The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

The mTOR signaling pathway mediates control of ribosomalprotein mRNA translation in rat liver

Ali K. Reiter, Tracy G. Anthony, Joshua C. Anthony,Leonard S. Jefferson, Scot R. Kimball∗

Department of Cellular and Molecular Physiology, The Pennsylvania State University College of Medicine,P.O. Box 850, Hershey, PA 17033, USA

Received 24 December 2003; received in revised form 14 April 2004; accepted 14 April 2004

Abstract

Previous studies have shown that oral administration of leucine to fasted rats results in a preferential increase in liver inthe translation of mRNAs containing an oligopyrimidine sequence at the 5′-end of the message (i.e. a TOP sequence). TOPmRNAs include those encoding the ribosomal proteins (rp) and translation elongation factors. In cells in culture, the prepon-derance of evidence suggests that translation of TOP mRNAs is regulated by the mammalian target of rapamycin (mTOR), aprotein kinase that signals through ribosomal protein S6 kinase (S6K1) to rpS6. However, the results of previous studies wererecently challenged by several reports suggesting that translation of TOP mRNAs is independent of mTOR, S6K1, and S6phosphorylation. The purpose of the present study was to evaluate the role of mTOR in the stimulation of TOP mRNA trans-lation by leucine in vivo. Fasted rats were treated with the mTOR inhibitor, rapamycin, prior to oral administration of leucine.It was found that rapamycin severely attenuated leucine-induced signaling through mTOR in liver. In addition, rapamycinprevented the enhanced translation of TOP mRNAs in rats administered leucine, as assessed by a decrease in the proportionof TOP mRNAs associated with polysomes (i.e. those mRNAs being actively translated). Instead, in rapamycin-treated rats,ribosomal protein mRNAs accumulated in the fraction containing monosomes (mRNA bound to one ribosome). The resultssuggest that in liver in vivo, mTOR-dependent signaling is critical for maximal stimulation of TOP mRNA translation.© 2004 Elsevier Ltd. All rights reserved.

Keywords: Leucine; Translation initiation; Eukaryotic initiation factor eIF4E

1. Introduction

The RNA content of the liver varies in relation tothe pattern of nutrient intake, rising in response toa meal and then declining during the postprandialperiod (Potter, Baril, Watanabe, & Whittle, 1968).

∗ Corresponding author. Tel.:+1-717-531-8970;fax: +1-717-531-7667.

E-mail address: skimball@psu.edu (S.R. Kimball).

The increase in RNA following a meal reflects anutrient-induced stimulation of ribosome biogene-sis. Nutrients act to stimulate ribosome biogenesisby causing an increase in portal blood insulin con-centration and by affecting directly signaling path-ways within the liver. The direct action of nutrientsis mediated in part by amino acids, particularlythe branched-chain amino acid, leucine (Anthony,Anthony, Yoshizawa, Kimball, & Jefferson, 2001). Akey signaling pathway affected by both insulin and

1357-2725/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.biocel.2004.04.004

2170 A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

leucine involves the protein kinase referred to as themammalian target of rapamycin (Gingras, Raught, &Sonenberg, 2001). Activation of the mTOR signalingpathway by insulin and leucine leads to increasedphosphorylation and activation of the ribosomal pro-tein (rp)S6 kinase, S6K1, which facilitates ribosomebiogenesis through increased translation of mRNAscontaining a terminal oligopyrimidine (TOP) se-quence at the 5′-end of the transcript (Meyuhas &Hornstein, 2000). Messages that contain a TOP se-quence include most, and perhaps all, of the ribosomalprotein mRNAs (Meyuhas, 2000).

A number of studies conducted in intact animalshave demonstrated activation of the mTOR signalingpathway in the liver following either intake of a pro-tein containing meal (Kimball, Jefferson, & Davis,2000) or oral administration of leucine (Anthonyet al., 2000). One study (Anthony et al., 2001) ex-amined the distribution of ribosomal protein mRNAsbetween nonpolysomal (i.e. inactive) and polysomal(i.e. actively translated) fractions following the ad-ministration of leucine to fasted rats and concludedthat activation of the mTOR signaling pathway wasresponsible for mediating the increased translationof TOP-containing messages. A similar conclusionhad been reached from studies involving mitogenstimulation of cells in culture (Jefferies, Reinhard,Kozma, & Thomas, 1994; Kawasome et al., 1998;Shima et al., 1998; Terada et al., 1994). However,the conclusion has been brought into question byrecent studies showing a divergence between TOPmRNA translation and mTOR-mediated signaling(Barth-Baus et al., 2002; Stolovich et al., 2002; Tanget al., 2001). In recent studies, cells in culture stimu-lated with either mitogens (Stolovich et al., 2002) oramino acids (Barth-Baus et al., 2002) manifested in-creased accumulation of TOP mRNAs in polysomesindependent of mTOR mediated signaling or rpS6phosphorylation. Because the results of the latterstudies using cells in culture disagree with the conclu-sion drawn from the earlier in vivo studies (Anthonyet al., 2001), we sought to reassess the requirement ofmTOR-mediated signaling for enhanced translation ofTOP mRNAs in rat liver following oral administrationof leucine. The results demonstrate that the mTORinhibitor rapamycin blocks leucine-induced signalingto rpS6 phosphorylation and this is associated with ashift of TOP mRNAs out of polysomes. Thus, in rat

liver in vivo, mTOR-mediated signaling appears to benecessary for the maintenance or stimulation of TOPmRNA translation.

