Electrophysiology of Cardiac Cells

8/8/2019 Electrophysiology of Cardiac Cells

1/13

277:433-444, 1999. Am J Physiol Heart Circ PhysiolVunjak-Novakovic and L. E. FreedN. Bursac, M. Papadaki, R. J. Cohen, F. J. Schoen, S. R. Eisenberg, R. Carrier, G.

You might find this additional information useful...

41 articles, 18 of which you can access free at:This article citeshttp://ajpheart.physiology.org/cgi/content/full/277/2/H433#BIBL

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

[PDF] [Full Text] [Abstract], February 1, 2004; 286 (2): H507-H516. Am J Phys iol Heart Ci r c Phys iol

M. Radisic, L. Yang, J. Boublik, R. J. Cohen, R. Langer, L. E. Freed and G. Vunjak-NovakovicMedium perfusion enables engineering of compact and contractile cardiac tissue

[PDF] [Full Text] [Abstract], March 1, 2005; 288 (3): H1278-H1289. Am J Phys iol Heart Ci r c Phys iol

M. Radisic, W. Deen, R. Langer and G. Vunjak-Novakovicarray perfused with culture medium containing oxygen carriersMathematical model of oxygen distribution in engineered cardiac tissue with parallel channel

[PDF] [Full Text] [Abstract], November 1, 2005; 130 (5): 1358-1363. J. Thorac. Cardiovasc. Surg.

O. Ishii, M. Shin, T. Sueda and J. P. Vacantimatrix-like topographyIn vitro tissue engineering of a cardiac graft using a degradable scaffold with an extracellular

[PDF] [Full Text] [Abstract], August 29 , 2007; 362 (1484): 1357-1368.Phil Trans R Soc B

M Radisic, H Park, S Gerecht, C Cannizzaro, R Langer and G Vunjak-NovakovicBiomimetic approach to cardiac tissue engineering

[PDF] [Full Text] [Abstract], January 1, 2008; 294 (1): H58-H65. Am J Phys iol Heart Ci r c Phys iol

J. W. Lin, L. Garber, Y. R. Qi, M. G. Chang, J. Cysyk and L. TungRegion of slowed conduction acts as core for spiral wave reentry in cardiac cell monolayers

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

Physiology .. RatsEngineering .. Biomedical EngineeringPhysiology .. Heart MusclePhysiology .. Cardiac Muscle

including high-resolution figures, can be found at:Updated information and serviceshttp://ajpheart.physiology.org/cgi/content/full/277/2/H433

can be found at: AJP - Heart and Circulatory PhysiologyaboutAdditional material and informationhttp://www.the-aps.org/publications/ajpheart

This information is current as of May 24, 2010 .

http://www.the-aps.org/.ESSN: 1522-1539. Visit our website at0363-6135,Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyri ght 2005 by the American Physiological Society. ISSN:

intact animal to the cellular, subcellular, and molecular levels. It is published 12 times a year (monthly) by the American Physiologicallymphatics, including experimental and theoretical studies of cardiovascular function at all levels of organization ranging from the

publishes original investigations on the physiology of the heart, blood vessels, and AJP - Heart and Circulatory Physiology
http://ajpheart.physiology.org/cgi/content/full/277/2/H433#BIBLhttp://ajpheart.physiology.org/cgi/reprint/286/2/H507http://ajpheart.physiology.org/cgi/content/full/286/2/H507http://ajpheart.physiology.org/cgi/content/full/286/2/H507http://ajpheart.physiology.org/cgi/content/abstract/286/2/H507http://ajpheart.physiology.org/cgi/content/abstract/286/2/H507http://ajpheart.physiology.org/cgi/content/full/286/2/H507http://ajpheart.physiology.org/cgi/reprint/286/2/H507http://ajpheart.physiology.org/cgi/reprint/288/3/H1278http://ajpheart.physiology.org/cgi/content/full/288/3/H1278http://ajpheart.physiology.org/cgi/content/full/288/3/H1278http://ajpheart.physiology.org/cgi/content/abstract/288/3/H1278http://ajpheart.physiology.org/cgi/content/abstract/288/3/H1278http://ajpheart.physiology.org/cgi/content/full/288/3/H1278http://ajpheart.physiology.org/cgi/reprint/288/3/H1278http://jtcs.ctsnetjournals.org/cgi/reprint/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/full/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/full/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/abstract/130/5/1358http://jtcs.ctsnetjournals.org/cgi/reprint/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/abstract/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/full/130/5/1358http://jtcs.ctsnetjournals.org/cgi/reprint/130/5/1358http://rstb.royalsocietypublishing.org/cgi/reprint/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/full/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/full/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/abstract/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/full/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/reprint/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/abstract/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/full/362/1484/1357http://ajpheart.physiology.org/cgi/reprint/294/1/H58http://ajpheart.physiology.org/cgi/content/full/294/1/H58http://ajpheart.physiology.org/cgi/content/full/294/1/H58http://ajpheart.physiology.org/cgi/content/abstract/294/1/H58http://ajpheart.physiology.org/cgi/content/abstract/294/1/H58http://ajpheart.physiology.org/cgi/content/full/294/1/H58http://ajpheart.physiology.org/cgi/reprint/294/1/H58http://highwire.stanford.edu/lists/artbytopic.dtlhttp://highwire.stanford.edu/lists/artbytopic.dtlhttp://ajpheart.physiology.org/cgi/content/full/277/2/H433http://www.the-aps.org/publications/ajphearthttp://www.the-aps.org/http://www.the-aps.org/http://www.the-aps.org/http://www.the-aps.org/publications/ajphearthttp://ajpheart.physiology.org/cgi/content/full/277/2/H433http://highwire.stanford.edu/lists/artbytopic.dtlhttp://ajpheart.physiology.org/cgi/reprint/286/2/H507http://ajpheart.physiology.org/cgi/content/full/286/2/H507http://ajpheart.physiology.org/cgi/content/abstract/286/2/H507http://ajpheart.physiology.org/cgi/reprint/288/3/H1278http://ajpheart.physiology.org/cgi/content/full/288/3/H1278http://ajpheart.physiology.org/cgi/content/abstract/288/3/H1278http://jtcs.ctsnetjournals.org/cgi/reprint/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/full/130/5/1358http://jtcs.ctsnetjournals.org/cgi/content/abstract/130/5/1358http://rstb.royalsocietypublishing.org/cgi/reprint/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/full/362/1484/1357http://rstb.royalsocietypublishing.org/cgi/content/abstract/362/1484/1357http://ajpheart.physiology.org/cgi/reprint/294/1/H58http://ajpheart.physiology.org/cgi/content/full/294/1/H58http://ajpheart.physiology.org/cgi/content/abstract/294/1/H58http://ajpheart.physiology.org/cgi/content/full/277/2/H433#BIBL


