecardiac marker myosin heavy chain (MHC; MF20) anddid not show any spontaneous beating, whereas all theday-matched embryonic ventricle explants were beating.After 4 days in culture, however, a significant portion ofTbx18+ explants (around 50%) started to beat spontane-ously with a rate significantly faster than that of day-matched embryonic ventricle explants. Furthermore,4-day-old Tbx18+ explants started to express MF20 andHCN4 abundantly, but they did not express Nkx2.5,characteristic of the myocardium composing the sinusvenosus.Although these data point to Tbx18 as a promising

marker for selection of SAN precursors, its use hasa major drawback in that it is an intracellular markerand thus is not suitable for selecting cells withoutprevious genetic manipulation of the organism (mouse).This drawback hampers the use of Tbx18 as a selectionmarker of human SAN precursors. In 2006, Hirataand coworkers showed that expression of CD166, anadhesion molecule also called activated leukocyte celladhesion (ALCAM or DS-GRASP), marked the tubularheart and the sinus venosus at an early stage ofdevelopment (E8.5), whereas its expression broadenedto other organs at later stages (Hirata et al., 2006). Werecently demonstrated that in mice, during cardiacdevelopment at E10.5, CD166 expression almost com-pletely overlaps with that of HCN4; at later stages ofdifferentiation (E12.5), HCN4 and CD166 still colocalizein the SAN region, but CD166 expression widens also toother cardiac regions and to extracardiac organs as well(Scavone et al., 2013). These data indicate that, withina well defined developmental window, CD166 can indeedidentify pacemaker cells, in particular SAN precursors,even though it is not strictly a cardiac marker.

IV. Mouse Stem Cells

The possibility of regenerating the heart has becomefeasible with the discovery of the plasticity of stem cells.Spontaneously contracting cardiomyocytes were firstdetected on dimethyl sulfoxide–induced differentiationof mouse embryonic carcinoma cells (Edwards et al., 1983)and on spontaneous differentiation of mouse embryonicstem cells (mESCs) (Wobus et al., 1991). Although theability of pluripotent stem cells to differentiate intofunctional cardiomyocytes is well established, both invitro and in vivo, their clinical research and therapeuticuse are hampered by their teratogenic potential and byethical issues (Mummery et al., 2003; Behfar et al.,2007; Zhang et al., 2008).

Later, several types of stem or progenitor cells werealso isolated from various postnatal tissue or organs ofmesodermal origin. Mesoderm-derived adult stem cells,such as cardiac stem cells, mesenchymal stem cells (MSCs),skeletal myoblasts, hematopoietic stem cells, and endo-thelial progenitor cells, have been reported to possess thepotential of differentiating into cardiomyocytes eitherwhen injected in vivo into the heart of animal modelswith a myocardial infarction (Makino et al., 1999; Orlicet al., 2001; Kajstura et al., 2005) or when cocultured invitro in strict association with neonatal cardiomyocytes(Badorff et al., 2003; Matsuura et al., 2004; Koyanagiet al., 2005). Since the cardiogenic potential of these cellswas often assessed only at the transcription level, bylooking at the appearance of mRNAs for early cardiacmarkers, such as Nkx2.5, GATA-4, and SMA, and becausein some cases results were contradictory or data could notbe reproduced (Balsam et al., 2004; Murry et al., 2004),the initial excitement faded when it became clear that theappearance of a few mesodermal and cardiac markers didnot automatically translate to the induction of a realcardiogenic program resulting in the development offunctional cardiomyocytes.

As of today, controversy continues about the differ-entiation potential of some of these adult cells (Joggerstand Hatzopoulos, 2009). Some stem cells or progenitorcells were actually shown to fuse both in vitro and invivo with the adjacent cardiomyocytes (Matsuura et al.,2004; Avitabile et al., 2011) rather than differentiate.Even those adult stem cells that were actually able todifferentiate in vitro into cardiomyocytes usually re-quired either coculture with isolated cardiomyocytes(hematopoietic stem cells and MSCs) or treatment withagents such as methylation inhibitors and histonedeacetylase inhibitors (MSCs, cardiac stem cells) (Makinoet al., 1999; Badorff et al., 2003; Beltrami et al., 2003). Sofar, only three types of adult mouse stem cells haveshown functional differentiation into spontaneously beat-ing cardiomyocytes: mouse MSCs (Makino et al., 1999;Planat-Benard et al., 2004), mouse dedifferentiated fat cells(Jumabay et al., 2010), and mouse mesoangioblasts derivedfrom ventricular vessels and aorta (Barbuti et al., 2010).


Makino et al. (1999) showed for the first time thatclonally expanded MSCs isolated from mouse bonemarrow could differentiate into beating cardiomyo-cytes after induction with 5-azacytidine. MSCs devoidof hematopoietic cell contamination by serial passagingwere treated with 3 mM 5-azacytidine. After 1 week,cells lost the fibroblast-like morphology and acquiredan elongated shape; by 3 weeks in culture, cells formeda myotube-like morphology and were spontaneouslybeating, a feature that was maintained for the next5 weeks. These MSC-derived cardiomyocytes expressedseveral cardiac proteins, such as atrial natriuretic peptide,GATA4, Nkx2.5, Mef2C, b-MHC, and, to a lesser extent,a-MHC and a-actin. The great majority of these cardio-myocytes exhibited action potentials typical of sinoatrial-like pacemaker cells, even though with further maturationin culture around 30%–40% of cells had ventricular-likeaction potentials (Makino et al., 1999).In 2004, Planat-Bénard and colleagues demonstrated

that a small fraction (0.02%–0.07%) of the heterogeneouscell population composing the vascularized stroma ofadipose tissue can spontaneously differentiate intobeating cardiomyocytes. These adipose tissue–derivedcells, beyond expressing the typical cardiac markersNkx2.5, GATA4, Mef2C, and atrial natriuretic peptide(ANP), and the sarcomeric proteins b-MHC, atrial andventricular myosin light chain-2 (MLC-2v, MLC-2a),fired spontaneous action potentials showing a prominentdiastolic depolarization and were also responsive toadrenergic and muscarinic stimulation (Planat-Benardet al., 2004).Also from adipose tissue, but this time from the

adipocyte fraction, came the first report of the sponta-neous differentiation of pacemaker-like cells. Jumabayet al. (2010) reported that 10%–15% of dedifferentiatedfat cells spontaneously differentiated into beating cardiacmyocytes. A fraction of these cardiomyocytes showedsome important functional and electrical featurestypical of pacemaker cells in that they were indeed ableto fire spontaneous action potentials and to respond toadrenergic agonists with an increased rate. Moreover,these cardiomyocytes displayed synchronous sarcoplasmicCa2+ release, which indirectly indicates the presence ofa well organized sarcoplasmic reticulum and expressionof the proteins and pumps necessary for proper calciumhandling. Although a detailed molecular analysis forsinoatrial genes was not carried out, dedifferentiated fatcell–derived cardiomyocytes expressed high levels ofMef2C mRNA, a direct activator of HCN4 transcription,but only a faint signal for Nkx2.5, a known inhibitor ofHCN4 and sinoatrial development in general (see sectionII) (Jumabay et al., 2010).The last type of adult mouse stem cell to show a

differentiation potential toward the sinoatrial-like line-age was the cardiac mesoangioblast isolated frommouse aorta and ventricles (Barbuti et al., 2010). Thesemesoangioblasts displayed the highest cardiogenic

potential, with spontaneous differentiation rates in therange of 50%–80%, depending on the clone.Mesoangioblastsexpressed several early cardiac markers (Mesp1, Nkx2.5,GATA4, Tbx5, Cx43, and ANF), together with stem cellmarkers (Sca-1, c-kit) already at an undifferentiatedstage. Interestingly, these progenitors also expressedGATA6, Isl-1, Tbx2, and Tbx3, transcription factorsspecifically involved in the embryonic development ofthe cardiac conduction system and SAN (Davis et al.,2001; Christoffels et al., 2010). On spontaneous differen-tiation, triggered by lowering the serum concentration inthe culture medium, mesoangioblast-derived cardiomyo-cytes expressed cardiac-specific proteins, ion channels,and ion currents and were indeed able to fire eithertriggered or spontaneous action potentials with differentwaveforms (Galvez et al., 2008; Barbuti et al., 2010).Around 30% of differentiating mesangioblasts had themolecular and functional feature of sinoatrial cells.Spontaneously beating mesoangioblast-derived cardio-myocytes expressed mainly the HCN4 channels thatgenerated an If current with kinetics resembling thoseof mature SAN cells. Furthermore, like SAN cells, theydid not express the inward rectifying IK1 current andexpressed the low-conductance junctional protein Cx45.Finally, the spontaneous rate of mesoangioblast-derivedpacemaker cells was modulated by the autonomic agonistsisoproterenol and acetylcholine (Barbuti et al., 2010).

A. Evidence for Possibility of Generating Sinoatrial-Like Cells from Murine Embryonic Stem Cells

While the cardiac differentiation potential of non-ESCs is still debated, the potential of mESCs to generatespontaneously beating cardiomyocytes was establishedalmost 30 years ago and has been used by hundreds oflaboratories around the world with quite similar results.The first ESC lines were generated in 1981 by Evansand Kaufman (1981); these blastocyst-derived ESC linesspontaneously differentiate in vitro to form cystic EBs,cell aggregates recapitulating the initial stages of in vivoembryonic development that show a high frequency ofcardiac differentiation (Doetschman et al., 1985).

With the rapid progression of research on ESCs, pro-tocols for efficient cardiac differentiation of ESCs havebeen exploited. The best described and most widely usedmethod for generating spontaneously beating cardiomyo-cytes consists of culturing cells in hanging drops contain-ing a defined number of undifferentiated mESCs (a highyield of cardiomyocytes is usually obtained plating dropscontaining between 300 and 500 cells) that by gravity,after 2 days, form EBs. After 4 to 5 days of culture insuspension, EBs are plated on gelatin-coated dishes andwithin 1 to 2 days, foci of spontaneous beating cellsstart to appear. Following this protocol, 80%–90% ofthe EBs display areas characterized by spontaneousbeating (Wobus et al., 1991). This protocol representsthe gold standard for cardiac differentiation of mESCs;it works independently of the mESC line used and of the


laboratory applying it, and this reproducibility is whatset mESCs apart from adult stem cells.Wobus et al. (1991) provided the first evidence that

the spontaneously beating portions of the EBs generatespontaneous action potentials synchronous with cellcontraction and with the slow diastolic depolarizationtypical of pacemaker cells. Like pacemaker cells, theseESC-derived cardiomyocytes were modulated by auto-nomic neurotransmitters through the physiologicb-adrenergic-AC-cAMP pathway). Subsequent workhas shown that as EBs mature, cardiomyocytes becomeheterogeneous with respect to action potential wave-form; in particular, whereas cardiomyocytes at an earlydifferentiation stage (9–11 days) preferentially displaypacemaker-like action potentials, at later stages ofdifferentiation (16–20 days) atrial-ventricular- andsinoatrial-like action potentials can be distinguished(Maltsev et al., 1993; Hescheler et al., 1997). Abi-Gerges et al. (2000) reported for the first time that 65%of beating cardiomyocytes isolated from EBs at earlydifferentiation stages displayed a robust If, withproperties similar to those of sinoatrial cells. Withfurther development, the percentage of cells displayingIf current decreased to 45%, but the current densityincreased. Although spontaneous activity and thepresence of If are important features of pacemakercells, since embryonic and neonatal ventricular car-diomyocytes can also spontaneously generate actionpotentials and express the If (Cerbai et al., 1999), thesefeature are not sufficient to catalog a cell as sinoatrial-like; thus, several other conditions must be considered.As described in section II, expression of HCN4 andCx45 characterizes both sinoatrial precursors andsinoatrial cells throughout development and postna-tally (Alcolea et al., 1999; Garcia-Frigola et al., 2003;Christoffels et al., 2010). In mature SAN cells, expres-sion of HCN1 has also been reported (Ishii et al., 1999;Liu et al., 2007; Brioschi et al., 2009), whereas HCN2and Cx43 are prevalent in neonatal ventricular cells(Yasui et al., 2001).Van Kempen et al. (2003) provided the first evidence

that ESC-derived cardiomyocytes express HCN1 andHCN4 mRNA. Using a genetically modified ESC line inwhich GFP expression is driven by the a-MHC pro-moter. Yanagi et al. (2007) confirmed at the proteinlevel that mESC-derived spontaneously beating cellsdiffusely express the pacemaker channels HCN1 andHCN4 and also express the T-type calcium channelsCaV3.1 and 3.2, characteristic of SAN myocytes(Marionneau et al., 2005). Our group has shown thatnot only do mESCs express both HCN4 and HCN1channels, these channels are functionally coexpressedin caveolin-3-positive cardiomyocytes (Barbuti et al.,2009), a feature previously observed in rabbit sinoatrialcardiomyocytes (Barbuti et al., 2007). Electrophysiologicanalysis revealed that mESC-derived cardiomyocytescan be grouped into two distinct populations, one with

a fast-activating If current and the other with slow-activating If. The existence of these two populations ofpacemaker cells has been previously reported in adultmouse SAN cells (Mangoni and Nargeot, 2001), evidencethat further reinforces the similarity between ESC-derived automatic cells and mature murine pacemakercells.