2. Materials and methods

2.1. Animals

Male Sprague–Dawley rats weighing∼200 g werefasted 16 h prior to the experiment. Rats were ran-domly assigned to one of four treatment groups. Therats assigned to the rapamycin-treatment group wereinjected via the tail vein with rapamycin (0.75 mg ra-pamycin/kg body weight) while rats not receiving ra-pamycin were injected via the tail vein with an equiv-alent volume of vehicle (0.155 mM NaCl, 2% (v/v)ethanol). Two hours later, one-half of the rats in eachgroup were administered eitherl-leucine (1.35 g/kgbody weight) or an equivalent volume of saline(0.155 mM) by oral gavage. Fifty minutes followingthe oral gavage, rats were injected with a floodingdose (1.0 ml/100 g of body weight) ofl-[2,3,4,5,6-3H]phenylalanine (150 mM containing 3.70 GBq/L). Tenminutes following injection of the radiolabeled aminoacid, rats were sacrificed via decapitation. Trunkblood was collected for measurement of the leucineconcentration and specific radioactivity and liver wasexcised and processed as described below.

2.2. Measurement of protein synthesis

The hepatic rate of protein synthesis was ana-lyzed as previously described (Garlick, McNurlan, &Preedy, 1980). Briefly, following excision, a portionof the liver was weighed and homogenized using aDounce homogenizer in 7 volumes of buffer con-sisting of (in mM) 20 N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (pH 7.4), 100 KCl, 0.2EDTA, 2 ethylene glycol-bis (�-aminoethyl ether)-N,N,N′,N′-tetraacetic acid, 1 dithiothretiol, 50 NaF,50 �-glycerophosphate, 0.1 phenylmethylsulfonylfluoride, 1 benzamidine, and 0.5 sodium vanadate. A250�l aliquot of homogenate was used to estimatethe hepatic rate of protein synthesis. The remaininghomogenate was centrifuged at 10,000× g for 10 minat 4◦C for analysis of translation initiation factorphosphorylation state. Global rates of protein syn-

A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179 2171

thesis were assessed as the rate of incorporation of[3H]phenylalanine into liver protein and normalizedto the specific radioactivity of [3H]phenylalanine inthe serum as described previously (Anthony et al.,2001). The elapsed time for incorporation was calcu-lated as the time from injection of the radiolabel tothe time the tissue was homogenized (∼13 min).

2.3. Quantitation of eIF4G and 4E-BP1 bound toeIF4E

eIF4E was immunoprecipitated from the 10,000×g supernatants using a monoclonal anti-eIF4E anti-body (Santa Cruz Biotechnology, Santa Cruz, CA) aspreviously described (Kimball, Jurasinski, Lawrence,& Jefferson, 1997). Western blot analysis was per-formed on the immunoprecipitates using a polyclonalanti-eIF4G antibody, a polyclonal antibody that rec-ognizes 4E-BP1 regardless of phosphorylation state(Calbiochem), and the eIF4E monoclonal antibody.Results were normalized to the amount of eIF4E inthe immunoprecipitate.

2.4. Phosphorylation of 4E-BP1

Hyperphosphorylation of 4E-BP1 was measured asa change in migration during SDS-polyacrylamide gelelectrophoresis as detected by Western blot analysisas described previously (Kimball et al., 1997). Briefly,an aliquot of the 10,000× g supernatant was boiledfor 10 min and centrifuged at 10,000× g for 30 min.The resultant supernatant was added to one volume ofSDS sample buffer and then subjected to Western blotanalysis using a polyclonal 4E-BP1 antibody.

2.5. Phosphorylation of S6K1

Phosphorylation of S6K1 was measured by two sep-arate methods. In the first method, hyperphosphoryla-tion of S6K1 was measured as a decrease in mobilityduring SDS-polyacrylamide gel electrophoresis as de-scribed previously (Gautsch et al., 1998). Briefly, analiquot of the 10,000×g supernatant was added to onevolume of SDS sample buffer. Western blot analysiswas then performed using a polyclonal S6K1 antibody(Santa Cruz Biotechnoloogy, Santa Cruz, CA). In thesecond method, phosphorylation of S6K1 on Thr389was measured by Western blot analysis using a poly-

clonal antibody that recognizes S6K1 only when itis phosphorylated on Thr389 (New England Biolabs,Beverly, MA).

2.6. Phosphorylation of rpS6

Phosphorylation of rpS6 was analyzed by Westernblot analysis using an equal mixture of polyclonal an-tibodies that recognize rpS6 when it is phosphory-lated on Ser235/236 and when it is phosphorylatedon Ser240/244 (Cell Signaling Technology, Beverly,MA).

2.7. Analysis of mRNA distribution in sucrosedensity gradient analysis

Sucrose density gradient analysis of mRNA polyso-mal distribution was performed as previously de-scribed (Anthony et al., 2001). Briefly, one gram ofliver was homogenized using a Dounce homogenizerin 3 volumes of buffer containing (in mM) 40 HEPES(pH 7.5), 100 KCl, and 5 magnesium chloride. Thehomogenate was then centrifuged at 3000× g for15 min at 4◦C. Nine volumes of supernatant wereadded to one volume of detergent (10% Triton X-100, 10% sodium deoxycholate), and the mix waslayered over a 10–70% sucrose gradient. The gradi-ents were centrifuged at 90,000× g in a BeckmanSW28 rotor for 210 min at 4◦C. Following centrifu-gation, the gradients were fractionated on an Iscogradient fractionator while the absorbance at 254 nmwas continuously recorded. Five-milliliter fractionswere collected, and total RNA was extracted fromthe fractions for RN-ase protection assays (RPAs) asdescribed previously (Anthony et al., 2001).

