cell cycle control during liver development in the rat...

Cell Cycle Control during Liver Development in the Rat:Evidence Indicating a Role for Cyclin D1Posttranscriptional Regulation1

Michael M. Awad and Philip A. Gruppuso2

Brown University School of Medicine, Providence, Rhode Island 02912,and Department of Pediatrics, Division of Pediatric Endocrinology andMetabolism, Rhode Island Hospital, Providence, Rhode Island 02903

AbstractHepatocytes are capable of marked changes inproliferation in response to various physiological andpathophysiological stimuli. Although the changes inadult hepatocyte growth regulation that accompanyreduction of liver mass, liver injury, and livercarcinogenesis have come under intense scrutiny, theregulation of hepatocyte growth during the latterstages of development is largely uncharacterized. Wehave examined hepatic cell cycle control in thedeveloping rat. Analysis of term (fetal day 21) liver andcultured, term hepatocytes revealed G0-G1 growth-arrested cells relative to preterm (fetal day 19) liver andisolated hepatocytes. G1 cyclin-dependent kinase(CDK) activity was correlated with growth arrest atterm in both in vivo and in vitro studies. The decline inCDK activity at term could not be attributed to achange in CDK protein content. Rather, the decline inCDK activity was associated with a concomitantdecline in cyclin D1 protein content. However, cyclinD1 mRNA levels did not correlate with protein levels.Cyclin D1 mRNA was present at a higher level in adultlivers, in which cyclin D1 protein was absent, than infetal livers. We also examined the phosphorylation(activation) state of p38 mitogen-activated proteinkinase, a potential hepatocyte-growth regulator andmodulator of cyclin D1 content. p38 activity wasinversely related to cyclin D1 content during liverdevelopment and regeneration. These data indicatethat a posttranscriptional mechanism regulating cyclinD1 content is involved in the temporary hepatocytegrowth arrest seen in the perinatal period and in themaintenance of adult hepatocytes in a quiescent state.We speculate that this posttranscriptional regulation

may be downstream from the p38 mitogen-activatedprotein kinase pathway.

IntroductionDuring the process of liver development, the transition fromfetal to adult life is accompanied by a striking change inhepatocyte proliferation. In rodents, the last 3 days beforebirth is accompanied by a burst of hepatocyte proliferationthat results in the tripling of liver mass and the replacementof the hematopoietic cell compartment by hepatocytes (1).This high rate of hepatocyte proliferation is in marked con-trast to the normally quiescent state seen in adult liver, inwhich only one in 20,000 hepatocytes undergoes mitosisunder normal circumstances. Yet adult rat hepatocytesmaintain the ability to proliferate after liver injury or reductionin liver mass (2). Such changes in growth require precise andresponsive cell cycle regulatory mechanisms.

Most of our current knowledge of these mechanisms inthe liver is derived from experiments on adult rats in whichliver mass is reduced by partial hepatectomy or injury,from models of hepatic carcinogenesis, or from studies onimmortalized hepatocyte cell lines (3, 4). However, mech-anisms regulating hepatocyte proliferation in these modelsmay not be representative of those that are active duringnormal liver development. For example, available dataindicate that growth factors such as hepatocyte growthfactor, transforming growth factor a, and epidermalgrowth factor play a significant role in liver regeneration,probably acting through the MAP3 kinase signal transduc-tion pathway (5). In contrast, data from our laboratoryindicate that fetal hepatocytes proliferate in the absence ofexogenous growth factors, and that the low, constitutivelevel of MAP kinase activation that is seen during lategestation may be growth factor-independent (6).

Studies using transgenic mice with homozygous gene de-letions have often been used to derive information on thedevelopmental role of various proteins. However such“knockout” experiments targeting cell cycle proteins haveprovided limited information. Deletions of most cyclins, mostnotably the G1 D- and E-type cyclins, result in few apparentdevelopmental abnormalities, none involving the liver (7, 8).Exceptions include deletions of the G2 cyclins A2 or B1,which are lethal early in embryogenesis. Germ-line deletionsof the CDKs have not been reported. Despite the importanceof the CKIs in postnatal carcinogenesis, deletions of these

Received 1/4/00; revised 3/17/00; accepted 4/17/00.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 indi-cate this fact.1 Supported by NIH Grants HD24455 and HD11343 and by the RhodeIsland Hospital Department of Pediatrics Research Endowment Fund.2 To whom requests for reprints should be addressed, at Department ofPediatrics, Rhode Island Hospital, 593 Eddy Street, Rhode Island Hospi-tal, Providence, RI 02903. Phone: (401) 444-5504; Fax: (401) 444-2534;E-mail: [email protected].

3 The abbreviations used are: MAP, mitogen-activated protein (kinase);CDK, cyclin-dependent kinase; CKI, CDK inhibitor; IP, immunoprecipita-tion; AEBSF, 4-(2-aminoethyl) benzenesulfonyl fluoride; RT-PCR, reverse-transcription PCR; GST, glutathione S-transferase.

325Vol. 11, 325–334, June 2000 Cell Growth & Differentiation

genes are generally well tolerated during development, lead-ing to only minor organomegaly of the pituitary, spleen, andthymus (9–11).

Limitations with these models include early embryonic le-thality prior to liver development and the lack of significantdevelopmental effects, most likely attributable to redun-dancy among cell cycle proteins. To circumvent theseissues, a transgenic mouse model was developed that spe-cifically overexpressed the CKI p21Cip1 in postnatal hepato-cytes (12). Hepatocyte proliferation was inhibited dramati-cally in the postnatal period, which resulted in a reduction inthe overall number of adult hepatocytes, aberrant tissueorganization, decreased liver growth, decreased somaticbody growth, and increased mortality. Significantly, thetransgenic p21 protein was demonstrated to be associatedwith most, if not all, of the cyclin D1-CDK4 complexes in liverbut not with other cyclin/CDK proteins, which emphasizesthe importance of functional cyclin D1-CDK4 complexes as apart of normal liver development.

To elucidate the role of specific proteins and complexesduring liver development, additional detailed studies of cellcycle protein expression and activity during the perinatalperiod are required. To this end, we have characterized thegrowth patterns of liver throughout development from lategestation through the adult period. Our earlier studies dem-onstrated an unusual ontogenic pattern of hepatocyte pro-liferation (13, 14). In vivo and correlative in vitro studiesshowed that the high rate of proliferation in preterm hepato-cytes is followed by an abrupt decline at term with subse-quent recovery of proliferation within 48 h of birth (13, 14).We have taken advantage of this observation by examiningthe content and activity of cell cycle constituents both duringthe period of temporary hepatocyte quiescence that occursat term and during the transition from proliferating neonatalhepatocytes to quiescent adult hepatocytes. In doing so, wedemonstrate G1 growth arrest in term fetal hepatocytes thatis accompanied by a concomitant decline in cyclin D1-associated CDK activity. Furthermore, we have obtainedevidence that this decrease is associated with inversechanges in cyclin D1 mRNA content, which indicates post-transcriptional regulation of cyclin D1 protein content in vivo.These results may pertain to mechanisms that maintain adulthepatocytes in a quiescent state. Our findings may, there-fore, relate to pathophysiological and physiological pertur-bations that are capable of reactivating hepatocyte growth inthe adult.