2/13

Cardiac muscle tissue engineering: toward anin vitro model for electrophysiological studies

N. BURSAC, 1,2 M. PAPADAKI, 1 R. J . COHEN, 1 F. J . SCHOEN, 3 S. R. EISENBERG, 2R. CARRIER, 1 G. VUNJAK-NOVAKOVIC, 1 AND L. E. FREED 11 Division of Health Sciences and Technology, Massachusetts Institute of Technology,Cambridge 02139; 2 Departm ent of Biom edical E ngineering, B oston University, Boston 02215; and 3 Departm ent of Pathology, Brigham an d Womens Hospital, B oston, M assachusetts 02115

Bursac , N . , M. Papada k i , R . J . Cohen , F. J . Schoen ,S. R. Eise nberg, R. Carr ier, G. Vunjak-Novakov ic , an dL. E. Free d. Cardiac muscle tissue engineering: toward an invitro model for electr ophysiological stu dies. Am. J . Physiol .277 ( Heart Circ. Physiol. 46): H433 H444, 1999.The objec-tive of this study was to establish a thr ee-dimensional (3-D)in vitro m odel system of car diac mu scle for electrophysiologi-cal studies. Primary neonatal r at ventricular cells conta ininglower or higher fractions of cardiac myocytes were cultured onpolymeric scaffolds in bioreactors to form regular or enr ichedcardiac muscle constructs, respectively. After 1 wk, all con-stru cts contained a peripheral t issue-like r egion (5070 m

thick) in which differentiated cardiac myocytes were orga-nized in m ultiple layers in a 3-D congur at ion. In dexes of cellsize (protein/DNA) an d m etabolic activity (tetr azolium conver-sion/DNA) were similar for constructs and neonatal ratventricles. Electrophysiological studies conducted using al inear array of extracel lular electrodes showed that theperipheral region of constructs exhibited relatively homoge-neous electrical properties and susta ined macroscopicallycontinu ous impu lse propagation on a centimeter-size scale.Electrophysiological properties of enriched constructs weresuperior to those of regular constructs but inferior to those of nat ive ventricles. These results demonstrat e th at 3-D cardiacmuscle constr ucts can be en gineered with cardiac-specicstr uctur al an d electr ophysiological propert ies and u sed for invitro impulse propagation studies.

myocyte; impulse propagation; electrophysiology; thr ee-dimensional

CULTURED CARDIAC MYOCYTES offer many advantages fordevelopmental, physiological, and pharmacological stud-ies of cardiac tissue because they allow for direct cellma nipulat ion an d control of environm enta l param eterswithout interference from the compensatory feedback mechanisms that exist in vivo. Compared with mono-layer cu l tu res , i t has been sugges ted tha t th ree-dimensional (3-D) multilayered cultur es ofcardiac myo-cytes more closely resemble intact cardiac tissue with

respect to cellular differentiation (8) and electricalproperties (38, 39). Three-dimensional cardiac myocytecul tures could t hus be u sed for in vit ro s tudies of cardiac tissue development and function and, if suffi-cient ly lar ge and functional, for in vivo cardia c repair.

Impulse propagation studies in cultures of cardiacmyocytes can improve our understanding of the electro-

physiological behavior of normal and pathological car-diac tissues. Such studies ar e current ly performed inone-dimensional cardiac strands and two-dimensional(2-D) isotropic, anisotropic, and photolithographicallypat terned monolayers using opt ical mapping tech-niques (9, 10, 27). Impu lse propagation st udies cann otbe performed in 3-D myocyte aggregates (17, 30) be-cau se of th eir sma ll size (100300 m) and isopotentialnature. Other 3-D cultures of cardiac myocytes grownon microcarrier beads (1, 31), collagen bers (1), syn-thetic, biodegradable polymeric templates (3, 12), or in

collagen gels (8) have not yet been evaluated electro-physiologically.The goal of th e presen t work wa s to establish a 3-D in

vitr o model system for impulse propagat ion stu dies incardiac muscle using tissue engineering principles.This approach relies on the use of pr imary cells inconjunction with biodegradable polymer scaffolds (13,18) and tissue culture bioreactors (11, 12). The polymerscaffold provides a 3-D substrate for cell attachmentand tissue formation, whereas the mixing of culturemedium in the bioreactor promotes mass transfer of nutr ients and gases to the forming t issue. Pr imaryneonata l rat ventricular cells were cultured on polymerscaffolds in biorea ctors to form tissu e const ru cts, which

were chara cterized histologically, biochemically, an delectrophysiologically and compared with neonatal andadult rat ventricular tissues.

MATERIALS AND METHODS

All experiment s involving an imals wer e performed a ccord-ing to the Inst i tut ional Committee on Animal Care of theMassa chusett s Inst itut e of Technology, which follows federa land state guidelines.

Cardiac myocyte preparation. Primary cultures of cardiacmyocytes were prepa red by enzymat ic digestion of ventr iclesobtained from neonatal (2 day old) Sprague-Dawley r ats(Taconic), as p revious ly described (44). Briey, ventr icles ( n50, 5 litters in 3 independent studies) were incubated with0.1% tr ypsin overnight a nd dissociat ed in four to ve sequen-tial st eps using 0.1% collagenase. Isolated cells were r esus-pended in culture m edium [DMEM, supplemented with 10%fetal bovine serum (FBS), 50 U/ml penici ll in and 10 mMHE PES , all obtain ed from GIBCO-BRL].

Two experimenta l groups wer e esta blished as follows (Fig.1 A ): 1 ) a regular g roup of vent r icu la r ce ll s i sola ted asdescribed a bove a nd 2 ) an enr iched group with a h igherfraction of car diac myocytes, prepa red from th e regular groupby centr ifugation at 600 rpm for 5 min, fol lowed by twopreplat ings, 75 min each (Fig. 1 A ); cells that remainedunattached after the second preplating were used. Cell yieldswere 6 10 6 and 5 10 6 cells/ventr icle for th e regular an d

The costs of publication of this a rticle were defrayed in pa rt by thepayment of page charges . The ar t icle must therefore be herebymar ked advertisement in accordan ce with 18 U.S.C. Section 1734solely to indicate th is fact.

