terminally differentiated skeletal myotubes arenotconfined...

Received 3/27/96; revised 5/15/96; accepted 6/4/96.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to mdi-cate this fact.1 The financial support of Telethon-Italy (Grant 467) is gratefully acknowl-edged. This work was supported in part by grants from the AssociazioneItaliana Ricerca sub Cancro and Ministero della Sanit#{225}.M. T. is the recip-iont of fellowships from the Finnish Academy of Sciences and the AlfredKordelin Foundation. D. P. is the recipient of a fellowship of the Fonda-zione A. Buzzati-Traverso.2 To whom requests for reprints should be addressed. Phone: 39-6-49852563; Fax: 39-6-4180526; E-mail: [email protected].

3 The abbreviations used are: TD, terminally differentiated; MSC, mousesatellite cells; HLH, helix-loop-helix; BrdUrd, 5-bromo-2’-deoxyurldine;Ara-C, cytosine �-r-arabinofuranoside; FBS, fetal bovine serum; SF, se-rum-free; MoAb, monoclonal antibody.4 M. Crescenzi, unpublished data.

Vol. 7, 1039-1050, August 1996 Cell Growth & Differentiation 1039

Terminally Differentiated Skeletal Myotubes Are Not Confinedto G0 but Can Enter G, upon Growth Factor Stimulation1

Marianne Tiainen, Deborah Pajalunga,Flavia Ferrantelli, Silvia Soddu, Giovanni Salvaton,Ada Sacchi, and Marco Crescenzi2Molecular Oncogenesis Laboratory, Regina Elena Cancer Center, viadelbe Messi d’Oro 156, 00158 Rome [M. T., D. P., F. F., S. S., A. S.,M. C.], and Institute of Histology and General Embryology, University ofRome “La Sapienza,” 00161 Rome [G. S.], Italy

AbstractTerminally differentiated cells are specialized cellsunable to proliferate that constitute most of themammalian body. Despite their abundance, littleinformation exists on the characteristics of cell cyclecontrol in these cells and the molecular mechanismsthat prevent their proliferation. They are generallybelieved to be irreversibly restricted to the G0 state. Inthis report, we define some features of a paradigmaticterminally differentiated system, the skeletal muscle,

by studying its responses to various mitogenic stimuli.We show that forced expression of a number of cellcycle-regulatory genes, including erbB-2, v-ras, v-myc,B-myb, Id-I, and E2F-1, alone or in combinations,cannot induce terminally differentiated skeletal musclecells (myotubes) to synthesize DNA. However, serum-stimulated myotubes display a typical immediate-earlyresponse, including the up-regulation of c-fos, c-jun,c-myc, and Id-I. They also elevate the expression ofcyclin Dl after 4 hours of serum treatmentAll these events take place in myotubes in a waythat is indistinguishable from that of quiescent,undifferentiated myoblasts reactivated by serum.Moreover, pretreatment with serum shortens the timerequired by EIA to induce DNA synthesis, confirming

that myotubes can partially traverse G1. Serum growthfactors do not activate late-G1 genes In myotubes,suggesting that the block that prevents terminallydifferentiated cells from proliferating acts in mid-G1.Our results show that terminally differentiated cells arenot confined to G� but can partially reenter G1 inresponse to growth factors; they contribute to a much-needed definition of terminal differentiation. The

important differences in the control of the cell cyclebetween terminally dIfferentiated and senescent cellsare discussed.

Introduction

TD3 cells can be defined as cells that have definitively losttheir ability to divide in the process of acquiring specializedfunctions. Examples of TD cells are neurons, skeletal andcardiac muscle cells, white fat adipocytes, and most hema-topoietic and epithelial cells in their final stages of differen-tiation. TD cells are distinct from resting or quiescent cells inthat growth arrest in the latter is, by definition, reversible.Although TD cells constitute the larger part of the mammalian

body, surprisingly little is known about their apparent inability

to respond to mitogenic stimuli and the mechanisms that

determine such unresponsiveness.Skeletal muscle cells constitute a classical example of

terminal differentiation (1). Proliferating precursors, or myo-blasts, express at least one muscle regulatory gene of theMyoD family (2). In vitro, myoblasts can be induced to dif-

ferentiate by growth factor withdrawal. Growth factor-starved myoblasts irreversibly exit from the cell cycle (corn-rnitment) and express a large number of muscle-specificgenes (biochemical differentiation), thus becoming TO myo-cytes (1). Genes expressed upon differentiation include myo-genin, a regulator of skeletal muscle differentiation belongingto the MyoD family, and a large number of structural genes,among which are myosin heavy chain and muscle creatinekinase (3). Myocytes eventually fuse with one another togenerate multinucleated cells termed myotubes (1).

We have recently demonstrated that TO skeletal musclecells and adipocytes can be forced to reenter the cell cycleby infection with wild-type and mutant adenovirus and con-sequent expression of the adenoviral oncogene E1A (4, 5).SV4O and polyoma T antigens share a number of functionswith E1A and can force at least one type of TD cell, themyotube, to reenter the cell cycle (6-8).� These DNA tumor-virus oncogenes are the only known gene products capableof reactivating the cell cycle in TD cells. Unlike most retroviraloncogenes, they have no defined cellular counterpart.

It has been found that TD myotubes up-regulate the c-myc

proto-oncogene when challenged by growth factors (9). This

indicates that they are not totally unresponsive to extracel-lular, proliferative signals. More recently, it has been shownthat MyoD, in addition to its function in regulating skeletalmuscle differentiation, is independently capable of inducing

1040 Cell Cycle Control in Terminal Differentiation

Table 1 Transfection of single genes into myotubes

Transfected gene Gene type C2�1��s MSCmyotubes

erbB-2

Ha-rasv-mycB-mybE2F-1Id-i

E1A/E1BMock

Receptor tyrosine kinase

GDP-binding protein; activator of the cell cycleNuclear viral oncogene; cell cycle regulator, activator of the cell cycleNuclear proto-oncogene; cell cycle regulatorTranscription factor� cell cycle regulator, activator of the cell cycleCell cycle regulator; Inhibitor of skeletal muscle differentiation

Adenovirus oncogenos; able to activato the cell cycle in cells

<0.1

<0.1<0.1<0.1

<0.1<0.1-1<0.1

<0.1

<0.1<0.1<0.1

<0.1<0.1

8<0.1

growth arrest (1 0, 1 1). This finding suggests that growth

arrest in TO cells may be controlled by the same genes thatdetermine their differentiation. The product of the retinoblas-toma gene, pAb, has been shown to give an important con-tribution to the establishment and maintenance of growtharrest in TO cells (1 2). pAb exerts these activities through

increased expression levels during terminal differentiation ofdifferent cell types and, possibly, in the skeletal muscle sys-tem, through direct binding of MyoD (1 3). Furthermore, p21,a general inhibitor of cell cycle-regulatory, cyclin-dependentkinases, has been shown recently to accumulate to highlevels in TO cells and is proposed to contribute to the es-tablishment and/or maintenance of their growth arrest (14-20). Oespite this progress, TO cells are still generically de-fined as cells that have ceased dividing and cannot berecalled into the cell cycle (21). This is only an operationaldefinition, with the additional weakness of relying on theabsence of a feature, rather than on its presence; TO cells arenever observed to divide in vitro or in vivo.

In an effort to achieve a better characterization of terminaldifferentiation, we tested the still largely assumed “unre-sponsiveness” of TO myotubes to proliferative stimuli. We

challenged purified myotubes with a number of activatedoncpgenes and cell cycle control genes as well as serum. Wefound that these TO cells do not respond with DNA synthesisto activated oncogenes, alone or in combinations, which arepowerful activators of the cell cycle when expressed in qui-escent (i.e., reversibly growth arrested) cells. Nor did TO cellssynthesize DNA as a result of the forced expression of somegenes involved in the control of the cell cycle. However,serum stimulation elicited up-regulation of a number of im-mediate-early genes in a manner that is indistinguishablefrom that of serum-stimulated, quiescent cells. At least onelater event, up-regulation of cyclin 01 , was also observed.

