Identification and functional significance of a new class of non-consensus-type plastid promoters

AF Martin, K Ball, LZ Gao, P Kumar and RJ Solaroheart myofibrils

Identification and functional significance of troponin I isoforms in neonatal rat

ISSN: 1524-4571 Copyright © 1991 American Heart Association. All rights reserved. Print ISSN: 0009-7330. Online

TX 72514Circulation Research is published by the American Heart Association. 7272 Greenville Avenue, Dallas,

1991, 69:1244-1252Circulation Research

http://circres.ahajournals.org/content/69/5/1244located on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://www.lww.com/reprintsReprints: Information about reprints can be found online at [email protected]. E-mail:

Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters http://circres.ahajournals.org//subscriptions/Subscriptions: Information about subscribing to Circulation Research is online at

by guest on July 10, 2011http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/

1244

Identification and Functional Significance ofTroponin I Isoforms in Neonatal Rat

Heart MyofibrilsAnne F. Martin, Karen Ball, Lizhu Gao, Pankaj Kumar, and R. John Solaro

We investigated the mechanism(s) responsible for differences in the effects of acidic pH on Ca2'activation of the activity of adult and neonatal rat heart myofilaments. Studies on preparationsof myofilaments reconstituted with adult troponin-tropomyosin (Tn-Tm) and either adult or

neonatal thick filaments indicated that the difference in effect of acidic pH is related todifferences in Tn-Tm and not other myofilament proteins. Immunoblotting analysis showedthat development of the rat heart myofibrils is associated with isoform switching from slowskeletal TnI to cardiac TnI and from a slow mobility isoform of TnT (TnTI) to a faster M,isoform (TnT2). Expression of slow skeletal TnI was associated with a relative insensitivity ofmyofilament Ca24 activation to deactivation by acidic pH. Moreover, the effect of acidic pH onCa2' activation of ATPase activity of soleus myofibrils, which contain cardiac TnC and slowskeletal TnI, was essentially the same as the effect of acidic pH on rat cardiac myofibrils in theearly neonatal period. Neonatal myofilaments also contained a relative abundance of a setof polypeptides copurifying with the thin filaments. We have identified these proteins ashistones. The relative amount of histones among a variety of preparations from different specieswas not correlated with the pH sensitivity of myofibrillar Ca2' activation. Shifts in TnTisoforms among these species were also not correlated with an altered response to acidic pH.Our data provide evidence in support of the hypothesis that the relative insensitivity of neonatalmyofilament activity to acidic pH is due to the presence of slow skeletal TnI in the thin-filamentregulatory complex. (Circulation Research 1991;69:1244-1252)

T he aim of work presented here was to deter-mine functional effects of isoform switchingof troponin (Tn) I in mammalian cardiac

myofibrils. TnI is a key inhibitory component of thethin-filament regulatory complex.' Together withTnC, the Ca2` receptor, and TnT, a tropomyosin(Tm) binding subunit, TnI confers thin-filament-related Ca2` sensitivity to the activity of striatedmyofilaments. Our hypothesis is that shifts in theisoform population of TnI alter Ca2` activation ofheart myofibrils, especially at acidic pH.

Evidence indicating that developing cardiac myo-fibrils may have altered thin-filament regulation

From the Department of Physiology and Biophysics (K.B., L.G.,R.J.S.), University of Illinois, College of Medicine, Chicago, Ill.,and the Departments of Pharmacology and Cell Biophysics(A.F.M.) and Physiology and Biophysics (P.K.), University ofCincinnati, College of Medicine, Cincinnati, Ohio.

Supported by National Institutes of Health grants RO1 HL-22231 (R.J.S.), P01 HL-22619-IIIB (R.J.S.), and T32 HL-07692(K.B.) and by American Heart Association grant 84-102 (A.F.M.)and grant SW-88-29 from the Ohio affiliate (A.F.M.).Address for reprints: R. John Solaro, PhD, Department of

Physiology and Biophysics (M/C 901), University of Illinois, Col-lege of Medicine, Box 6998, Chicago, IL 60680.

Received March 20, 1991; accepted June 19, 1991.

came from studies investigating the effect of acidicpH on Ca2' activation of dog2 and rat3 cardiacmyofibrillar force and ATPase activity. Results of thiswork showed that the decreased Ca24 sensitivity ofmyofibrillar activity associated with acidic pH wassignificantly less in the neonatal rat and dog heartthan in the adult heart. We2,3 suggested that themechanism for this altered response to acidic pH wasa shift in the population of isoforms of TnI. This ideawas based on several observations: 1) analysis of themyofibrils by sodium dodecyl sulfate-polyacrylamidegel electrophoresis (SDS-PAGE) showing multiplebands in the region of migration of TnI and areduced amount of adult cardiac TnI,23 2) alterationsin the relative amounts of adult TnI induced byhypothyroidism correlated with the relative effect ofacidic pH on myofilament Ca21 sensitivity,4 and 3)studies56 showing that the effect of acidic pH on Ca2+binding to pure TnC was much less than its effect onCa2' binding to TnC in the TnI-TnC complex.Although there is no evidence that the developing

mammalian heart expresses isoforms of TnC,2,7 thereis now clear evidence that, during fetal and neonataldevelopment of the rat heart, slow skeletal TnI