2.8. Preparation of RNA probes and RPAs

Preparation of RNA probes and RPA analysiswas performed as previously described (Anthonyet al., 2001). Briefly, full-length cDNAs for rpS4,S8, and L26 were kindly provided by Dr. Ira Wool(Department of Biochemistry and Molecular Biol-ogy, University of Chicago, Chicago, IL). Codingregions for albumin, rpS4, S8, L26 were subclonedinto the pBluescript II SK+ vector (Stratatgene) andlinearized to produce template DNA. A linearized�-actin DNA template was purchased from Ambion.

2172 A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

One microgram of DNA template was mixed with[32P]UTP (800 Ci/mmol, 20 mCi/ml; Amersham) andlimiting (0.1 mM) non-radiolabeled UTP in a 20-�ltranscription reaction (MAXIscript In Vitro Tran-scription Kit, Ambion). The reactions mixtures weretreated with DNase I, heat denatured, and then gel-purified to produce a single-stranded RNA probe. Fivemicrograms of total RNA were coimmunoprecipitatedwith �-actin, rpL26, S4, and S8 RNA probes in asingle tube by hybridizing overnight in a water bathat 42◦C in Hybridization Buffer (RPA III, Ambion).The following day samples were digested with RNaseA/TI, and the remaining double-stranded mRNAfragment was ethanol precipitated, heat denatured,and then loaded on to a 5% acrylamide/8 M urea slabgel. After electrophoresis, the gels were wrapped inplastic and exposed to X-ray film at−70◦C for up to4 h. Albumin mRNA content was measured using thesame procedure in a separate tube.

2.9. Statistical analysis

All statistical analyses were performed usingInStat version 3.0a for Macintosh (GraphPad SoftwareInc., San Diego, CA). A one-way ANOVA was per-formed in which treatment group was the independentvariable. If statistically significant differences weredetected, a Tukey post-hoc analysis was performedto assess differences between individual means. Thelevel of significance was set atP < 0.05 for allstatistical tests.

3. Results

A previous study reported that oral administrationof leucine to fasted rats resulted in a trend towardincreased protein synthesis in liver (Anthony et al.,2001). In agreement with that report, in the presentstudy, global rates of protein synthesis tended (P =0.07) to be higher in rats administered leucine com-pared to fasted rats (Fig. 1). Administration of ra-pamycin had no effect on protein synthesis in liversof fasted rats and did not prevent the trend towardenhanced protein synthesis in response to leucine ad-ministration. Overall, the results suggest that mTOR-mediated signaling contributes little to the global rate

0

Fig. 1. Effect of oral leucine administration and rapamycin onglobal rates of protein synthesis in rat liver. Rats were deprived offood for 16 h and then administered rapamycin (dark grey bars)or an equal volume of vehicle (light grey bars) via the tail vein.Two hours later the rats in each group were divided into twosubgroups that were administered either saline (fasted) or leucine(leucine; 1.35 mg/kg body weight) by oral gavage. Global rates ofprotein synthesis were measured using the flooding dose technique50 min after gavage as described under Section 2. The results areexpressed as mean ± S.E.M., n = 7.

of protein synthesis when studied over a short timeperiod such as 1 h.

Previously, we demonstrated that an oral bolusof leucine resulted in increased phosphorylation ofthe mTOR substrates, 4E-BP1 and S6K1, in liver(Anthony et al., 2001). To evaluate the requirementof mTOR-mediated signaling in the leucine-inducedhyperphosphorylation of 4E-BP1 and S6K1, we re-examined these parameters in the current study. Phos-phorylation of both 4E-BP1 and S6K1 at multipleresidues results in altered mobility of the proteinsduring SDS-polyacrylamide gel electrophoresis withmobility decreasing as phosphorylation increases.Thus, the most highly phosphorylated form of eitherprotein corresponds to the slowest migrating band.The hyperphosphorylated form of 4E-BP1, referredto as the �-form, corresponds to the most highlyphosphorylated form of the protein and, in contrastto hypophosphorylated forms, is unable to bind toeIF4E. Based on the proportion of 4E-BP1 in the�-form, it is evident from the data shown in Fig. 2Athat rapamycin treatment caused a marked reduction

A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179 2173

in the basal state of phosphorylation of the protein.As previously observed (Anthony et al., 2001), oraladministration of leucine resulted in a significantincrease in 4E-BP1 hyperphosphorylation to 180%of the basal value. The leucine-induced stimulationof 4E-BP1 phosphorylation was completely blockedby rapamycin with the amount of the protein in the�-form being reduced to near the level seen in therapamycin-treated controls.