Our results begin to functionally define TO cells and to drawa map of mitogenically inactive signal transduction pathwaysin these cells. More importantly, they establish that TO cellsare not confined to G0 but can enter G1 as well as theirnon-TO counterparts.

Results

Overexpression of a Number of Retroviral Oncogenesand Cell Cycle Regulators Cannot Force TD Cells to Syn-thesize DNA. Sparse reports exist in the literature, suggest-

ing that retroviral oncogenes such as src (22), ras (23), and

myc (24) cannot reactivate the cell cycle in TO cells. Despite

this, we wished to probe more systematically the signaltransduction pathways of TO cells and determine whetheroverexpression of oncogenes and/or cell cycle regulatorscould reinduce DNA synthesis. To this end, we set up lipo-some-mediated transfection of C2C1 2-derived myotubes orprimary MSC-derived myotubes. Preliminary experiments

with a �-galactosidase-encoding plasmid determined thatapproximately 4% of C2C1 2-derived myotubes and up to25% of MSC-derived myotubes can be typically transfected(data not shown), in good agreement with the literature (25).The undifferentiated cells present in the myotube culturescould also be transfected, somewhat less efficiently than themyotubes.

We selected a number of proto-oncogenes or activated

oncogenes that are prototypes of different classes, or knownpowerful transformers/cell cycle activators, or both (Table 1).

The cell cycle regulator E2F-1 was chosen because it is ableto drive resting cells into S phase (26, 27) and also becauseits sequestration appears to be an important means by whichpAb exerts a negative control on the cell cycle (28). Finally,the HLH protein Id-i was selected because it plays a role inthe reactivation of the cell cycle in quiescent cells (29, 30), itis up-regulated early upon serum stimulation of serum-

deprived cells (29), and could conceivably interfere with themuscle differentiation program by interacting with key basic-HLH factors (31 , 32).

Plasmids encoding these proteins were transfected one ata time into myotubes and undifferentiated cells, BrdUrd wasadded to the culture medium, and the cells were fixed 48 or72 h after transfection. The transfected cultures were thensubjected to double immunofluorescence for MHC, a marker

of terminal differentiation in skeletal muscle, and BrdUrd, amarker of DNA synthesis. Exogenous gene expression was

confirmed in parallel dishes by immunofluorescence (Fig. 1).The average expression of the transfected gene(s) far ex-

ceeded that of the endogenous ones, when present (forexamples, see Id-i and E2F-i in Fig. 1). A plasmid carryingthe adenovirus E1A and E1B genes was used as positivecontrol throughout these experiments, since we had previ-ously shown that E1A can induce TO cells to reenter the cellcycle (4, 5). Fig. 2 shows that all of the transfected plasmidsstimulated DNA synthesis in undifferentiated, quiescentcells. In contrast, although many myotubes transfected withthe E1A/Ei B plasmid were labeled by BrdUrd, none of theother transfected genes could elicit DNA synthesis in thesecells (Table 1). Wherever possible, parallel dishes were also

#{149}1#{149}

�‘ . �. ,-� .----“-

Hoichit

,� .�. � �11. �

� .� � �... �

Phi.. con�ast

�

c-

ld.1

E2F-1

�.

BIdU

�‘

BrdU

.

%

.

� �

PPm contrast MHC BrdU

1 50

(, 100�

w>0

.0(U

0 50-

0

Fig. 2. Activity of cell cycle regulatory genes in quiescent, undifferenti-ated cells. Starved cultures of MSC were transfected with either puCi8(negative control) or one of the positive regulators of the cell cycle,incubated in the presence of BrdUrd, and stained. One thousand nucleiper transfection were scored for BrdtJrd incorporation. Results are pre-sented as a percentage of BrdUrd-positive nuclei above background, asrepresented by puCi 8 (which forms the backbone of several of the otherconstructs). The E1A/E1B expression vector, used as positive control,appears more active, at least in part, because it activates nuclei in myo-tubes as well as undifferentiated cells. This experiment was performedthree times. Although there was significant variability in the absolutemagnitude of cell cycle activation, the relative strength of each constructcompared to the others was constant. A representative experiment isshown here.

c�1 (0 0 .0,-. S �5 >� >, I

C.) I� �- �

� .0 (b � � (�10. � I > � W

V w

w

Cell Growth & Differentiation 1041

Fig. 1. Expression of cell cycleregulatory genes in MSC-derivedmyotubes and effects on DNAsynthesis. Upper row, /d-i -trans-fected myotubes. Arrowheadspoint at a transfected myotubeshowing two Id-i -positive nuclei(middle) that did not incorporateBrdUrd (BrdU, right). Nuclei arecounterstained with Hoechst dye(left). Middle row, E2F-i -trans-fected myotubes. A large myotubeshowing many E2F-i -positive nu-clei (middle) did not incorporateBrdUrd (BrdU, right), unlike someof the surrounding nuclei, belong-ing to undifferentiated cells (Phasecontrast, left). Notice that in thisand in the upper row, the endoge-nous proteins are not visible in thesurrounding, presumably untrans-fected nuclei. Lower row, E1NE1B-transfected myotubes. Thelarge myotube in the middle of themicroscopic field shows flattenedmorphology (left) and somewhatreduced MHC expression (middle),and all of its four nuclei areBrdUrd-positive. Bar, 25 �m.

double-stained for simultaneous detection of the foreign pro-

tein and BrdUrd to assess DNA synthesis directly in the

transfected cells (Fig. 1). Even when the available antibodies

did not allow this procedure, the high transfectability of

MSC-derived myotubes (8% of the myotubes became

BrdUrd-positive following transfection of E1AIE1B; Table 1)

ensured that even low-efficiency reactivation of DNA synthe-

sis would have been detected.

The inability of E2F-i to force DNA synthesis in myotubes

was rather surprising, in view of the supposedly direct acti-

vation of genes involved in DNA synthesis by this transcrip-

tion factor. To confirm that the E2F-1 expression vector we

used can efficiently stimulate DNA replication in quiescent

fibroblasts, NIH3T3 cells were transfected with this vector or,

as an inert control, with a deleted, silent mutant of MyoD

(DM:143-162; Aef. 10). The transfected cells were then se-

rum starved to induce quiescence, incubated with BrdUrd,

and double-stained for the transfected gene and BrdUrd.

Table 2 shows that E2F-1, but not DM:143-162, efficiently

induced DNA synthesis in the transfected cells, although the

percentage of E2F-positive cells was lower than in myo-

tubes. These results demonstrate that the expression vector

we used is capable of forcing DNA synthesis in permissive,

quiescent cells. We also asked whether E2F-i overexpres-

sion could possibly activate, in myotubes, events other than

DNA synthesis, such as cyclin A expression and/or pRb

phosphorylation. E2F-1 was transfected into MSC-derived

myotubes, and 48 h later, immunofluorescence was used to

assess the expression of cyclin A in the E2F-i -positive cells.

Similarly, the phosphorylation status of pAb was determined

in the E2F-1 -transfected myotubes by its sensitivity to a

hypotonic wash (33). Fig. 3 shows that transfection of E2F-i

into myotubes caused neither cycbin A expression nor pRb

phosphorylation. Thus, TD myotubes appear to be totally

S

b

E2F

C

cyclin A

E2F pRb

erbB-2 E2F-1

Fig. 4. Coexpression of cotransfected genes in a myotube. MSC myo-tubes were cotransfected with the erbB-2 and E2F-i constructs andsubjected to double immunofluorescence. Both antigens are present inthe same myotube; erbB-2 is on the membrane, while E2F-1 is in thenuclei.