Martin et al TnI Isoforms in Development 1245

(ssTnl) is expressed.8'9 Yet, whether the alteredeffects of acidic pH on Ca2' activation of the myo-fibrils from the newborn heart are due to TnI switch-ing is not clear. Immature hearts have differences inisoforms of other myofilament proteins such as myo-sin heavy chain,10 Tm,11 and TnT.812 We2 have alsoshown that there are differences in the relativeamounts of thin-filament-associated proteins hereto-fore unidentified.Our objective in doing the present experiments

was to identify these thin-filament proteins and toprovide clearer evidence on the correlation betweenTnI isoform switching and developmental differencesin effects of acidic pH on myofibrillar Ca2' activation.Our results indicate that these functional differencesare localized to Tn-Tm and that the thin-filament-associated proteins are, in fact, histones that copurifywith the myofibrils. Our results support the hypoth-esis that, among the changes in myofilament proteinsthat occur during development, shifts in TnI isoformsare most closely associated with the altered responseof myofilament activity to acidic pH.

Materials and MethodsPreparations

Myofibrils. Sprague-Dawley rats and New ZealandWhite rabbits were used for all preparations. Ratswere anesthetized with ether, and rabbits were anes-thetized by intraperitoneal or intravenous injectionof pentobarbital. After deep anesthesia, hearts wererapidly removed, placed in ice-cold saline, andcleaned free of atria and connective tissue, andventricular tissue was used for preparation of cardiacmyofibrils. If ventricular tissue was not used imme-diately, the tissue was quick-frozen in liquid nitrogenand stored at -80°C. Isolated cardiac myofibrils wereprepared by Triton X-100 extraction as describedpreviously.13 The procedure was modified by additionof the proteolytic enzyme inhibitors pepstatin (0.5gg/ml), leupeptin (0.5 ,ug/ml), and phenylmethylsul-fonyl fluoride (0.2 mM) at all steps of the isolation.Protein concentrations were determined by themethod of Lowry et al.14 Assays using the myofibrilswere completed within 24 hours of preparation.

Preparations ofmyofibnillarproteins. Myofibrils lack-ing Tn-Tm (desensitized myofibrils) were preparedas described previously.15 Isolated myofibrils werewashed extensively and dialyzed against a buffercontaining 1 mM dithiothreitol, 0.1 mM NaN3, and 5mM Tris, pH 8.0. This resulted in removal of Tn andTm and loss of Ca21 sensitivity. Tn-Tm complex wasisolated from beef heart ventricular muscle as de-scribed by Stull and Buss.16 Reconstituted prepara-tions were formed as previously described17 by incu-bating the desensitized myofibrils with Tn-Tm at aratio of 2: 1 (wt/wt) for 15 minutes in a solution of 250mM KCl, 2 mM MgCl2, and 30 mM imidazole, pH7.0. After dilution to 60 mM KCI, the hybrid prepa-rations were collected and washed by centrifugation

and resuspension. TnC was prepared and covalentlyattached to Sepharose 4B as previously described.18

ProceduresMyofibrillar A TPase activity. Ca2'-dependent

MgATPase activity of the isolated myofibrils wasmeasured by determination of inorganic phosphaterelease during a 10-minute incubation at 30°C. Otherconditions were 2 mM Mg'2, 60 mM imidazole, 5 mMMgATP2-, and an ionic strength of 120 mM. Ca'+concentration was varied over a broad pCa (-log ofthe molar free Ca2 concentration) range and assayswere run at pH 7.0 and 6.5. Total concentrations ofCaCl2, EGTA, KCI, MgCl2, and ATP were calculatedas previously described.4"5 The reactions werestarted with ATP and stopped with ice-cold 10%trichloroacetic acid. The precipitate was removed bycentrifugation, and inorganic phosphate in the super-natant fraction was determined as described byCarter and Karl.19 The pCa-ATPase activity rela-tions were fit to an empiric form of the Hill equationusing a nonlinear least-squares regression. Statisticaldifferences in the curves were evaluated using Stu-dent's t test.

Gel electrophoresis and immunoblot analysis. Myo-fibrillar preparations were analyzed by SDS-PAGEas described by Laemmli.20 In some cases, afterSDS-PAGE, proteins were electroblotted onto nitro-cellulose for immunoblotting as described by Towbinet al.2'Binding of myofibrillar proteins to Sepharose-TnC.