0363-6135/99 $5.00 Copyright 1999 t he American Physiologica l Society H433


3/13

Fig. 1. A: model system for t issue engineering. Cells from neonata l rat ventricles were seeded onto polymerscaffolds an d cultur ed for 7 days t o form regular and cardiac myocyte-enriched constructs. B : elect rophysiologicalsetup. Tissue constructs were studied u sing an extracellular microelectrode arr ay ( inset ) under controlledenvironmental conditions in a 37C/5% CO 2 perfused chamber. Stimulation was bipolar, and extracellularrecordings were unipolar with reference to a Ag-AgCl electrode placed 3.5 cm away from the microelectrode array.

H434 CARDIAC MUSCLE TISSUE ENGINEERING


4/13

enriched group, respectively. Cell viability was 91 3%, asassessed by trypan blue exclusion.

Monolayer studies. Cells from the regular and enrichedgroups were cultured in monolayers at a cell density of 1.310 4 cells/cm 2 in 12-well dishes, T75 asks, and on glasscoverslips to assess spontaneous contractions and biochemi-cal and immu nohistochemical pa ra meter s, respectively. After2 days of sta tic cultu re, monolayers were placed on an orbital

sha ker set to 75 rpm. Medium wa s completely replaced on d ay3 and by 50% on d a y 5 . Spontaneous contract ions wereassessed by videomicroscopy, by man ua lly countin g the n um -ber of beats per minute using ve randomly selected elds(0.3 0.4 mm 2 each) per plate and six plates per experimen-tal group, on d a y s 3 , 5 , a n d 7 . Cells in T75 asks wereremoved after 7 days by a 5-min incubation with 0.05%trypsin-EDTA (GIBCO-BRL) and counted, and a suspensionof 2 10 6 cells/ml was stored at 20C for determination of DNA and pr otein content s a nd lactate dehydrogenase (LDH)activity per cell. Cells on glass coverslips were xed withHistoCHOICE (Amr esco) for immu nohistochemical a na lysis.

3-D t issue culture s tudies . Cel ls from the regular andenriched groups were cultur ed on polyglycolic acid (PGA)scaffolds, which a re highly porous (97%) meshes of rand omlyentangled 13-m bers formed as 5 2-mm (diameterthickness) disks (Fig. 1 A; Ref. 13). Br iey, scaffolds wereprewetted in culture medium, positioned on thin stainlesssteel wires using segments of silicone tu bing, and xed to asi l icone stopper placed in the mouth of a spinner ask (8scaffolds per ask) (12). Flasks were lled with 120 ml of cultur e medium, placed in a h um idied 37C, 5% CO 2 incuba-tor with the side ar m caps loosened to permit gas exchan ge,and mixed at 50 rpm using a magnetic s t ir bar. After 24 h,asks were inoculated with cells (8 10 6 cells per scaffold).Culture medium was r eplaced by 100% on day 3 and by 50%on day 5 . Cell-polymer constructs ( n 22, from 3 independ entstudies) were harvested after 7 days for morphometric,histological, biochemical, an d electr ophysiological assess-ments.

Ventricular tissues. To verify the a na lytical met hods, evalu-at e the developmenta l stat e of car diac myocytes in const ru cts,and establish basel ine values for parameters s tudied inengineered constructs that were not readily found in thel iterature, two control groups were examined. Adult ven-tricles ( n 10) were obtained from 3- to 4-mo-old Sprague-Dawley rats following anesthesia by intramuscular injectionof 65 m g/kg keta mine a nd 5 m g/kg xylazine (Sigma). Heartswere rapidly removed, and ventricular sections were excisedfrom 1 mm below the atr ioventr icular groove to 12 mmabove the apex. For electrophysiological studies, full-thick-ness pieces of the ventricular wall ( 9 7 mm 2, 2 4 m mthick) were then prepared by making two longitudinal cutsparallel to the base-apex line. Neonatal ventricles ( n 10,from 3 litters) were obtained from 2-day-old rats followingdecapitat ion. For electr ophysiological stu dies, full-th ickn esspieces of th e ventr icular wall ( 6 4 m m 2, 1.5 2.5 mm thick)were prepared by bisecting the ventricle. Smaller pieces of the adult and neonatal ventricles (713 mg wet wt) were usedfor biochemical and histological assessments. The propertiesof neonatal and adult ventricles were compared with those of constructs without a priori assumption that the engineeredtissue resembled either of the native ventricular tissues.

Histological and imm unohistochemical assessments. Cellson glass coverslips were incubated for 30 min with mousean tisar comeric tr opomyosin monoclonal a nt ibody (clone CH1,Sigma) diluted 1:100 in PBS conta ining 0.5% Tween 20 an d1.5% horse serum and then for 30 min with a secondaryant ibody (Vectasta in), diluted 1:200. Coverslips were then

incubated with avidin-biot in complex r eagent and 3,3 -diam inobenzidine (Sigma). Ten r an domly selected elds (0.30.4 mm 2 each) from six coverslips from each group wereanalyzed using videomicroscopy and NIH Image 1.60 soft-ware to estimate cardiac myocyte fraction as a percentage of cell area sta ined positively for tr opomyosin.

Ventr icles and 7-day const ru cts were xed in 2% glutar alde-hyde for 10 min, rinsed in PBS, and immersed in 10% neutral

buffered Form alin (Sigma). Samples were embedded in p ar af-n, sect ioned at 5 m, and stained with hematoxylin andeosin (H E) for general evalua tion and Massons tr ichr omestain for collagen assessment. Immunohistochemical stain-ing for t ropomyosin was u sed to assess t he fraction of cardiacmyocytes in constru cts. Sections were incubat ed with 1 m g/mltrypsin (Sigma) at 37C for 15 m in and 0.3% hydrogenperoxide for 30 min, blocked with horse ser um for 30 m in, andincubated with antisarcomeric t ropomyosin as describedabove. A humidied chamber was used for al l incubationsteps. Sections wer e count erst ained with Mayer s hema toxy-lin (Sigma) and coverslipped using glycerol mounting media(Sigma). Specicity of staining for tropomyosin was con-rmed by staining for sarcomeric -actin, another myocyte-specic protein, u sing oth erwise iden tical met hodology. Con-

struct macroscopic architecture was assessed from stainedtissue sections using videomicroscopy and NIH Image 1.60software.