Hoichit pRb

Fig. 3. Absence of late G1 effects of E2F-1 expression in MSC-derivedmyotubes. First (upper) row, phase contrast view (a) of E2F-1 -transfectedmyotubes. A trinucleated, E2F-1 -positive myotube (b) does not showcyclin A staining (C), which is visible in an undifferentiated cell in the samefield. Second row (from the top), positive control of cyclin A expression:myotubes in which the cell cycle has been reactivated by adenovirusinfection (5). A phase contrast view of adenovirus-infected myotubes (d)that express cyclin A (t) is shown. Hoechst staining of the same field (e) is

also shown. Third row, E2F-1 -transfected myotubes, hypotonic-washmobilization of pRb: g, phase contrast; h, a trinucbeated, E2F-1 -positivemyotube shows pRb staining (,)after a hypotonic wash, indicating hy-pophosphorybation of pRb. Fourth row, adenovirus-infected myotubes: j,phase contrast; k, Hoechst staining. A mubtinucbeated myotube (largearrowhead) shows no pRb staining in all but one of its nuclei (I), whereasa trinucleated myotube (small arrowhead) is still pRb positive (I). The poorappearance of the cells in the last two rows is due to the hypotonictreatment. Bar, 25 �.tm.


Table 2 Effects of transfec tion of E2F-i or MyoD mu tant DM:143-162 in quiescent NIH3 T3 fibroblasts and TD myotubes

Cell type % E2F-1 � cells % BrdUrJVE2F-1 � cells % MyoD� cells % BrdUrd�/MyoD� cells

MSC myotubes 27 <0.5 ND” NDNIH3T3 4 75 5 7

aND, not done.

�-

�

t�.

, -.�. 1’-

v%% I

‘.� I.v, #{149} ‘p.a . #{149} S

#{149}

�

::�.&a II

impermeable to E2F-1 activity, underscoring the peculiarity

of their cell cycle control.

Next, we asked whether combinations of the above genescould overcome the cell cycle block in TD cells. We trans-footed MSC-derived myotubes with combinations of the

genes listed in Table i and subjected the transfected cells tothe experimental protocol described above. Table 3 shows

that none of the pairs of genes tested could reactivate the

cell cycle in myotubes. These results expand the boundaries

Table 3 Transfection of pairs o f genes into myotubes

Transfected genes % BrdUrd MSC myotubes

v-myc + ras <0.1v-myc + B-myb <0.1v-myc + Id-i <0.1

ras + Id-i <0.1

erbB-2 + E2F-i <0.1E1AJE1B + Id-i -5

of the ability of myotubes to resist reentry into the prolifera-

tive state. As a positive control, the E1A/El B-expressing

plasmid was cotransfected along with that encoding Id-i (the

latter being a presumably irrelevant partner). That cotrans-

fection of two plasmids, in our system, results in the coex-

pression of the two respective proteins in the same myotubewas demonstrated in the favorable case of erbB-2 andE2F-i , which are easily detectable by double immunofluo-

rescence and localize to different subcellular compartments.

Most cells expressing either protein also expressed the other

(Fig. 4).

Serum-stimulated Myotubes Enter G1. To begin to ex-

plore the mechanisms that prevent TO cells from prolifer-

ating, we wished to assess whether the lack of response to

growth factors of myotubes was absolute or limited to

DNA synthesis. We set up cultures of C2C12 myotubes

largely purified from contaminating, undifferentiated myo-

blasts by a treatment with Ara-C, which kills undifferenti-

ated cells. The myotubes were mitogenically stimulated by

replacing SF medium with medium containing iO% FBS.

RNA was then extracted at various times after serum

addition. For comparison, similar RNA samples were ex-

tracted from the C2Qi6 subclone of C2Ci2 cells that hadbeen kept in SF medium for 36 h, long enough to inducequiescence, but not significant differentiation, and then

serum stimulated. Fig. 5A shows a series of Northern blots

w

. R.#{149}��:!�.

V

c-foe (A)

c-jun (B)

c-myc (D)

Id-I (A)

(C)

112 1 2 4 8 16 24 36 48 72

differentiated: hr after 10% serum

prolif.

d

PCNA (B)

B-myb (C)

RbI (B)

p21 (B)

0 112 1 2 4 8 16 24

differentiated: hr after 10% serum

prolif.

C�- _ __ � � c-fos

0 1/2 1 2 4

prolif.

�--�- �

differentiated:hr after 10% serum

0 16 24

hr in 10% serum


a b

(A)

,.w.�..... (B)

cycllnDl(D) #{149}N�cycllnE(C) �

-,. cyclln A (B) #{149}0 ND � � �:#{216}$ #{149}� (E)

M�M�b�WWW cdk4(A)

C2C12myotubes

�. �0 16 24

hr in 10% serum

prolif. prolif.

quiescent

myoblasts

� cyclin Dlw

cyclin A

Fig. 5. Time course expression of cell cycle-related genes in C2C1 2 myotubes upon serum stimulation. a, gene expression in myotubes. Each blot shouldbe compared with its respective glyceraldehyde-3-phosphate dehydrogenase hybridization (b), indicated by the capital letters in parentheses that followthe names of the genes. c, Western blot analysis of c-fos expression in C2C1 2 myotubes stimulated with serum for the indicated times. d, Westem blotanalysis of cyclin Di and cyclin A in C2C12 myotubes or C2Q16 quiescent myoblasts stimulated with serum for the indicated times. prolif., proliferating,undifferentiated myoblasts.

derived from myotube samples. In response to serum

stimulation, myotubes up-regulated a number of genes,among which were the c-fos, c-myc, and c-jun proto-

oncogenes. These are defined as immediate-early genes,

since they are rapidly expressed, independently of protein

synthesis, following stimulations that drive quiescent cells

into G1 (34-36). Id-i was also promptly up-regulated, its

mRNA reaching maximum steady-state levels within 1 to

2 h, and remaining essentially constant thereafter. Id-i is

a regulator of and necessary for reentry into the cell cycle

(29, 30). It is also up-regulated during G0-G1 transition as

rapidly as some of the immediate-early genes. Fig. 6A

shows that quiescent, nondifferentiated myoblasts up-

regulated c-fos, c-myc, c-jun, and Id-i with very similar

kinetics upon serum stimulation. To determine whether

early-gene mRNA accumulation promoted expression of

the relative proteins, we analyzed c-fos protein levels in

Ara-C-purified myotubes. c-fos protein increased after se-

rum stimulation, reaching a peak within 2 h (Fig. 5C). Id-i

protein was also accumulated in response to serum (see

a

S. �#{248}�V#{149}�r.

I

!�; :� � � ‘ _____________

� �

(F)

.....I...

#{149}IOOO.�Os (H)

� (I)

112 1 2 4 8 16 24

Os. quiescent myoblasts:hrafterlO% serum

�a,.


Fig. 6. Expression of cell cycle-related genes in quiescent, undifferentiated C2Q16 myobbasts (a and b) or NIH3T3 fibroblasts (c) upon serum stimulation.Northem blot analyses as in Fig. 5.

... .M�c-fos (H)

c-Jun(F) bc-myc (I)

Id-I (I)

cyclin Dl (I)

cyclin E (H)

cyclin A (I)

PCNA (G)

B-myb (G)

Rbl (F)

prolif.

112 1 2 4 8 16 24

quiescent myoblasts:hr after I 0% serum

p21 (F) prolif.