Interactions of myofibrillar proteins with TnC weremeasured by incubating urea extracts with TnC co-valently attached to Sepharose. The myofibrils weredissolved in urea-Tris buffer (9 M urea, 100 mM Tris,and 30 mM ,B-mercaptoethanol, pH 8.0) and dialyzedagainst urea-Tris buffer for 5 hours. The extract wasthen mixed overnight with Sepharose-TnC in thepresence of either 1 mM CaCl2 or 10 mM EGTA.The mixture was centrifuged to pellet the Sepharose-TnC, and samples of the supernatant fraction weredialyzed against 1% SDS and analyzed by SDS-PAGE. Binding of TnC to myofibrillar proteins wasalso measured by gel-overlay experiments using 12511labeled TnC according to the method described byCarlin et al.22

Protein sequencing and amino acid composition. Thelow molecular weight proteins present in myofibrilsfrom neonatal rat heart were separated electro-phoretically and transferred to polyvinylidene diflu-oride membrane as described by Matsudaira.23 Theelectroblot was performed at 300 mA for 45 minutesat 40C. The membrane was stained briefly with 0.1%Coomassie blue in 50% methanol, and bands corre-sponding to the four prominent low molecular weightpolypeptides (Figures 2-4, P2-P5) and P1, whichmigrated behind cardiac TnI (cTnI), were sequencedon a gas phase sequencer (model 470A, AppliedBiosystems, Inc., Foster City, Calif.). The resultantphenylthiohydantoin derivatives were analyzed on aWaters Nova-Pak C18 reverse-phase column. All



1246 Circulation Research Vol 69, No 5 November 1991

100.0 A

JO.O1~~~.T.

60.0-- 0

60.0 ..40.07.

20.0. -O*pH00.

. 7.5 7.0 6.5 6.0 5.5 5.0 4

pCa

100.0

80.0

60.0

40.0

20.0

8.0 7.5 7.0 6.5 6.0 5.5 5.0

c_i-

0R

..5

pCa

i.--F-

100.0-

80.0

60.0

40.0

20.0

U .U 7.0 ' 6.08.0 7.5 7.0 5. . .S

pCa pCa

FIGURE 1. Graphs comparing the effect of acidic pH on Ca2l-dependent actomyosin ATPase activity of adult (panel A) and7-day-old (panel C) rat cardiac myofibrils with preparations in which the native troponin-tropomyosin has been exchanged withbeef heart troponin-tropomyosin in the adult (panel B) and neonatal (panel D) preparations. ATPase activity is normalized tomaximal activity. The preparations were incubated atpH 7.0 (o) andpH 6.5 (0) in 5 mMMgATP2-, 2 mMMg 2+, 20 mMKCl,60mM imidazole, 1 mMEGTA, and 0.2 mgproteinlml with varyingpCa values. Data are shown as mean±SEMfor three to fiveexperiments. Curves were generated by nonlinear least-squares regression fitting of the data to the Hill equation. The effect ofacidicpHproduced equally significant shifts (p<O.Ol) in the pCa-ATPase relations in experiments depicted in panels A, B, and D. ThepCa-ATPase relations depicted in panel C were not significantly different. See "Materials and Methods" for details.

amino acid analyses were determined using the Wa-ters Pico-Tag system. Amino acid analyses and pro-tein sequences were performed by the Protein Chem-istry Core Facility of the Department of Pharmacologyand Cell Biophysics, University of Cincinnati.

ResultsA distinctive feature of the Ca2' regulation of

myofibrils from neonatal rat hearts is that they aremore resistant to deactivation by acidic pH thanadult myofibrils.2-4 An important question is whetherthis developmental difference in the effect of acidicpH resides in variations in thin- or thick-filamentproteins. We approached this question by exchangingneonatal with adult thin-filament regulatory proteins.First, we extracted the Tn-Tm complex from neona-tal and adult myofibrils using a low ionic strengthextraction procedure that was a slight modification ofthe procedure described by Lehman.24 As previouslyreported,15 the Mg2+-ATPase activity of these prep-arations was turned on and independent of Ca2 .

These "desensitized" myofibrils were reconstitutedwith Tn-Tm prepared from adult beef heart. TnIfrom beef heart closely resembles that of the adult ratheart.25 Thus, the reconstituted preparations had the

same thin-filament regulatory complex, but they con-tained thick filaments native to adult and neonatalmyofilaments. Relations between pCa and Mg2`-ATPase activity of reconstituted native and adult andneonatal myofibrils at pH 7.0 and pH 6.5 are shownin Figure 1. The data indicated that the relativeinsensitivity of the neonatal myofibrils to acidic pHwas not present in hybrid myofibrils containing neo-natal thick filaments and adult Tn-Tm. Substitutionof beef adult Tn-Tm for native adult Tn-Tm did notinfluence the effect of pH on myofilament Ca2'activation (Figures 1A and 1B). Substitution of adultTn-Tm for neonatal Tn-Tm conferred pH sensitivityof Ca2' activation similar to that seen in native adultpreparations (Figures 1C and 1D). We conclude thatthe differences in activity of neonatal and adultmyofibrils are most likely due to differences in Tn-Tm.What variations in thin filaments could account for

the differences in Ca2' and pH dependence betweenadult and neonatal myofibrils? As previously re-

ported by us2 and by Toyota et al,7 it is apparent that,throughout development of the heart, a single TnCisoform is expressed. On the other hand, there isclear evidence for expression of isoforms of TnI and

I-50

_J

F:

0

-J

x

>R

C C

Y7.0

- OpH 7.0 /,..*pH6&S

D S

-OpH 7.0*pHS.5

. - . . . . .