Transm ission electron m icroscopy. Samples were xed inKarnovskys reagent (0.1 M sodium cacodylate with 2%para forma ldehyde and 2.5% gluta raldehyde, pH 7.4), post-xed in 2% osmium tetroxide, dehydrated in ethanol inpropylene ox ide , and embedded in Poly /Bed812 (Poly-sciences). Sections were cut at 60 nm, stained with leadcitrate and uranyl acetate, and examined using a t ransmis-sion electron microscope (JEOL-100CX, JEOL).

Media analysis. Physiological ran ges of P O2 (115130mmHg), P CO 2 (4855 mmHg), and pH (7.217.33) weremainta ined for t he du rat ion of cultivation, as measur ed by ablood gas analyzer (IL 1610, Instrumentation Laboratory).

Glucose and lactate concentrations were measured using aglucose/lactat e an alyzer (2300 Sta tP lus, YSI). The activity of LDH in the culture media was monitored using a LDH-Lreagent kit (Chiron Diagnostics). Media samples were soni-cated using a Sonic Dismembrat or (Vibra-Cell, Sonics andMaterials), and absorbance was measured at 340 nm (Spec-tr onic 1001 , Milton Roy) against cell-free medium. An LDHactivity of 1 U/l corresponded to 3,600 cells in monolayers.

DNA and protein assays. DNA and protein assays wereperformed on engineered constr ucts an d n ative ventriclesusing modications of previously described methods (7).Sam ples were homogenized in buffer (1 N ammoniu m hydr ox-ide/2% Triton X-100, 0.04 ml/mg wet wt) for 1 min. For theDNA assay, homogenates were incubated at 37C for 10 min,diluted with assa y buffer (100 mM Na Cl, 1 mM EDTA, 10 mM

Tris, pH 7.00), and centrifuged. DNA contents of superna-tan ts were determined u sing a spectrouorometer (PTI) andcalf thymus DNA as a standa rd (7). DNA contents m easuredfor regular and enriched monolayers were comparable (7.10.2 pg/cell) and consistent with p ublished values (7).

For protein assays, the viscosi ty of h omogenates wasredu ced by several pass ages thr ough a 26-gau ge needle.Aftercentrifugation, protein concentration was measured in thesuperna tan t us ing a Bio-Rad DC prote in assay k it and amicroplate spectrophotometer (MR5000, Dynatech). Regularand enriched monolayers had comparable protein contents(290 pg/cell), r esulting in protein-to-DNA ra tios of 41 m g/mgthat were consistent with published values (29).

H435CARDIAC MUSCLE TISSUE ENGINEERING


5/13

Metabolic activity assays. Meta bolic activities of cells with incons t ructs and vent r icu la r t i ssues were assessed by theuptake and enzymat ic reduct ion of the te t razolium dye3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide(MTT; Sigma). Samples (215 mg wet wt) were r insed withPBS and incubated with MEM (GIBCO-BRL) without phenolred an d 0.5 mg/ml MTT for 4 h on an orbital sha ker a t 37Cand 60 rpm. Medium was replaced with an equal volume of 0.1 N HCl in absolute isopropanol and pipet ted direct lythrough the constructs to solubilize the resulting formazancrystals. After 10 min of incubation at 37C, the absorbancewas read at 570 nm, using a microplate spectrophotometer.

Electrophysiological assessm ent. An electr ophysiologicalsystem was custom-designed to enable stimulation an d record-ing of unipolar extracel lular potentials in constructs andventricular tissues u nder controlled environment al condi-t ions using a l inear array of microelectrodes (Fig. 1 B ). Acylindrical Plexiglas cham ber was tightly tted inside anelectrically grounded brass casing placed on a 37C heater(VWR). The brass case distributed the heat evenly throughthe chamber and served as an e lect ros ta t ic sh ie ld . Thechamber was gassed with a pr ewarmed mixtur e of 5% CO 2 inair a nd lled with 50 ml of cultur e medium (DMEM with 15mM HEP ES, 4.5 g/l glucose), which was recirculated (at 60ml/min for constructs and 120 ml/min for ventricular tissues)using a pulseless gear pump (Cole-Parmer). Temperature andpH were maintained at 37 0.1C and 7.32 0.02, respec-tively.

A photomicrogra ph of the m icroelectr ode arr ay is sh own inFig. 1 B . All microelectr odes were ma de of insu lated t un gstenwire and had uninsulated tips with diameters of 50 6 m(Microprobe). Two electr odes for bipolar stimu lation werepositioned 200 m apart and connected to a programmablecardiac stimulator (SEC-3102, Nihon Kohden). Eight record-ing electr odes were positioned 500 m a par t in a linear a rr ay,1 .5 to 5 mm from the s t imula t ing s ite . Exact d is tancesbetween electrodes were measured using a microscope andNIH 1.60 image an alysis softwar e. Shielded cables conn ectedrecording electrodes to bioelectric ampliers (AB.601G, Ni-

hon Kohden). A reference Ag-AgCl electr ode (WPI) wa s placedin th e medium 3.5 cm away from th e microelectr ode arra y.Samples were placed in a tissue holder 23 mm under t he

surface of the cultur e m edium, secured u sing Teon screws,and left to equil ibrate for 15 min. An XYZ m echanicalmicropositioner (Taurusr, WPI) was used to gradually ad-vance the m icroelectrode arr ay toward either th e top sur faceof th e constr uct or th e epicar dial sur face of th e ventr icle, andpacing impulses were simultaneously applied (35 V, 1-mspulses at a ra te of 60 beats/min). The position of th e arr ay wasxed a t the poin t where the ampli tudes of th e recordedresponses appeared maximal, and a recording protocol wasperformed as follows.