C. .#{149}...� I.,,.. a

c-myc

Id-I

,m. cyclln Dl

0 112 1 2 4 8 16 24

quiescent NIH3T3:hr after 10% serum

prolif.

cyclln E

RbI

GAPDH

kDa

-29

Id-1(E)

MHC(D)


I NDISS#{149}#{149}OS#{149}#{149}ND MyoD(E)

WwW’� 4WWW WV myogenln(D)

MCK (D)

Id-I -

I � 4 hr

prollf. I differentIated

- 18.4

-14.3

Fig. 7. Id-i protein expression. Id-i expression was analyzed by West-em blotting in proliferating cells and in myotubes untreated (0 h) andtreated for 4 h with 10% FBS.

Fig. 7). These results show that, in myotubes, serum elicits

accumulation of early gene proteins as well as mRNAs.

In contrast to the early genes, the mRNA levels of most

genes whose transcription is normally up-regulated later in

G1 remained unmodified in the myotubes (Fig. 5A). These

include cyclin E and A, the cyclin-dependent kinase cdk4,

B-myb, and PCNA. A slight increase in the mRNA levels of

these genes at 16-24 h is most likely due to the residual

undifferentiated cells present in our cultures. The high mRNA

levels of the cell cycle regulators Rb and p21 , whose accu-

mulation is strongly enhanced during muscle differentiation,

remained also unaltered in serum-stimulated myotubes. For

comparison, the expression of some of these genes in se-

rum-stimulated, quiescent myoblasts and quiescent NIH3T3

fibroblasts is shown in Fig. 6, A and C, respectively. Quies-

cent myobbasts and fibroblasts entered S phase with similar

kinetics, as determined by time course BrdUrd incorporation

analysis (data not shown).

At variance with other G1 cyclins, strongly increased cyclinDi mRNA levels were observed in myotubes, 4 to 8 h after

serum addition (Fig. 5A). This increase was similar in timing

and extent to that displayed by reactivated myoblasts and

fibroblasts (Fig. 6, A and C). Importantly, the late accumula-

tion of cyclin Di indicates that at least some G1 events takeplace normally in myotubes several hours after the beginning

of serum stimulation. To determine whether cyclin Di was

also accumulated as protein in myotubes, we performed a

time course Western blot analysis of purified, serum-stimu-

bated myotubes. Fig. 5D shows that cyclin Di accumulates

slowly in muscle cells. Even after 24 h of treatment with

serum, its levels remained well below those of proliferating

myoblasts. In contrast, serum-stimulated myoblasts showed

cyclin Dl levels as high as those found in exponentially

growing cells (Fig. 5D). To confirm that cyclin Di was pro-

0 1/2 1 2 4 8 16 24 36 48 72

differentiated: hrafter 10% serum

prolif.

Fig. 8. Expression of Id-i and muscle-specific genes in myotubes in thecourse of prolonged treatment with 10% FBS. Legend as in Fig. 5a. Theappropriate gbyceraldehyde-3-phosphate dehydrogenase hybridizationsare in Fig. 5b. ND, not done.

duced by the myotubes, rather than any residual myoblasts

in the culture, we probed the same blots with an anti-cyclin

A antibody. Whereas quiescent myoblasts up-regulated cy-

din A expression during serum treatment, there was no

detectable up-regulation of the same protein in the myotubes

(Fig. 5D). This shows that myoblast contamination was neg-

ligible in our cultures and demonstrates that TD myotubes

can accumulate cyclin Di in response to growth factors,

albeit at a slow rate.

The expression of the HLH protein Id-i is usually down-

regulated early during muscle differentiation. Its forced ex-

pression in myoblasts has been shown to inhibit differentia-

tion through heterodimerization with necessary basic-HLH

proteins (32). The high Id-i expression levels of proliferating

myoblasts have been proposed to inhibit untimely differen-

tiation (31). Thus, the swift accumulation of Id-i mRNA after

serum stimulation suggested that it might induce at least

partial dedifferentiation in the form of deinduction of MyoD-

family dependent, muscle-specific genes. Western blot anal-

ysis confirmed that Id-i protein is indeed accumulated in

serum-stimulated C2Ci 2 myotubes, at levels comparablewith those found in myoblasts (Fig. 7). This prompted us to

analyze expression of muscle-specific genes. Surprisingly,

even after a 72-h exposure to serum, no significant variations

were detected in the mRNA levels of the regulatory genes

MyoD and myogenin, nor in those of the structural genes

myosin heavy chain and muscle creatine kinase (Fig. 8). Thus,

although Id-i protein can be expressed in myotubes, there

appears to be a mechanism to prevent it from interfering with

the differentiation program.

Serum Stimulation Anticipates Entry into S Phase ofEIA-reactivated Myotubes. To confirm by an independent

experiment that TD myotubes can partially traverse G1 , we

exploited the ability of the adenoviral oncogene E1A to re-

activate the cell cycle in TD cells. We reasoned that, if serum-

stimulated myotubes progressed through G1, this might re-

suit in a shortening of the time needed by E1A to drive them

into S phase. To test this hypothesis, we stably transfected

C2C12 cells with a construct expressing a chimeric protein,

EiA-ER, containing the NH2-terminal half of E1A fused to the

100#{149}

80

+� 60

�

� 40

20

0�

-U- SF

-0- FBS stimul.

11 13 15 17 19 21 23 25 27 29 31

hrs from EIA activation

Fig. 9. Entry into S phase of C2(E1A-ER) cells. C2(E1A-ER) cells werepretreated for 6 h with medium containing 10% FBS (FBS stimul.) or nottreated (SF), before activation of E1A by the addition of f3-estradiol in SFmedium. The dotted lines indicate the time points at which 50% of thetotal number of myotubes that would eventually synthesize DNA becameBrdUrd positive. One experiment, representative of three, is shown.


estrogen receptor hormone-binding domain. This construct

allows us to induce EiA activity by the addition of �-estradioIto the tissue culture medium (37). We isolated a clone of

C2C12 cells, C2(EiA-ER), that expresses high levels of

EiA-ER and can differentiate into rnyotubes that remain

rnitogenically inactive in the absence of 13-estradiol. How-ever, after the addition of f3-estradiol, more than 80% of the

myotubes synthesizes DNA, as assessed by BrdUrd incor-

poration.We set up two identical series of C2(Ei A-ER)-derived

myotubes. One set was stimulated for 6 h with growth me-

dium containing 10% FBS, then switched to SF medium

containing p-estradiol and BrdUrd. The second series re-

mained in SF medium until the addition of p-estradiol and

BrdUrd, at the same time as the first set. Dishes from each

series were fixed at various times after the beginning ofestrogen stimulation and analyzed by immunofluorescence

to detect MHC and BrdUrd. Percentages of BrdUrd-positive

myotubes were scored blindly at all time points. Fig. 9 showsthat serum-stimulated cells entered S phase earlier than

unstimulated cells; 50% of the myotubes kept in SF mediumreached S phase in little more than i 7 h, whereas cells

stimulated with i 0% FBS for 6 h prior to E1A activation tookslightly less than i 6 h to reach the same point. These exper-

iments were repeated three times and consistently showed a

difference of about i h between prestimulated and unstimu-

lated rnyotubes. However, there was no difference in thetotal number of myotubes eventually undergoing DNA syn-

thesis. This eliminated the possibility that serum, in conjunc-tion with E’lA, might recruit more myotubes into S phase. We

interpret these data as showing that serum-stimulated myo-tubes traverse part of G1 . Thus, after EiA activation, the time

necessary to drive them through the first part of the cell cycleis saved. These results provide independent evidence that

serum-stimulated rnyotubes do reenter the G1 phase of the

cell cycle.