5.0 4.5U.U l- c OF- * -. * . . * . . .V.U nn. ^ -,i, a e0 a

--T



Martin et al TnI Isoforms in Development 1247

A MYOSIN

HEAVY CHAIN01_-----

F * .1

~ ~ ~ 4

- - _- - _ _ ~~~~~~~~~ACTIN_ + .v = ~~~~~~~~~TNTNTM

4-

4 a

_-4pm

-b - TNI

_--0LC1

-m~ _e - LC2

Ua

U,,_3

- -

*-W

F S 1D 5D 12D 211 ADULI

B

cTNI

ssTNIfsTNI

FIGURE 2. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis andWestem blot analysis of myofibrillarproteinsfrom rat cardiac muscle duringdevelopment from neonatal day 1 toadult andfrom tibialis anteriorfast (F)and soleus slow (S) rat skeletal muscle.TNT, troponin T; TM, tropomyosin;cTNI, cardiac troponin I; LCI, myosinlight chain 1; LC2, myosin light chain2; ssTNI, slow skeletal troponin I;fsTNI, fast skeletal troponin I; TNT-i,slow mobility isoform of TNT; TNT-2,faster M, isoform of TNT; D, days afterbirth. Panel A: Coomassie blue-stained protein profile. Panel B: West-em blot analysis of TNI isoforms.Panel C: Western blot analysis of TNTisoforms.

F S 1D 5D 12D 21D ADULT

C

a -_ TNT-1e 0-0-TNT-2

5D 12D 21D ADULT

TnT.8,9,12 In the results shown in Figure 2, we haveanalyzed neonatal and adult heart preparations bySDS-PAGE (Figure 2A) and Western blots with a

monoclonal antibody to rabbit skeletal TnI (Figure2B) that reacts with all variants of TnI-ssTnl, fastskeletal TnI, and cTnI. This monoclonal antibodywas generously provided by Dr. James Potter of theUniversity of Miami, School of Medicine. The SDS-PAGE profile of myofibrillar proteins did not change

after phosphorylation of the myofibrils with cAMP-dependent protein kinase (data not shown). Figure2C also illustrates a similar analysis using a monoclo-nal antibody to TnT.26 These results show that duringneonatal development of rat heart myofibrils, there isa switch in the isoform population of TnI and TnTconsistent with earlier reports using ventricular tis-sue. Early in neonatal life ssTnl was predominantlyexpressed, but by day 21, cTnI was the predominant

a

1-Ao*

cTNT 1D

._1-__!_s

..I s,

,1,

1-em .l-



1248 Circulation Research

MYOSINHEAVY CHAIN

P" W-

ACTIN

TNTTM

cTNI -

LCLLm~2~

P1

cTNI

a~

P3P4

P5

ADULT

isoform. In the case of TnT, an embryonic isoform(TnT,) with slower mobility than the adult isoform(TnT2) was abundant early in neonatal life up to day12, when TnT2 became predominant. The resultsshown in Figure 2 indicate that the time frame of thedevelopmental changes in the expression of Tnl andTnT isoforms in the rat heart are similar. We4 havepreviously shown that the relative abundance of cTnIis correlated with the shift in pCa50 associated with a

0.5-unit drop in pH.Electrophoretic profiles of myofibrillar proteins

(Figure 2A) show the presence of two prominentbands in neonatal heart, running with a mobility justbehind adult eTnI. Moreover, there are several poly-peptides with mobilities that are present with greaterabundance in the neonatal myofibrils. Evidence pre-sented in Figure 3 indicates that these polypeptides(P1-P5) interact with TnC. In these experiments,myofibrils were dissolved in urea and incubated withSepharose-TnC. Under these conditions TnI in theextract will bind with the Sepharose-TnC in thepresence of calcium.18After sedimenting the Sepha-rose-TnC, we examined the supernatant fraction bySDS-PAGE (Figure 3). The results indicated thatTnI interacts with the Sepharose-TnC, as expected.However, the gels also show that the low molecularweight proteins also bind to Sepharose-TnC. Bindingof the polypeptides to TnC occurred independent ofthe Ca2' ion concentration.We now present evidence that the polypeptides

P1-P5 are, in fact, histones that copurify with themyofibrils. We identified the proteins by amino acidsequencing (Table 1) and by comparison with elec-

FIGURE 3. Binding of myofibrillar-- proteins including polypeptides (P1-e'_t P5) to Sepharose-troponin C.

Coomassie-stained profiles of adult_, _ (left side) and 12-day-old (right side)

myofibrillar proteins were separated bya- sodium dodecyl sulfate-polyacrylamide

gel electrophoresis. TNT, troponin T;TM, tropomyosin; cTNI, cardiac tropo-nin I; LC1, myosin light chain 1; LC2,myosin light chain 2; Before, samples

_ dissolved in urea as described in "Ma-terials and Methods" and then pro-

cessed for electrophoresis; After, sam-

ples dissolved in urea and incubatedwith Sepharose-troponin C and thenprocessed for electrophoresis as de-

scribed in 'Materials and Methods."