Sponta neous beatin g, if presen t, was recorded for 35 min .After 15 min, monopha sic pacing pulses (1-ms dur at ion) were

applied at a r ate of 60 beats/min, star ting at a pa cing voltageof 0.1 V, which was t hen increased in 0.1-V incremen ts u nt ilthe sample was captured (i.e., until each pacing impulse wasfollowed by a recorded tissue response). The correspondingpacing voltage, dened as the excitation threshold, repre-sented the lowest stimulus that produced a stable propaga-t ion (for at least 1 min at a rate of 60 beats/min) over thelength of the recording arr ay. For the next 2030 min, thesample was continuously paced at 60 beats/min using pacingamplitudes 1.5 times higher than the excitation threshold,and responses were recorded every 45 min for a period of 1min. The pacing rate was then increased every 5 min by 30beats/min, and responses at each rate were recorded for thelast 40 s, similar to the protocol in Ref. 37. The maximumpacing frequen cy at wh ich t he sam ple could be capt ur ed for atleast 5 min was dened a s th e maximum capture ra te. Afterreaching the maximum captu re ra te, stimulation was st oppedfor 10 min and then reapplied at 30 and 60 beats/min for 5min each to check for reproducibility of the recorded wave-forms . At the end of the exper iment , double and t r ipleextrastimuli and rapid stimulation at frequencies above themaximum capture rate were applied in an attempt to inducearrhythmia.

All recorded signals were amplied and band-pass lteredbetween 0.3 and 1,000 Hz. The unltered n oise level was 35V, peak to peak , with virt ua lly no 60-Hz componen t. Ana logrecordings were digitized at a sampling rate of 3 kHz using a16-bit an alog-to-digital boar d (AT-MIO-16X, Na tional Inst ru -ments), real-time displayed using LabView data acquisitionsoftware, a nd stored and analyzed using MATLAB (TheMathworks).

Activation times at each recording electrode were deter-mined as the minima of ve-point derivat ives (2) of thelow-pass ltered signals. The stimulus-activation time inter-vals at each electr ode (condu ction t imes) were plotted again stthe corresponding distances and tted by linear regression.The conduction velocity of a propagated beat was calculatedas the inverse slope of the best linear t (16). The peak-to-peak (p-p) am plitudes of th e responses were determ ined from

linearly detrended signals around the activation times. Re-cording si tes with very low or fract ionated (polyphasic)activity were ignored.

For each tissue sample, p-p amplitudes at each electrodeand conduction velocities were averaged from recordingsmad e dur ing the init ial 20 min of pacing at 60 beat s/min (i.e.,over at least 200 beats). Conduction veloci ty, maximumamplitude, and average amplitude were calculated, respec-tively, as th e aver ages of condu ction velocities, ma ximum p-pamplitudes, and all p-p amplitudes from all samples within agroup. The m aximum and average am plitudes, r espectively,represented local and spatially averaged properties of con-structs or ventricles.

Statistics. Data were ca lcu la ted as means S E a n danalyzed using either a paired t -test or one-way ANOVA

Table 1. Morphom etric and biochem ical param eters in 7-day constru cts

Group

Construct Weight and Dimensions Cell Numberin Construct,

10 6

Cell Number perConstru ct Lost Dur ing

Cultivation, 10 6

Glucose Consu mpt ionRate per Construct,

g l 1 day 1

Lactate ProductionRate per Construct,

g l 1 day 1Wet weight , mg Thickness , mm Circular area , mm 2

R egu la r 35 .45 3.33 1.34 0.11 20.14 0.82 3.75 0.59 5.36 0.19 2.21 0.02 1.43 0.12(20.95 0.65)

E nr iched 31. 08 1.71 1.30 0.07 22.92 0.58* 3.00 0.46 4.63 0.31 5.37 0.95* 2.51 0.11*(17.68 1.12)

Values are means SE ; n 5 constr ucts. Par enth etical values are cumu lative lactate deh ydrogenase a ctivity per constr uct, in U/l. Regularand enriched groups were prepared as dened in MATERIALS AND METHODS . *Sta tistically signicant difference between regular and enrichedgroups.



6/13

followed by Fish er s prote cted least s ignican t differen ce posthoc test. To determine time-dependence trends for beatingrat es in monolayer cultures, a un ivariate r epeated-measur esANOVA was used. Differences were considered statisticallysignicant when P 0.05. All calculations were performedusin g SuperANOVA III for Macintosh.

RESULTS

Monolayer cultures. After 24 h of culture, cardiacmyocytes from both the regular and enriched groupsstar ted to contract spontaneously and by d a y 3 4formed synchronously contracting networks. Rates of

contra ction decreased signicantly between culturedays 3 and 7 (P 0.05) in monolayers from both group s.At day 7 , enriched monolayers h ad signicantly highercardiac myocyte fractions and contra ction rates tha nregular monolayers (60.5 1.5 vs. 43.8 0.5% of th ecul ture area, P 0.04, and 169 8 vs . 132 10beats/min, P 0.01), which is consistent with previousreport s (22).

Construct morphology. After 7 days of culture, cell-polymer constru cts appeared discoid [ 5 1.3 mm(diameter thickness); Table 1]. The peripheral zonewas 50 70 m t hick (Fig. 2 A ) and consisted of more cell

layers in the enriched than in the regular group (7 1vs. 5 1 layers, respectively). Cells in this outermostzone formed a continuous, 3-D tissuelike structure byattaching to other cells, spreading along the randomlyoriented PGA bers, and forming bridges between thebers (Fig. 2, A and C ). Distinct cardiac bundles,spa tially oriented groups of cells ( 100 m in size), an dinterstit ial collagen septa were n ot observed. Ran-domly oriented cells in the periphera l zone exhibited avariety of shapes, from elongated cells spread on thepolymer bers to round una ttached cells, as assessed

histologically. The majority of the cells expressed themuscle-specic proteins sarcomeric tropomyosin (Fig.2, C and D ) an d sarcomeric -actin (data not shown).Immediately below the peripheral zone was a 60- to70-m-thick r egion consisting mainly of cells th at didnot express t ropomyosin. At t he const ruct center, cellswere sparsely distributed and either elongated, express-ing tropomyosin, or round, with pyknotic nuclei andacidophilic cytoplasm (Fig. 2 B ).

Cross striations were present in cells in the periph-eral zone of the constru cts as well as in neonatal a ndadu lt ventr icles, as a ssessed immu nohistochemically

Fig. 2. Histology and immun ohistochem-istry of enriched constructs and nativeventricles. A: construct periphery (hema-toxylin and eosin, H E). B : constructcenter (H E). C : construct periphery(tropomyosin). D : const ruc t per iphery(tropomyosin). E : n eon a t a l ven t r i cl e(tropomyosin). F : adult ventricle (tropo-myosin). A C : magnication, 400; bar,50 m. D F : magnication, 1,000; bar,20 m . Brown color indicates tr opomyosin-positive cells. Asterisks denote polymerbers; arrows point to cross str iations.