Hyperphosphorylation of 4E-BP1 results in a re-lease of the protein from the 4E-BP1·eIF4E com-plex permitting eIF4E to bind to eIF4G to form theactive eIF4G·eIF4E complex (reviewed in Raught,Gingras, & Sonenberg, 2000). To assess changes ineIF4E association with 4E-BP1 and eIF4G, eIF4E andprotein complexes containing eIF4E were immuno-

Fig. 2. Rapamycin attenuates the effect of leucine administrationon 4E-BP1 phosphorylation and association of eIF4E with 4E-BP1and eIF4G. Rats were administered rapamycin and/or leucine asdescribed in the legend to Fig. 1. (A) 4E-BP1 hyperphosphoryla-tion was assessed as a change in the amount of the protein presentin the more slowly migrating �-form. Results are presented as thepercent of the protein present in the �-form relative to the totalamount of the protein present in all three forms and are expressedas mean ± S.E.M. Light grey bars, control rats; dark grey bars,rapamycin-treated rats. A representative blot is shown in the inset.The position of the �-, �-, and �-forms of 4E-BP1 are denotedto the right of the inset. F, fasted rats; FR, fasted rats treatedwith rapamycin; L, rats administered leucine; LR, rats treated withrapamycin and subsequently administered leucine. ∗P < 0.001vs. rapamycin-treated, fasted rats; ∗∗P < 0.001 vs. fasted rats;†P = 0.001 vs. rats administered leucine (n = 7). (B) The as-sociation of eIF4E with 4E-BP1 was measured by Western blotanalysis of eIF4E immunoprecipitates as described under Section2. Results were normalized for the amount of eIF4E present inthe immunoprecipitate and are expressed as mean ± S.E.M. Lightgrey bars, control rats; dark grey bars, rapamycin-treated rats. Arepresentative blot is shown in the inset. The position of the �-and �-forms of 4E-BP1 are denoted to the left of the inset. F,fasted rats; FR, fasted rats treated with rapamycin; L, rats admin-istered leucine; LR, rats treated with rapamycin and subsequentlyadministered leucine. ∗P < 0.02 vs. fasted rats; †P = 0.005 vs.rats administered leucine (n = 7). (C) The association of eIF4Ewith eIF4G was measured by Western blot analysis using the sameimmunoprecipitates used in panel (B). Results were normalizedfor the amount of eIF4E present in the immunoprecipitate and areexpressed as mean ± S.E.M. Light grey bars, control rats; darkgrey bars, rapamycin-treated rats. A representative blot is shownin the inset. F, fasted rats; FR, fasted rats treated with rapamycin;L, rats administered leucine; LR, rats treated with rapamycin andsubsequently administered leucine. ∗P < 0.005 vs. fasted rats;†P = 0.002 vs. rats administered leucine (n = 7).

2174 A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

precipitated from liver using a monoclonal anti-eIF4Eantibody and the immunoprecipitate was analyzed byWestern blot analysis for eIF4E, 4E-BP1, and eIF4Gcontent. As shown in Fig. 2B, oral administration of

leucine caused a small, but not significant, decrease inthe amount of 4E-BP1 bound to eIF4E at the time pointstudied herein. Rapamycin administration increasedthe amount of 4E-BP1 bound to eIF4E in both fastedand leucine-treated rats. In contrast, leucine adminis-tration tended (P = 0.06) to enhance the associationof eIF4E with eIF4G (Fig. 2C). Administration of ra-pamycin to either fasted or leucine-treated rats reducedthe amount of eIF4E bound to eIF4G to below thecontrol value. Therefore, in liver, rapamycin-treatmentresults in increased eIF4E association with 4E-BP1and decreased assembly of the eIF4G·eIF4E complex.

Phosphorylation of S6K1, as assessed by re-duced mobility during SDS-polyacrylamide gel elec-trophoresis, was also enhanced by oral administrationof leucine to fasted rats (Fig. 3A), whereas adminis-tration of rapamycin to fasted rats resulted in a shift

Fig. 3. Rapamycin blocks the stimulation of S6K1 and rpS6 phos-phorylation associated with oral leucine administration. Rats wereadministered rapamycin and/or leucine as described in the leg-end to Fig. 1. (A) S6K1 hyperphosphorylation was assessed aschanges in the amount of the protein present in the more slowlymigrating �- and �-forms. Results are presented as the percentof the protein present in the �- and �-forms relative to the totalamount of the protein present in all three forms and are expressedas mean ± S.E.M. Light grey bars, control rats; dark grey bars,rapamycin-treated rats. A representative blot is shown in the in-set. The position of the �-, �-, and �-forms of S6K1 are denotedto the right of the inset. F, fasted rats; FR, fasted rats treatedwith rapamycin; L, rats administered leucine; LR, rats treated withrapamycin and subsequently administered leucine. ∗P < 0.001vs. rapamycin-treated, fasted rats; ∗∗P < 0.05 vs. fasted rats;†P = 0.001 vs. rats administered leucine (n = 5–7). (B) Phos-phorylation of S6K1 on Thr389 was measured by Western blotanalysis using an anti-phospho-S6K1 (Thr389) antibody as de-scribed under Section 2. Results are expressed as mean ± S.E.M.Light grey bars, control rats; dark grey bars, rapamycin-treatedrats. A representative blot is shown in the inset. F, fasted rats; FR,fasted rats treated with rapamycin; L, rats administered leucine;LR, rats treated with rapamycin and subsequently administeredleucine. ∗∗P < 0.005 vs. fasted rats; †P = 0.003 vs. rats adminis-tered leucine (n = 6–7). (C) Phosphorylation of rpS6 on Ser 235,236, 240, and 244 was measured by Western blot analysis using amixture of anti-phospho-rpS6 (Ser235/236) and anti-phospho-rpS6(Ser240/244) antibodies as described under Section 2. Results areexpressed as mean ± S.E.M. Light grey bars, control rats; darkgrey bars, rapamycin-treated rats. A representative blot is shownin the inset. F, fasted rats; FR, fasted rats treated with rapamycin;L, rats administered leucine; LR, rats treated with rapamycin andsubsequently administered leucine. ∗∗P < 0.05 vs. fasted rats;†P < 0.05 vs. rats administered leucine (n = 3).

