| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biochemistry and Molecular Biology (K.M.C., R.E.R.), Louisiana State University Health Sciences Center, Shreveport, Louisiana, 71130-3932; and Institute of Animal Science (I.B.), The Volcani Center, 50250 Bet-Degan, Israel
Address all correspondence and requests for reprints to: Robert E. Rhoads, Department of Biochemistry and Molecular Biology, Louisiana State University Health Sciences Center, Shreveport, Louisiana, 71130-3932. E-mail: rrhoad{at}lsuhsc.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-, ß-, and 
-caseins, -lactalbumin, ß-lactoglobulin, and whey acidic protein (WAP). Both development and lactation are regulated by a variety of hormones. Insulin, glucocorticoids, and prolactin play major roles in the accumulation of milk protein mRNAs (2). The functions of individual hormones in the onset and maintenance of lactation have been extensively studied at several levels, but one of the most intriguing aspects is the synergy exhibited by combinations of hormones (e.g. Ref. 3). At present, the mechanisms responsible for this synergy are only partially understood.
Transcription of the ß-casein gene is highly dependent on prolactin and less so on glucocorticoids, whereas the reverse is true for the WAP gene (reviewed in Ref. 2). Hydrocortisone further increases prolactin-stimulated ß-casein mRNA accumulation but has no effect in cells grown with insulin alone. The principal mechanism by which prolactin activates milk protein gene transcription is via the Jak2-Stat5 pathway. Milk protein synthesis is also regulated by changes in mRNA stability. Addition of prolactin to rat mammary gland explants grown in the presence of insulin plus hydrocortisone increases the stability of casein mRNA 17- to 25-fold (4). For cultured mouse mammary epithelial cells growing on collagen gels, addition of all three lactogenic hormones increases the rate of ß-casein gene transcription by 2.5-fold but increases the accumulation of ß-casein mRNA by 73-fold. Conversely, insulin has a major effect on transcription of the ß-casein gene but little effect on ß-casein mRNA stability (5). The increase in casein mRNA stability may result from increased polyadenylation. Kuraishi et al. (6) found that the poly(A) tract of mouse ß- and -casein mRNAs is shorter after weaning but longer when pups are again allowed to nurse. The length of the poly(A) tract is correlated with casein mRNA half-life both in vitro (6) and in vivo (7). Furthermore, there is an increase in the mRNA for poly(A) polymerase (PAP) during pregnancy and lactation, in parallel with the increase in poly(A) length of casein mRNA, suggesting that the poly(A) tract length is determined by PAP gene expression.
Various types of evidence have suggested that milk gene expression may also be regulated at the translational level. Travers et al. (8) found that involution of the rat mammary gland could be induced by either litter removal or deprivation of GH and prolactin. Litter removal for 48 h caused a decrease in casein mRNA levels and an increased association of casein mRNA with the monosome fraction, whereas deprivation of hormones caused no change in the association of casein mRNA with the polysome fraction. Baruch et al. (9) expressed in transgenic mice a chimeric gene derived from the promoter of ovine ß-lactoglobulin and the structural gene of human serum albumin. When mammary explants from virgin mice were cultured in the absence of hormones, human serum albumin mRNA continued to accumulate, but synthesis of the corresponding protein required insulin and prolactin. This phenomenon was pursued in mouse mammary epithelial cells (CID 9) cultured in the presence of insulin, where it was observed that addition of prolactin gradually increased the rate of protein synthesis over 24 h (10). This study also implicated a role for phosphorylated heat- and acid-stable protein regulated by insulin (PHAS-I) in prolactin-stimulated translation. PHAS-I binds to and sequesters initiation factor eIF4E, the cap-binding protein, but phosphorylation of PHAS-I in response to a variety of hormones and growth factors causes release of eIF4E and stimulation of cap-dependent translation (11). Addition of prolactin to CID 9 cells increased phosphorylation of PHAS-I and its release from eIF4E (10).