DiscussionIn this report, we present evidence that TD cells, as repre-sented by myotubes, are extremely refractory to full cell cyclereentry, even under the influence of such potent stimulatorsas cooperating, activated, transforming oncogenes or keycell cycle regulators like E2F-i . However, we show that

serum stimulation can bring about partial reentry of TD cellsinto the cell cycle and induce myotubes to leave G0 andtraverse the early G1 phase.

We expressed a variety of positive regulators of the cellcycle in TO myotubes. In general, although the endogenousgene products were not detected in myoblasts or in myo-tubes in our experimental conditions, the exogenous pro-teins could be readily revealed. This demonstrates expres-sion at levels exceeding those sufficient for the function ofeach protein in proliferating myoblasts. The same expressionconstructs forced cell cycle reentry of quiescent, undifferen-

tiated cells despite their lower transfectability, compared tomyotubes. Nevertheless, myotubes did not undertake DNAsynthesis, although their partial response to serum impliesthat at least some signal transduction pathways are func-tional in these cells. There are very few reports on the (ab-sence of) mitogenic effects of the expression of oncogenesor cell cycle regulators in already TD cells (22-24). A system-atic assessment of the activity of the known regulators of cellproliferation in TO cells should allow us to better define the

nature and extent of the block(s) that prevents these cellsfrom dividing. From this point of view, it is interesting thatmyotubes did not synthesize DNA in response to overex-pression of E2F-i . This transcription factor is known to reg-ulate the expression of a number of key S-phase genes (38)and to be bound and inactivated by pRb (39). It has beensuggested that the binding of E2F family members might bethe main if not the only means by which pRb exerts its keynegative control on the cell cycle. Indeed, forced expression

of E2F-i is sufficient to drive quiescent cells into S phase,presumably by overwhelming pRb control (26, 27). AlthoughE2F-i exerts at least some of its functions directly at theG1-S boundary (38), its exogenous expression in TD cells isnot sufficient to drive them, unlike quiescent fibroblasts, intoS phase. In myotubes, E2F-i is also incapable to induce

cyclin A expression or pRb phosphorylation. This exemplifiesthe uniqueness of the cell cycle control in TO cells; theremust be preconditions for E2F-i -mediated reactivation ofthe cell cycle that are not fulfilled in TO cells. It is worth notingthat since EiA does induce DNA synthesis in rnyotubes, therelease of E2F cannot be sufficient to account for the activityof E1A in such cells.

Although myotubes appear unresponsive to most prolifer-

ative stimuli, we wished to assess whether they are normallyabsolutely incapable to leave G0 or the proliferation block liesdownstream of the G0-G1 boundary. Our mRNA expressionstudies show that TO skeletal muscle cells are capable of anostensibly normal, immediate-early response to serumgrowth factors. Our results agree with and expand on the

previously known fact that c-myc expression can be up-regulated by growth factors in myotubes (9). The present

data demonstrate that c-myc does not constitute an isolatedcase but is part of a full-fledged response, which also in-


5 A. Felsani, personal communication.

volves c-fos, c-jun, and Id-i . Altogether, our data indicatethat the signal transduction pathways leading to the imme-

diate-early response are intact in TD cells. Importantly, we

show that at least c-fos and Id-i proteins are expressed bymyotubes at levels comparable to those of proliferating myo-

blasts. These molecules are necessary for cell cycle progres-

sion and may allow myotubes to advance further through G1.

In fact, at least one event, up-regulation of cyclin Di , takesplace in myotubes as late as 4 hours after the beginning of

serum stimulation, similarly to resting myoblasts and fibro-blasts. This shows that at least one chain of eventsprogresses, unaffected by negative regulators, for a consid-erable duration and correctly times a mid-G1 occurrence.Although the timely activation of cyclin Di mRNA expressionis sufficient to indicate that myotubes can reach that point in

mid-G1 where cyclin Di is sharply up-regulated, we alsolooked for the corresponding protein. We found that cyclinDi protein is accumulated in myotubes but at a markedly

slower rate than in serum-stimulated, quiescent myoblasts.

Even after 24 h of serum stimulation, myotubes expressed far

less cyclin Di than proliferating myoblasts. This raises theinteresting possibility that prevention of cyclin Di accumu-lation is one of the mechanisms that prohibit myotube pro-

liferation. Indeed, our Northern blot data suggest that theprogression of myotubes through G1 is arrested at the timeof cyclin Di expression or shortly thereafter.

We show that serum stimulation shortens the time neces-

sary for EiA to drive myotubes into S phase by approxi-

mately i h. This confirms by independent means that serum-

stimulated myotubes indeed traverse G1 . The i -h extent ofsuch progression through G1 suggested by our experiment isonly a minimal estimate. EiA itself is known to shorten the

time necessary for entry into S phase by circumventing someearly Gi events (40). Thus, the i -h difference we observed

constitutes the further shortening, afforded by serum, of a G1phase already abbreviated by EiA.

The results presented here are compatible with the recentview that inhibitors of cyclin-dependent kinases such as p2imight be at least in part responsible for the inability of TD

cells to proliferate. Immediate-early genes, most likely, are

not under the control of this family of inhibitors and are,

therefore, normally modulated. We show that, in rnyotubes,

some up-regulation of cyclin Di , an early Gi cyclin, takesplace in the presence of significant amounts of al least one ofits known kinase partners, cdk4.5 In contrast, later events do

not occur, suggesting a block in the pathway downstream of

cyclin Di that could be mediated in part by kinase inhibitors.

However, as discussed below, the recessiveness of growtharrest in TD cells suggests that the absence of a positiveregulator of the cell cycle, rather than the presence of an

inhibitor, constitutes the ultimate reason for the definitive

growth arrest of these cells.

Several other points deserve discussion. It has been pro-posed that cellular senescence resembles a state of terminaldifferentiation (41). Whereas both conditions are character-

ized by definitive growth arrest, heterocaryon studies clearly

show that the growth arrest of senescent cells is dominant(4i), but that of TO cells is recessive (42, 43), over prolifer-

ation of young cells. Our data show, at the gene expressionlevel, that there are significant differences between senes-

cent and TO cells. Unlike senescent cells (44), myotubes arefully capable to activate transcription of c-fos in response to

serum. Following serum stimulation, TO cells can also up-regulate and sustain expression of Id-i, whose mRNA isexpressed only at low levels in senescent human fibroblasts(45). Conversely, whereas senescent cells show quasi-con-

stitutive, elevated cyclin Di and cyclin E levels (46), myo-tubes demonstrated normal inducibility of cyclin Di mRNA

but did not express cyclin E under the conditions tested.

Altogether, these data provide strong confirmation of the oldcell-fusion evidence; at the level of expression of cell cycle

regulatory genes, the control of growth arrest is profoundly

different in senescent and TO cells.

We show in this report that Id-i is expressed in myotubes,as both mRNA and protein, at levels comparable with thoseof proliferating myoblasts. Id-i has been shown to preventmuscle differentiation by titrating out E2A-encoded proteins

(32). Since the latter are necessary partners of the MyoD

family proteins, Id-i expression should result in their func-

tional inactivation. MyoD family members are believed to be

necessary not only for the establishment, but also for themaintenance, of skeletal muscle differentiation, based on thecontinued expression of one or more of these genes through-out adult life and their direct involvement in the regulation of

a number of muscle-specific, structural gene promoters (2).

However, despite prolonged, serum-induced expression ofId-i , we found no evidence of down-regulation of muscle

regulatory or structural genes. A number of explanations arepossible. The MyoD family proteins and/or E2A gene prod-ucts, as postranslationally modified in the TD environment,

might have higher affinities for one another than for Id-i.Alternatively, since myotubes are capable of expressing high

levels of Id-i , they might have means to functionally macti-vate this protein and prevent an inappropriate, potentially

disastrous shut-off of the differentiation program. Anotherhypothesis is that the products of the MyoD gene family arenot absolutely required for the maintenance of skeletal mus-

cle differentiation, which might be sustained by other regu-

lators such as, for example, members of the functionally

analogous MEF-2 gene family (47, 48).