A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179 2175

in distribution of the protein into the hypophosphory-lated �-form. Furthermore, rapamycin blocked com-pletely the leucine-induced phosphorylation of S6K1such that essentially all of the protein was present in

Fig. 4. Rapamycin attenuates the leucine-induced redistribution of ribosomal protein mRNAs into polysomes. Rats were administeredrapamycin and/or leucine as described in the legend to Fig. 1. Livers were harvested and polysomes were resolved by sucrose densitygradient centrifugation as described under Section 2. During fractionation, sucrose density gradients were separated into four fractions asthe absorbance at 254 nm was recorded (left panels). RNA was isolated from each fraction and subsequently analyzed by ribonucleaseprotection assay for the content of rpL26, rpS4, rpS8, �-actin, and albumin mRNAs as described under Section 2. Results of a typicalset of ribonuclease protection assays are shown in the panels to the right. Fraction numbers are denoted at the bottom of the figure. Thestudy was repeated three times and two to three animals per condition were analyzed.

the hypophosphorylated form in rapamycin-treatedrats. In cells in culture, phosphorylation of Thr389 onS6K1 is associated with activation of the kinase andis blocked by rapamycin (Pullen & Thomas, 1997).

2176 A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

Thus, phosphorylation of Thr389 may be a more di-rect indicator of S6K1 activation and mTOR-mediatedsignaling than changes in electrophoretic migrationof total S6K1. In the present study, phosphorylationof S6K1 (Thr389) was assessed by Western blotanalysis using an antibody specific for the phospho-rylated form of the protein. As shown in Fig. 3B,leucine administration resulted in a nine-fold increasein phosphorylation of S6K1 on Thr389. Rapamycinhad no significant effect on S6K1 (Thr389) phospho-rylation in fasted rats, but prevented completely thephosphorylation associated with leucine administra-tion. To confirm that the observed changes in S6K1phosphorylation were associated with altered activ-ity, phosphorylation of the S6K1 substrate, rpS6, wasmeasured by Western blot analysis using phospho-specific antibodies. It was found that leucine admin-istration was associated with a three-fold increase inrpS6 phosphorylation. As observed for phosphoryla-tion of S6K1 on Thr389, rapamycin had no significanteffect on rpS6 phosphorylation in fasted rats, but re-pressed the increase caused by leucine. Overall, theresults demonstrate that oral administration of leucinepromotes signaling through mTOR in liver and thatrapamycin strongly attenuates the effect.

To assess the effect of leucine and rapamycin onTOP mRNA translation, the distribution of mRNAsencoding rpS4, rpS8, and rpL26 in sucrose densitygradients was compared to that of �-actin and albuminmRNAs. As reported previously, in liver from fastedrats, the mRNAs encoding rpS4, rpS8, and rpL26 ex-hibited a predominantly non-polysomal distribution(i.e. fractions 1 and 2), with only a small portion ofeach mRNA associated with polysomes (Fig. 4). Incontrast, essentially all of the �-actin mRNA was as-sociated with polysomes. The distribution of albuminmRNA was intermediate between ribosomal proteinand �-actin mRNAs, with approximately 50% of themRNA associated with polysomes. Administration ofleucine to fasted rats had little or no effect on theamount of �-actin or albumin mRNA associated withpolysomes, but caused a dramatic shift of rpS4, rpS8,and rpL26 mRNAs from nonpolysomal to polysomalfractions. Rapamycin had no detectable effect on thedistribution of �-actin or albumin mRNA in eitherfasted or leucine-treated rats and had little or no ef-fect on the distribution of ribosomal protein mRNAsin fasted animals. In contrast, rapamycin strongly at-

tenuated the leucine-induced shift of ribosomal pro-tein mRNAs into polysomal fractions. However, inrapamycin-treated rats administered leucine, the pro-portion of ribosomal mRNAs in fraction 2, i.e. thefraction containing 40S and 60S ribosomal subunits aswell as 80S monosomes, compared to fraction 1 wasincreased.

4. Discussion

A number of studies have reported a correlation be-tween mTOR mediated signaling and translation ofmRNAs containing a 5′-TOP sequence (Jefferies et al.,1994, 1997; Levy, Avni, Hariharan, Perry, & Meyuhas,1991; Meyuhas, Avni, & Shama, 1996). Moreover,until recently, the enhanced TOP mRNA translationassociated with mTOR activation was thought to oc-cur through activation of S6K1 and phosphorylationof rpS6 (reviewed in Meyuhas & Hornstein, 2000). Inthis regard, a correlation between rpS6 phosphoryla-tion and TOP mRNA translation has previously beenobserved in cells in culture and in rat liver (Anthonyet al., 2001; Anthony, Reiter, Anthony, Kimball, &Jefferson, 2001; Jefferies et al., 1994, 1997). How-ever, results from recent studies using cells in culture(Barth-Baus et al., 2002; Stolovich et al., 2002; Tanget al., 2001) have brought into question the role ofmTOR-mediated signaling and S6K1 activation in theregulation of TOP mRNA translation. For example,stimulation of mouse embryonic stem cells with eithermitogens or amino acids reportedly enhances the pro-portion of TOP mRNAs in polysomes to the same ex-tent in cells lacking S6K1 as in wildtype cells (Barth-Baus et al., 2002; Stolovich et al., 2002; Tang et al.,2001). Furthermore, rapamycin-treatment of cells didnot significantly alter the proportion of TOP mRNAsin polysomes compared to untreated cells (Barth-Bauset al., 2002; Stolovich et al., 2002; Tang et al., 2001).Together, the results of the studies by Meyuhas andcoworkers (Barth-Baus et al., 2002; Stolovich et al.,2002; Tang et al., 2001) suggest that TOP mRNAtranslation does not depend on mTOR-mediated sig-naling or rpS6 phosphorylation.