12D

trophoretic profiles of calf thymus histones (Figure4). The sequence obtained from low molecularweight bands P2 and P3 was homologous to histoneH3 and histone H2B, respectively (Table 1, Figure 4).No sequence was obtained from P1, P3, or P5, whichhave electrophoretic mobilities similar to those ofhistones Hi, H2a, and H4, respectively, all of whichhave blocked N-termini.27 The Lys/Arg ratio fromamino acid analysis is consistent with the assumptionthat P1 P3, and P5 are histones Hi, H2a, and H4.27We also tested the interaction of a calf thymushistone preparation with [1251I]TnC by gel-overlayexperiments (data not shown). The results of thesestudies confirmed the results shown in Figure 3showing that TnC binds to the histones as well as toTnI. The binding of histones to TnC explains our

initial observation that these proteins were associ-ated with the thin filament.2

TABLE 1. Analysis of Low Molecular Weight Polypeptides P1-P5

Protein Amino acid sequence Histone

P1 Blocked N-terminal HI?P2 Ala-Arg-Thr-Lys-GIn-Thr-Ala-Arg-Lys- H3

Ser-Thr-Gly-Gly-Lys-Ala-Pro-Arg-Lys-Glu-Leu

P3 Pro-Glu-Pro-Ala-Lys-Ser-Ala-Pro-Lys- H2bLys-Gly-Ser-Lys

P4 Blocked N-terminal H2a?P5 Blocked N-terminal H4?

See Figures 2 and 4 for sodium dodecyl sulfate-polyacrylamidegel electrophoresis

Vol 69, No 5 November 1991

~_

I*4~



Martin et al Tnl Isoforms in Development 1249

a_t MYOSINHEAVY CHAIN

F

a- ACTIN

-I-TNT

TM

--_

P1

~ _ oeTNI

_ _ ~~~~LC1

L>C2

-P2 - P5

_-

HISTONES 5D ADULT

FIGURE 4. Comparison of electrophoretic profiles of calfthymus histones and neonatal and adult cardiac myofibrillarpreparations separated on sodium dodecyl sulfate-polyacryl-amide gels. TNT, troponin T; TM, tropomyosin; PF-P5,polypeptides; cTNI, cardiac troponin 1; LCI, myosin lightchain 1; LC2, myosin light chain 2; D, days after birth. SeeTable 1 and text for further discussion.

Does the presence of histones affect the pCa-ATPase relation and its sensitivity to acidic pH? Wehave addressed this issue by comparing myofibrillarpreparations that have different histone contents butno difference in TnI composition. These experimentsindicate that the relative amount of the histonespresent in various preparations does not correlatewith the relative insensitivity to acidic pH. In theresults depicted in Figure 5, for example, cardiacmyofibrils prepared from 22-day-old neonates andadult rat hearts showed a substantial difference in theabundance of histones. Yet the effect of acidic pH on

the pCa50 for activation of myofibrillar ATPase activ-ity was shifted to the same extent in both prepara-tions. Similar results were obtained by comparisonsof SDS-PAGE profiles and effects of acidic pH inmyofibrillar fractions from fetal and adult guinea pighearts,28 5-day-old and adult rabbit hearts,29 and1-day-old calf and adult beef hearts (authors' unpub-

lished results). In general, these experiments indicatethat the relative amount of the histones present invarious preparations has little or no effect on thesensitivity of Ca2' activation of myofibrils to changesin pH. In a second approach, we added exogenouscalf thymus histones to adult myofibrils in a ratiosimilar to that apparent in the neonatal preparations.The normalized pCa-myofibrillar ATPase relationsat pH 7.0 were unaffected by addition of exogenoushistones. Similar experiments at pH 6.5 were difficultto interpret in that the addition of the histonesresulted in a partial loss of Ca' regulation. Yet,myofibrillar preparations containing endogenous his-tones were well regulated by Ca`. This suggests thathistones added exogenously may bind to differentsites than histones copurifying with the myofibrils.

Results presented so far indicate that the relativeinsensitivity of neonatal myofibrils to deactivation byacidic pH is best correlated with a shift in isoformpopulation of TnI from one containing a predom-inance of eTnI to one in which ssTnl is the majorform. A straightforward prediction is that the effectof acidic pH on Ca' activation of slow skeletalmyofibrils should be more similar to the effect onneonatal than adult cardiac myofibrils. Althoughthere are differences in the isoforms of Tm and TnTthat are present in neonatal rat heart and slowskeletal myofibrils, both contain cTnC and ssTnl.30 Ina test of the effect of acidic pH on calcium activationof slow skeletal myofibrils (data not shown), ourexperiments showed that the pCa-ATPase activityrelation was not significantly different when mea-sured at pH 7.0 and pH 6.5.