Materials and MethodsCells. The mouse myoblast line C2C12 (49), a clone derived from the C2cell line (50) was cuitured in DMEM containing 10% FBS and antibiotics.Differentiation was induced by plating the cells into collagen-coateddishes and switching them to SF medium: DMEM supplemented withRedu-Ser (Upstate Biotechnology Incorporated, Lake Placid, NY) to a final

concentration of 5 pg/mb human insulin, 5 pg/mI human (hobo) transfenin,and 5 ng/mI sodium selenite. Fifty �u.i Ara-C was added to SF medium toeliminate undifferentiated cells. In our conditions, Ara-C-purified myo-tubes routinely contain more than 90% of the nuclei in the culture, as

assessed by staining nuclei with the Hoechst 33258 dye. The C2Q16clone of C2C1 2 was chosen among a number of randomly isolated onesfor its ability to become resting within 36 h from the switch to SF mediumwithout undergoing significant differentiation before 48 h. Primary satellitecells were isolated from the hind limb muscles of young C57BV1 0 mice,maintained in cuiture exactly as described by Rando and Bbau (51), and


6 T. Babdari, unpublished data.7 M. Crescenzi and S. Abem#{225},unpublished data.

induced to differentiatefor 3-4 days in DMEM containing 2% donor horseserum (differentiation medium). Mouse NIH3T3 fibroblasts were routinelycultured in DMEM containing 10% bovine calf serum.

Plasmlds. The following expression vectors were used for all trans-fections: erbB-2, LTR/erbB-2 (52); Ha-ras, T24/LTR, derived from pT24C3(53) by insertion of a retroviral long terminal repeat for enhanced expres-sion6; v-myc, Babe-c/v-myc (24); B-myb, pSV-B-myb (54); E2F-i , pCMV-E2F-1 (26); Id-i, pE:ld(s) (31); and E1A/E1B, pXhol.C (55).

Transfections and DNA Synthesis Assessment. For transfections ofcell cycle regulators, C2C12 cells were seeded into collagen-coated,35-mm dishes at 2 x 1O� cells/dish and switched to SF medium with 50�LM Ara-C after attachment. MSC were seeded into collagen-coated,35-mm dishes at 2 x i05 cells/dish and switched to differentiation me-dium. Differentiated C2C12 or MSC were subjected to liposome-mediated

transfection according to a modification of a published protocol (56).Myotubes were washed once with Opti-MEM (Life Technologies, Inc.,Gaithersburg, MD), then transfections were performed in Opti-MEM usingUpofectin (Life Technologies, Inc.), according to the manufacturer’s in-structions. Best transfection efficiency and survival were obtained with 4f.L9 of circular plasmid and 8 �l of Upofectin per 35-mm dish. The cellswere kept for 6 h in the presence of the Upofectmn/DNA mixture, thenswitched to their culture medium. To determine the effects of transfectedgenes on DNA synthesis, BrdUrd was added to the cuitures 24 h after thebeginning of transfections. The cultures were then fixed 24 or 48 h laterand subjected to double, immunofluorescent detection of MHC andBrdUrd. In each dish, at least 500 MHC-positive cells were scored, andthe percentage of BrdUrd-positive myotubes was calculated. To deter-mine the effects of the cell cycle regulators on quiescent, undifferentiatedcells, MSC were plated exactly as described above; they were starved indifferentiation medium for 39 h only, to increase the number of quiescent

but not-yet-differentiated cells. Transfections and BrdUrd addition wereperformed as described above. The cells were stained to detect BrdUrdincorporation at 48 h from transfection. In each dish, 1000 nuclei werescored, irrespective of their belonging to myotubes or undifferentiated

cells.E2F-1 transfection into NIH3T3 cells was performed as follows. On the

day of transfection, 1.5 x iO� cells were seeded into 60-mm diameterdishes and transfected a few hours later by the calcium-phosphate co-precipitation method with 10 �g of pCMv-E2F-1 or DM:i43-162 (control)plasmid DNA and 10 �g of salmon sperm DNA. Sixteen h later, theprecipitate was removed, and fresh medium was added. The day after, thetransfected cells were seeded into 35-mm dishes to obtain a high cell

density and switched to medium containing 0.5% calf serum after attach-ment. To determine the effects of the transfected genes on DNA synthe-sis, BrdUrd was added to the cultures 96 h after the beginning of thetransfection. The cuitures were then fixed 20 h later and subjected to

double immunofluorescent detection of the transfected gene product andBrdUrd. In each dish, 100 E2F-1- or DM:143-162-positive cells were

counted, and the percentage of BrdUrd-positive cells was calculatedamong them.

The ES.HE1 4 plasmid (37), encoding a chimeric protein containing the150 NH2-terminal amino acids of adenovlrus-2 E1A fused to the estrogenreceptor hormone-binding domain, was stably transfected into undiffer-entiated C2C12 myoblasts by the calcium-phosphate method. ES.HEI4was transfected along with the selectable pRSVneo plasmid, at a 20:1molar ratio. Twenty G41 8-resistant clones were isolated and tested for theability of myotubes derived from them to synthesize DNA after �-estradioIinduction. One clone, C2(ES.HE14)cl.13 [here referred to as C2(E1A-ER)],was selected and used for the following experiments. C2(E1A-ER) cellswere seeded at 5 x 10� cells/collagen-coated, 35-mm dish and switchedto SF medium with Ara-C, as described above. Differentiated cells were

stimulated with 1 �u�i p-estradiol in SF medium eitherdirectly or after a 6-hstimulation with 10% FBS. Replica dishes were then fixed at various timepoints and stained to detect MHC and BrdUrd, as above. For each timepoint, 200 myotubes were scored by a researcher blind to the labels of thedishes.

Immunofluorescence and Western Blotting. The following first an-

tibodies were used: erbB-2, W6/iO0 A4 MoAb (57); Id-i , rabbit antiserum

(30); an affinity-purified fraction of this antiserum was used for Westernblots; E2F-i , rabbit antiserum, kind gift of A. Giordano (JeffersonCancer Institute, Philadelphia, PA); cyclin A, rabbit antiserum, a gift ofG. Draetta (Istituto Oncobogico Europeo, Milan, Italy); pRb, G3-245MoAb (PharMingen, San Diego, CA); MyoD, rabbit antiserum raisedagainst whole mouse MyoD protein7; El A, M73 MoAb (OncogeneScience, Uniondale, NY); and myosin heavy chain, MF2O MoAb (58).For second antibodies, MoAbs were detected by affinity-purified, an-timouse IgG goat serum (Organon Teknika, West Chester, PA). Reac-tion of rabbit antisera was detected by affinity-purified, antirabbit lgGgoat serum (Organon Teknika). Second antisera to mouse or rabbitimmunogbobubins were conjugated with rhodamine or fluorescein forimmunofluorescence, or peroxidase for Western blotting. In situ anal-ysis of pRb phosphorylation by selective hypotonic elution was per-formed exactly as described (33). Nuclei were stained after immuno-fluorescence treatments by incubating the cells for 3 mm with a 1 pg/mIsolution of Hoechst 33258 dye in PBS. Western blot samples werenormalized so that lysates of cells possessing the same total numberof nuclei were loaded in each well (to compensate for the higher protein

content in myotubes, compared to myoblasts). Western blots weredeveloped using the ECL kit (Amersham, Little Chalfont, England).