It is important, however, to mention that differentresults have also been published concerning theexpression of HCN isoforms in ESC-derived cardio-myocytes. Two studies reported that mESC-derivedcardiomyocytes predominantly express the HCN2 andHCN3 isoforms both at the mRNA and protein levelwith faint or null expression of HCN4 and HCN1(White and Claycomb, 2005; Qu et al., 2008). Whetherthese differences arise from the different ESC linesused or from the selection of particular subtypes ofcardiomyocyte is presently unknown.

B. Improvements in the Generation ofSinoatrial-Like Cells

If the fact that mESCs can spontaneously differentiateinto beating EBs is well established, so is the fact thatwith time in culture, the number of beating EBsdecreases (Wobus et al., 1991; Barbuti et al., 2009). Thisdecrease in the number of pacemaker cardiomyocytesprompted researchers to find new approaches for enrich-ing cultures in cardiomyocytes. Different approacheshave been pursued in this direction: 1) cell-engineeringapproaches that use fluorescence molecules under thetranscriptional control of cardiac specific promoters;2) pharmacologic approaches based on the addition ofspecific molecules to the culture medium to drive specificcardiac differentiation; 3) selection of sinoatrial cellsbased on expression of endogenous markers.

In an attempt to isolate cardiac precursors, in 2003Hidaka et al. ( 2003) generated a mESC line expressingenhanced GFP (EGFP) under the control of the Nkx2.5gene. Although Nkx2.5 is a transcription factor involvedin chamber specification, the authors showed that earlyselection of EGFP (Nkx2.5)+ cells allowed the isolation ofa population of mixed cardiomyocytes that, based onelectrophysiologic features, represent ventricular-atrialand sinoatrial-like cardiomyocytes. This can be explainedby the fact that homologous recombination used togenerate the EGFP (Nkx2.5) clone inactivated one allele;indeed, these ESC cells express only half of the Nkx2.5expressed by “wild-type” ESC cells. As expected by therole of Nkx2.5 in cardiac development, after 28 days ofculture, only a small fraction of EGFP-selected cells stillshowed spontaneous activity. Despite the fact that thisapproach was not devised to isolate pacemaker cardio-myocytes, it demonstrated the possibility of actuallyenriching the cardiomyocyte population as evidencedby the fact that 98% of flow cytometry-selected cellsstained positive for myosin heavy chain and tropomyosin(Hidaka et al., 2003). In this same work, the authors also


demonstrated that a pharmacologic approach can alterthe proportion of a specific subpopulation of cardiomyo-cytes; for example, exogenous administration of retinoicacid (1027 M) in the culture medium during EB dif-ferentiation preferentially induced the atrial gene pro-gram (Hidaka et al., 2003).A second approach to isolate a specific subtype of

cardiomyocyte from ESCs was pursued by Gassanov andcolleagues (2004), who generated an mESC line in whichthe ANP promoter drove EGFP expression. Analysis ofEGFP-positive cells revealed the presence of cells with atriangle, spindle, and round morphology in the pro-portion of around 20%, 60%, and 20%, respectively.Triangle-shaped cells were either quiescent or fired ata relatively slow rate (around 1 Hz) and displayed actionpotentials with morphology and characteristics (restingpotential, overshoot, and duration) typical of atrial-likecardiomyocytes. All spindle-shaped cells fired insteadspontaneous action potentials at significantly higher rate(close to 3 Hz) and showed the marked slow diastolicdepolarization typical of sinus node cells. In agreementwith the action potential features, spindle-shaped cellshad a larger If current that activated with faster kineticsand at more positive potentials than triangle-shapedcells.A step closer to the isolation of a cell population with

pacemaker-like properties was accomplished by gener-ating an ESC line with EGFP under the transcriptionalcontrol of the a-MHC promoter (Kolossov et al., 2005).During mouse cardiac development, a-MHC is indeedhighly expressed in the atria and in the forming sinusnode but only faintly in the ventricles (Lyons et al.,1990). Using this ESC line, Kolossov et al. (2005) foundthat the selection of cells based on EGFP expressionyielded two populations of cells: 1) a round-cell populationwith strong EGFP expression that, based on multielec-trode array analysis, was recognized as the leadingpacemaker cells within the EBs; 2) and a population oftriangle-shaped cells displaying weak EGFP expressionand electrophysiologic properties of atrial-like cells. Asexpected, cells with ventricular-like action potential didnot express EGFP. Both EGFP bright and dim cellsexpressed a robust If current at early stages with kineticscompatible with the pacemaker activity (V1/2 = 278 mV),but at later differentiation stages spontaneous rate ofatrial-like cells declined in parallel with a decrease inIf density. Further evidence that EGFP bright round-shaped cell represent bona fide sinoatrial-like cellscame from the analysis of the IK1 current. In comparisonwith atrial-like cells, not only a significant smaller frac-tion of EGFP-bright cells expressed IK1 (30 versus.80%), but current density was also significantly lower(Kolossov et al., 2005). This work represents the firstproof of the possibility to specifically select or enrich apopulation of sinoatrial-like cardiomyocytes from mESCwith electrical features able to sustain spontaneouselectrical activity (Fig. 2).

Using a similar approach aimed at identifying a purepopulation of cardiac pacemaking-conduction system cells,White and Claycomb (2005) have generated a mESCline expressing the reporter genes LacZ and enhancedred fluorescence protein under the control of the minKpromoter (minK-LacZ) and the chicken GATA6 enhancer(cGATA6-enhanced red fluorescence protein), respec-tively. MinK, a b-subunit functionally interacting withHERG or KvLQT1 channels to give rise to native IKr andIKs currents, respectively, is expressed in the sinoatrial-conduction system regions both during development andin the adult heart (Kupershmidt et al., 1999). GATA6 hasbeen instead found in the developing AV conductionsystem (Davis et al., 2001; Adamo et al., 2004). Usingthese cell lines, the authors found that beating EBsshowed cells coexpressing GATA6 and minK near thecontracting area and that their separation preventedthe contraction of the EB, suggesting that they were thepacemaker region. From an electrophysiologic point ofview, GATA6/minK cells displayed heterogeneous prop-erties because even though they expressed If current,both sinoatrial- and atrial-like action potentials wererecorded. This heterogeneity becomes even clearer whenthe molecular profile of a pure population of GATA-6–expressing cells, selected for their resistance to neomycin(resulting from a neomycin-resistance cassette underthe GATA6 enhancer), was analyzed; cGATA-6 cellsexpressed moderate to high levels of ventricular andatrial Nkx2.5, Cx43, HCN2, and MLC-2a mRNAs;very low levels of the atrial ANF and Cx40 mRNAs;moderate levels of the conduction system mRNA Tbx3,Cx45, and CaV1.3; very low levels of the sinoatrialmRNA HCN4 and HCN1; curiously, very low levelsof minK mRNA were detected (White and Claycomb,2005).

As we have pointed out, HCN4 is a fundamental andspecific functional marker of the developing and matureSAN and conduction system (Garcia-Frigola et al., 2003;Morikawa et al., 2010). Morikawa et al. (2010) and, morerecently, our group (Scavone et al., 2013), generated ESClines in which EGFP expression was driven by the HCN4promoter. Differentiation of these cell lines gave rise toEBs with a strong EGFP signal in the spontaneouslybeating regions; these same EGFP+ regions expressedHCN4 and other proteins, such as caveolin-3, HCN1, andCaV3.2, characteristic of mature SAN cells. Althoughpromising, the selection strategy based on the activity ofthe HCN4 promoter failed to yield a pure or enrichedpopulation of sinoatrial-like cells; in fact, most of theEGFP+ cells, selected by flow cytometry, were quiescent.A possible explanation could be that, since EGFP wasclearly detectable only starting from day 7 of differenti-ation and peaked around day 13, HCN4 started to beexpressed also by other cell types (such as neuronalprecursors), as demonstrated during normal embry-onic development (Garcia-Frigola et al., 2003); indeed,this seems to be the case because the EGFP-selected


population expressed low levels of the neuronal markernestin (Morikawa et al., 2010). Moreover, since HCN4 isalso expressed in FHF progenitors, sorting of HCN4+

cells obtained either from mouse embryos or humanESC results in a high percentage of ventricular- andatrial-like cells and only a minor percentage of nodal-like cells (Spater et al., 2013). These data clearly showthe difficulty of finding a single selection marker that byitself recognizes a specific cell type, mainly becausemany genes are turned on and off at different phases ofcardiac development (Christoffels et al., 2010).More recently, Hashem and Calycomb (2013) engi-

neered mESC with a vector containing the neomycin-resistance gene under the control of a promoter region ofthe Shox2 gene. This approach allowed them to obtainShox2-expressing cells by selection with a culture me-dium supplemented with a high concentration of neo-mycin. The surviving cells, despite expressing cardiacmarkers Tbx5, GATA4, a-cardiac actin a-skeletal actin,Mlc-2a, Mlc-2v, and desmin, expressed several genes ofthe sinoatrial/conduction system program, includingTbx2 and Tbx3, GATA6, Cx45, Cx30.2, CaV1.3, CaV3.1,HCN4, and HCN2. At the same time, the ventricular andatrial genes Nkx2.5 and ANF were only modestlyexpressed. However, no functional data have yet beenshown that could ultimately identify these cells assinoatrial-like (Table 1).

As good as these selection methods are, they requirethe manipulation of the cell genome that in the best casescenario requires inactivation of one of the allele pairs,causing a haplo insufficiency and, in the worst case,the random insertion of exogenous DNA that makes themethod unsafe for future clinical use. Apart from theapproaches just described, nongenomic pharmacologicapproaches have also been developed to improve cardiacdifferentiation from ESCs. Although not initially thoughtto push specifically toward the pacemaker lineage, someof these approaches ended up showing a specific enrich-ment in sinoatrial-like cells.