ResultsTerm Rat Liver Is Arrested in G0-G1 Phase of the CellCycle. To determine the phase of the cell cycle in whichterm hepatocytes are growth-arrested, flow cytometry ofcells recovered from preterm, term, and adult rat liver sec-tions was performed (Fig. 1A). Term liver obtained 6 h afterbirth demonstrated a high proportion of cells in G0-G1 phase,almost none in S-phase and few in G2-M relative to pretermliver cells. Adult liver cells also contained a large fraction ofcells in G0-G1 and almost none in S-phase. There was alarger proportion of cells derived from adult liver that were

identified as G2-M. This finding could be accounted for bythe tetraploid nature of nonproliferating adult hepatocytes.

To confirm the cell cycle status of term hepatocytes, flowcytometry was performed on freshly isolated preterm andterm hepatocytes (Fig. 1B). One-quarter of preterm hepato-cytes were found to be traversing S or G2-M phase. Incontrast, fewer than 10% of term hepatocytes were found tobe in S phase or G2-M. Nonetheless, both sets of analyseswere interpreted as showing G0-G1 hepatocyte growth arrestat term. Our previous findings comparing bromodeoxyuri-dine incorporation in cultured preterm versus term hepato-cytes (14) showed that DNA synthesis was asynchronous inpreterm hepatocyte cultures. In contrast, term hepatocytesshowed minimal DNA synthesis on the first day in culture andentered S-phase synchronously on the second day in cul-ture. These prior results, which indicated growth arrest thatwas relieved when cells were isolated, are consistent with thepresent flow cytometry findings that indicate G0-G1 growtharrest at term.

G0-G1 Growth Arrest at Term Is Associated with a De-crease in G1 CDK Activity. Passage through the G1 phaseof the cell cycle is dependent on activation of the CDKsCDK4 and/or CDK6 and the resulting phosphorylation of theretinoblastoma gene product pRb (15). Therefore, we meas-ured CDK4 and CDK6 activity in preterm and term whole liverhomogenates and cultured hepatocyte lysates. CDK4 andCDK6 activities were determined using an in vitro IP kinaseassay with GST-pRb as the kinase substrate. CDK4 activity

Fig. 1. Flow cytometry analysis of the cell cycle status of developinghepatocytes. Flow cytometry was performed on cells isolated from fixedliver (A), and on freshly isolated hepatocytes (B). Graphs represent theflow cytometry results as DNA content versus number of cells [left-mostgray peak, G0-G1; central black peak (when present), S phase; right-mostgray peak, G2-M]. Percentages below each graph, the proportion of cellsin each phase of the cell cycle. These results are representative of tripli-cate experiments.

326 Cell Cycle Control during Liver Development

decreased dramatically in term liver when compared withpreterm liver (Fig. 2A). Activity in adult liver was negligible.CDK6 activity paralleled the pattern for CDK4 (data notshown). However, signal intensities were very low for reasonsthat are not clear.

Lysates of preterm and term hepatocytes cultured for up to78 h were also assayed for CDK4 activity (Fig. 2B). In both ofthem, activity within 6 h of isolation was low. Activity inpreterm hepatocytes began to rise between 6 and 12 h inculture and reached peak CDK4 activity at 30–36 h in culture.In contrast, term hepatocytes did not demonstrate increasedCDK4 activity until 54 h in culture, peaking near the end ofthe experiment.

Activity of CDK4 and CDK6 is dependent on associationwith G1 cyclins, most notably cyclin D1 (16). To assess theinvolvement of cyclin D1 in active G1 CDK complexes, weimmunoprecipitated cyclin D1 from preterm, term, and adult

whole liver homogenates and used GST-pRb phosphoryla-tion as a measure of cyclin D1-associated kinase activity(Fig. 3). Kinase activity was high in preterm liver and nearlyundetectable in term and adult liver.

Hepatocyte Quiescence at Term Is Associated withDecreased Nuclear Cyclin D1 and Cyclin E Protein Con-tent. The activity of G1 CDKs requires the presence of nu-clear complexes of cyclin D1 with CDK4 and CDK6 (16). Todetermine whether the marked decrease in perinatal CDKactivity might be attributable to a decrease in content of oneor of more than one component, we measured both cyclin D1and CDK4/CDK6 nuclear protein content and complex for-mation. Nuclear extracts were prepared from preterm, term,and adult livers. Nuclear CDK4 content was determined bydirect Western blot analysis. Nuclear CDK4 protein levelsshowed a modest decline throughout the developmentalperiod studied (Fig. 4A). CDK6 protein levels could not beassessed by direct immunoblotting, presumably attributableto either low sensitivity of the CDK6 antibodies or low CDK6content. Therefore, these analyses were performed using IPfollowed by Western blotting. CDK6 nuclear protein contentdeclined as gestation proceeded (Fig. 4B). Levels in adultliver were similar to those seen in growth-arrested term liver.In both cases, CDK content did not decline to a degree thatcorrelated with the changes in kinase activity.

Coimmunoprecipitation assays of CDK6 with cyclin D1from perinatal and adult whole liver homogenates were per-formed to assess the level of G1 cyclin-CDK complexes.Significant levels of cyclin D1-associated CDK6 were de-tected in preterm liver. In contrast, cyclin D1-immunoprecipi-table CDK6 was nearly undetectable in term fetal and adultlivers (Fig. 5). In parallel experiments, cyclin D1-associatedCDK4 could not be detected in any samples. Again, our datadid not provide an explanation for this, although it is possiblethat the epitope recognized by the cyclin D1 antibody ismade inaccessible by binding to CDK4.

Fig. 2. CDK4 activity in vivo and in vitro. IP kinase assays were per-formed on whole-liver homogenates (A) and lysates of fetal hepatocytescultured for the indicated times [B, E19 (filled bars) and E21 (unfilled bars)].Activity was measured as incorporation of 32P from [g-32P]ATP into GST-pRb fusion protein. Representative autoradiograms are shown (lowerpanels) with corresponding densitometry (graphs). Error bars, SE. Resultswere confirmed in three additional experiments. C1, minus antibody con-trol; C2, minus sample control; C3, minus substrate control.

Fig. 3. Cyclin D1-associated kinase activity. IP kinase assays for GST-pRb phosphorylation were performed on E19, E21, and adult whole liverhomogenates immunoprecipitated with cyclin D1 antibodies. Densitom-etry is shown as the mean plus SE. Lower panel, a representative auto-radiogram. Control conditions are those described for Fig. 2. Similarresults were obtained in two additional experiments.