Northern Blot Analysis. C2Ci2 cells were plated at 5 x 106 cells/

collagen-coated, 150-mm dish and induced to differentiate as above inthe presence of Are-C. Total cellular RNA was extracted from myotubesin SF medium (time 0) or stimulated for various lengths of time by 10%FBS in DMEM. Samples were loaded (25 pg/lane) on formaldehyde gels,then separated, transferred, blotted, and hybridized according to stand-aid protocols (59).

AcknowledgmentsWe are indebted to the following people who donated plasmids, probes,and antibodies: A. Felsani, F. Peverali, G. Draetta, D. Spitkovsky, P.Jansen-D#{252}rr,J. Nevins, A. Giordano, 0. Segatto, T. Baldari, and G.Raschebl#{225}.We thank F. Tat#{243}for critical reading of the manuscript.

References1 . Okazaki, K., and Holtzer, H. Myogenesis: fusion, myosin synthesis, andthe mitotic cycle. Proc. NatI. Acad. Sci. USA, 56: 1484-1490, 1966.

2. Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M., Benezra,R., Blackwelb, K. T., Tumer, D., Rupp, R., Hoblenberg, S., Zhuang, Y., and

Lassar, A. The myoD gene family: nodal point during specification of themuscle cell lineage. Science (Washington DC), 251: 761-766, 1991.

3. Olson, E. N. MyoD family: a paradigm for development? Genes 0ev., 4:1454-1461, 1990.

4. Crescenzi, M., Soddu, S., Sacchi, A., and Tatb, F. Adenovirus infectioninduces reentry into the cell cycle of terminally differentiated skeletalmuscle cells. Ann. N. Y. Acad. Sci., 752: 9-18, 1995.

5. Crescenzi, M., Soddu, S., and Tat#{243},F. Mitotic cycle reactivation interminally differentiated cells by adenovirus infection. J. Cell. Physiol.,162: 26-35, 1995.

6. Cardoso, M. C., Leonhardt, H., and Nadal-Ginard, B. Reversal ofterminal differentiation and control of DNA replication: cyclin A and cdk2

specifically localize at subnuclear sites of DNA replication. Cell, 74: 979-992, 1993.

7. Endo, T., and Nadal-Ginard, B. SV4O large T antigen induces reentry ofterminally differentiated myotubes into the cell cycle. In: L H. Kedes andF. E. Stockdale (eds.),. Cellular and Molecular Biology of Muscle Devel-opment, pp. 95-104. New York: Alan R. Liss, Inc., 1989.

8. Yaffe, D., and Gershon, D. Multinucleated muscle fibres: induction ofDNA synthesis and mitosis by polyoma virus infection. Nature (Lond.),215: 421-424, 1967.

9. Endo, T., and Nadal-Ginard, B. Transcriptional and posttranscriptionalcontrol of c-myc during myogenesis: its mRNA remains inducible in dif-


28. Weinberg, R. A. The retinoblastoma protein and cell cycle control.

Cell, 81: 323-330, 1995.

ferentiated cells and does not suppress the differentiated phenotype. Mol.Cell. Biol., 6: 1412-1421, 1986.

10. Crescenzi, M., Fleming, T. P., Lassar, A. B., Weintraub, H., and

Aaronson, S. A. MyoD induces growth arrest independent of differentia-tion in normal and transformed cells. Proc. NatI. Acad. Sci. USA, 87:

8442-8446, 1990.

1 1 . Sorrentino, V., Pepperkok, R., Davis, R. L, Ansorge, W., and

Philipson, L Cell proliferation inhibited by MyoDi independently of myo-genic differentiation. Nature (Lond.), 345: 813-815, 1990.

12. Schneider, J. W., Gu, W., Zhu, L, Mahdavi, V., and Nadal-Ginard, B.Reversal of terminal differentiation mediated by p107 in Rb’ musclecells. Science (Washington DC), 264: 1467-1471 ,1994.

13. Gu, W., Schneider, J. W., Condorelli, G., Kaushal, S., Mahdavi, V., and

Nadal-Ginard, B. Interaction of myogenic factors and the retinoblastomaprotein mediates muscle cell commitment and differentiation. Cell, 72:

309-324, 1993.

14. Guo, K., Wang, J., Andres, V., Smith, R. C., and Walsh, K. MyoD-

induced expression of p21 inhibits cyclin-dependent kinase activity uponmyocyte terminal differentiation. Mol. Cell. Biol., 15: 3823-3829, 1995.

15. Habevy, 0., Novitch, B. G., Spicer, D. B., Skapek, S. X., Rhee, J.,Hannon, G. J., Beach, D., and Lassar, A. B. Correlation of terminal cell

cycle arrest of skeletal muscle with induction of p21 by MyoD. Science(Washington DC), 267: 1018-1021, 1995.

16. Jiang, H., Lin, J., Su, Z-z., Collart, F. R., Huberman, E., and Fisher, P.B. Induction of differentiation in human promyelocytic HL-60 leukemia

cells activates p21, WAF1/CIP1 ,expression in the absence of p53. On-

cogene, 9: 3397-3406, 1994.

17. Missero, C., Calautti, E., Eckner, R., Chin, J., Tsai, L H., Livingston,

D. M., and Dotto, G. P. Involvement of the cell-cycle inhibitor Cipl/WAF1

and the E1A-associated p300 protein in terminal differentiation. Proc.NatI. Acad. Sci. USA, 92: 5451-5455, 1995.

18. Parker, S. B., Eichele, G., Zhang, P.,, Rawls, A., Sands, A. T., Bradley,A., Olson, E. N., Harper, J. W., and Elledge, S. J. p53-independent

expression of p21 cipi in muscle and other terminally differentiated cells.

Science (Washington DC), 267: 1024-1027, 1995.

19. Skapek, S. S., Rhee, J., Spicer, D. B., and Lassar, A. B. Inhibition of

myogenic differentiation in proliferating myoblasts by cyclin Dl -depend-

ent klnase. Science (Washington DC), 267: 1022-1 024, 199S.

20. Steinman, R. A., Hoffman, B., Iro, A., Guillouf, C., Uebermann, D. A.,and El-Houseini, M. E. Induction of p21 (WAF-1/CIP1) during differentia-

tion. Oncogene, 9: 3389-3396, 1994.

21 . Baserga, R. The biology of cell reproduction. Cambridge: HarvardUniversity Press, 1985.

22. Falcone, G., Alem#{225},S., and TatO, F. Transcription of muscle-specific

genes is repressed by reactivation of pp60v� in postmitotic quail myo-

tubes. Mob. Cell. Biol., 11: 3331-3338, 1991.

23. Gosset, L. A., Zhang, W., and Olson, E. N. Dexamethasone-depen-dent inhibition of differentiation of C2 myoblasts bearing steroid-inducibleN-ms oncogenes. J. Cell Biol., 106: 2127-2137, 1988.

24. Crescenzi, M., Crouch, D. H., and TatO, F. Transformation by mycprevents fusion but not biochemical differentiation of C2C1 2 myoblasts:mechanisms of phenotypic correction in mixed cuiture with normal cells.J. Cell Biol., 125: 1137-1145, 1994.

25. Wolff, J. A., Dowty, M. E., Jiao, S., Repetto, G., Berg, R. K., Ludtke,J. J., and Williams, P. Expression of naked plasmids by cultured myotubesand entry of plasmids into T tubules and caveolae of mammalian skeletalmuscle. J. Cell Sci., 103: 1249-1259, 1992.

26. Johnson, D. G., Schwarz, J. K., Cress, D. W., and Nevins, J. R.Expression of transcription factor E2F1 induces quiescent cells to enter Sphase. Nature (Lond.), 365: 349-352, 1993.