As already described, the first demonstration camefrom the work of Hidaka et al. (2003) using retinoic acidto push cardiac differentiation toward the atrial lineage.A similar attempt, but in the pacemaker direction, wasdone by Gassanov et al. (2004) using an ANP-EGFP ESCline. They found that incubating ESCs from the beginningof the differentiation protocol to day 14 with the cytokineendothelin-1 (ET-1) dose dependently increased the pro-portion of the spindle-shaped EGFP+ cells correspondingto sinoatrial-like cells. With 100 nM ET-1, differentiationof sinoatrial-like cells increased significantly, becoming30% of the total EGFP+ cardiac population; at the sametime, triangle-shaped atrial-like cells decreased from 60%to 40%. ET-1 treatment also increased the expressionlevel of Cx45 and Cx40 but not that of the ventricular

Fig. 2. mESC-derived cardiomyocytes selected based on a-MHC–driven EGFP expression present different morphologies (A) and the levels offluorescence expression (B). Electrophysiologic analysis revealed that EGFP bright cells display pacemaker-like action potentials (C). Calibration barin (A), 10 mm. Reproduced with permission from Kolossov et al., 2005.


isoform Cx43 (Gassanov et al., 2004). The ET-1 effectwas very specific because coincubating cells with theselective ET-A and ET-B receptor antagonists BQ123[cyclo(D-Trp-D-Asp-Pro-D-Val-Leu)] and BQ788 (N-[(cis-2,6-dimethyl-1-piperidinyl)carbonyl]-4-methyl-L-leucyl-1-(methoxycarbonyl)-D-tryptophyl-D-norleucine) completelyprevented the ET-1–mediated increase in sinoatrial-likecells. This work provides the first proof of the possibilityof pharmacologically pushing mESC differentiation to-ward the nodal phenotype, even though the enrichmentwas still poor in terms of possible applications.Later, Kleger et al. (2010), while studying the role of

Ca2+-activated potassium channels during cardiac de-velopment, found that mESC treatment with EBIO(1-ethyl-2-benzimidazolinone), a drug that keeps previouslyactivated Ca2+-activated potassium (SK) channels in theopen state, increased 4- to 5-fold their cardiac commit-ment. Based on immunofluorescence and flow cytometryanalyses, they found that cells expressing a-actininincreased from 15% in control conditions to 65% with1 mM EBIO; similarly, troponin-positive cells increasedfrom 11% to 60%. Also, almost 60% of these EBIO-treated cardiomyocytes displayed pacemaker-like actionpotentials and expressed the If, compared with 7% ofnontreated cardiomyocytes. Molecular analysis showedthat most of the EBIO-treated cardiomyocytes showeda robust expression of HCN4 on the plasma membraneand of Cx30.2 between cells. The enrichment in sinoatrial-like cells is further supported by the fact that EBIO-treatment caused a significant upregulation of Tbx2,Tbx3, Shox2, and Isl-1, transcription factors deeplyinvolved in the specification of the cardiac conductionsystem. These effects were mediated by the SK4 isoform,the most abundantly expressed isoform in the atriumand conduction system, and in differentiating mESC.The specific involvement of SK4 was demonstrated usingclotrimazole, a specific SK4 inhibitor, and using a specificsmall hairpin RNA to knock down expression of thechannel. In fact both interventions were effective inpreventing EBIO-mediated cardiac differentiation andin preventing upregulation of HCN4 channels (Klegeret al., 2010).Finally, Wiese et al. (2011) tested on differentiating

ESCs the effect of suramin, a compound known to interactwith specific growth factors and cytokine receptors and,at high concentrations, to induce differentiation of heartstructures in Xenopus embryos (Grunz, 1992). Transienttreatment with 0.5 mM suramin from day 5 to day 7during EB differentiation increased the yield of sinoatrial-like cells that, as assessed by electrophysiologic analysisof action potential waveform, constituted between 50%and 70% of the total cardiomyocytes. In agreement withfunctional data, suramin treatment doubled the expres-sion levels of HCN4 and Tbx3 two well known markers ofsinus node cells. At the same time, suramin prevented thedifferentiation of mESC toward the neuronal and skeletalmuscle phenotypes (Wiese et al., 2011).

This and the other pharmacologic approaches, evenif they do not provide a sufficiently pure population ofsinus node–like cells, have the great advantage of notrequiring manipulation of the ESC genome and so, ifcombined with a proper selection or isolation method,may become attractive from a clinical point of view.

We have stated at the beginning of section IV.B thatESCs have the advantage of promptly differentiatinginto cardiomyocytes, but they are also particularlyinteresting for regenerative applications because oftheir high self-renewal capacity that ensures an un-limited growth potential in vitro. Their proliferation anddifferentiation potential, however, represent also thegreatest disadvantage of ESC because of the teratogenicrisk that accompanies pluripotency. For this reason,even though the proposed pharmacologic approachesrepresent quite effective methods to enrich ESC-derivedcardiomyocytes in sinoatrial-like cells, they do not pre-vent the contamination of undifferentiated, potentiallyteratogenic cells. On the other hand, modifications of thecell genome so as to express reporter genes specifically inthe lineage of interest allow specific selection of the cell ofinterest but are of little interest with respect to futureclinical applications. The ideal approach would consist ofspecific selection of the cell lineage of interest usingnative antigens to be used for cell sorting. Unfortunately,as of today, proteins that recognize specifically sinoatrialcells or even cardiomyocytes in general are not known.

So far, the only two surface markers found to beexpressed both in the heart and in ESC-derived cardio-myocytes are CD166 (or ALCAM or DS-GRASP) (Hirataet al., 2006; Rust et al., 2009; Lin et al., 2012; Scavoneet al., 2013) and CD172a (or signal-regulatory protein a[SIRPA]) (Dubois et al., 2011). The first evidence of theexpression of CD166 in the developing heart of mice camefrom work that compared the transcriptome of cardiacprogenitor cells with that of noncardiac cells (Masinoet al., 2004). Using transgenic mice expressing enhancedyellow fluorescent protein under the transcriptionalcontrol of an enhancer element of the transcription factorNkx2.5, they isolated by flow cytometry an almost purepopulation of cardiac progenitors at three developmentalstages: E7.75, E8.5, and E9.5. CD166 was one of thosegenes specifically enriched in cardiac progenitors andclearly marked the linear heart of E8.5 embryos (Masinoet al., 2004). In 2007, Murakami et al. (2007) demon-strated that CD166 expression can be used to select fromE8.5 yolk sack cardiovascular precursors that in cultureform clusters of cells showing spontaneous beating. Basedon these data and on our data showing that CD166 wasspecifically coexpressed with HCN4 at early develop-mental stages (E10.5) but not at later stages, our grouprecently produced a protocol that allows the isolation ofa pure population of SAN precursors from differentiatingESCs (Scavone et al., 2013). CD166+ ESCs, differentiatedaccording to the hanging drop method, were sortedbetween day 6 and day 8 of differentiation, the critical


time window in which CD166 was highly specific forsinoatrial precursors. Cells sorted by a fluorescence-activated cell sorter were reaggregated by gravity in low-adhesion dishes for 24 hours and then plated. The resultingCD166-selected cells formed a spontaneously beatingsyncytium that can be cultured for up to 1 month. Alreadyfrom the beginning, CD166+ cells expressed high levels ofthe transcription factors that govern SAN development,such as Tbx18, Shox2, Tbx3, isl-1, and several structuraland functional proteins that drive SAN cell function such asthe pacemaker channels HCN4 and HCN1, the calciumchannels CaV1.3 and CaV3.2, the connexin isoforms Cx30.2and Cx45, and the skeletal isoform of troponin I slow(ssTnI). At the same time, CD166+ cells expressed lowlevels of the atrial and ventricular genes Nkx2.5, HCN2,Cx43 Myh6, and Mlc2v (Table 1). Besides these molecularfeatures, when CD166-selected cells were compared withnative mouse sinoatrial cells, their functional propertieswere quite similar, in particular with respect to thecontribution of the If and L- and T-type calcium currentsto their electrical activity. Interestingly, CD166-selectedcells after 3 weeks of culture coexpress HCN4 andcaveolin3 and assume the spindle-shaped morphologytypical of native sinoatrial cells, a feature never observedbefore in vitro in stem cell–derived cardiomyocytes (Fig. 3).The only other nongenomic approaches to select an

enriched population of cardiomyocytes from pluripotentstem cells used tetramethylrhodamine methyl ester

perchlorate (TMRM), fluorescent molecules specificallymarking mitochondria (Hattori et al., 2010) and SIRPA(or CD172) (Dubois et al., 2011). Both methods yieldeda highly pure population of cardiomyocytes but withouta clear specificity for a particular subtype. TMRM in factwas effective in selecting cardiomyocytes from all the speciesanalyzed (mouse, marmoset, rat, and human), but the factthat it recognized Nkx2.5+/a-actinin+ cardiomyocytes atvarious developmental stages suggests a lack of specificityfor sinoatrial-like pacemaker cells (Hattori et al., 2010).Similarly, since SIRPA is expressed both in human atrialand ventricular cardiomyocytes from the fetal stages toadulthood, it may not be a selection marker for sinoatrial-like cardiomyocytes. Moreover, SIRPA was not detected inthe mouse heart, indicating that its expression is notconserved during evolution (Dubois et al., 2011). Un-fortunately, because functional data were not providedin either the TMRM or SIRPA reports, it remains to beestablished if one of these selection methods may beeffective at a particular differentiation time point inenriching the population of sinoatrial-like cells, so CD166-based selection appears at present to be the best methodavailable.

V. Human Stem Cells

The human equivalent of mESCs, human embryonicstem cells (hESCs), were first obtained in 1998 by

Fig. 3. (A) Schematic representation of the protocol used to obtain CD166-selected sinoatrial-like cardiomyocytes form mESC. (B) Top: Representativeaction potential recordings from an mESC-derived CD166-selected cells after 20 days of culture (left) and from an isolated adult sinoatrial cell (right)showing waveform similarity. Bottom: Confocal images of a portion of an aggregate of CD166-derived cells after 25 days in culture stained with anti-HCN4 and anti-caveolin3 antibodies; it is possible to appreciate the presence of HCN4/caveolin labeled cells with the typical SAN morphology.


Thomson and colleagues from human embryos pro-duced by in vitro fertilization (Thomson et al., 1998). Inthe following few years, translation and adaptation ofthe differentiation protocol developed with mESCssucceeded in obtaining human cardiomyocytes fromhESCs. At present, however, a universally reproduc-ible protocol, like the hanging drop method developedfor mESCs, is still lacking. In 2001, Kehat at al. (2001)demonstrated for the first time that hESCs can developspontaneously contracting regions that can be main-tained in culture for at least 5 weeks. At early stages,contracting regions were composed mainly of round-shaped mononucleated cells of around 10–30 mm indiameter, similar to early pacemaker cells found inmESCs(Gassanov et al., 2004; Kolossov et al., 2005). Even thoughthese hESC-derived cardiomyocytes expressed severalcardiac markers (GATA4, Nkx2.5, cTnI, cTnT, MLC-2aand -2v, a-MHC, and atrial natriuretic factor), presentedspontaneous and repetitive calcium transients, and werealso able to respond appropriately to adrenergic andmuscarinic stimulation, no information on specific sub-types of action potentials were provided (Kehat et al.,2001).In 2003, two studies separately showed that, similar

to mESC, hESCs differentiate into cardiomyocytes withaction potentials typical of ventricular-, atrial-, and alsopacemaker-like cells (He et al., 2003; Mummery et al.,2003) (Fig. 4). In 2004, Satin et al. measured for the firsttime ionic currents and ion channel expression fromhESC. They found that spontaneously beating hESC-derived cardiomyocytes express robust INa and If currentsbut not the inward rectifying current IK1. Curiously, theystated that the automaticity of these cells was mainlydependent on INa because whereas application of 3mMTTX slowed the beating rate and 10 mM TTX stopped itcompletely, neither nifedipine nor 2 mM Cs2+ (whichblock ICaL and If, respectively) had any effect on the actionpotential rate. mRNA analysis revealed that the hESC-derived cardiomyocytes abundantly expressed the typicalventricular isoforms NaV1.5, CaV1.2, and HCN2 of thesodium, calcium, and f channels, respectively, but theydid not express the typical sinoatrial isoforms CaV1.3 andHCN4, suggesting that the analysis focused on immatureventricular cardiomyocytes rather than nodal-like pace-maker cells (Satin et al., 2004). Later, Sartiani et al.(2007) carried out a study comparing spontaneouslybeating hESC-derived cardiomyocytes at early (15–30 days) and late (55–110 days) stages of differentiation.They found that early cardiomyocytes express a robustIf current that decreased in amplitude and activatedmuch slower after maturation. Analysis of the mRNAexpression indicated that maturation induced a decreasedexpression of HCN1 and HCN4 with time, whereas whileHCN2, which was the most highly expressed isoform,remained constant. In contrast to what was previouslyreported by Satin et al. (2004), inhibition of If by 10 mMzatebradine significantly reduced beating rate, meaning

that this current plays an important role in the diastolicdepolarization of these cells. From the overall electro-physiologic analysis and from the ion channel expressionpattern, and in agreement with previous reports, itemerges that hESC differentiated more toward the atrialand ventricular lineage than toward a sinoatrial-likephenotype (Sartiani et al., 2007).