327Cell Growth & Differentiation

The decrease in cyclin D1-immunoprecipitable CDK6 pro-tein despite a modest decline in nuclear CDK6 levels sug-gested a block in cyclin-CDK complex formation. This ismost commonly attributable to the presence of a CKI or theabsence of cyclin D1 available for complex formation (17).Cyclin D1 protein levels were analyzed by Western blot ofunfractionated liver homogenates and nuclear extracts. Cy-clin D1 levels in whole liver homogenates were highest onE17, markedly reduced in the perinatal period (E21, P1), and

slightly increased in the immediate postnatal period (Fig. 6A).Cyclin D1 was undetectable in adult liver homogenates. Asimilar pattern was observed for cyclin D1 protein levels fromnuclear extracts (Fig. 6B). As for whole liver homogenates, nocyclin D1 of Mr 34,000 could be detected in adult nuclearextracts. However, several higher molecular weight immu-noreactive bands ranging from Mr 36,000 to 46,000 wereconsistently observed in adult nuclear extracts. Attempts torecover these forms by IP and to detect them in immunoblotshave been unsuccessful. Thus, their relationship to cyclin D1is uncertain.

Cyclin E-CDK2 complexes are required for cell cycle pro-gression through late G1 and for the G1-S transition (18).Cyclin E expression is dependent on and succeeds the ex-pression of cyclin D1 and subsequent activation of CDK4and/or CDK6 early in G1 phase (19). As an additional indica-tor of cyclin D1 down-regulation, cyclin E levels were as-sessed by Western blot analysis of nuclear extracts. Cyclin Eprotein levels were maximal in preterm liver, declined mark-edly in term liver, and were negligible in adult liver (Fig. 6C).

Fig. 4. CDK protein content in liver nuclear extracts. In A, nuclear ex-tracts prepared from livers of various developmental ages were analyzedby Western blot for CDK4. The result shown was replicated in a second,independent experiment. In B, parallel analyses were done for CDK6 usingIP followed by Western immunoblot. Upper panel, densitometric analysisof the autoradiogram shown in the lower panel. Error bars, SE; C1, noantibody control. Similar results were obtained in two replicate experi-ments.

Fig. 5. Cyclin-CDK complexes in preterm, term, and adult liver. Whole-liver homogenates were immunoprecipitated with anticyclin D1 and thenanalyzed by Western immunoblotting for CDK6. The graph shows densi-tometric analysis of the resulting autoradiogram (inset). Error bars, SE.

Fig. 6. Ontogeny of hepatic G1-cyclin protein content. Densitometry ofWestern immunoblotting analysis of cyclin D1 from whole-liver homoge-nates (A), cyclin D1 from nuclear extracts (B), and cyclin E from nuclearextracts (C) is shown with accompanying representative autoradiogram(insets). P.25, 6 h after birth; P.75, 18 h after birth. Results for all of theexperiments were replicated three times.


Cyclin D1 Down-Regulation Occurs in Term Hepato-cytes in Vivo. In the immediate prenatal period, ;50% of thecells in liver are of hematopoietic origin (1). To determine whichcell types contribute to the perinatal decrease in liver cyclin D1protein content, cyclin D1 immunohistochemistry of pretermand term liver was performed. In preterm liver, .85% of livercells demonstrated intense staining for cyclin D1 (Fig. 7A). Incontrast, ,15% of cells were cyclin D1-positive in term liver,and staining was significantly less intense in positive cells (Fig.7B). Individual hepatocytes in preterm and term liver were iden-tified by morphological phenotype and counted for cyclin D1staining. Results showed that the decrease in cyclin D1 stainingin hepatocytes from preterm to term paralleled that of all of thecells (85 versus 15%).

Discordance between Cyclin D1 mRNA Content andCyclin D1 Protein Levels during Liver Development. Cy-clin expression has been shown to be regulated at transcrip-tional, posttranscriptional, and posttranslational levels (20–23). We, therefore, examined whether mRNA levels parallelthe decrease in liver cyclin D1 and cyclin E protein levels asfetuses approach term. Total RNA from preterm, term, andadult livers were analyzed by Northern blot (Fig. 8A). Surpris-ingly, the 4.4-kb RNA detected was most highly abundant inthe three adult liver RNA preparations that we analyzed. Thesame RNA species was present in very low levels in the RNApreparations from preterm and term fetal liver. This pattern ofexpression could not be accounted for by RNA loading.Given the very low cyclin D1 mRNA level in fetal liver, weproceeded to confirm these results by relative quantitativeRT-PCR using the primer-dropping method (Fig. 8A). Thismethod has the advantage of higher sensitivity and specific-ity, assured by confirming the sequence of the cyclin D1 PCRproduct. Results showed that cyclin D1 mRNA levels weresimilar in preterm and term fetal liver. Again, the increase inadult liver was found in triplicate analyses. In addition, weanalyzed RNA preparations from a more complete panel ofliver samples (Fig. 8C). Results showed that a marked in-crease in cyclin D1 mRNA content occurred between the endof gestation and the end of the first postnatal week, by whichtime adult cyclin D1 mRNA levels had been attained. Tofurther confirm the validity of the analyses, we also examinedcyclin D1 expression after partial hepatectomy. As expected,cyclin D1 mRNA was markedly induced 24 h after partialhepatectomy compared with sham operation. In contrast to

cyclin D1, cyclin E mRNA paralleled cyclin E protein levels,being highest in preterm liver, lower in term liver, and nearlyundetectable in adult liver (Fig. 8D).

Temporal Association between p38 MAP Kinase Activ-ity and Cyclin D1 Protein during Liver Development andRegeneration. On the basis of available data (see “Discus-sion”), we hypothesized a potential role for the p38 MAPkinase pathway in the regulation of cyclin D1. Although our invivo liver development model is not conducive to studies thatcould place cyclin D1 downstream from p38, we did takeadvantage of our prior characterization of hepatocyte prolif-eration and cell cycle events during the perinatal period. Theactivity state of p38 was determined indirectly by Westernimmunoblotting with antibodies that recognize the active,phosphorylated form of the enzyme. Results (Fig. 9) showedan increase in phospho-p38 at term. This was not associatedwith a change in total p38 content. At 6 and 24 h after partialhepatectomy, p38 was under-phosphorylated comparedwith its state in livers from sham-operated animals (Fig. 9). By48 h after partial hepatectomy, p38 was phosphorylated to asimilarly high degree in both regenerating liver and liver fromsham-operated animals. These results demonstrate that p38activation, as determined by its phosphorylation state, isinversely related to cyclin D1 content during normal devel-opment. The finding that it is inactivated after partial hepa-tectomy supports the hypothesis that it functions as a tonicgrowth inhibitor in adult liver.