DiscussionThere is now clear evidence indicating that there is

expression of isoforms of TnI during development ofthe heart. Using antibodies to fast skeletal TnI as aprobe, Toyota and Shimada3' first suggested thepossible expression of a skeletal variant of TnI inembryonic chick heart muscle. More recent immuno-logical studies89 have also provided strong evidencethat distinct isoforms of TnI are expressed duringdevelopment of the rat heart. These studies showthat there is a switch in isoform population fromssTnl to cTnI. Results presented here confirm andextend these findings in that we show that thesechanges occur in the myofibrils themselves. More-over, the temporal pattern of expression of eTnI andthe identification of a neonatal/embryonic isoform ofTnI as ssTnl is now strongly supported by molecularcloning studies of the rat cTnI32,33 and ssTnI34 genes.Martin and Orlowski32 showed approximately a 15-fold induction of expression of cTnI mRNA in theadult compared with the 14-day embryonic rat heart.Murphy et a133 reported that steady-state levels ofcTnI mRNA increased throughout fetal and earlyneonatal development of the heart. Murphy et a133have also shown that the nucleotide sequence of aclone of TnI obtained from neonatal cardiac musclehad more than 99% homology with rat ssTnI.34 This




FIGURE 5. Graph comparingthe effect of acidic pH on thepCa-ATPase activity relation ofmyofibrils prepared from 22-day-old (22D) and adult (A) rat car-diac myofibrils. ATPase activitywas normalized relative to maxi-mal activity and was measured asdescribed in the legend to Figure2. Inset: Protein profiles showingincreased relative levels of his-tones in 22-day-old comparedwith adultpreparations. Data aremean±SEM for n=3. Curveswere generated by nonlinearleast-squares regression fitting ofthe data to the Hill equation.Acidic pH produced equally sig-nificant shifts (pc0.01) in thepCa-ATPase relations in experi-ments on adult and 22-day-oldmyofibrils.

ssTnl cDNA was shown to hybridize with mRNAfrom slow skeletal TnI as well as mRNA from fetaland neonatal rat heart. The time course of expressionof the ssTnl and cTnI mRNA was similar to thepattern of expression of these proteins.33 Takentogether, these data indicate that, during the fetaland neonatal period, regulation of expression ofssTnl and cTnI is oppositely directed.Our results support the idea that this TnI isoform

switching largely accounts for the differences be-tween adult and neonatal myofibrils with regard toinhibition of Ca> activation by acidic pH. Experi-ments reported here (Figure 1) and in our earlierstudies26,15 indicate that the differential effects ofacidic pH on neonatal and adult myofilaments are

not related to shifts in myosin heavy chain isoforms.Although shifts in TnT isoforms may account fordifferences in Ca>' activation at a particular pH,29935they do not correlate with the differences in the effectof acidic pH on Ca>2 activation. For example, thereare prominent differences in the TnT isoform popu-lation of myofibrils from 5-day-old neonatal andadult rabbit hearts,29 but there is no difference in theeffect of acidic pH on myofibrillar Ca> activation.Moreover, our previous studies5 indicated that,whereas TnI binding to TnC alters its Ca> affinityand response to acidic pH, there are no furtherchanges when TnT is incorporated into the complex.The affinity of TnC for TnI is reduced at acidic pH,5and we think that this might be the main reason forthe reduced affinity of TnC for Ca>2 at acidic pH.Our hypothesis is that ssTnl binds cTnC with an

affinity that is different from cTnI and is affected byacidic pH to a lesser extent. Of course, this does notrule out the possibility that acidic pH affects otherdomains in the thin-filament Ca2+-signaling process.

Although there is relatively high homology among

regions of ssTnl and cTnl, there are differences inprimary structure that might account for the differ-ential effects of acidic pH on Ca' activation ofmyofibrils from neonatal/fetal ventricular or soleusmuscle and adult rat ventricular muscle.25,31' Themost prominent difference between skeletal TnI andcTnl is in the amino-terminal region, where cTnI hasa 32-amino acid stretch that is absent in skeletalTnI.25,30 This region of cTnI contains two sequentialserine residues at positions 22 and 23, which are

potential sites for phosphorylation.25 This region ofthe molecule most likely affects the Ca2+-bindingproperties of TnC. Earlier studies in our laboratory36showed that the phosphorylation of this region of themolecule reduces the pCa50 for Ca>2 binding to theregulatory binding site on cTnC. However, wel' havealso shown that the effect of acidic pH on the pCa50for activation of myofibrillar ATPase activity was thesame in myofibrils containing dephosphorylated andphosphorylated TnI. Therefore, this domain mightnot be an important determinant of the differentialeffects of acidic pH on activation of myofilamentscontaining ssTnl and cTnI. Other domains thatmay be important in these differences are the TnT-binding domains, the TnC-binding domains, and theinhibitory domain that reacts with actin-Tm. Al-though in some cases these domains are assigned byanalogy to their counterpart in fast skeletal TnIrather than by direct examination of the cardiacisoforms, it is likely that the general function of a

particular domain is conserved among variants ofTnI. All of these domains show diversity that may berelated to the difference we report here. Yet, themost likely regions that may be affected differently byacidic pH are in the TnC-binding domains and theinhibitory domain. These regions are highly con-