In a recent work, the same group has shown thatalthough mRNA of HCN4 decreased during maturationof hESC-derived cardiomyocytes, HCN4 protein actuallyincreased (Bosman et al., 2013). After 60 days ofdifferentiation, HCN4 protein signal is widespread ina-actinin-positive cardiomyocytes; after 110 days ofdifferentiation, HCN4 is expressed at high density ina restricted number of cells, where it colocalizes withcaveolin-3, a feature typical of sinoatrial cells andsuggesting that the number of sinoatrial-like cellsmay be scarce, but present, in differentiating hESC.Recently, Weisbrod et al. (2013) have a provided a de-tailed characterization of the pacemaking mechanismsin early (days 11–21) hESC-derived cardiomyocytes.They found that although If is expressed in most beatingcells, inhibition of this conductance by the specific blockerszatebradine and ZD 7288 (4-ethylphenylamino-1,2-dimethyl-6-methylaminopyrimidinium chloride) sig-nificantly decreased the rate in only a subset (58%).Furthermore, they showed that pacemaker activity wasblocked by inhibition of ICaL by nifedipine but was notsensitive to block of sodium channels with 10 mM TTX.This evidence and the fact that they found a relativelyhigh expression of HCN2, HCN4, CaV1.3 NCX-1, andTbx3, which are abundant in human SAN and perinodalareas (Chandler et al., 2009), suggest that these cellsare early pacemaker or sinoatrial-like cells rather thanimmature atrial or ventricular cardiomyocytes.

Weisbrod et al. (2013) also found that all the hESC-derived spontaneously beating cells at early stages ofdifferentiation functionally expressed the Ca-activatedpotassium channel SK4, and specific inhibition of thischannel by either clotrimazole or TRAM-34 completelystopped spontaneous beating. Even though at presentno data on the expression of SK4 in native SAN cellsare available, the fact that in mESC activation of thischannel with EBIO significantly increased differentiationof sinoatrial-like cells (Kleger et al., 2010) and the fact thatat later stages of hESC differentiation only 30% of thebeating cells are responsive to TRAM-34 (Weisbrod et al.,2013) suggest that expression of SK4 channels mayprovide a novel way to distinguish functionally sinoatrial-like cells from immature working cardiomyocytes.

Regardless of the relatively low abundance of sinoatrial-like cells derived from hESC, it is important to mentionthat when the beating portions of hESC-derived EB werecoupled in vitro to neonatal ventricular cardiomyocytes orinjected in vivo in the hearts of swine (Kehat et al., 2004)and guinea pig (Xue et al., 2005), a regular spontaneousrhythm was induced. Even though the injected cellular


substrates were not selected and thus quite heterogeneousregarding cell composition, and despite the fact that theionic bases underlying the pacemaking activity were notaddressed, it is likely that sinoatrial-like cells contributedto the generation of spontaneous rhythmic activity, asevidenced by the fact that perfusion of the f-channelblocker ZD 7288 reduced the spontaneous rate andperfusion of isoproterenol was effective in increasingrate (Kehat et al., 2004; Xue et al., 2005).

A. Human Embryonic Stem Cell–DerivedCardiomyocyte Selection and Sinoatrial-LikeCell Enrichment

Not only do hESCs differ from mESCs in the lack ofa fully reproducible differentiation protocol for obtainingbeating cardiomyocytes, but also the difficulty of gener-ating hESC lines expressing reporter genes underspecific cardiac promoters is holding back the specificselection of sinoatrial-like cells. In 2007, Huber et al.(2007) showed that the generation of stable hESC linesexpressing EGFP under the transcriptional control ofthe human myosin light chain-2v promoter allowed theisolation of a population composed 95% of cardiomyo-cytes. Interestingly, EGFP expression was maintainedalso when EBs were generated from hESC lines culturedfor many passages, indicating the lack of transgeneshutdown, a limiting factor in the genetic modificationof hESC (Huber et al., 2007). In parallel, Anderson

and colleagues (2007) generated hESC lines expressingGFP and puromycin resistance under the transcrip-tional control of the a-MHC (MYH6) promoter. After atleast 11 days of differentiation, selection with puromycingenerated a population consisting 91% of spontaneouslybeating cardiomyocytes, which were kept in culture forup to 6 weeks and were responsive to isoproterenolstimulation. It is interesting to note that whereas theselection of a-MHC-EGFP mESC-derived cells resultsin the specific enrichment of atrial- and sinoatrial-likecells (Kolossov et al., 2005), hESC-derived puromycin(a-MHC)-selection results in a heterogeneous popula-tion of cardiomyocytes displaying ventricular-, atrial-,and sinoatrial-like action potentials (Anderson et al.,2007).

The aim of the preceding studies was to enrich ageneral cardiomyocyte population. Regarding enrich-ment methods for nodal-like cells, an advantage at leastin terms of potential therapeutic applications (e.g.,biologic pacemakers) is that the required number ofcells for implantation will be far smaller than for cardiacrepair. Zhu et al. (2010) focused their attention onenriching and selecting such a population of sinoatrial-like cardiomyocytes. The selection procedure was basedon exposure of differentiating cardiomyocytes to a lenti-viral vector containing the EGFP reporter under thecontrol of the cGATA-6 promoter, which allows thevisualization of nodal-like cells (White and Claycomb,2005; Barbuti et al., 2010). Their data clearly demon-strate that cGATA6-GFP+ cardiomyocytes presented sev-eral properties typical of nodal-like cells. From a molecularpoint of view, cGATA6-GFP+ cells expressed high levelsof HCN4 together with other cardiac markers (cardiactroponin T, sarcomeric actin, b-MHC), whereas froma functional point of view, 95% of EGFP+ cells displayednodal-like action potentials and expressed a robustIf current. Furthermore, like nodal cardiomyocytes,cGATA6-GFP+ cells displayed a low proliferative potential.The enrichment procedure was instead based on theinhibition of the pathway activated by the growth factorneuregulin 1b and its receptors ErbB (b/ErbB). Treatmentof hESC cells at day 5 of differentiation with antibodiesagainst neuregulin 1b or with the ErbB antagonistAG1478 (N-(3-chlorophenyl)-6,7-dimethoxyquinazolin-4-amine) caused a premature appearance of spontane-ously beating cells, increased by almost 3-fold thenumber of cardiomyocytes with nodal-like action poten-tials (from around 20% to 50%–60%) and induced asignificant upregulation of genes characteristic of sino-atrial cells such as Tbx3, HCN4, and CACNA1G (theCaV3.1 T-type calcium channel) while concurrentlydecreasing the working myocardium genes CACNA1C(CaV1.2 L-type calcium channel) and NaV1.5 sodiumchannel (Zhu et al., 2010). To date, this approach remainsthe most effective in generating sinoatrial-like cells fromhESC. Unfortunately, as is the case with many of theapproaches applied to mESC (see section IV), the selection

Fig. 4. Action potentials in hESC-derived beating cardiomyocytes andisolated human fetal ventricular and atrial cells. Reproduced with permissionfrom Mummery et al., 2003.


strategy suffers from the use of a lentiviral infectionprocedure that might not be suitable for future thera-peutic applications (despite the small number of cellsimplanted), although it would be entirely appropriate forin vitro drug screening applications. On the other hand,more recent approaches avoid random integration, forexample, the use of zinc-finger technology to targetthe AAVS1 locus on chromosome 19 (Hockemeyer et al.,2009).Cell-selection approaches using extracellular anti-

gens have also been exploited using both hESC andhuman iPSC, even though these strategies were notdevised for selecting pacemaker-like cardiomyocytes.The selection procedure using CD166 that works sowell in selecting sinoatrial-like cells from differentiat-ing mESC (Scavone et al., 2013) was actually first usedon hESCs for enriching the population of cardiomyo-cytes (Rust et al., 2009) and later on human iPSCs (Linet al., 2012). Rust and colleagues, for example, havesorted CD166+ cells from human ESCs after 12 days ofdifferentiation, a stage at which expression of CD166and other cardiac markers (Nkx2.5, Gata4, Isl-1, Mef2C,Tbx5, and a-MHC) was highest. Interestingly, CD166-selected cells expressed high levels of the HCN1 andHCN4 pacemaker genes and of the CACNA1C L-typecalcium gene (see Table 1). Even though a detailedanalysis of specific sinoatrial genes and electrophys-iologic data were not provided, the evidence thatreaggregated human CD166-selected cells keep beat-ing spontaneously for at least 3 weeks suggests thatCD166 may represent a good marker of sinoatrial-likecells (Rust et al., 2009). CD166 has also been used toselect a population of cells enriched in cardiomyocytesfrom differentiating human iPSCs previously selectedfor the cardiovascular marker KDR; also in this case,however, despite the evidence that more than 90% ofCD166+ cells expressed the cardiac troponin T protein(but also smooth muscle actin), no functional data onaction potential waveform or ionic currents were pro-vided (Lin et al., 2012).Two other extracellular proteins possibly suitable as

selection markers for sinoatrial-like cells are HNK-1(also known as Leu-7 or CD57) and VCAM1 (or CD106).HNK-1 has been shown to transiently mark the sinusvenosus and other precursors of the cardiac conductionsystem during human and rat development (Blom et al.,1999; Wenink et al., 2000); interestingly, in humans,HNK-1-positive precursors give rise also to the myocar-dium surrounding the pulmonary veins, a locus ofteninvolved in the generation of abnormal ectopic atrialautomaticity (Blom et al., 1999). Besides this evidence,it is not presently known whether it can be used to selectpacemaker-like cells from human pluripotent stem cells.The second possible marker is VCAM1. Using hiPSC-derived cardiomyocytes, Uosaki et al. (2011) found thatVCAM1 was the only protein among other 242 markersto be coexpressed with the cardiac isoform of troponin T

(TNNT2) and, indeed, VCAM1-positive cells, selected atdifferentiation day11, expressed several cardiac markersand displayed both ventricular- and pacemaker-likeaction potentials.

It is worth noting that all the markers mentioned(CD166, HNK-1, and VCAM1) are not typical cardiacproteins; instead, they are nonspecific adhesion mole-cules transiently expressed during development. Thisevidence suggests that timing of selection may be asimportant as the marker(s) used for selection.

In the future, it will be interesting to understand ifa combination of the pharmacologic and selectionapproaches described herein could succeed in isolatinga pure population of sinoatrial-like cells from plurip-otent cells of human origin.