In the present study, mTOR-mediated signaling in-creased in response to oral administration of leucineas assessed by phosphorylation of downstream targetsof mTOR including 4E-BP1, S6K1, and rpS6. On the

A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179 2177

other hand, mTOR-mediated signaling was repressedby rapamycin-treatment as indicated by decreased4E-BP1, S6K1, and rpS6 phosphorylation. Consistentwith previous reports (Anthony et al., 2001), leucineadministration resulted in enhanced loading of ribo-somes onto TOP mRNAs. However, in contrast tothe recent studies using cells in culture (Stolovichet al., 2002; Tang et al., 2001) rapamycin preventedcompletely the leucine-induced shift of ribosomalprotein mRNAs into polysomes. Instead, a significantproportion of the mRNAs encoding rpS4, rpS8, andrpL26 were present in fraction 2 following leucineadministration in the presence of rapamycin. Fraction2 contains 40S and 60S ribosomal subunits as well as80S monomers, i.e. 40S and 60S subunits joined in acomplex. Thus, in the presence of rapamycin, leucinestill stimulates the binding of the 40S subunit to TOPmRNAs and may promote the association of a com-plete ribosome on the message. However, the processis inefficient, and multiple ribosomes do not associatewith TOP mRNAs, i.e. they are not in polysomes.

A possible explanation for the failure of TOP mR-NAs to accumulate in polysomes is that in the absenceof S6K1 activation and rpS6 phosphorylation, the ini-tiation of translation on mRNAs with a TOP sequenceis inefficient, such that fewer 40S ribosomes associatewith the mRNA. The location of rpS6 near the mRNAbinding site (reviewed in Jefferies & Thomas, 1996)would support such an idea. An alternative explana-tion is that rather than interfering with the binding ofthe 40S ribosome to TOP mRNAs, dephosphorylationof rpS6 either prevents assembly of an active ribosomeat the AUG start translation site or stalls the ribosomeat the start site. In such a model, two independent sig-nals would be required for optimal TOP mRNA trans-lation. An mTOR-independent signal would promotethe binding of the 40S ribosomal subunit to TOP mR-NAs and a second, mTOR-dependent signal wouldenhance the initial steps in translation elongation onTOP mRNAs. The 5′-untranslated region of ribosomalprotein mRNAs is typically short. For example, the5′-untranslated region of the human S4 and S8 mR-NAs are 24 and 23 nucleotides, respectively (Meyuhas,2000). Considering that a ribosome located at the AUGstart codon occupies approximately 20 nucleotides up-stream (reviewed in Mathews, Sonenberg, & Hershey,2000), if a ribosome was stalled at the AUG start codonof the S4 or S8 mRNA, it would cover essentially all

of the 5′-untranslated region, making it unlikely thata second ribosome could bind to the cap structure atthe 5′-end of the message. Thus, one interpretation ofthe results of the present study is that reduced signal-ing through mTOR causes the ribosome to pause at,or near, the AUG start codon of TOP mRNAs.

The apparent discrepancy between the results ofthe present study and previous ones concluding thatmTOR-dependent signaling was not required for TOPmRNA translation (Barth-Baus et al., 2002; Stolovichet al., 2002; Tang et al., 2001), may be explainedin part by differences in experimental design. Forexample, in previous studies (Stolovich et al., 2002;Tang et al., 2001), cells in culture were deprived ofall amino acids followed by readdition of a com-plete amino acid mixture to the cells. In contrast,in the present study, amino acids were continuallypresent at basal values, even in fasted animals. Thus,in the studies using cells in culture, severe aminoacid deprivation may have resulted in modulationof both mTOR-dependent and -independent intracel-lular signaling pathways, e.g. the pathway leadingto activation of the eIF2� kinase, mGCN2 (e.g. seeKimball, Horetsky, & Jefferson, 1998). In contrast,we have previously reported that fasting does not in-duce eIF2� phosphorylation in rat liver (Yoshizawa,Kimball, & Jefferson, 1997). Another difference be-tween the present study and previous ones is that inearlier studies (Stolovich et al., 2002; Tang et al.,2001), amino acids and rapamycin were adminis-tered simultaneously whereas in the present studyrapamycin was administered 2 h prior to oral adminis-tration of leucine. In cells treated simultaneously withamino acids and rapamycin, no effect of rapamycinon polysomal aggregation of TOP mRNAs is ob-served in the first 15–30 min (Stolovich et al., 2002;Tang et al., 2001). However, upon more prolongedexposure, i.e. 1–2 h, rapamycin clearly reduces theproportion of TOP mRNAs present in the polysomalfraction (Tang et al., 2001). This observation is con-sistent with the results of the present study showingthat rapamycin administered 2 h prior to oral admin-istration of leucine attenuates the leucine-inducedredistribution of TOP mRNAs into polysomes.