DiscussionExtraordinary progress has been made over the last decadein understanding the means by which cell replication is con-trolled. With regard to liver biology, many studies have fo-cused on the model of liver regeneration after partial hepa-tectomy (24–26). However, relatively few studies havefocused on mechanisms of cell cycle control under normalphysiological conditions associated with marked changes inrates of cell proliferation, such as those seen during theperinatal and neonatal-to-adult transitions in the rat. Hepa-tocyte proliferation proceeds rapidly during late gestationand the immediate neonatal period in the rat, with a period oftemporary quiescence occurring at term (12–14). A transitionto a quiescent adult hepatocyte phenotype takes place bythe end of the first postnatal week. The present studies were

Fig. 7. Cyclin D1 indirect fluores-cent immunohistochemistry of E19(A) and E21 (B) liver sections(3200).


undertaken with the premise that hepatocyte cell cycle reg-ulation during normal development involves mechanismsthat are relevant to other aspects of liver biology.

On the basis of the results from flow cytometry analyses, wefocused on the G1 phase of the cell cycle as the control point forperinatal hepatocyte growth arrest. The difference in flow cy-tometry results when comparing the whole liver, and cell cultureexperiments may be accounted for by the presence of a sig-nificant hematopoietic component in the fetal whole-liver sam-ples (1, 27), which is nearly absent in primary fetal hepatocytecultures (13). Alternatively, hepatocyte isolation may have af-fected the cell cycle status of isolated hepatocytes. Of note, we

found a significant proportion of cells in adult liver with 4 N DNAcontent. This is attributable to the increased ploidy (binucleatediploid and mononulceate tetraploid) typically seen in postnatalhepatocytes (27). These cells were indistinguishable from G2-Mphase cells in our analyses.

Our data indicate that, in particular, cyclin D1 can beassigned a key G1-phase regulatory role in perinatal hepa-tocyte growth arrest. IP kinase assays showed growth-associated changes in the activity of the early G1-phasecyclin D-dependent kinases, CDK4 and CDK6, without whichG1-phase cannot progress (28). Our in vivo analyses weresupported by in vitro results showing that preterm (E19)hepatocytes possess significant CDK4 activity during thefirst day in culture, whereas CDK4 activation was delayed by36–48 h in term hepatocytes. This pattern is consistent withour earlier findings (14), which suggest that term hepatocytesare under a growth-inhibitory influence in vivo and, oncerelieved from this influence, spontaneously enter the cellcycle in culture without exposure to serum or growth factors.

In term livers, pRb kinase activity in cyclin D1-immunopre-cipitated complexes was similarly diminished. Potential mech-anisms for the inhibition of CDK activity include the absence ofG1 cyclins and/or CDKs, prevention of complex formation bythe action of CKIs, or inhibition of the kinase activity of formedcyclin D1/CDK complexes by CKIs (29). CDK4 and CDK6 pro-teins could be detected in nuclear extracts at levels that wouldnot be limiting for the formation of cyclin-CDK complexes. Thisis consistent with the manner of posttranslational regulation bywhich CDKs are regulated in other systems. It should be noted,however, that the presence of G1-phase CDKs in adult livernuclear preparations has not been described previously. Thisobservation was unexpected given the view that quiescent,mature hepatocytes are generally considered to be arrested inG0. Persistent hepatic CDK expression in the adult may relate tothe observation that hepatocytes have considerable potentialfor rapid reactivation of growth after growth stimuli or liver injury(5, 25, 30).

In contrast to G1 CDK levels, cyclin D1 protein levels variedconsiderably and in parallel with hepatocyte proliferation,being lowest in term and adult liver preparations. Similarly,cyclin E levels correlated with hepatocyte growth arrest. Thisfinding is consistent with the role of cyclin E as a late G1-Scyclin that follows and is dependent on early G1 cyclin D-associated complex activity (19, 31, 32). Whereas there is asignificant decline in the proportion of hematopoietic cellsfrom preterm to term liver (1, 27), this did not account for thedecrease in hepatic cyclin D1 content based on the results ofimmunohistochemistry.

Published studies (33, 34) have demonstrated the involve-ment of the CKIs p21 and p27 in regulation of CDK activityduring liver regeneration. It is likely that these growth mod-ulators play a role in normal liver development. However,their ability to regulate CDK activity presumes the presenceof the requisite cyclin required for any particular CDK. Wehave performed preliminary studies on the expression ofCKIs during the perinatal period.4 Multiple CKIs show

4 P. A. Gruppuso, unpublished observations.

Fig. 8. G1-cyclin mRNA levels. Total RNA isolated from rat liver wasanalyzed for cyclin expression levels. Northern analysis was performed forcyclin D1 using subcloned PCR product as a probe (A, upper panel). rRNAlevels on the ethidium bromide-stained gel were used as a loading control(A, lower panel). To confirm these results, multiplex semiquantitative RT-PCR was performed to assess cyclin D1 mRNA content at the threedevelopmental time points analyzed by Northern blot (B). An additionalexperiment (C) was performed in which cyclin D1 mRNA content in singlesamples from an E19 fetal rat; from rat pups on postnatal (P) days 1, 4, 7,14, and 28; and from a normal adult rat (Ad). This experiment also includedanalysis of RNA preparations from two adult rats that underwent partialhepatectomy (PH) or sham operation (S) 24 h before sacrifice. For theselast three samples, the number of PCR cycles was reduced to keepamplification in the linear range. D, an experiment in which a parallelanalysis was carried out for cyclin E.


changes in their expression during the 48 h before, and theweek after, parturition. Whereas these findings may indicatea role for CKI expression in the control of hepatocyte prolif-eration during development, it is unlikely that they supplantthe role of cyclin D1 down-regulation at times of growtharrest, because cyclin D1 content would be limiting at thesetimes. This is in contrast to the acute growth stimulation seenduring liver regeneration after partial hepatectomy. Underthese circumstances, CKIs of the Cip/Kip family might berequired for cyclin/CDK complex formation (35).

Analysis of steady-state mRNA levels was used to deter-mine whether the regulation of G1 cyclins was transcriptionalor posttranscriptional. Whereas cyclin E mRNA levels paral-leled the profile for cyclin E protein content, cyclin D1 mRNAlevels did not decrease in maturing liver. In fact, cyclin D1message levels were elevated in adult liver relative to earlierdevelopmental time points. As was the case for the highadult liver CDK4 and CDK6 content, this result was unex-pected given the quiescent state of adult liver. Again, thisindicates that the resting state of quiescent adult rat hepa-tocytes differs markedly from the G0 state seen in otherwell-characterized cells, such as fibroblast cell lines, in whichexpression of G1 cell cycle proteins is absent or markedlydiminished (36–42).

Posttranscriptional regulation of cyclin D1 has been de-scribed previously in NIH3T3 cells and in immortalized bron-chial epithelial cell lines (20–22). It has been suggested thatthis is mediated in NIH3T3 cells at the translational level bycyclin D1 message interaction with the eukaryotic initiationfactor, eIF4E. Cyclin D1 posttranscriptional regulation in an invivo model, liver regeneration after partial hepatectomy in therat, has been proposed (26) but not demonstrated.