B. Human Induced Pluripotent Stem Cellsand Automaticity

Human induced pluripotent stem cells have beenproduced for a number of years (Yu et al., 2007), andfurther progress has been made in recent years in termsof the methods and diversity of tissue sources, as well asin driving these cells down a cardiac lineage. A detaileddiscussion of protocols and potential pitfalls in theproduction and use of human iPSCs is beyond the scopeof this article but has been the subject of recent reviewsand editorials (Yoshida and Yamanaka, 2010, 2011;Mummery, 2011; Xu et al., 2011; Sinnecker et al., 2012).With respect to the specific issue of electrophysiologiccharacteristics of these cells, there also have been severalrecent reviews (Poon et al., 2011; Hoekstra et al., 2012;Zeevi-Levin et al., 2012). In the following discussion,we focus specifically on reports concerning aspects ofautomaticity and the presence of nodal-like cells inthese preparations since that is most relevant to thecurrent topic.

Zhang et al. (2009) used OCT4, SOX2, NANOG, andLIN28 transgenes to reprogram human IPSCs obtainedfrom fetal and newborn tissue samples and comparedthe results both molecularly and electrophysiologicallywith the H1 and H9 hESCs lines. They reported recordingaction potentials with ventricular, atrial, and nodal-likecharacteristics from both preparations, and the relativepercentage of each cell type was similar in hESCs andhiPSCs. In both, most of the cells were ventricular-like(;60% of ESCs and ;72% of iPSCs) and nodal-like cellsrepresented 10%–20% of the population (13% and 20% inthe 2 iPSC lines studied and 11 and 20% in the two ESClines). The spontaneous rate also was similar in bothpreparations, as well as chronotropic responsiveness toisoproterenol. However, although they measured variouscardiac and sarcomeric markers, they did not test for theexpression of sinoatrial-specific markers.

In 2011, Ma et al. (2011) conducted a similar studyon iPSCs derived from a human fibroblast cell line,although without a direct comparison with hESC lines.They also reported a mix of ventricular, atrial, and


nodal-like cells in similar proportion to the earlier study(54% ventricular, 22% nodal). As with the prior study,the definition of nodal-like cells was based entirely onaction potential parameters. Although they did alsomeasure individual currents, including the If, they appearto have restricted such studies to the ventricular-like cellsin their population.Lee et al. (2011) also compared hESC and two hiPSC

cell lines (IMR90 and KS1) but they focused on theCa2+-handling characteristics of the cells, a key as-pect of sinoatrial automaticity. They found that theCa2+-handling characteristics of hiPSCs were immaturerelative to those of hESCs. In particular, calcium transientswere smaller and had slower upstroke velocities. Mostimportantly, ryanodine was less effective in reducingthe calcium transients in the hiPSCs. Surprisingly, Leeand colleagues reported a higher percentage of nodal-like cells (again based on action potential character-istics) than earlier studies, both in hiPSCs and hESCs(45% and 43%, respectively).A recent study reported on the derivation and char-

acterization of cardiomyocytes from iPSCs from kerati-nocytes obtained from adult human hair samples(Mandel et al., 2012). Besides the advantage of a readilyaccessible tissue source from adults, this study com-pared automaticity and beat rate variability in hiPSCsand hESCs. They found the beating rates of both prep-arations were similar and were stable during the 15-daystudy period. They also reported that both preparationsexhibited similar intrinsic beat rate variability, as pre-viously reported in human populations.Overall, the current knowledge concerning the nodal-

like nature of cardiomyocytes derived from hiPSCs, andin particular hiPSCs from adult tissue sources, is limited.The results suggest that cells with nodal-like actionpotentials are certainly present in these populations, butit remains to be determined whether these are cells oftrue nodal lineage or simply immature “ventricular-like”cells with pacemaker activity. The one report of immaturecalcium handling in hiPSCs (Lee et al., 2011) also raisesa cautionary note that merits further study.It should be noted that a number of other studies

have used patient-specific iPSCs to study cardiomyocytesfrom diseased individuals, including LEOPARD syn-drome (multiple lentigines syndrome), familial dilatedcardiomyopathy, and long QT syndrome (Carvajal-Vergara et al., 2010; Sun et al., 2012; Terrenoire et al.,2013). This is a promising future direction in terms offurthering understanding these diseases and perhapsdeveloping more targeted therapeutic approaches but doesnot directly relate to the issue of producing an enrichedpopulation of nodal-like cells. Similarly, once nodal-likecells (of either iPS or ES origin) can be routinely prepared,opportunities also arise for studying acquired sinusdysfunction. However, full realization of this potentialmay require production of more complex in vitro prep-arations involving mixed cell types (i.e., fibroblasts and

atrial-like cells along with nodal-like cells) since in situnodal function depends on such heterogeneous structuresand pathology involves both structural and electricalremodeling (Kohl et al., 2005; Oren and Clancy, 2010;Hao et al., 2011).

VI. Conclusions

Much of the research on the derivation of cardiomyo-cytes from various stem cell sources has been directedat ventricular-like cells, which in fact represents themajority of the cells in the typical heterogeneous cellpopulation. Certainly, such cells have tremendous thera-peutic potential in terms of improving cardiac mechanicalfunction in the diseased heart. Research on pacemakingcharacteristics has been more recent and to a large extentdirected toward the potential use of such cell preparationsas source material for creating a biologic alternative toelectronic pacemakers. Unfortunately, since our knowl-edge on the genetic pathway leading to SAN formationhas been partly clarified only in the last decade, there hasnot always been a distinction made between pacemaker-like and nodal-like cells, and so it remains uncertain theextent to which these preparations, which have reported“nodal-like” action potentials in anywhere from a fewpercent to as high as 45% of the cells, truly contain cells ofnodal lineage. This distinction is not trivial with respect tothe biologic pacemaker application since such a pacemakerbased on immature but automatic cells of ventricularlineage might subsequently become excitable but non-spontaneous mature ventricular cells.

Furthermore, because most of the research and proto-cols for stem cell differentiation toward nodal lineage havebeen developed in mouse cells, only a few studies haveattempted to select or enrich specifically for human cells ofnodal lineage, based on intracellular or surface markers.These results are promising, but much more progressneeds to be made both on translating mouse research tohuman and in determining the optimal markers to use,as well as developing methods to enrich the prepara-tions for these markers. If these efforts prove successful,there would be a diverse range of applications for a humanpurified cell line of nodal lineage that recapitulates thekey characteristics of SAN pacemaking.

Authorship Contributions

Wrote or contributed to the writing of the manuscript: Barbuti,Robinson.

ReferencesAbi-Gerges N, Ji GJ, Lu ZJ, Fischmeister R, Hescheler J, and Fleischmann BK (2000)Functional expression and regulation of the hyperpolarization activated non-selective cation current in embryonic stem cell-derived cardiomyocytes. J Physiol523:377–389.

Adamo RF, Guay CL, Edwards AV, Wessels A, and Burch JB (2004) GATA-6 geneenhancer contains nested regulatory modules for primary myocardium and theembedded nascent atrioventricular conduction system. Anat Rec A Discov Mol CellEvol Biol 280:1062–1071.

Aird WC (2011) Discovery of the cardiovascular system: from Galen to WilliamHarvey. J Thromb Haemost 9 (Suppl 1):118–129.

Alcoléa S, Théveniau-Ruissy M, Jarry-Guichard T, Marics I, Tzouanacou E, ChauvinJP, Briand JP, Moorman AF, Lamers WH, and Gros DB (1999) Downregulation of


connexin 45 gene products during mouse heart development. Circ Res 84:1365–1379.

Anderson D, Self T, Mellor IR, Goh G, Hill SJ, and Denning C (2007) Transgenicenrichment of cardiomyocytes from human embryonic stem cells. Mol Ther 15:2027–2036.

Avitabile D, Crespi A, Brioschi C, Parente V, Toietta G, Devanna P, Baruscotti M,Truffa S, Scavone A, Rusconi F, et al. (2011) Human cord blood CD34+ progenitorcells acquire functional cardiac properties through a cell fusion process. Am JPhysiol Heart Circ Physiol 300:H1875–H1884.

Badorff C, Brandes RP, Popp R, Rupp S, Urbich C, Aicher A, Fleming I, Busse R,Zeiher AM, and Dimmeler S (2003) Transdifferentiation of blood-derived humanadult endothelial progenitor cells into functionally active cardiomyocytes. Circu-lation 107:1024–1032.

Balsam LB, Wagers AJ, Christensen JL, Kofidis T, Weissman IL, and Robbins RC(2004) Haematopoietic stem cells adopt mature haematopoietic fates in ischaemicmyocardium. Nature 428:668–673.

Barbuti A, Crespi A, Capilupo D, Mazzocchi N, Baruscotti M, and DiFrancesco D(2009) Molecular composition and functional properties of f-channels in murineembryonic stem cell-derived pacemaker cells. J Mol Cell Cardiol 46:343–351.

Barbuti A, Galvez BG, Crespi A, Scavone A, Baruscotti M, Brioschi C, Cossu G,and DiFrancesco D (2010) Mesoangioblasts from ventricular vessels can differen-tiate in vitro into cardiac myocytes with sinoatrial-like properties. J Mol CellCardiol 48:415–423.

Barbuti A, Terragni B, Brioschi C, and DiFrancesco D (2007) Localization off-channels to caveolae mediates specific beta2-adrenergic receptor modulation ofrate in sinoatrial myocytes. J Mol Cell Cardiol 42:71–78.

Baruscotti M, Barbuti A, and Bucchi A (2010) The cardiac pacemaker current. J MolCell Cardiol 48:55–64.

Baruscotti M, DiFrancesco D, and Robinson RB (1996) A TTX-sensitive inward so-dium current contributes to spontaneous activity in newborn rabbit sino-atrialnode cells. J Physiol 492:21–30.

Baruscotti M, Westenbroek R, Catterall WA, DiFrancesco D, and Robinson RB (1997)The newborn rabbit sino-atrial node expresses a neuronal type I-like Na+ channel.J Physiol 498:641–648.

Behfar A, Perez-Terzic C, Faustino RS, Arrell DK, Hodgson DM, Yamada S, PuceatM, Niederländer N, Alekseev AE, Zingman LV, et al. (2007) Cardiopoietic pro-gramming of embryonic stem cells for tumor-free heart repair. J Exp Med 204:405–420.

Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, Chimenti S, Kasahara H,Rota M, Musso E, Urbanek K, et al. (2003) Adult cardiac stem cells are multipotentand support myocardial regeneration. Cell 114:763–776.

Blom NA, Gittenberger-de Groot AC, DeRuiter MC, Poelmann RE, Mentink MM,and Ottenkamp J (1999) Development of the cardiac conduction tissue in humanembryos using HNK-1 antigen expression: possible relevance for understanding ofabnormal atrial automaticity. Circulation 99:800–806.

Bogdanov KY, Vinogradova TM, and Lakatta EG (2001) Sinoatrial nodal cell rya-nodine receptor and Na(+)-Ca(2+) exchanger: molecular partners in pacemakerregulation. Circ Res 88:1254–1258.

Bosman A, Sartiani L, Spinelli V, Del Lungo M, Stillitano F, Nosi D, Mugelli A,Cerbai E, and Jaconi M (2013) Molecular and functional evidence of HCN4 andcaveolin-3 interaction during cardiomyocyte differentiation from human embryonicstem cells. Stem Cells Dev 22:1717–1727.

Boyett MR, Dobrzynski H, Lancaster MK, Jones SA, Honjo H, and Kodama I (2003)Sophisticated architecture is required for the sinoatrial node to perform its normalpacemaker function. J Cardiovasc Electrophysiol 14:104–106.

Boyett MR, Honjo H, and Kodama I (2000) The sinoatrial node, a heterogeneouspacemaker structure. Cardiovasc Res 47:658–687.

Boyett MR, Inada S, Yoo S, Li J, Liu J, Tellez J, Greener ID, Honjo H, Billeter R, LeiM,, et al. (2006) Connexins in the sinoatrial and atrioventricular nodes. Adv Car-diol 42:175–197.