In agreement with the results of a previous study(Anthony et al., 2001), in the present study, oral ad-ministration of leucine had no significant effect onglobal rates of protein synthesis. This finding con-

2178 A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179

trasts with the observation that leucine significantlyincreases global rates of protein synthesis in skele-tal muscle (Anthony, Anthony, Kimball, Vary, &Jefferson, 2000). Moreover, unlike liver, in skeletalmuscle, rapamycin attenuates, but does not prevent,the increase in global rates of protein synthesis thatoccurs after leucine administration (Anthony et al.,2000). The present results in rat liver also differfrom those of a previous study examining the effectof feeding on protein synthesis in liver and skeletalmuscle of neonatal pigs (Kimball et al., 2000). In thatstudy, feeding significantly increased protein synthe-sis in liver and muscle and rapamycin attenuated theeffect in both tissues. Similar to results observed inadult rats administered leucine, feeding a completemeal to neonatal pigs caused a larger increase in therate of protein synthesis in muscle than liver, andrapamycin did not completely prevent the stimulationof global rates of protein synthesis in muscle. Thus,the differences observed among the aforementionedstudies likely reflect both tissue- and age-dependentvariations in the response to amino acid availabil-ity. In support of this idea, the magnitude of thefeeding-induced stimulation of protein synthesis inboth liver and muscle of neonatal pigs declines withage (Burrin, Davis, Fiorotto, & Reeds, 1992; Davis,Fiorotto, Nguyen, & Reeds, 1993). Finally, the greatereffect of rapamycin in muscle compared to liver couldoccur if the translation of TOP mRNAs represented agreater proportion of total mRNA translation in mus-cle compared to liver. This idea is an interesting topicfor future studies.

Acknowledgements

We thank Lynne Hugendubler and Sharon Rannelsfor technical assistance. This project was supported byNational Institutes of Health grant DK-15658.

References

Anthony, J. C., Anthony, T. G., Kimball, S. R., Vary, T. C., &Jefferson, L. S. (2000). Orally administered leucine stimulatesprotein synthesis in skeletal muscle of post-absorptive ratsin association with increased eIF4F formation. Journal ofNutrition, 130, 139–145.

Anthony, J. C., Yoshizawa, F., Anthony, T. G., Vary, T. C.,Jefferson, L. S., & Kimball, S. R. (2000). Leucine stimulatestranslation initiation in skeletal muscle of post-absorptive ratsvia a rapamycin-sensitive pathway. Journal of Nutrition, 130,2413–2419.

Anthony, T. G., Anthony, J. C., Yoshizawa, F., Kimball, S. R., &Jefferson, L. S. (2001). Oral administration of leucine stimulatesribosomal protein mRNA translation but not global rates ofprotein synthesis. Journal of Nutrition, 131, 1171–1176.

Anthony, T. G., Reiter, A. K., Anthony, J. C., Kimball, S. R., &Jefferson, L. S. (2001). Deficiency of dietary EAA preferentiallyinhibits mRNA translation of ribosomal proteins in the liverof meal-fed rats. American Journal of Physiology, 281, E430–E439.

Barth-Baus, D., Stratton, C. A., Parrott, L., Myerson, H., Meyuhas,O., & Templeton, D. J. et al., (2002). S6 phosphorylation-independent pathways regulate translation of 5′-terminaloligopyrimidine tract-containing mRNAs in differentiatinghematopoietic cells. Nucleic Acid Research, 30, 1919–1928.

Burrin, D. G., Davis, T. A., Fiorotto, M. L., & Reeds, P. J. (1992).Hepatic protein synthesis in suckling rats: Effects of state ofdevelopment and fasting. Pediatric Research, 31, 247–252.

Davis, T. A., Fiorotto, M. L., Nguyen, H. V., & Reeds, P. J.(1993). Enhanced response of muscle protein synthesis andplasma insulin to food intake in suckled rats. American Journalof Physiology, 265, R334–R340.

Garlick, P. J., McNurlan, M. A., & Preedy, V. R. (1980). Arapid and convenient technique for measuring the rate ofprotein synthesis in tissue by injection of [3H]phenylalanine.Biochemistry Journal, 192, 719–723.

Gautsch, T. A., Anthony, J. C., Kimball, S. R., Paul, G. L., Layman,D. K., & Jefferson, L. S. (1998). Availability of eIF4E regulatesskeletal muscle protein synthesis during recovery from exercise.American Journal of Physiology, 274, C406–C414.

Gingras, A.-C., Raught, B., & Sonenberg, N. (2001). Regulationof translation initiation by FRAP/mTOR. Genes Development,15, 807–826.

Jefferies, H. B. J., & Thomas, G. (1996). Ribosomal protein S6phosphorylation and signal transduction. In J. W. B. Hershey, M.B. Mathews, & N. Sonenberg (Eds.), Translational control (pp.389–409). Cold Spring Harbor: Cold Spring Harbor LaboratoryPress.

Jefferies, H. B. J., Reinhard, C., Kozma, S. C., & Thomas,G. (1994). Rapamycin selectively represses translation of the“polypyrimidine tract” mRNA family. Proceedings of NationalAcademic Sciences of the United States of America, 91, 4441–4445.