7/13

(Fig. 2, D F ). The presence of su bcellular elementscharacteristic of cardiac myocytes, including myola-ments with well-dened sarcomeres, z-lines, glycogen

granules (Fig. 3 A ), and mitochondria (Fig. 3 B ) in theoutermost layer of constru cts, wa s demonstra ted bytr an smission electr on m icroscopy (TEM). Cell-to-cellconnect ions were demonstrated by the presence of desm osomes (Fig. 3 C ) and int ercalated disks (Fig. 3 D ).

Construct composition. After 3 days, the respectivenu mbers of viable cells presen t in enriched and regularconst ru cts were 66 an d 57% of th ose seeded at t i m e 0 , a scalculated from medium LDH levels. LDH release be-tween culture days 3 and 7 was one-third of thatbetween days 0 and 3 , indicating that the cell deathrate decreased with cultivation time. At 7 days, cellnumbers in enriched and regular constructs were 38an d 47% of the respective nu mbers seeded a t t ime 0 , asdetermined by the DNA content of constructs . Forcompar ison, 7-day cell monolayers from both groupscontained 61 6% of the initially plated cells. Thenumber of cells seeded at t ime 0 (8 million per PGAdisk) could be accounted for by sum ming cell number sin constructs at day 7 (determin ed from DN A cont ent)an d in t he medium over 7 da ys (calculated from cumu-lative LDH activity/construct) (Table 1), implying thatno signicant cell proliferation occurr ed during thecultivation period. Glucose consumption and lactateproduct ion ra tes were h igher in enr iched than inregular constructs ( P 0.005, Table 1), whereas thelactate-to-glucose molar ra tios were similar for both

construct groups (1.00 0.20 and 1.30 0.11, r espec-tively).

Ventricular tissues from neonatal and adult rats had

respectively six- and threefold higher DNAcontents perunit wet weight (an index of cellularity) than engi-neered constructs from either group ( P 0.01, Fig. 4 A ),which is consistent with the relatively acellular appear-ance of the construct centers (Fig. 2 B ). Relative cellsize, assessed from the ratio of total protein to DNA,was comparable for cells in constructs, neonatal ven-tr icles, an d monolayers an d lower th an for cells in adultventricles ( P 0.01) (Fig. 4 B ). This nding was consis-tent with the relative cross-sectional areas of cells inconstructs and neonatal and adult ventricles observedhist ologically (Fig. 2, D F , respectively). The MTTconversion per un it DNA (an index of metabolic activ-ity) was similar for constructs and neonatal ventriclesand was slightly higher in a dult vent ricles (Fig. 4 C ).

Constru ct electroph ysiology. Sponta neous, m acro-scopic contractions of engineered constructs were visu-ally observed in asks between days 2 and 4 of cult iva-tion, which indicated the presence of intercellularcommun icat ion. At day 7 the majority of constru cts an dnat ive ventricles exhibited tran sient spontan eous beat-ing lasting for 1 10 contra ctions (Fig. 5 A ), which ma yhave resulted from reentrant or triggered activity (4).Electrical stimulat ion r esulted in impu lse propagationin the peripheral cardiac tissue-like zone of the con-stru cts. In contra st, impulses failed to propagate whenthe electrodes were advan ced toward t he centra l acellu-

Fig. 3. Transmission electron micro-graph of an enriched constr uct. A: well-organized myolaments (M) wi thclearly dened sarcomeres, z-lines (Z),and abunda nt glycogen gran ules (Gly).Magnication, 26,000; bar, 1 m. B :several mitochondria (Mito) located be-tween myolaments . Magnicat ion ,

36,000; bar, 400 nm . C : desm osomes(Des) at lateral border of adjacent car-diac myocytes. Magnication, 18,000;bar, 250 nm . D : intercalated disk (ID)with visible desmosomes. Magnica-tion, 18,000; bar, 250 nm.



8/13

lar region of the constructs. All 7-day constructs wereelectrically excitable and could be captured over a widera nge of pacing frequen cies (up to 270 beats/min , Fig. 5, B- D ). Step increases in constru ct pacing frequencyresulted in transient decreases in conduction velocity tosteady-state values (data not shown). Rapid stimula-tion indu ced short ta chyar rh ythmias with rat es close tothe maximum capture ra tes in 3 of 6 enr iched con-structs, 2 of 6 regular constructs (Fig. 5 E ), 1 of 10 adultventricles, and 0 of 10 neonatal ventricles. A separateexperiment showed that constructs remained electri-

cal ly exci table for up to 4 wk of culture (data notshown).

Representative examples of impulse propagation inan enriched construct , a neonatal ventr icle, and anadult ventricle are shown in Fig. 6, A C , respectively.Propagat ing extracellular wa veforms in const ructs a ndnative tissues showed fairly smooth, biphasic shapeswith distinct downward deections that enabled con-dent determination of activation times a nd impliednondecrement al, ma croscopically continuous propaga-tion with out wa ve collisions (34). Notches in extr acellu-lar waveforms occasionally observed in constructs (seeE3 in F ig. 6 A ) may h ave reected asynchronous excita-tion in a djacent groups of cells cau sed by th e presenceof empty space, polymer bers, and/or necrotic tissue(36). Conduction times in ventr icles and constru ctsincreased linearly with distance over 5 mm (Fig. 6 D ,

Fig. 4. Cellularity, hypertrophy, and metabolic activity indexes of constructs and ventricles. A: DNA conten t per unit wet weight (ww). B : protein per unit DNA. C : MTT conversion per unit DNA. Dat arepresent means SE of 5 sam ples. *Sta tistical difference betweenconstructs and native ventricles; stat ist ical difference betweenneonatal and adult ventricles.

Fig. 5. E lectrophysiological recordin gs. A: short transient spontan e-ous beating at a rate of 70 beats/min, which lost regularity after 6 s. B , C , and D : steady-sta te responses after 4 min of pacing at rat es of 80, 150, and 200 beat s/min, respectively. E : short , 5-s tachyarrhyth-mia at 190 beat s/min, apparent ly induced by 4 rapid st imuli at 250beats/min. S and R indicate the stimulus spike and the constructresponse, respectively. All t racings were recorded from enrichedconstr ucts using th e same electrode.