Brioschi C, Micheloni S, Tellez JO, Pisoni G, Longhi R, Moroni P, Billeter R, Barbuti A,Dobrzynski H, Boyett MR, et al. (2009) Distribution of the pacemaker HCN4 channelmRNA and protein in the rabbit sinoatrial node. J Mol Cell Cardiol 47:221–227.

Bucchi A, Baruscotti M, Robinson RB, and DiFrancesco D (2003) I(f)-dependentmodulation of pacemaker rate mediated by cAMP in the presence of ryanodine inrabbit sino-atrial node cells. J Mol Cell Cardiol 35:905–913.

Camelliti P, McCulloch AD, and Kohl P (2005) Microstructured cocultures of cardiacmyocytes and fibroblasts: a two-dimensional in vitro model of cardiac tissue.Microsc Microanal 11:249–259.

Carvajal-Vergara X, Sevilla A, D’Souza SL, Ang YS, Schaniel C, Lee DF, Yang L,Kaplan AD, Adler ED, Rozov R, et al. (2010) Patient-specific induced pluripotentstem-cell-derived models of LEOPARD syndrome. Nature 465:808–812.

Cerbai E, Pino R, Sartiani L, and Mugelli A (1999) Influence of postnatal-developmenton I(f) occurrence and properties in neonatal rat ventricular myocytes. CardiovascRes 42:416–423.

Chandler NJ, Greener ID, Tellez JO, Inada S, Musa H, Molenaar P, Difrancesco D,Baruscotti M, Longhi R, Anderson RH, et al. (2009) Molecular architecture of thehuman sinus node: insights into the function of the cardiac pacemaker. Circulation119:1562–1575.

Chow M, Boheler KR, and Li RA (2013) Human pluripotent stem cell-derived car-diomyocytes for heart regeneration, drug discovery and disease modeling: from thegenetic, epigenetic, and tissue modeling perspectives (Abstract). Stem Cell ResTher 4:97.

Christoffels VM, Smits GJ, Kispert A, and Moorman AF (2010) Development of thepacemaker tissues of the heart. Circ Res 106:240–254.

Cohen IS and Robinson RB (2006) Pacemaker current and automatic rhythms: to-ward a molecular understanding, in Basis and Treatment of Cardiac Arrhythmias(Clancy CE and Kass RS, eds) pp 41–72, Springer-Verlag, Berlin.

Davis DL, Edwards AV, Juraszek AL, Phelps A, Wessels A, and Burch JB (2001)A GATA-6 gene heart-region-specific enhancer provides a novel means to mark

and probe a discrete component of the mouse cardiac conduction system. MechDev 108:105–119.

DeHaan RL (1965) Development of pacemaker tissue in the embryonic heart. Ann NY Acad Sci 127:7–18.

Delorme B, Dahl E, Jarry-Guichard T, Briand JP, Willecke K, Gros D, and Théveniau-Ruissy M (1997) Expression pattern of connexin gene products at the early de-velopmental stages of the mouse cardiovascular system. Circ Res 81:423–437.

DiFrancesco D (2010) The role of the funny current in pacemaker activity. Circ Res106:434–446.

DiFrancesco D, Ducouret P, and Robinson RB (1989) Muscarinic modulation of car-diac rate at low acetylcholine concentrations. Science 243:669–671.

Doetschman TC, Eistetter H, Katz M, Schmidt W, and Kemler R (1985) The in vitrodevelopment of blastocyst-derived embryonic stem cell lines: formation of visceralyolk sac, blood islands and myocardium. J Embryol Exp Morphol 87:27–45.

Dubois NC, Craft AM, Sharma P, Elliott DA, Stanley EG, Elefanty AG, Gramolini A,and Keller G (2011) SIRPA is a specific cell-surface marker for isolating car-diomyocytes derived from human pluripotent stem cells. Nat Biotechnol 29:1011–1018.

Edwards MK, Harris JF, and McBurney MW (1983) Induced muscle differentiationin an embryonal carcinoma cell line. Mol Cell Biol 3:2280–2286.

Espinoza-Lewis RA, Yu L, He F, Liu H, Tang R, Shi J, Sun X, Martin JF, Wang D,Yang J, et al. (2009) Shox2 is essential for the differentiation of cardiac pacemakercells by repressing Nkx2-5. Dev Biol 327:376–385.

Evans MJ and Kaufman MH (1981) Establishment in culture of pluripotential cellsfrom mouse embryos. Nature 292:154–156.

Galvez BG, Sampaolesi M, Barbuti A, Crespi A, Covarello D, Brunelli S, Dellavalle A,Crippa S, Balconi G, Cuccovillo I, et al. (2008) Cardiac mesoangioblasts are com-mitted, self-renewable progenitors, associated with small vessels of juvenile mouseventricle. Cell Death Differ 15:1417–1428.

Garcia-Frigola C, Shi Y, and Evans SM (2003) Expression of the hyperpolarization-activated cyclic nucleotide-gated cation channel HCN4 during mouse heart de-velopment. Gene Expr Patterns 3:777–783.

Gaskell WH (1882) Preliminary Observations on the Innervation of the Heart of theTortoise. J Physiol 3:369–379.

Gassanov N, Er F, Zagidullin N, and Hoppe UC (2004) Endothelin induces differ-entiation of ANP-EGFP expressing embryonic stem cells towards a pacemakerphenotype. FASEB J 18:1710–1712.

Gorza L, Schiaffino S, and Vitadello M (1988) Heart conduction system: a neural crestderivative? Brain Res 457:360–366.

Grunz H (1992) Suramin changes the fate of Spemann’s organizer and preventsneural induction in Xenopus laevis. Mech Dev 38:133–141.

Hao X, Zhang Y, Zhang X, Nirmalan M, Davies L, Konstantinou D, Yin F, DobrzynskiH, Wang X, Grace A, et al. (2011) TGF-b1-mediated fibrosis and ion channelremodeling are key mechanisms in producing the sinus node dysfunction associ-ated with SCN5A deficiency and aging. Circ Arrhythm Electrophysiol 4:397–406.

Hashem SI and Claycomb WC (2013) Genetic isolation of stem cell-derivedpacemaker-nodal cardiac myocytes. Mol Cell Biochem 383:161–171.

Hattori F, Chen H, Yamashita H, Tohyama S, Satoh YS, Yuasa S, Li W, YamakawaH, Tanaka T, Onitsuka T, et al. (2010) Nongenetic method for purifying stem cell-derived cardiomyocytes. Nat Methods 7:61–66.

He JQ, Ma Y, Lee Y, Thomson JA, and Kamp TJ (2003) Human embryonic stem cellsdevelop into multiple types of cardiac myocytes: action potential characterization.Circ Res 93:32–39.

Hescheler J, Fleischmann BK, Lentini S, Maltsev VA, Rohwedel J, Wobus AM,and Addicks K (1997) Embryonic stem cells: a model to study structural andfunctional properties in cardiomyogenesis. Cardiovasc Res 36:149–162.

Hidaka K, Lee JK, Kim HS, Ihm CH, Iio A, Ogawa M, Nishikawa S, Kodama I,and Morisaki T (2003) Chamber-specific differentiation of Nkx2.5-positive cardiacprecursor cells from murine embryonic stem cells. FASEB J 17:740–742.

Hirata H, Murakami Y, Miyamoto Y, Tosaka M, Inoue K, Nagahashi A, Jakt LM,Asahara T, Iwata H, Sawa Y, et al. (2006) ALCAM (CD166) is a surface marker forearly murine cardiomyocytes. Cells Tissues Organs 184:172–180.

Hockemeyer D, Soldner F, Beard C, Gao Q, Mitalipova M, DeKelver RC, Katibah GE,Amora R, Boydston EA, Zeitler B, et al. (2009) Efficient targeting of expressed andsilent genes in human ESCs and iPSCs using zinc-finger nucleases. Nat Biotechnol27:851–857.

Hoekstra M, Mummery CL, Wilde AA, Bezzina CR, and Verkerk AO (2012) Inducedpluripotent stem cell derived cardiomyocytes as models for cardiac arrhythmias(Abstract). Front Physiol 3:346.

Hoogaars WM, Engel A, Brons JF, Verkerk AO, de Lange FJ, Wong LY, BakkerML, Clout DE, Wakker V, Barnett P, et al. (2007) Tbx3 controls the sinoatrialnode gene program and imposes pacemaker function on the atria. Genes Dev 21:1098–1112.

Hoogaars WM, Tessari A, Moorman AF, de Boer PA, Hagoort J, Soufan AT,Campione M, and Christoffels VM (2004) The transcriptional repressor Tbx3delineates the developing central conduction system of the heart. Cardiovasc Res62:489–499.

Huber I, Itzhaki I, Caspi O, Arbel G, Tzukerman M, Gepstein A, Habib M, YankelsonL, Kehat I, and Gepstein L (2007) Identification and selection of cardiomyocytesduring human embryonic stem cell differentiation. FASEB J 21:2551–2563.

Ishii TM, Takano M, Xie LH, Noma A, and Ohmori H (1999) Molecular character-ization of the hyperpolarization-activated cation channel in rabbit heart sinoatrialnode. J Biol Chem 274:12835–12839.

Itzhaki I, Maizels L, Huber I, Zwi-Dantsis L, Caspi O, Winterstern A, Feldman O,Gepstein A, Arbel G, Hammerman H, et al. (2011) Modelling the long QT syndromewith induced pluripotent stem cells. Nature 471:225–229.

Joggerst SJ and Hatzopoulos AK (2009) Stem cell therapy for cardiac repair: benefitsand barriers. Expert Rev Mol Med 11:e20.

Joyner RW and van Capelle FJ (1986) Propagation through electrically coupled cells.How a small SA node drives a large atrium. Biophys J 50:1157–1164.


Ju YK and Allen DG (1999) How does beta-adrenergic stimulation increase the heartrate? The role of intracellular Ca2+ release in amphibian pacemaker cells. J Physiol516:793–804.

Jumabay M, Zhang R, Yao Y, Goldhaber JI, and Boström KI (2010) Spontaneouslybeating cardiomyocytes derived from white mature adipocytes. Cardiovasc Res 85:17–27.

Kajstura J, Rota M, Whang B, Cascapera S, Hosoda T, Bearzi C, Nurzynska D,Kasahara H, Zias E, Bonafé M, et al. (2005) Bone marrow cells differentiate incardiac cell lineages after infarction independently of cell fusion. Circ Res 96:127–137.

Kamino K (1991) Optical approaches to ontogeny of electrical activity and relatedfunctional organization during early heart development. Physiol Rev 71:53–91.

Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M, Gepstein A, Livne E, BinahO, Itskovitz-Eldor J, and Gepstein L (2001) Human embryonic stem cells can dif-ferentiate into myocytes with structural and functional properties of car-diomyocytes. J Clin Invest 108:407–414.

Kehat I, Khimovich L, Caspi O, Gepstein A, Shofti R, Arbel G, Huber I, Satin J,Itskovitz-Eldor J, and Gepstein L (2004) Electromechanical integration of car-diomyocytes derived from human embryonic stem cells. Nat Biotechnol 22:1282–1289.

Keith A and Flack M (1907) The form and nature of the muscular connections be-tween the primary divisions of the vertebrate heart. J Anat Physiol 41:172–189.

Kleger A, Seufferlein T, Malan D, Tischendorf M, Storch A, Wolheim A, Latz S,Protze S, Porzner M, Proepper C, et al. (2010) Modulation of calcium-activatedpotassium channels induces cardiogenesis of pluripotent stem cells and enrichmentof pacemaker-like cells. Circulation 122:1823–1836.

Kohl P, Camelliti P, Burton FL, and Smith GL (2005) Electrical coupling of fibro-blasts and myocytes: relevance for cardiac propagation. J Electrocardiol 38(4,Suppl)45–50.