100.0

>

cE0

80.0

60.0

40.0

20.0

0.0

pCa



Martin et al Tnl Isoforms in Development 1251

served among cTnls from a variety of species but doshow structural diversity with their counterparts inssTnI, in particular in the N-terminal TnC-bindingdomain.25,303334 On the other hand, a second TnC-binding domain, residues 129-149, shows less diver-sity between ssTnl and cTnI.We have shown that histones are present in our

myofibrillar preparations. Yet, our data indicate thatthe relative abundance of histones in the neonatalmyofibrillar preparations is not correlated with theeffects of acidic pH on the pCa50 for myofibrillar Ca21activation. It is apparent though that histones maymodify the actin-myosin interaction. There is evi-dence that histones bind to actin and inhibit actomy-osin ATPase activity.37,38 Therefore, it is important torealize that histones are present in some prepara-tions, that they potentially bind to TnC, and thataddition of exogenous histones to adult myofibrillarpreparations may affect the ATPase rate. We do notknow exactly how to interpret the discrepancy be-tween our findings with exogenous and endogenoushistones. The result does indicate that exogenouslyadded histones may bind to different sites than thehistones that are present in the myofibrillar fractionsprepared by homogenization and differential centrif-ugation. Although there is some evidence for cyto-plasmically located histones,39 we do not yet knowwhether the histones are present in vivo in themyofibrillar space.An important question is whether the difference in

effect of acidic pH on neonatal myofibrils is reflectedin the response of intact cell and heart to changes inoxygenation and pH. Our previous studies3 indicatedthat the initial transient decrease in force associatedwith hypercapnic acidosis in adult heart papillarymuscle was significantly less in the neonatal heartpreparations. In preparations from both the neonateand adult, the Ca21 transient was increased duringthis response; thus, Ca2' availability was not limiting.We concluded that the altered response of the neo-natal myofilaments to acidosis might be reflected inthe intact myocardium during this condition. Studiescomparing adult and neonatal rabbit heart prepara-tions have indicated that this difference in responsemay also contribute to a difference in the buffercapacity of the adult and neonatal preparations.40There are also species differences in the time

course of developmental expression of ssTnl andcTnI. For example, the amount of cTnI relative toTnI expressed in newborn rabbit hearts (data notshown) is similar to that in 10-12-day-old rat hearts.It is important to note that myofilaments from 1-day-old rabbit hearts show a decrease in the effect ofacidic pH on Ca2' activation. However, by 5 days ofage, the newborn rabbit heart expresses little ssTnl,and the effect of acidic pH is essentially the same asthat obtained with the adult rabbit heart.

AcknowledgmentsWe thank Kimberly Palmiter and David Robinson

for providing excellent technical assistance. We also

thank Linda Alaniz, Janice Gentry, and LaVerne Ri-chardson for help with preparation of the manuscript.

References1. Solaro RJ, Pan B-S: Control and modulation of contractile

activity of cardiac myofilaments, in Sperelakis N (ed): Physi-ology and Pathophysiology of the Heart, ed 2. Boston, KluwerAcademic Publishers, 1988, pp 291-303

2. Solaro RJ, Kumar P, Blanchard EM, Martin AF: Differentialeffects of pH on calcium activation of adult and perinatal doghearts. Circ Res 1986;58:721-729

3. Solaro RJ, Lee JA, Kentish JC, Allen DG: Effect of acidosison ventricular muscle from adult and neonatal rats. Circ Res1988;63:779-787

4. Dieckman LJ, Solaro RJ: Effect of thyroid status on thin-filament Ca2' regulation and expression of troponin I inperinatal and adult rat hearts. Circ Res 1990;67:344-351

5. El-Saleh SC, Solaro RJ: Troponin I enhances acidic pH-induced depression of Ca2' binding to the regulatory sites inskeletal troponin C. J Biol Chem 1988;263:3274-3278

6. Solaro RJ, El-Saleh SC, Kentish JC: Ca2', pH, and theregulation of cardiac myofilament force and ATPase activity.Mol Cell Biochem 1989;89:163-168

7. Toyota N, Shimada Y, Bader D: Molecular cloning andexpression of chicken cardiac troponin C. Circ Res 1989;65:1241-1246

8. Sabry MA, Dhoot GK: Identification and pattern of expres-sion of a developmental isoform of troponin I in chicken andrat cardiac muscle. J Muscle Res Cell Motil 1989;10:85-91

9. Saggin L, Gorza L, Ausoni S, Schiaffino S: Troponin Iswitching in the developing heart. J Biol Chem 1989;264:16299-16302

10. Chizzonite RA, Zak R: Regulation of myosin isoenzymecomposition in fetal and neonatal rat ventricle by endogenousthyroid hormones. J Biol Chem 1984;259:12628-12632

11. Humphreys JE, Cummins P: Regulatory proteins of the myo-cardium. J Mol Cell Cardiol 1984;16:643-657

12. Saggin L, Ausoni S, Gorza L, Sartore S, Schiaffino S: TroponinT switching in the developing rat heart. J Biol Chem 1988;263:18488-18492

13. Pagani ED, Solaro RJ: Methods for measuring functionalproperties of sarcoplasmic reticulum and myofibrils in smallsamples of myocardium, in Schwartz A (ed): Methods inPharmacology. New York, Plenum Publishing Corp, 1984, vol5, pp 44-61

14. Lowry DH, Rosebrough NJ, Farr AL, Randall RJ: Proteinmeasurement with the Folin phenol reagent. J Biol Chem1951;193:165-175