9/13



10/13

regr ession coefficients 0.98), implying similar conduc-tion velocities between adjacent electrodes and thu srelatively homogeneous electrical properties through-out the cardiac tissue-like zone.

Condu ction velocities descended in th e followingorder: adult ventricles, neonat al ventricles, enrichedconstructs, and regular constructs (Table 2, Fig. 6 D )(P 0.001). Lower conduction velocities measured inneonatal than in adult ventricles were consistent with

previously pu blished va lues (40, 41). Condu ction veloci-ties measured in enriched constructs were 30 and50% as h igh as t hose in adult a nd neonat al ventricles,respectively ( P 0.001) and were 27% higher th an inregular constru cts ( P 0.02). Excitat ion thr esholdswere higher in engineered constructs than in nativeventricles ( P 0.001, Table 2) and were lower inneonatal than in adult ventricles ( P 0.01). E xcita tionthresholds of constructs and ventr icles thus var iedinversely with cellular ity indexes (Table 2, Fig. 4 A ).

Maximum and average amplitudes were s igni-cantly lower in constru cts tha n in native ventricles(P 0.0001, Table 2). Maximum amplitudes were1.7-fold higher in enriched than in regular constructs

(P 0.002), wherea s a verage am plitu des did n ot differsignicantly. The ran ge of r ecorded amplitudes was3-fold higher in constructs than in native ventricles

(data not shown). Maximum capture ra tes differedsignicantly ( P 0.001) among groups a nd descendedin the following order: neonatal ventricles, adult ven-tricles, enriched constru cts, and regular constru cts(Table 2). The higher ma ximum captu re ra tes in neona-tal ventricles than in adult ventricles were consistentwith the higher res t ing hear t ra tes a nd higher toler-an ce to ischemia previously reported for neonata l ven-tricles (47).

DISCUSSION

The present s tudy demonstrates that 3-D cardiacmuscle constructs with cardiac-specic structural andelectrophysiological properties can be engineered invitro using isolated cells and biodegradable polymerscaffolds. In part icular, constr ucts contained a periph-

eral cardiac tissue-like zone in which differentiatedcardiac myocytes were organized in multiple layers andattached to other cells and/or polymer bers in a 3-Dcongur ation. Impu lse propagation st udies carried outusing a n arr ay of extr acellular microelectrodes demon-strated tha t the peripheral cardiac tissue-like zone of constru cts su stained ma croscopically cont inuous im-pulse pr opagation (Fig. 6 A ) t h a t d ep en d ed on t h efra ction of seeded cardiac myocytes (Table 2). Fu nc-

tiona l constru cts m ay th us en able in vitro electrophysi-ological studies that may complement those currentlycar ried out u sing th in ventr icular slices (5, 14, 35) andmonolayers of cardiac myocytes (9).

Structurally, constructs were 5 1.3-mm (diam eterthickness) disks and contained a 50- to 70-m-thick out er car diac tissue-like zone composed of cells tha texpressed sarcomeric tropomyosin (Fig. 2 C ) and con-tained myolaments, desmosomes, a nd intercalateddisks (Fig. 3 D ). For comparison, a recently reported (8)hear t muscle model system based on car diac myocyte-populated 3-D collagen gels (15 mm long 8 mmwide 180 m thick) contained several layers of differentiated cells at the edges and less concentratedcells central ly. The small thickness of the cardiactissue-like zone in constr ucts (Fig. 2 A ) and collagengels (8) can be attributed to the low survival rate of meta bolically deman ding cardiac myocytes located moreth an 50 m from a source of gas exchan ge (15).

The molar ra t ios of lactate to glucose of 1 .01.3indicated a erobic cell meta bolism in t he const ru cts (21).Compared with the regular group, enriched constructshad higher glucose consu mption r at es (Table 1), prob-ably due to th e r elatively h igher fraction of myocytes.The absence of cell proliferation in constructs (Table 1)was consistent with the previous ndings tha t neonata lventricular cardiac myocytes lose their ability to prolif-erate after 23 days in vitro (45), whereas broblastsprolifera te slowly in 3-D cultu res (20).

Electroph ysiologically, impu lse pr opagat ion in con-structs was s tudied on a macroscopic level using alinear array of extracellular electrodes (Fig. 1 B ). Int er-electr ode distan ces of 500 m were selected on th e basis

Fig. 6. Impulse propagation (representat ive 66 ms) in a const ruct from the enriched group ( A ), neonat al ventr icle( B ), and adult ventricle ( C ). The extracellular waveform is shown propagating from the electrode closest to thestimu lus site (E1) toward t he furth est electrode still in th e sample (E6). Simulta neous deections at t he beginningof traces (S) represent st imulus artifacts. Amplitude ra nges for each electrode were a djusted to best displayrecorded response waveforms. D : plot of conduction t imes vs. distances relative to E 1, which was assigned thecoordinates (0,0). Conduction velocities were calculated as inverse slopes of best t lines.

Table 2. Electrophysiological param eters in 7-day constru cts and nat ive ventricles

Group nExcitation

Threshold, VConduction

Velocity, cm/sMaximum

Amplitude, m VAverage

Amplitude, m VMaximum Capture

Rate, beats/min

ConstructsRegula r* 6 2.70 0.24 9.35 0.27 0.52 0.05 0.26 0.09 111.7 9.5Enr iched* 6 2.97 0.30 11.89 0.46 0.90 0.14 0.43 0.14 175.0 21.3

VentriclesNeona t a l 10 0.74 0.20 21.82 1.48 31.91 3.53 18.34 4.31 475.0 25.0Adult 10 1.34 0.17 31.69 4.44 25.82 2.81 14.62 3.59 281.2 21.0

Data represent means SE ; n no. of constructs or ventricles. *Signicant difference between constructs and ventricles; signicantdifference between enr iched and r egular constructs; signicant difference between neonat al an d adu lt ventr icles.