Kolossov E, Lu Z, Drobinskaya I, Gassanov N, Duan Y, Sauer H, Manzke O, Bloch W,Bohlen H, Hescheler J, et al. (2005) Identification and characterization of embry-onic stem cell-derived pacemaker and atrial cardiomyocytes. FASEB J 19:577–579.

Koyanagi M, Haendeler J, Badorff C, Brandes RP, Hoffmann J, Pandur P, Zeiher AM,Kühl M, and Dimmeler S (2005) Non-canonical Wnt signaling enhances differen-tiation of human circulating progenitor cells to cardiomyogenic cells. J Biol Chem280:16838–16842.

Kryukova YN, Protas L, and Robinson RB (2012) Ca2+-activated adenylyl cyclase 1introduces Ca2+-dependence to b-adrenergic stimulation of HCN2 current. J MolCell Cardiol 52:1233–1239.

Kupershmidt S, Yang T, Anderson ME, Wessels A, Niswender KD, Magnuson MA,and Roden DM (1999) Replacement by homologous recombination of the minK genewith lacZ reveals restriction of minK expression to the mouse cardiac conductionsystem. Circ Res 84:146–152.

Lakatta EG and DiFrancesco D (2009) What keeps us ticking: a funny current,a calcium clock, or both? J Mol Cell Cardiol 47:157–170.

Lakatta EG, Vinogradova TM, and Maltsev VA (2008) The missing link in the mysteryof normal automaticity of cardiac pacemaker cells. Ann N Y Acad Sci 1123:41–57.

Lee YK, Ng KM, LaiWH, Chan YC, Lau YM, Lian Q, Tse HF, and Siu CW (2011) Calciumhomeostasis in human induced pluripotent stem cell-derived cardiomyocytes. Stem CellRev 7:976–986.

Liang X, Wang G, Lin L, Lowe J, Zhang Q, Bu L, Chen Y, Chen J, Sun Y, and EvansSM (2013) HCN4 dynamically marks the first heart field and conduction systemprecursors. Circ Res 113:399–407.

Lin B, Kim J, Li Y, Pan H, Carvajal-Vergara X, Salama G, Cheng T, Li Y, Lo CW,and Yang L (2012) High-purity enrichment of functional cardiovascular cells fromhuman iPS cells. Cardiovasc Res 95:327–335.

Liu H, Espinoza-Lewis RA, Chen C, Hu X, Zhang Y, and Chen Y (2012) The role ofShox2 in SAN development and function. Pediatr Cardiol 33:882–889.

Liu J, Dobrzynski H, Yanni J, Boyett MR, and Lei M (2007) Organisation of themouse sinoatrial node: structure and expression of HCN channels. Cardiovasc Res73:729–738.

Lyons GE, Schiaffino S, Sassoon D, Barton P, and Buckingham M (1990) De-velopmental regulation of myosin gene expression in mouse cardiac muscle. J CellBiol 111:2427–2436.

Ma J, Guo L, Fiene SJ, Anson BD, Thomson JA, Kamp TJ, Kolaja KL, Swanson BJ,and January CT (2011) High purity human-induced pluripotent stem cell-derivedcardiomyocytes: electrophysiological properties of action potentials and ionic cur-rents. Am J Physiol Heart Circ Physiol 301:H2006–H2017.

Maier SK, Westenbroek RE, Yamanushi TT, Dobrzynski H, Boyett MR, CatterallWA, and Scheuer T (2003) An unexpected requirement for brain-type sodiumchannels for control of heart rate in the mouse sinoatrial node. Proc Natl Acad SciUSA 100:3507–3512.

Makino S, Fukuda K, Miyoshi S, Konishi F, Kodama H, Pan J, Sano M, Takahashi T,Hori S, Abe H, et al. (1999) Cardiomyocytes can be generated from marrow stromalcells in vitro. J Clin Invest 103:697–705.

Malan D, Friedrichs S, Fleischmann BK, and Sasse P (2011) Cardiomyocytes obtainedfrom induced pluripotent stem cells with long-QT syndrome 3 recapitulate typicaldisease-specific features in vitro. Circ Res 109:841–847.

Maltsev VA, Rohwedel J, Hescheler J, and Wobus AM (1993) Embryonic stem cellsdifferentiate in vitro into cardiomyocytes representing sinusnodal, atrial andventricular cell types. Mech Dev 44:41–50.

Mandel Y, Weissman A, Schick R, Barad L, Novak A, Meiry G, Goldberg S, Lorber A,Rosen MR, Itskovitz-Eldor J, et al. (2012) Human embryonic and induced plurip-otent stem cell-derived cardiomyocytes exhibit beat rate variability and power-lawbehavior. Circulation 125:883–893.

Mangoni ME, Couette B, Bourinet E, Platzer J, Reimer D, Striessnig J, and NargeotJ (2003) Functional role of L-type Cav1.3 Ca2+ channels in cardiac pacemakeractivity. Proc Natl Acad Sci USA 100:5543–5548.

Mangoni ME and Nargeot J (2001) Properties of the hyperpolarization-activatedcurrent (I(f)) in isolated mouse sino-atrial cells. Cardiovasc Res 52:51–64.

Marionneau C, Couette B, Liu J, Li H, Mangoni ME, Nargeot J, Lei M, Escande D,and Demolombe S (2005) Specific pattern of ionic channel gene expression asso-ciated with pacemaker activity in the mouse heart. J Physiol 562:223–234.

Masino AM, Gallardo TD, Wilcox CA, Olson EN, Williams RS, and Garry DJ (2004)Transcriptional regulation of cardiac progenitor cell populations. Circ Res 95:389–397.

Matsuura K, Nagai T, Nishigaki N, Oyama T, Nishi J, Wada H, Sano M, Toko H,Akazawa H, Sato T, et al. (2004) Adult cardiac Sca-1-positive cells differentiate intobeating cardiomyocytes. J Biol Chem 279:11384–11391.

Mattick P, Parrington J, Odia E, Simpson A, Collins T, and Terrar D (2007)Ca2+-stimulated adenylyl cyclase isoform AC1 is preferentially expressed in guinea-pigsino-atrial node cells and modulates the I(f) pacemaker current. J Physiol 582:1195–1203.

Mommersteeg MT, Domínguez JN, Wiese C, Norden J, de Gier-de Vries C, Burch JB,Kispert A, Brown NA, Moorman AF, and Christoffels VM (2010) The sinus venosusprogenitors separate and diversify from the first and second heart fields early indevelopment. Cardiovasc Res 87:92–101.

Mommersteeg MT, Hoogaars WM, Prall OW, de Gier-de Vries C, Wiese C, Clout DE,Papaioannou VE, Brown NA, Harvey RP, Moorman AF, et al. (2007) Molecularpathway for the localized formation of the sinoatrial node. Circ Res 100:354–362.

Moorman AF and Christoffels VM (2003) Cardiac chamber formation: development,genes, and evolution. Physiol Rev 83:1223–1267.

Moretti A, Bellin M, Welling A, Jung CB, Lam JT, Bott-Flügel L, Dorn T, Goedel A,Höhnke C, Hofmann F, et al. (2010) Patient-specific induced pluripotent stem-cellmodels for long-QT syndrome. N Engl J Med 363:1397–1409.

Morikawa K, Bahrudin U, Miake J, Igawa O, Kurata Y, Nakayama Y, Shirayoshi Y,and Hisatome I (2010) Identification, isolation and characterization of HCN4-positive pacemaking cells derived from murine embryonic stem cells during cardiacdifferentiation. Pacing Clin Electrophysiol 33:290–303.

Mummery C (2011) Induced pluripotent stem cells—a cautionary note. N Engl J Med364:2160–2162.

Mummery C, Ward-van Oostwaard D, Doevendans P, Spijker R, van den Brink S,Hassink R, van der Heyden M, Opthof T, Pera M, de la Riviere AB, et al. (2003)Differentiation of human embryonic stem cells to cardiomyocytes: role of coculturewith visceral endoderm-like cells. Circulation 107:2733–2740.

Murakami Y, Hirata H, Miyamoto Y, Nagahashi A, Sawa Y, Jakt M, Asahara T,and Kawamata S (2007) Isolation of cardiac cells from E8.5 yolk sac by ALCAM(CD166) expression. Mech Dev 124:830–839.

Murry CE, Soonpaa MH, Reinecke H, Nakajima H, Nakajima HO, Rubart M,Pasumarthi KB, Virag JI, Bartelmez SH, Poppa V, et al. (2004) Haematopoieticstem cells do not transdifferentiate into cardiac myocytes in myocardial infarcts.Nature 428:664–668.

Nawathe PA, Kryukova Y, Oren RV, Milanesi R, Clancy CE, Lu JT, Moss AJ,Difrancesco D, and Robinson RB (2013) An LQTS6 MiRP1 mutation suppressespacemaker current and is associated with sinus bradycardia. J Cardiovasc Elec-trophysiol 24:1021–1027.

Novak A, Shtrichman R, Germanguz I, Segev H, Zeevi-Levin N, Fishman B, MandelYE, Barad L, Domev H, Kotton D, et al. (2010) Enhanced reprogramming andcardiac differentiation of human keratinocytes derived from plucked hair follicles,using a single excisable lentivirus. Cell Reprogram 12:665–678.

Oren RV and Clancy CE (2010) Determinants of heterogeneity, excitation and con-duction in the sinoatrial node: a model study. PLOS Comput Biol 6:e1001041.

Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R,Nadal-Ginard B, Bodine DM, et al. (2001) Bone marrow cells regenerate infarctedmyocardium. Nature 410:701–705.

Patten BM (1949) Initiation and early changes in the character of the heart beat invertebrate embryos. Physiol Rev 29:31–47.

Planat-Bénard V, Menard C, André M, Puceat M, Perez A, Garcia-Verdugo JM,Pénicaud L, and Casteilla L (2004) Spontaneous cardiomyocyte differentiationfrom adipose tissue stroma cells. Circ Res 94:223–229.

Poon E, Kong CW, and Li RA (2011) Human pluripotent stem cell-based approaches formyocardial repair: from the electrophysiological perspective. Mol Pharm 8:1495–1504.

Potapova I, Plotnikov A, Lu Z, Danilo P Jr, Valiunas V, Qu J, Doronin S, ZuckermanJ, Shlapakova IN, Gao J, et al. (2004) Human mesenchymal stem cells as a genedelivery system to create cardiac pacemakers. Circ Res 94:952–959.

Qu Y,Whitaker GM, Hove-Madsen L, Tibbits GF, and Accili EA (2008) Hyperpolarization-activated cyclic nucleotide-modulated ‘HCN’ channels confer regular and faster rhyth-micity to beating mouse embryonic stem cells. J Physiol 586:701–716.

Robertson C, Tran DD, and George SC (2013) Concise review: maturation phases ofhuman pluripotent stem cell-derived cardiomyocytes. Stem Cells 31:829–837.

Robinson RB and Siegelbaum SA (2003) Hyperpolarization-activated cation currents:from molecules to physiological function. Annu Rev Physiol 65:453–480.

Robinson RB, Yu H, Chang F, and Cohen IS (1997) Developmental change in thevoltage-dependence of the pacemaker current, if, in rat ventricle cells. PflugersArch 433:533–535.

Rosen MR, Brink PR, Cohen IS, Danilo P Jr, Robinson RB, Rosen AB, and SzabolcsMJ (2008) Regenerative therapies in electrophysiology and pacing. J Interv CardElectrophysiol 22:87–98.

Ruhparwar A, Tebbenjohanns J, Niehaus M, Mengel M, Irtel T, Kofidis T, PichlmaierAM, and Haverich A (2002) Transplanted fetal cardiomyocytes as cardiac pace-maker. Eur J Cardiothorac Surg 21:853–857.