27. Shan, B., and Lee, W-H. Deregulated expression of E2F-l inducesS-phase entry and leads to apoptosis. Mol. Cell. Biol., 14: 8166-8173.1994.

29. Barone, M. V., Pepperkok, R., Peverabi, F. A., and Philipson, L Idproteins control growth induction in mammalian cells. Proc. NatI. Acad.

Sci. USA, 91: 4985-4988, 1994.

30. Peverali, F. A., Ramqvist, T., Saffrich, R., Pepperkok, R., Barone, M.V., and Philipson, L Regulation of G1 progression by E2A and Id helix-loop-helix proteins. EMBO J., 13: 4291-4301 ,1994.

31 . Benezra, R., Davis, R. L, Lockshon, D., Turner, D. L, and Weintraub,H. The protein Id: a negative regulator of helix-loop-helix DNA bindingproteins. Cell, 61: 49-59, 1990.

32. Jen, V., Weintraub, H., and Benezra, R. Overexpression of Id proteininhibits the muscle differentiation program: in viva association of Id withE2A proteins. Genes Dev., 6: 1466-1479, 1992.

33. Mittnacht, S., and Weinberg, A. A. G1/S phosphorylation of the reti-noblastoma protein is associated with an altered affinity for the nuclearcompartment. Cell, 65: 381-393, 1991.

34. Kelly, K., Cochran, B. H., Stiles, C. D., and Leder, P. Cell-specificregulation of the c-myc gene by lymphocyte mitogens and platelet-derived growth factor. Cell, 35: 603-610, 1983.

35. Lamph, W. W., Wamsley, P., Sassone-Corsi, P., and Verma, I. M.Induction of proto-oncogene JUN/AP-1 by serum and TPA. Nature

(Lond.), 334: 629-631, 1988.

36. Muller, R., Bravo, R., and Burckhardt, J. Induction of c-fos gene andprotein by growth factors precedes activation of c-myc. Nature (Lond.),312: 716-720, 1984.

37. Spitkovsky, D., Steiner, P., Lukas, J., Lees, E., Pagano, M., Schulze,A., Joswig, S., Picard, D., Tommasino, M., Eilers, M., and Jansen-DOrr, P.Modulation of cyclmn gene expression by adenovirus E1A in a cell line withE1A-dependent conditional proliferation. J. Virol., 68: 2206-2214, 1994.

38. Nevins, J. R. E2F: a link between the Rb tumor suppressor protein and

viral oncoproteins. Science (Washington DC), 258: 424-429, 1992.

39. Chellappan, S. P., Hiebert, S., Mudryj, M., Horowitz, J. M., andNevins, J. R. The E2F transcription factor is a cellular target for the RBprotein. Cell, 65: 1053-1 061 ,1991.

40. Hara, E., Ohshima, T., Ishii, T., Sugino, W., Tsutsui, K., Nakada, S.,Tsuchida, N., and Oda, K. Mechanism of induction of cellular DNA syn-

thesis by the adenovirus E1A 12S cDNA product. Exp. Cell Res., 198:250-258, 1992.

41 . Goldstein, S. Replicative senescence: the human fibroblast comes of

age. Science (Washington DC), 249: 1 129-1 133, 1990.

42. Clegg, C. H., and Hauschka, S. D. Heterokaryon analysis of muscle

differentiation: regulation of the postmitotic state. J. Cell Biob., 105: 937-947, 1987.

43. Ringertz, N. R., and Savage, R. E. Cell hybrids. New York: AcademicPress, 1976.

44. Seshadri, T., and Campisi, J. Repression of c-fos transcription and an

altered genetic program in senescent human fibroblasts. Science (Wash-ington DC), 247: 205-209, 1990.

45. Hara, E., Yamaguchi, T., Nojima, H., Ide, T., Campisi, J., Okayama, H.,

and Oda, K. Id-related genes encoding helix-loop-helix proteins are re-quired for G1 progression and are repressed in senescent human fibro-

blasts. J. Biol. Chem., 269: 2139-2145, 1994.

46. Dulic, V., Drullinger, L F., Lees, E., Reed, S. I., and Stein, G. H. Alteredregulation of G1 cyclins in senescent human diploid fibroblasts: accumu-lation of inactive cyclin E-Cdk2 and cyclin Dl -Cdk2 complexes. Proc.NatI. Acad. Sci. USA, 90: 1 1034-1 1038, 1993.

47. Duprey, P., and Lesens, C. Control of skeletal muscle-specific tran-

scription: involvement of paired homeodomain and MADS domain tran-scription factors. Int. J. Dev. Biol., 38: 591-604, 1994.

48. Gosset, L A., Kelvin, D. J., Stemberg, E. A., and Olson, E. N. A newmyocyte-specific enhancer-binding factor that recognizes a conserved

element associated with multiple muscle-specific genes. Mol. Cell. Biol.,9: 5022-5033, 1989.

49. Blau, H. M., Pavlath, G. K., Hardeman, E. C., Chiu, C-P., Silberstein,L, Webster, S. G., Miller, S. C., and Webster, C. Plasticity of the differ-

entiated state. Science (Washington DC), 230: 758-766, 1985.

1�0 Cell Cycle Control in Terminal Differentiation

50. Yaffe, D., and Saxel, 0. Serial passaging and differentiation of myo-

genic cells isolated from dystrophic mouse muscle. Nature (Land.), 270:

725-727, 1977.

51 . Rando, T. A., and Blau, H. M. Primary mouse myoblast purification,characterization, and transplantation for cell-mediated gene therapy. J.Cell Biol., 125: 1275-1287, 1994.

52. Di Flore, P. P., Pierce, J. H., Fleming, T. P., Hazan, R., UlIrich, A., King,

C. R., Schlessinger, J., and Aaronson, S. A. Overexpression of the humanEGF receptor confers an EGF-dependent transformed phenotype to NIH3T3 cells. Cell, 51: 1063-1070, 1987.

53. Santos, E., Tronick, S. R., Aaronson, S. A., Pulciani, S., and Barbacid,M. T24 human bladder carcinoma oncogene is an activated form of thenormal human homologue of BALB- and Harvey-MSV transforminggenes. Nature (Land.), 298: 343-347, 1982.

54. Sala, A., and Calabretta, B. Regulation of BALB/c 3T3 fibroblastproliferation by B-myb Is accompanied by selective activation of cdc2

and cyclin Dl expression. Proc. NatI. Acad. Sci. USA, 89: 10415-

10419, 1992.

55. Bemards, R., Houweling, A., Schrier, P. I., Bos, J. L, and van der Eb,

A. J. Characterization of cells transformed by Ad5/Ad12 hybrid eariyregion I plasmids. Virology, 120: 422-432, 1982.

56. Jiao, S., Schultz, E., and Wolff, J. A. Intracerebral transplants ofprimary muscle cells: a potential “platform” for transgene expression inthe brain. Brain Res., 575: 143-147, 1992.

57. Digiesi, G., Giacomini, P., Fraioli, R., Mariani, M., Nicotra, M. R.,

Segatto, 0., and Natali, P. G. Production and characterization of murinemAbs to the extracellubar domain of human neu oncogene product

GP185HER2. Hybridoma, 11: 519-527, 1992.

58. Bader, D., Masaki, T., and Fischman, D. A. Immunochemical analysisof myosin heavy chain during avian myogenesis in viva and in vitro. J. Cell

Biol., 95: 763-770, 1982.

59. Sambrook, J., Fritsch, E. F., and Maniatis, T. Molecular Cloning. A

Laboratory Manual. Cold Spring Harbor, New York: Cold Spring HarborLaboratory, 1989.

terminally differentiated skeletal myotubes arenotconfined...

Documents