11/13

of previously reported in vivo and ex vivo epicardialmapping studies (6, 46, 48). Bipolar point stimulationan d u nipolar recording (16, 25) in th e custom-designedtest chamber (Fig. 1 B ) did not ad versely affect sam pleswith respect t o their electrical properties (waveformshapes were stable) or structure (no apparent tissuedam age was observed h istologically). Automated dat aanalysis was facilitated by the high average signal-to-noise ratios (of 10 and 470 for constructs and nativeventricles, respectively). Whereas 1- to 5-V amplitude,1-ms dur at ion electrical pulses wer e sufficient t o indu ceimpulse propagation in slices of ventricles and in theperipheral zone of 7-day constructs, it was difficult tooverdrive 7-day conuen t monolayers of neona ta l car-diac myocytes even when using stimuli of twice thisamplitude and duration. In addition, impulse propaga-tion in monolayers could not be asses sed usin g extra cel-lular electr odes becau se of fractionat ion a nd low ampli-tudes of recorded waveforms. These ndings may bedue to 3-D electrotonic interactions between cells (9)an d relatively high cell density ar ound th e stimulat ing

and recording electrodes in 3-D constructs comparedwith 2-D monolayers.The inferior electrophysiological properties of con-

structs compared with native ventricles (Table 2) canbe a ttr ibuted t o differences in their ma croscopic tissuearchitecture. In particular, the relatively high excita-tion th resholds (24) and low response a mplitudes wereassociated with low construct cellularity (Fig. 4 A ). Lowmaximum capture rates and conduction velocities inconstructs probably resulted from decreased cell cou-pling, the pr esence of inter cellular clefts, and geometr iccurrent -to-load mismat ches (due to t issue discontinu i-ties) (9, 26). Other mechanisms tha t could contribute toinferior constr uct electrophysiological properties in-

clude cell depolarization, reduced excitability, andslower repolar ization result ing from injur y durin g isola-tion an d/or cultivat ion (32, 39). Int ra cellular recordin gswould be necessary to test t he pr oposed mechanisms.

Compared with enriched constructs, lower conduc-tion velocities, maximum captur e rates, and ampli-tudes in regular constructs probably resulted from 1 )th e higher fra ction of noncardiomyocytic cells, whichwould be expected to form high-resistan ce junctionswith cardiac myocytes (28) and act as passive currentsinks (9), and 2 ) the thinner cardiac tissue-like zone(Table 1). Lower ma ximu m captu re ra tes in th e regularthan enr iched cons t ruct s cou ld a lso be due to therelatively longer duration of cellular action potentials

(as pr eviously observed in brotic compared with nor-ma l car diac tissu e; Ref. 42).Neonatal and adult ventricular t issues did not ex-

hibit sponta neous beat ing ex vivo in a previous (39) orthe present study. In contrast, enzymatically isolatedventricular cardiac myocytes cultured in monolayersare known to revert to a less differentiated phenotype,depolarize, and regain spontan eous cont ractile activityfor as yet unknown reasons (39). In the present study,visible spontaneous contractions in constructs ceasedafter 4 days of cul t ivat ion. This nding might beattributed to gradual depolarization and decoupling of

cardiac myocytes due to injury during cul t ivat ion.However, it is more likely th at th e cult ivation of cardia cmyocytes on 3-D biomaterial scaffolds in tissue culturebioreactors (Fig. 1 A ) promoted differentiated cellularphenotype and function. In support of this hypothesis,Sperelakis (38) showed th at 3-D a ggregates composedof electr ically different iated cardiac m yocytes did notcontr act sponta neously but responded to electricalstimulation.

The aim of the present s tudy was to demonstratebasic cardiac-specic features in constru cts and toevaluate construct structure and electrophysiologicalproperties on a ma croscopic (tissue) level, rat her tha non a cellular level. In ongoing work, we ar e expandingour electrophysiological stud ies to include whole cellclamp and sharp microelectrode intracellular record-ings a nd assessment of the spat ial distribution of thegap junctional protein connexin 43 (23). We are alsoattempting to culture constructs with a thicker cardiactissu e-like zone by direct per fusion of const ru cts dur ingcultivation (to improve ma ss t ran sfer) an d by cocultur-

ing cardiac m yocytes with microvascular endothelialcells (as a rst step toward inducing vascularization).In conclusion, card iac-specic featu res of engineer ed

car diac muscle const ru cts were demonstra ted str uctur-ally and electrophysiologically an d wer e r elated t o thecellular composition of constructs. The 3-D multilayerstructure in conjunction with macroscopic impulsepropagation in engineered constructs can offer advan-tages for in vitro studies of cardiac muscle. In addition,stru ctu ra lly an d functionally improved 3-D engineeredcardiac muscle constructs could be eventually appliedin vivo. To date, attempts to regenerate cardiac tissuehave involved the injection of different muscle celltypes (33, 43) or small t issue fragments (19) into thehear t. Implan ta tion of cardiac muscle constru cts with adened shape instead of isolated cells could potentiallyimprove the efficiency and localization of tissue repair.

N. Bursac and M. Papa daki contr ibuted equally to this study.We thank R. Langer for advice, R. Padera for help with animal

surgery, H. Shing for carrying out the transmission electron micros-copy, Y. Lee for help establishing the electrophysiological recordingsystem, and J. Merok, H. Cho, and P. Gupta for help with biochemicalassays.

This work was supported by National Aeronautics and SpaceAdministrat ion Gra nt NAG9-836.

Address for reprint requests a nd other correspondence: L. E.Freed, Massachusetts Institute of Technology, Div. of Health Sciencean d Techn ology, MIT, Bldg. E25 342, Cambr idge, MA02139 (E-ma il:lfreed @mit .edu ).

Received 7 October 1998; accepted in n al form 8 Mar ch 1999.

REFERENCES

1. Akins, R. E. , N. A. Schroedl , S. R. Gonda, an d C. R. Hartzel l .Neonata l ra t heart cells cultured in simulated microgravity. InVitro Cell Dev. Biol . 33: 337343, 1997.

2. Blan cha rd, S. , W. Smith , R. Damian o, D. Molter, R. Ideke r,a n d J . L o w e . Four digital algorithms for activation detectionfrom unipolar epicardial electrograms. IEEE Trans. Biomed. Eng . 36: 256261, 1989.

3. C a rr i e r, R ., M . P a p a d a k i , M . R u p n i c k , F. J . S c h o e n , N .Bursac, R. Langer, L. E. Freed, and G. Vunjak-Novakovic.Cardiac tissue engineering: cell seeding, cultivation parameters



12/13


13/13

Documents

Electrophysiology of Cardiac Cells