Rust W, Balakrishnan T, and Zweigerdt R (2009) Cardiomyocyte enrichment fromhuman embryonic stem cell cultures by selection of ALCAM surface expression.Regen Med 4:225–237.

Sartiani L, Bettiol E, Stillitano F, Mugelli A, Cerbai E, and Jaconi ME (2007) De-velopmental changes in cardiomyocytes differentiated from human embryonic stemcells: a molecular and electrophysiological approach. Stem Cells 25:1136–1144.

Satin J, Kehat I, Caspi O, Huber I, Arbel G, Itzhaki I, Magyar J, Schroder EA,Perlman I, and Gepstein L (2004) Mechanism of spontaneous excitability in humanembryonic stem cell derived cardiomyocytes. J Physiol 559:479–496.


Scavone A, Capilupo D, Mazzocchi N, Crespi A, Zoia S, Campostrini G, Bucchi A,Milanesi R, Baruscotti M, Benedetti S, et al. (2013) Embryonic stem cell-derivedCD166+ precursors develop into fully functional sinoatrial-like cells. Circ Res 113:389–398.

Shi W, Wymore R, Yu H, Wu J, Wymore RT, Pan Z, Robinson RB, Dixon JE, McKinnonD, and Cohen IS (1999) Distribution and prevalence of hyperpolarization-activatedcation channel (HCN) mRNA expression in cardiac tissues. Circ Res 85:e1–e6.

Sinnecker D, Dirschinger RJ, Goedel A, Moretti A, Lipp P, and Laugwitz KL (2012)Induced pluripotent stem cells in cardiovascular research. Rev Physiol BiochemPharmacol 163:1–26.

Später D, Abramczuk MK, Buac K, Zangi L, Stachel MW, Clarke J, Sahara M,Ludwig A, and Chien KR (2013) A HCN4+ cardiomyogenic progenitor derived fromthe first heart field and human pluripotent stem cells. Nat Cell Biol 15:1098–1106.

Stieber J, Herrmann S, Feil S, Löster J, Feil R, Biel M, Hofmann F, and Ludwig A(2003) The hyperpolarization-activated channel HCN4 is required for the genera-tion of pacemaker action potentials in the embryonic heart. Proc Natl Acad SciUSA 100:15235–15240.

Sun N, Yazawa M, Liu J, Han L, Sanchez-Freire V, Abilez OJ, Navarrete EG, Hu S,Wang L, Lee A, et al. (2012) Patient-specific induced pluripotent stem cells asa model for familial dilated cardiomyopathy. Sci Transl Med 4:130ra47.

Tawara S (1906) Das Reizleitungssystem des Saügethierherzens, Gustav Fischer,Jena.

Terrenoire C, Wang K, Tung KW, Chung WK, Pass RH, Lu JT, Jean JC, Omari A,Sampson KJ, Kotton DN, et al. (2013) Induced pluripotent stem cells used to revealdrug actions in a long QT syndrome family with complex genetics. J Gen Physiol141:61–72.

Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS,and Jones JM (1998) Embryonic stem cell lines derived from human blastocysts.Science 282:1145–1147.

Uosaki H, Fukushima H, Takeuchi A, Matsuoka S, Nakatsuji N, Yamanaka S,and Yamashita JK (2011) Efficient and scalable purification of cardiomyocytesfrom human embryonic and induced pluripotent stem cells by VCAM1 surfaceexpression. PLoS ONE 6:e23657.

van Kempen M, van Ginneken A, de Grijs I, Mutsaers N, Opthof T, Jongsma H,and van der Heyden M (2003) Expression of the electrophysiological system duringmurine embryonic stem cell cardiac differentiation. Cell Physiol Biochem 13:263–270.

van Kempen MJ, Fromaget C, Gros D, Moorman AF, and Lamers WH (1991) Spatialdistribution of connexin43, the major cardiac gap junction protein, in the de-veloping and adult rat heart. Circ Res 68:1638–1651.

Van Mierop LH (1967) Location of pacemaker in chick embryo heart at the time ofinitiation of heartbeat. Am J Physiol 212:407–415.

Vedantham V, Evangelista M, Huang Y, and Srivastava D (2013) Spatiotemporalregulation of an Hcn4 enhancer defines a role for Mef2c and HDACs in cardiacelectrical patterning. Dev Biol 373:149–162.

Verkerk AO, van Borren MM, Peters RJ, Broekhuis E, Lam KY, Coronel R, deBakker JM, Tan HL, and Wilders R (2007a) Single cells isolated from humansinoatrial node: action potentials and numerical reconstruction of pacemakercurrent. Conf Proc IEEE Eng Med Biol Soc 2007:904–907.

Verkerk AO and Wilders R (2014) Pacemaker activity of the human sinoatrial node:effects of HCN4 mutations on the hyperpolarization-activated current. Europace16:384–395.

Verkerk AO, Wilders R, van Borren MM, Peters RJ, Broekhuis E, Lam K, Coronel R,de Bakker JM, and Tan HL (2007b) Pacemaker current (I(f)) in the human sino-atrial node. Eur Heart J 28:2472–2478.

Vinogradova TM, Lyashkov AE, Zhu W, Ruknudin AM, Sirenko S, Yang D, Deo S,BarlowM, Johnson S, Caffrey JL, et al. (2006) High basal protein kinase A-dependentphosphorylation drives rhythmic internal Ca2+ store oscillations and spontaneousbeating of cardiac pacemaker cells. Circ Res 98:505–514.

Vitadello M, Vettore S, Lamar E, Chien KR, and Gorza L (1996) Neurofilament MmRNA is expressed in conduction system myocytes of the developing and adultrabbit heart. J Mol Cell Cardiol 28:1833–1844.

Weisbrod D, Peretz A, Ziskind A, Menaker N, Oz S, Barad L, Eliyahu S, Itskovitz-Eldor J, Dascal N, Khananshvili D, et al. (2013) SK4 Ca2+ activated K+ channel isa critical player in cardiac pacemaker derived from human embryonic stem cells.Proc Natl Acad Sci USA 110:E1685–E1694.

Wenink AC, Symersky P, Ikeda T, DeRuiter MC, Poelmann RE, and Gittenberger-deGroot AC (2000) HNK-1 expression patterns in the embryonic rat heart distinguishbetween sinuatrial tissues and atrial myocardium. Anat Embryol (Berl) 201:39–50.

White SM and Claycomb WC (2005) Embryonic stem cells form an organized, func-tional cardiac conduction system in vitro. Am J Physiol Heart Circ Physiol 288:H670–H679.

Wiese C, Grieskamp T, Airik R, Mommersteeg MT, Gardiwal A, de Gier-de Vries C,Schuster-Gossler K, Moorman AF, Kispert A, and Christoffels VM (2009) Forma-tion of the sinus node head and differentiation of sinus node myocardium areindependently regulated by Tbx18 and Tbx3. Circ Res 104:388–397.

Wiese C, Nikolova T, Zahanich I, Sulzbacher S, Fuchs J, Yamanaka S, Graf E,Ravens U, Boheler KR, and Wobus AM (2011) Differentiation induction of mouseembryonic stem cells into sinus node-like cells by suramin. Int J Cardiol 147:95–111.

Wobus AM, Wallukat G, and Hescheler J (1991) Pluripotent mouse embryonic stemcells are able to differentiate into cardiomyocytes expressing chronotropic responsesto adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 48:173–182.

Xu H, Yi BA, and Chien KR (2011) Shortcuts to making cardiomyocytes.Nat Cell Biol13:191–193.

Xue T, Cho HC, Akar FG, Tsang SY, Jones SP, Marbán E, Tomaselli GF, and Li RA (2005)Functional integration of electrically active cardiac derivatives from geneticallyengineered human embryonic stem cells with quiescent recipient ventricular car-diomyocytes: insights into the development of cell-based pacemakers. Circulation111:11–20.

Yamamoto M, Dobrzynski H, Tellez J, Niwa R, Billeter R, Honjo H, Kodama I,and Boyett MR (2006) Extended atrial conduction system characterised by theexpression of the HCN4 channel and connexin45. Cardiovasc Res 72:271–281.

Yanagi K, Takano M, Narazaki G, Uosaki H, Hoshino T, Ishii T, Misaki T,and Yamashita JK (2007) Hyperpolarization-activated cyclic nucleotide-gatedchannels and T-type calcium channels confer automaticity of embryonic stem cell-derived cardiomyocytes. Stem Cells 25:2712–2719.

Yang X, Pabon L, and Murry CE (2014) Engineering adolescence: maturation ofhuman pluripotent stem cell-derived cardiomyocytes. Circ Res 114:511–523.

Yasui K, Liu W, Opthof T, Kada K, Lee JK, Kamiya K, and Kodama I (2001) I(f)current and spontaneous activity in mouse embryonic ventricular myocytes. CircRes 88:536–542.

Yazawa M, Hsueh B, Jia X, Pasca AM, Bernstein JA, Hallmayer J, and DolmetschRE (2011) Using induced pluripotent stem cells to investigate cardiac phenotypesin Timothy syndrome. Nature 471:230–234.

Yoshida Y and Yamanaka S (2010) Recent stem cell advances: induced pluripotent stemcells for disease modeling and stem cell-based regeneration. Circulation 122:80–87.

Yoshida Y and Yamanaka S (2011) iPS cells: a source of cardiac regeneration. J MolCell Cardiol 50:327–332.

Younes A, Lyashkov AE, Graham D, Sheydina A, Volkova MV, Mitsak M, VinogradovaTM, Lukyanenko YO, Li Y, and Ruknudin AM, et al. (2008) Ca(2+) -stimulated basaladenylyl cyclase activity localization in membrane lipid microdomains of cardiac si-noatrial nodal pacemaker cells. J Biol Chem 283:14461–14468.

Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, Nie J,Jonsdottir GA, Ruotti V, and Stewart R, et al. (2007) Induced pluripotent stem celllines derived from human somatic cells. Science 318:1917–1920.

Zahanich I, Sirenko SG, Maltseva LA, Tarasova YS, Spurgeon HA, Boheler KR, SternMD, Lakatta EG, and Maltsev VA (2011) Rhythmic beating of stem cell-derivedcardiac cells requires dynamic coupling of electrophysiology and Ca cycling. J MolCell Cardiol 50:66–76.

Zeevi-Levin N, Itskovitz-Eldor J, and Binah O (2012) Cardiomyocytes derivedfrom human pluripotent stem cells for drug screening. Pharmacol Ther 134:180–188.

Zhang H, Lau DH, Shlapakova IN, Zhao X, Danilo P, Robinson RB, Cohen IS, Qu D,Xu Z, and Rosen MR (2011) Implantation of sinoatrial node cells into canine rightventricle: biological pacing appears limited by the substrate. Cell Transplant 20:1907–1914.

Zhang J, Wilson GF, Soerens AG, Koonce CH, Yu J, Palecek SP, Thomson JA,and Kamp TJ (2009) Functional cardiomyocytes derived from human inducedpluripotent stem cells. Circ Res 104:e30–e41.

Zhang P, Li J, Tan Z, Wang C, Liu T, Chen L, Yong J, Jiang W, Sun X, and Du L, ,et al. (2008) Short-term BMP-4 treatment initiates mesoderm induction in humanembryonic stem cells. Blood 111:1933–1941.

Zhu W-Z, Xie Y, Moyes KW, Gold JD, Askari B, and Laflamme MA (2010)Neuregulin/ErbB signaling regulates cardiac subtype specification in differentiat-ing human embryonic stem cells. Circ Res 107:776–786.


Documents

Stem Cell Derived Nodal-Like Cardiomyocytes as a …pharmrev.aspetjournals.org/content/pharmrev/67/2/368...aimed at preferentially driving stem cells down a nodal lineage. Finally,