15. Blanchard EM, Solaro RJ: Inhibition of the activation andtroponin Ca2' binding of dog cardiac myofibrils by acidic pH.Circ Res 1984;55:382-391

16. Stull JT, Buss JE: Phosphorylation of cardiac troponin bycyclic adenosine 3' :5'-monophosphate-dependent proteinkinase. J Biol Chem 1977;252:851-857

17. Garvey JL, Kranias EG, Solaro, RJ: Phosphorylation of C-protein, troponin I, and phospholamban in isolated rabbithearts. Biochem J 1988;249:709-714

18. Solaro RJ, Moir AJG, Perry SV: Phosphorylation of troponinI and the inotropic effect of adrenalin in the perfused rabbitheart. Nature 1976;262:615-617

19. Carter SG, Karl DW: Inorganic phosphate assay with mala-chite green: An improvement and evaluation. J BiochemBiophys Methods 1982;7:7-13

20. Laemmli UK: Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 1970;227:680-685

21. Towbin H, Staehelin T, Gordon J: Electrophoretic transfer ofproteins from polyacrylamide gels to nitrocellulose sheets:Procedure and some applications. Proc Natl Acad Sci U S A1979;76:4350-4354

22. Carlin RK, Grab DJ, Siekevitz P: Function of calmodulin inpostsynaptic densities: III. Calmodulin-binding proteins of thepostsynaptic density. I Cell Biol 1981;89:449-455




23. Matsudaira P: Sequence from picomole quantities of proteinselectroblotted onto polyvinylidene difluoride membranes. JBiol Chem 1987;262:10035-10038

24. Lehman W: Thick filament linked calcium regulation in ver-

tebrate striated muscle. Nature 1978;274:80-8125. Leszyk J, Dumaswala R, Potter JD, Collins JH: Amino acid

sequence of bovine cardiac troponin I. Biochemistry 1988;27:2821-2827

26. Jin JP, Lin JJC: Rapid purification of mammalian cardiactroponin T and its isoform switching in rat hearts duringdevelopment. J Biol Chem 1988;263:7309-7315

27. Elgin SCR, Weintraub H: Chromosomal proteins and chro-matin structure. Annu Rev Biochem 1975;22:725-774

28. Dieckman LJ: Transitions in isoforms of troponin I andmyofibrillar pH sensitivity in the developing heart (PhDthesis). Cincinnati, Ohio, University of Cincinnati, 1989, pp

25-2629. McAuliffe JJ, Gao L, Solaro RJ: Changes in myofibrillar

activation and troponin C Ca2+-binding associated with tropo-nin T isoform switching in developing rabbit heart. Circ Res1990;66:1204-1216

30. Wilkinson JM, Grand RJA: Comparison of amino acidsequence of troponin I from different striated muscles. Nature1978;271:31-35

31. Toyota N, Shimada YJ: Differentiation of troponin in cardiacand skeletal muscles in chicken embryos as studied by immu-nofluorescence microscopy. Cell Biol 1981;91:487-504

32. Martin AF, Orlowski J: Molecular cloning and developmentalexpression of the rat cardiac-specific isoform of troponin. JMol Cell Cardiol 1991;23:583-588

33. Murphy AM, Jones L, Sims HF, Strauss AW: The molecularcloning of rat cardiac troponin I and analysis of troponin Iisoform expression in developing rat heart. Biochemistry 1991;30:707-712

34. Koppe RI, Hallauer PL, Karpate G, Hastings KEM: cDNAclone and expression analysis of rodent fast and slow skeletalmuscle troponin I mRNAs. J Biol Chem 1989;2664:14327-14333

35. Tobacman LS, Lee R: Isolation and functional comparison ofbovine cardiac troponin T isoforms. J Biol Chem 1987;262:4059-4064

36. Solaro RJ, Robertson SP, Johnson JD, Holroyde MJ, PotterJD: TNI phosphorylation: A unique regulator of the amountsof Ca2' required to activate cardiac myofibrils. Cold SpringHarb Symp Protein Phosphorylation 1981;8:901-911

37. Groschel-Stewart U, Lehmann-Klose S, Magel E, BaumggartP: Histones bind to contractile proteins and inhibit actomyosinATPase. Biochem Int 1989;18:519-524

38. Der Terrossian E, Pruliere G, d'Albis A: Purification of atroponin-like factor from pig platelet. FEBS Lett 1983;152:202-206

39. Zlatanove JS, Srebreva LN, Banchev TB, Tasheva BT: Cyto-plasmic pool of histone HI in mammalian cells. I Cell Sci1990;96:461-468

40. Nakanishi T, Seguchi M, Tsuchiya T, Yasukouchi S, Takao A:Effect of acidosis on intracellular pH and calcium concentra-tion in the newborn and adult rabbit myocardium. Circ Res1990;67:111-123

KEY WORDS * development * calcium * thin filamenttroponin * histones



Identification and functional significance of a new class of non-consensus-type plastid promoters

Documents

Transcript of Identification and functional significance of a new class of non-consensus-type plastid promoters