Jefferies, H. B., Fumagalli, S., Dennis, P. B., Reinhard, C., Pearson,R. B., & Thomas, G. (1997). Rapamycin suppresses 5′TOPmRNA translation through inhibition of p70s6k. EMBO Journal,16, 3693–3704.

Kawasome, H., Papst, P., Webb, S., Keller, G. M., Johnson, G.L., & Gelfand, E. W. et al., (1998). Targeted disruption ofthe p70S6k defines its role in protein synthesis and rapamycinsensitivity. Proceedings of National Academic Sciences of theUnited States of America, 95, 5033–5038.

Kimball, S. R., Jurasinski, C. V., Lawrence, J. C., & Jefferson,L. S. (1997). Insulin stimulates protein synthesis in skeletal

A.K. Reiter et al. / The International Journal of Biochemistry & Cell Biology 36 (2004) 2169–2179 2179

muscle by enhancing the association of eIF-4E and eIF-4G.American Journal of Physiology, 272, C754–C759.

Kimball, S. R., Horetsky, R. L., & Jefferson, L. S. (1998).Implication of eIF2B rather than eIF4E in the regulation ofglobal protein synthesis by amino acids in L6 myoblasts.Journal of Biological Chemistry, 273, 30945–30953.

Kimball, S. R., Jefferson, L. S., & Davis, T. A. (2000). Feedingstimulates protein synthesis in muscle and liver of neonatalpigs through an mTOR-dependent process. American Journalof Physiology, 279, E1080–E1087.

Levy, S., Avni, D., Hariharan, N., Perry, R. P., & Meyuhas, O.(1991). Oligopyrimidine tract at the 5′ end of mammalianribosomal protein mRNAs is required for their translationalcontrol. Proceedings of National Academic Sciences of theUnited States of America, 88, 3319–3323.

Mathews, M. B., Sonenberg, N., & Hershey, J. W. B. (2000).Origins and principles of translational control. In N. Sonenberg,J. W. B. Hershey, & M. B. Mathews (Eds.), Translationalcontrol of gene expression (pp. 1–31). Cold Spring Harbor:Cold Spring Harbor Laboratory Press.

Meyuhas, O. (2000). Synthesis of the translational apparatusis regulated at the translational level. European Journal ofBiochemistry, 267, 6321–6330.

Meyuhas, O., & Hornstein, E. (2000). Translational control of TOPmRNAs. In N. Sonenberg, J. W. B. Hershey, & M. B. Mathews(Eds.), Translational control of gene expression (pp. 671–693).Cold Spring Harbor: Cold Spring Harbor Laboratory Press.

Meyuhas, O., Avni, D., & Shama, S. (1996). Translational controlof ribosomal protein mRNAs in eukaryotes. In J. W. B.Hershey, M. B. Mathews, & N. Sonenberg (Eds.), Translationalcontrol (pp. 363–388). Cold Spring Harbor: Cold Spring HarborLaboratory Press.

Potter, V. R., Baril, E. F., Watanabe, M., & Whittle, E. D. (1968).Systemic oscillations in metabolic functions in liver from rats

adapted to controlled feeding schedules. Federal Proceedings,27, 1238–1245.

Pullen, N., & Thomas, G. (1997). The modular phosphorylationand activation of p70s6k. FEBS Letters, 410, 78–82.

Raught, B., Gingras, A. -C., & Sonenberg, N. (2000). Regulationof ribosomal recruitment in eukaryotes. In N. Sonenberg, J. W.B. Hershey, & M. B. Mathews (Eds.), Translational controlof gene expression (pp. 245–293). Cold Spring Harbor: ColdSpring Harbor Laboratory Press.

Shima, H., Pende, M., Chen, Y., Fumagalli, S., Thomas, G., &Kozma, S. C. (1998). Disruption of the p70(S6k)/p85(S6k)gene reveals a small mouse phenotype and a new functionalS6 kinase. EMBO Journal, 17, 6649–6659.

Stolovich, M., Tang, H., Hornstein, E., Levy, G., Cohen, R., &Bae, S. S. et al., (2002). Transduction of growth or mitogenicsignals into translational activation of TOP mRNAs is fullyreliant on the phosphatidylinositol 3-kinase-mediated pathwaybut requires neither S6K1 nor rpS6 phosphorylation. Molecularand Cellular Biology, 22, 8101–8113.

Tang, H., Hornstein, E., Stolovich, M., Levy, G., Livingstone, M.,& Templeton, D. et al., (2001). Amino acid-induced translationof TOP mRNAs is fully dependent on phosphatidylinositol 3-kinase-mediated signaling, is partially inhibited by rapamycin,and is independent of S6K1 and rpS6 phosphorylation.Molecular and Cellular Biology, 21, 8671–8683.

Terada, N., Patel, H. R., Takase, K., Kohno, K., Nairn,A. C., & Gelfand, E. W. (1994). Rapamycin selectivelyinhibits translation of mRNAs encoding elongation factors andribosomal proteins. Proceedings of National Academic Sciencesof the United States of America, 91, 11477–11481.

Yoshizawa, F., Kimball, S. R., & Jefferson, L. S. (1997).Modulation of translation initiation in rat skeletal muscle andliver in response to food intake. Biochemical and BiophysicalResearch Communications, 240, 825–831.