The mitotic cyclins A and B have been well characterizedwith regard to their modification and subsequent degrada-tion by ubiquitin-mediated proteolysis on the completion ofthe cell cycle (43–45). More recently, the ubiquitin-proteo-some pathway has been assigned a role in the control ofcyclin D and cyclin E content (46–48). Whereas the Mr

34,000 cyclin D1 bands detected by Western immunoblot-ting disappeared over the course of postnatal development,several cyclin D1-immunoreactive bands ranging from Mr

36,000 to 46,000 were consistently observed in adult livernuclear extracts. It is possible that these higher molecularweight bands represent a modified form of cyclin D1. It ispossible that higher molecular weight cyclin D1 immunore-active proteins may be an indication that the observed post-transcriptional regulation of cyclin D1 content involves cyclinD1 ubiquitination during liver development and in the main-tenance of the quiescent state in adult hepatocytes.

Our results do not define a mechanism for the posttran-scriptional regulation of cyclin D1 during development. How-ever, our data do suggest that cyclin D1 posttranscriptionalregulation has a key growth-regulating role in liver develop-ment. This is consistent with the findings of Albrecht andHansen (49), who showed that overexpression of cyclin D1 inprimary cultures of adult rat hepatocytes was sufficient topromote progression through the G1 restriction point. Withregard to the upstream mechanisms controlling cyclin D1content, we were led to examine a possible role for the p38MAP kinase pathway based on studies that have defined thissignaling kinase as mediating growth inhibition via an effecton cyclin D1 (50). In addition, we were influenced by datashowing that p38 can mediate posttranscriptional regulatoryevents (51). Our preliminary data examining the regulation ofp38 activity demonstrate a pattern that is inversely related tocyclin D1 abundance. On the basis of this, we have per-formed subsequent studies that indicate that p38 activationin cultured fetal hepatocytes can down-regulate cyclin D1content.5 The direct demonstration that this is mediated byposttranscriptional events will require further investigation.Nonetheless, the present studies strongly support the phys-

5 M. M. Awad, H. Enslen, J. M. Boylan, R. J. Davis, and P. A. Gruppuso.Growth regulation via p38 mitogen-activated protein kinase, submitted forpublication.

Fig. 9. Regulation of p38 MAP kinase during liver development and liver regeneration. Liver homogenates were analyzed by Western immunoblotting usingprimary antibody specific for phosphorylated (active) p38 (P-p38) or total p38. Samples were obtained from fetal rats on E17, E19, or E21. Postnatal sampleswere obtained 1, 4, 7, and 28 days after birth (P1, P4, P7, P28). Additional samples were obtained at various times after partial hepatectomy (PH) or shamoperation (S).


iological relevance of cyclin D1 to physiological hepatocytegrowth regulation, including a contribution of posttranscrip-tional control.

Materials and MethodsAnimals. Pregnant Sprague Dawley rats (Charles River Breeding Labo-ratory, Wilmington, MA) of known gestational age (day of conceptiondesignated as E0, preterm as E19, and term as E21) were used for all ofthe studies. Given the importance of precisely determining timing at theend of gestation, term animals were identified by frequent observation foronset of parturition. Partial (two-thirds) hepatectomy was carried out onadult male rats (125–150 g) under methoxyfluorane anesthesia. Sham-operated animals underwent liver exteriorization without excision.

Hepatocyte Isolation and Primary Culture. Fetal and newborn rathepatocytes were isolated by collagenase digestion as described previ-ously (13). Immunocytochemical analyses (13) have demonstrated thatthese preparations consist of ;90% hepatocytes, with the remaining cellpopulation consisting of a mixture of nonparenchymal cell types. This levelof hepatocyte predominance persists for up to 78 h in culture under thedefined mitogen-free conditions used for all of the experiments.

Hepatocytes were cultured on Falcon Primaria plates (Becton Dickin-son, Franklin Lakes, NJ) at a density of 2 3 106 cells per 100-mm plate.Cells became attached within 2 h in MEM containing 5% fetal bovineserum and supplemented with nutrients and cofactors, as describedpreviously (13). After cell attachment, all of the studies were done indefined (serum-free), supplemented MEM.

Preparation of Hepatocyte Lysates, Whole Liver Homogenates,and Nuclear Extracts. Preparation of hepatocyte lysates was carried outas described previously (28). Briefly, cultured cells were rinsed twice with10 ml of cold PBS and then scraped into 2 ml of IP buffer [50 mM HEPES(pH 7.5), 150 mM NaCl, 1 mM EDTA, 2.5 mM EGTA, 1 mM DTT, and 0.1%Tween 20] containing 10% glycerol, 144 mM AEBSF, 10 mg/ml leupeptin,10 mg/ml aprotinin, 10 mM b-glycerophosphate, 1 mM NaF, and 0.1 mM

sodium orthovanadate. Lysates were then sonicated at 4°C (full microtippower twice for 10 s each time; Ultrasonic Homogenizer 4710 Series,Cole-Parmer, Chicago, IL) and clarified by centrifugation at 10,000 3 g for5 min at 4°C.

For preparation of whole liver homogenates, the pooled livers from onelitter were combined in 1 ml per 100 mg tissue cold IP homogenizationbuffer without Tween 20. Tissue was homogenized for 10 strokes at 700rpm using glass-teflon homogenization vessels. Tween 20 was thenadded to a final concentration of 0.1%. Homogenates were clarified bycentrifugation at 10,000 3 g for 10 min at 4°C, frozen immediately on dryice, and stored at 270°C.

For preparation of whole-liver nuclear extracts, 0.5–1.0 g of pooled liverwas homogenized in 5 ml of buffer A1 [15 mM HEPES (pH 7.5), 300 mM

sucrose, 60 mM KCl, 15 mM NaCl, 2 mM EDTA, 0.5 mM EGTA, 14 mM

2-mercaptoethanol, 10 mM NaF, 1 mM sodium orthovanadate, 144 mM

AEBSF, 10 mg/ml leupeptin, and 10 mg/ml aprotinin] with 5 strokes at 800rpm in glass-teflon homogenization vessels. Homogenates were allowedto settle on ice for 5 min, and then the top 4 ml was centrifuged at 700 3g for 5 min at 4°C. The resulting supernatant was resuspended in 5 mlbuffer A2 (A1 with 250 ml of NP40 per 50 ml) and centrifuged over 5 ml ofbuffer B [15 mM HEPES (pH 7.5), 30% sucrose, 60 mM KCl, 15 mM NaCl,2 mM EDTA, 0.5 mM EGTA, and 14 mM 2-mercaptoethanol] for 5 min at1500 3 g at 4°C. Pelleted nuclei were resuspended in IP buffer. Lysateswere clarified by centrifugation at 10,000 3 g for 15 min at 4°C. Sampleswere frozen on dry ice and stored at 270°C.

Flow Cytometry. For preparation of cells from fixed livers, tissue wassuspended in IP buffer diluted with an equal volume of pepsin solution[140 mM NaCl and 5 mg of pepsin per ml of solution (pH 1.5)], incubatedat 37°C for 30 min with high-speed vortexing every 5 min, and incubatedin 2.5 volumes trypsin solution {120 mg of trypsin per ml citrate buffer [14mM sodium citrate, 2 mM Tris, 0.4% (v/v) NP40] and 10 mM spermine (pH7.6)} for 10 min at 20°C. Trypsin digestion was stopped by incubation with0.6 volumes trypsin inhibitor solution (0.1 g of trypsin inhibitor, 0.02 g ofRNase A per 50 ml citrate buffer) for 10 min at 20°C. Cells were stainedwith 0.4 volumes propidium iodide solution containing 0.083 g of pro-pidium iodide and 0.23 g of spermine per 50 ml of citrate buffer) for 15 minin the dark at 20°C and then analyzed.

For preparation of cultured hepatocyte suspensions, cells werescraped from 100-mm plates into 2 ml of cold PBS, washed with 2 3 10ml of cold PBS, and resuspended in PBS. Cells (2.0 3 106) were pelletedand resuspended in 1.0 ml of propidium iodide solution [7.5 mM propidiumiodide, 0.1% (v/v) NP40, and 0.1% (w/v) sodium citrate] for 15 min in thedark at 20°C and then analyzed.

All of the flow cytometric analyses were performed on a FACScan FlowCytometry System (Becton Dickinson, Franklin Lakes, NJ).

IP and CDK Assays. IP and kinase assays were performed as de-scribed by Matsushime et al. (28) with minor modifications. Briefly, 4 mgof liver homogenate protein or 100 mg of hepatocyte lysate protein wereimmunoprecipitated for 2 h at 4°C with protein A-Sepharose beads cross-linked to saturating amounts of the indicated antibodies (52). For kinaseassays, immunoprecipitated proteins on beads were washed four timeswith 1 ml of IP buffer and twice with 50 mM HEPES (pH 7.5) containing 1mM DTT. The beads were suspended in 30 ml of kinase buffer [50 mM

HEPES (pH 7.5), 10 mM MgCl2, 1 mM DTT] containing substrate [1 mg ofsoluble GST-pRb fusion protein (Santa Cruz Biotechnology, Inc., SantaCruz, CA)], 2.5 mM EGTA, 10 mM b-glycerophosphate, 0.1 mM sodiumorthovanadate, 1 mM NaF, 20 mM ATP and 10 mCi [g-32P]ATP (3000Ci/mmol; NEN DuPont, Boston, MA). Results were validated by the use ofthree control conditions: omission of antibody in the IP reaction (noantibody control), omission of sample, and omission of kinase substrate.After incubation for 30 min at 30°C with occasional mixing, the sampleswere boiled in PAGE sample buffer containing SDS and were separated byPAGE. Phosphorylated proteins were visualized by autoradiography of thedried gels.

PAGE and Western Blot Analyses. Liver proteins were separated on12% SDS-polyacrylamide gels and transferred to polyvinylidene difluoridemembranes (Bio-Rad, Hercules, CA). CDKs and cyclins were detectedusing antibodies obtained from Santa Cruz Biotechnology, Inc. (SantaCruz, CA). In particular, the cyclin D1 antibody does not cross-react withcyclins D2 or D3. For Western immunoblot detection of phosphorylated(active) p38 MAP kinase and total p38 MAP kinase, primary antibodieswere obtained from New England Biolabs (Beverly, MA). For all of theWestern blots, we used peroxidase-conjugated secondary antibody fol-lowed by chemiluminescent detection with the ECL Plus detection system(Amersham, Inc., Piscataway, NJ).

Immunohistochemistry. Formalin-fixed liver sections (6-mm) werecovered with 10 mM sodium citrate buffer (pH 6.0) and were heat-treatedat 95°C twice for 5 min. Sections were treated with avidin/biotin blockingsolutions (Vector Laboratories, Burlingame, CA), and then with 5% normalhorse serum (Life Technologies, Inc., Gaithersburg, MD) in PBS (15 mineach at room temperature). Sections were then incubated with 20 mg/mlcyclin D1 primary antibody (sc-8396; Santa Cruz Biotechnology, Inc.) inPBS for 30 min at room temperature followed by incubation with horseantimouse secondary antibody (1:500 dilution; Vector Laboratories). Sig-nal was detected after incubation with fluorescein-streptavidin conjugate(Vector Laboratories).

Relative Quantitative RT-PCR. Total RNA was isolated from frozenliver samples by homogenization in guanidium isothiocyanate followed bycesium chloride density centrifugation (53). cDNA was generated using 3mg of total RNA in the Superscript Preamplification System for First-Strand cDNA Synthesis kit (Life Technologies, Inc.). Primer-dropping PCRwas performed as described previously (54). Primer sequences used fordetection of rat cyclin D1 transcripts were 59-GCGTACCCTGACAC-CAATCT-39 for the sense primer and 59-GCTCCAGAGACAAGAAAC-GG-39 for the antisense primer, resulting in a predicted PCR product sizeof 232 bp. These primers do not recognize other D-type cyclins. Primersused to detect rat cyclin E transcripts were 59-ATGTCCAAGTGGC-CTACGTC-39 for the sense primer and 59-CTTTCTTTGCTTGGGCTTT-G-39 for the antisense primer, resulting in a predicted PCR product size of375 bp. PCR products were sequenced to confirm identity (Yale UniversityKeck Biotechnology DNA Sequencing Laboratories, New Haven, CT).Control rat b-actin primers were obtained from Clontech, Inc. (Palo Alto,CA). Optimal PCR cycle numbers, required for exponential amplificationfor each primer set, were determined by preliminary range-finding exper-iments. Total amplification in each multiplex reaction was kept belowsaturation levels to permit the products to remain within the exponentialrange of the amplification curve and, thereby, provide semiquantitativedata. Gels were illuminated with UV light, photographed, and analyzed by


digital image analysis. All of the PCR experiments were performed intriplicate to verify results.

Northern Blot Analysis. Total RNA was isolated as described above.Total RNA (20 mg) was separated on a 1% formaldehyde-agarose gel,followed by transfer to nylon membranes (Amersham Inc., Piscataway,NJ). Cyclin D1 probe was generated by subcloning the PCR productobtained from RT-PCR as described above into the pCRII-TOPO vector,linearizing the plasmid with BamHI, and generating a riboprobe using theRiboprobe In Vitro Transcription System as described by the manufac-turer (Promega, Inc., Madison, WI). Probe was labeled with 59-[a-32P]CTPto a specific activity of 2.0 3 108 cpm/mg. Labeled probe was incubatedwith membrane at 65°C for 18 h in hybridization buffer (0.1% SDS, 50%formamide, 53 SSC, 50 mM NaPO4 (pH 6.8), 0.1% sodium pyrophos-phate, 53 Denhardt’s solution, and 50 mg/ml salmon sperm DNA) fol-lowed by two 5-min washes (13 SSC-0.1% SDS) at 65°C. Membraneswere exposed to film for autoradiography.

Data Analysis. Quantification of bands from kinase assays, Westernblots, IPs, RT-PCR, and Northern blots was performed by digital imageanalysis using a Hewlett-Packard ScanJet 6100C scanner and Gel-ProAnalyzer software (Media Cybernetics, Silver Spring, MD).

AcknowledgmentsWe thank Joan Boylan for her invaluable advice and assistance in the

performance of these studies. We greatly appreciate the work of TheresaBienieki, who performed the hepatocyte isolations. We also thank the staffin the Central Research Laboratories at Rhode Island Hospital for theirassistance in the performance of the flow cytometry studies.

References1. Greengard, O., Federman, M., and Knox, W. E. Cytomorphometry ofdeveloping rat liver and its application to enzymic differentiation. J. CellBiol., 52: 261–272, 1972.

2. Steer, C. J. Liver regeneration. FASEB J., 9: 1396–1400, 1995.

3. Diehl, A. M., and Rai, R. M. Liver regeneration. 3. Regulation of signaltransduction during liver regeneration. FASEB J., 10: 215–227, 1996.

4. Sherr, C. J. Cancer cell cycles. Science (Washington DC), 274: 1672–1677, 1996.

5. Fausto, N., Laird, A. D., and Webber, E. M. Liver regeneration. 2. Roleof growth factors and cytokines in hepatic regeneration. FASEB J., 9:1527–1536, 1995.

6. Boylan, J. M., and Gruppuso, P. Uncoupling of hepatic, epidermalgrowth factor-mediated mitogen-activated protein kinase activation in thefetal rat. J. Biol. Chem., 273: 3784–3790, 1998.

7. Fantl, V., Stamp, G., Andrews, A., Rosewell, I., and Dickson, C. Micelacking cyclin D1 are small and show defects in eye and mammary glanddevelopment. Genes Dev., 9: 2364–2372, 1995.

8. Ma, C., Papermaster, D., and Cepko, C. L. A unique pattern of pho-toreceptor degeneration in cyclin D1 mutant mice. Proc. Natl. Acad. Sci.USA, 95: 9938–9943, 1998.

9. Serrano, M., Lee, H., Chin, L., Cordon, C. C., Beach, D., and DePinho,R. A. Role of the INK4a locus in tumor suppression and cell mortality. Cell,85: 27–37, 1996.

10. Fero, M. L., Rivkin, M., Tasch, M., Porter, P., Carow, C. E., Firpo, E.,Polyak, K., Tsai, L. H., Broudy, V., Perlmutter, R. M., Kaushansky, K., andRoberts, J. M. A syndrome of multiorgan hyperplasia with features ofgigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice.Cell, 85: 733–744, 1996.

11. Yan, Y., Frisen, J., Lee, M. H., Massague, J., and Barbacid, M.Ablation of the CDK inhibitor p57Kip2 results in increased apoptosis anddelayed differentiation during mouse development. Genes Dev., 11: 973–983, 1997.

12. Wu, H., Wade, M., Krall, L., Grisham, J., Xiong, Y., and Van, D. T.Targeted in vivo expression of the cyclin-dependent kinase inhibitor p21halts hepatocyte cell-cycle progression, postnatal liver development andregeneration. Genes Dev., 10: 245–260, 1996.

13. Curran, T. R. J., Bahner, R. I. J., Oh, W., and Gruppuso, P. A.Mitogen-independent DNA synthesis by fetal rat hepatocytes in primaryculture. Exp. Cell Res., 209: 53–57, 1993.

14. Gruppuso, P. A., Awad, M., Bienieki, T. C., Boylan, J. M., Fernando,S., and Faris, R. A. Modulation of mitogen-independent hepatocyte pro-liferation during the perinatal period in the rat. In Vitro Cell. Dev. Biol.Anim., 33: 562–568, 1997.

15. Roberts, J. M., Koff, A., Polyak, K., Firpo, E., Collins, S., Ohtsubo, M.,and Massague, J. Cyclins, Cdks, and cyclin kinase inhibitors. Cold SpringHarbor Symp. Quant. Biol., 59: 31–38, 1994.

16. Grana, X., and Reddy, E. P. Cell cycle control in mammalian cells: roleof cyclins, cyclin dependent kinases (CDKs), growth suppressor genesand cyclin-dependent kinase inhibitors (CKIs). Oncogene, 11: 211–219,1995.

17. Sherr, C. J., and Roberts, J. M. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev., 9: 1149–1163, 1995.

18. Ohtsubo, M., Theodoras, A. M., Schumacher, J., Roberts, J. M., andPagano, M. Human cyclin E, a nuclear protein essential for the G1-to-Sphase transition. Mol. Cell. Biol., 15: 2612–2624, 1995.

19. Lundberg, A. S., and Weinberg, R. A. Functional inactivation of theretinoblastoma protein requires sequential modification by at least twodistinct cyclin-cdk complexes. Mol. Cell. Biol., 18: 753–761, 1998.

20. Choi, Y. H., Lee, S. J., Nguyen, P., Jang, J. S., Lee, J., Wu, M. L.,Takano, E., Maki, M., Henkart, P. A., and Trepel, J. B. Regulation of cyclinD1 by calpain protease. J. Biol. Chem., 272: 28479–28484, 1997.

21. Hashemolhosseini, S., Nagamine, Y., Morley, S. J., Desrivieres, S.,Mercep, L., and Ferrari, S. Rapamycin inhibition of the G1 to S transitionis mediated by effects on cyclin D1 mRNA and protein stability. J. Biol.Chem., 273: 14424–14429, 1998.

22. Langenfeld, J., Kiyokawa, H., Sekula, D., Boyle, J., and Dmitrovsky, E.Posttranslational regulation of cyclin D1 by retinoic acid: a chemopreven-tion mechanism. Proc. Natl. Acad. Sci. USA, 94: 12070–12074, 1997.

23. Rosenwald, I. B., Lazaris, K. A., Sonenberg, N., and Schmidt, E. V.Elevated levels of cyclin D1 protein in response to increased expression ofeukaryotic initiation factor 4E. Mol. Cell. Biol., 13: 7358–7363, 1993.

24. Michalopoulos, G. K., and DeFrances, M. C. Liver regeneration. Sci-ence (Washington DC), 276: 60–66, 1997.

25. Fausto, N., and Webber, E. M. Control of liver growth. Crit. Rev.Eukaryotic Gene Expr., 3: 117–135, 1993.

26. Albrecht, J. H., Hu, M. Y., and Cerra, F. B. Distinct patterns of cyclinD1 regulation in models of liver regeneration and human liver. Biochem.Biophys. Res. Comm., 209: 648–655, 1995.

27. Knook, D. L., and Hollander, C. F. Embryology and Aging of the RatLiver. In: P. M. Newberne and W. H. Butler (eds.), Rat Hepatic Neoplasia,pp. 8–41. Cambridge, Massachusetts: The MIT Press. 1978.

28. Matsushime, H., Quelle, D. E., Shurtleff, S. A., Shibuya, M., Sherr,C. J., and Kato, J. Y. D-type cyclin-dependent kinase activity in mamma-lian cells. Mol. Cell. Biol., 14: 2066–2076, 1994.

29. Sherr, C. J., and Roberts, J. M. CDK inhibitors: positive and negativeregulators of G1-phase progression. Genes Dev., 13: 1501–1512, 1999.

30. Mead, J. E., Braun, L., Martin, D. A., and Fausto, N. Induction ofreplicative competence (“priming”) in normal liver. Cancer Res., 50: 7023–7030, 1990.

31. Jaumot, M., Estanyol, J. M., Serratosa, J., Agell, N., and Bachs, O.Activation of cdk4 and cdk2 during rat liver regeneration is associated withintranuclear rearrangements of cyclin-cdk complexes. Hepatology, 29:385–395, 1999.

32. Jiang, H., Chou, H. S., and Zhu, L. Requirement of cyclin E-Cdk2inhibition in p16INK4a-mediated growth suppression. Mol. Cell. Biol., 18:5284–5290, 1998.

33. Albrecht, J. H., Poon, R. Y., Ahonen, C. L., Rieland, B. M., Deng, C.,and Crary, G. S. Involvement of p21 and p27 in the regulation of CDKactivity and cell cycle progression in the regenerating liver. Oncogene, 16:2141–2150, 1998.

34. Ehrenfried, J. A., Ko, T. C., Thompson, E. A., and Evers, B. M. Cellcycle-mediated regulation of hepatic regeneration. Surgery, 122: 927–935, 1997.


35. Missero, C., Di, C. F., Kiyokawa, H., Koff, A., and Dotto, G. P. Theabsence of p21Cip1/WAF1 alters keratinocyte growth and differentiationand promotes ras-tumor progression. Genes Dev., 10: 3065–3075, 1996.

36. Tsao, Y. P., Kuo, S. W., Li, S. F., Liu, J. C., Lin, S. Z., Chen, K. Y., andChen, S. L. Differential regulation of cyclin A, cyclin B and p21 concen-trations in a growth-restricted human fibroblast cell line. Biochem. J., 312:693–698, 1995.

37. Herzinger, T., and Reed, S. I. Cyclin D3 is rate-limiting for the G1/Sphase transition in fibroblasts. J. Biol. Chem., 273: 14958–14961, 1998.

38. Lucibello, F. C., Sewing, A., Brusselbach, S., Burger, C., and Muller,R. Deregulation of cyclins D1 and E and suppression of cdk2 and cdk4 insenescent human fibroblasts. J. Cell Sci., 105: 123–133, 1993.

39. L’Allemain, G., Lavoie, J. N., Rivard, N., Baldin, V., and Pouyssegur,J. Cyclin D1 expression is a major target of the cAMP-induced inhibitionof cell cycle entry in fibroblasts. Oncogene, 14: 1981–1990, 1997.

40. Blain, S. W., Montalvo, E., and Massague, J. Differential interaction ofthe cyclin-dependent kinase (Cdk) inhibitor p27Kip1 with cyclin A-Cdk2and cyclin D2-Cdk4. J. Biol. Chem., 272: 25863–25872, 1997.

41. Meyer, D. H., Bachem, M. G., and Gressner, A. M. Bidirectionaleffects of Kupffer cells on hepatocyte proliferation in vitro. FEBS Lett.,283: 150–154, 1991.

42. Hirama, T., and Koeffler, H. P. Role of the cyclin-dependent kinaseinhibitors in the development of cancer. Blood, 86: 841–854, 1995.

43. Glotzer, M., Murray, A. W., and Kirschner, M. W. Cyclin is degradedby the ubiquitin pathway. Nature (Lond.), 349: 132–138, 1991.

44. King, R. W., Glotzer, M., and Kirschner, M. W. Mutagenic analysis ofthe destruction signal of mitotic cyclins and structural characterization ofubiquitinated intermediates. Mol. Biol. Cell, 7: 1343–1357, 1996.

45. Udvardy, A. The role of controlled proteolysis in cell-cycle regulation.Eur. J. Biochem., 240: 307–313, 1996.

46. Diehl, J. A., Zindy, F., and Sherr, C. J. Inhibition of cyclin D1 phos-phorylation on threonine-286 prevents its rapid degradation via the ubiq-uitin-proteasome pathway. Genes Dev., 11: 957–972, 1997.

47. Clurman, B. E., Sheaff, R. J., Thress, K., Groudine, M., and Roberts,J. M. Turnover of cyclin E by the ubiquitin-proteasome pathway is regu-lated by cdk2 binding and cyclin phosphorylation. Genes Dev., 10: 1979–1990, 1996.

48. Elledge, S. J., and Harper, J. W. The role of protein stability in the cellcycle and cancer. Biochim. Biophys. Acta, 1377: M61–M701998.

49. Albrecht, J. H., and Hansen, L. K. Cyclin D1 promotes mitogen-independent cell cycle progression in hepatocytes. Cell Growth Differ., 10:397–404, 1999.

50. Lavoie, J. N., L’Allemain, G., Brunet, A., Muller, R., and Pouyssegur,J. Cyclin D1 expression is regulated positively by the p42/p44MAPK andnegatively by the p38/HOGMAPK pathway. J. Biol. Chem., 271: 20608–20616, 1996.

51. Lee, J. C., Laydon, J. T., McDonnell, P. C., Gallagher, T. F., Kumar, S.,Green, D., McNulty, D., Blumenthal, M. J., Heys, J. R., and Landvatter,S. W. A protein kinase involved in the regulation of inflammatory cytokinebiosynthesis. Nature (Lond.), 372: 739–746, 1994.

52. Boylan, J. M., and Gruppuso, P. A. A comparative study of the hepaticmitogen-activated protein kinase and Jun-NH2-terminal kinase pathwaysin the late-gestation fetal rat. Cell Growth Diff., 7: 1261–1269, 1996.

53. Braun, L., Gruppuso, P., Mikumo, R., and Fausto, N. Transforminggrowth factor b 1 in liver carcinogenesis: messenger RNA expression andgrowth effects. Cell Growth Diff., 1: 103–111, 1990.

54. Wong, H., Anderson, W. D., Cheng, T., and Riabowol, K. T. MonitoringmRNA expression by polymerase chain reaction: the “primer-dropping”method. Anal. Biochem., 223: 251–258, 1994.


cell cycle control during liver development in the rat...

Documents