These previous studies in whole animals, mammary gland explants, and cultured cells suggest, but do not prove, an effect of lactogenic hormones on the process of translation itself. Due to the long time courses over which these experiments have been conducted (days to weeks), it is not possible to determine the extent to which changes in milk protein synthesis are due to mRNA levels, elaboration of endoplasmic reticulum, or other long-term cellular alterations rather than a direct translational effect. We therefore undertook a study of cultured mammary epithelial cells to test for changes in the protein synthesis rate immediately after hormone addition (0–30 min), verifying that there were no changes in mRNA levels. The results indicated that insulin alone stimulates the rate of total protein synthesis whereas prolactin alone has no effect, but insulin plus prolactin stimulate protein synthesis more than the sum of either hormone alone. These effects are even more striking for ß-casein synthesis. The mRNA-specific translational enhancement occurs through lengthening of the poly(A) tract via the cytoplasmic polyadenylation element binding protein (CPEB), not through a change in PHAS-I phosphorylation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We investigated protein synthesis during differentiation of CID 9 cells on Matrigel in the presence of either insulin alone or insulin plus prolactin by pulse-labeling for 30 min with [35S]Met each day. Western blotting indicated that - and ß-caseins began to accumulate by d 1 to d 2 in the presence of insulin plus prolactin but not insulin alone, whereas immunoprecipitation of proteins pulse labeled with [35S]Met indicated an increase in the instantaneous rate of - and ß-casein synthesis from d 3 to d 5 (data not shown). The increase in ß-casein synthesis was correlated with an increased accumulation of ß-casein mRNA. We chose d 5 cells cultured in the presence of insulin and prolactin for further experiments because milk proteins are clearly expressed.
To test whether lactogenic hormones affected milk protein synthesis at the translational level, it was necessary to measure their effect over a period of time too short for changes in mRNA levels. (The absence of mRNA changes was confirmed in experiments presented below.) We deprived the cells of insulin and prolactin for various lengths of time and then measured protein synthesis for the first 30 min after readdition of hormones (Fig. 1
). After 1, 3, or 6 h of hormone deprivation, there was no significant difference in the rate of protein synthesis between cells cultured for 30 min with no hormone (open bars), with prolactin alone (cross-hatched bars), with insulin alone (shaded bars), or with both hormones (striped bars). After 9 h of hormone deprivation, protein synthesis was stimulated by readdition of insulin, reaching a plateau after 14 h of hormone removal (1.66-fold over no hormone readdition). There was no increase in protein synthesis caused by readdition of prolactin alone. However, the combination of insulin plus prolactin stimulated protein synthesis more than the sum of insulin alone and prolactin alone (1.59-fold stimulation at 14 h compared with insulin alone, 2.64-fold compared with no hormone).
|
, striped bars at 14 h vs. 1 h). This may reflect hypersensitization of signal transduction components when hormones are removed. Attenuation of hormonal responses upon long-term exposure to either insulin (e.g. Ref. 15) or prolactin (e.g. Ref. 16) has been documented. When cells were subsequently maintained in the presence of insulin plus prolactin for 30 h, the protein synthesis rate returned to that observed with no deprivation of hormones (data not shown), supporting the explanation that protein synthesis is attenuated after long-term exposure to hormones.
It is conceivable that the increase in protein labeling could merely reflect a different rate of [35S]Met uptake in the presence of hormones, rather than a true increase in the translational rate. To test this, we measured the trichloroacetic acid-soluble radioactivity in CID 9 cells over a 30-min time course after addition of [35S]Met to the medium, comparing the four hormone treatments used in Fig. 1. The results indicated that [35S]Met uptake was the same for all treatments ± 8.2% (data not shown).
The dose response of total protein synthesis to prolactin, with and without insulin, is shown in Fig. 2A. Addition of prolactin alone failed to stimulate protein synthesis over a range of 0.05–0.8 µg/ml (open bars). Addition of insulin alone increased protein synthesis 1.8-fold compared with no hormone (striped vs. open bars at far left). However, addition of a constant amount of insulin and increasing amounts of prolactin stimulated protein synthesis 3.0-fold up to 0.5 µg/ml prolactin, after which it decreased (striped bars). This latter effect is similar to the biphasic action of prolactin in other systems, e.g. the human chorionic gonadotropin-stimulated secretion of progesterone by Leydig cells (17).
|
). Most bands increased in intensity after insulin treatment. When both hormones were added, some bands remained the same compared with insulin alone and some increased further, e.g. at approximately 46 kDa and approximately 30 kDa. To determine the effects of lactogenic hormones on the synthesis of individual proteins, we used antibodies to milk proteins (-casein, ß-casein, and WAP) and a typical non-milk protein, glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Fig. 3B). Synthesis of the three milk proteins was stimulated by insulin alone and was further stimulated by insulin plus prolactin. Synthesis of GAPDH, on the other hand, was stimulated by insulin alone, but no further stimulation occurred when prolactin was included.
|
. Prolactin alone failed to stimulate ß-casein synthesis (open bars). Insulin alone increased ß-casein synthesis 9.5-fold compared with no hormone (striped vs. open bars at far left). Addition of a constant amount of insulin and increasing amounts of prolactin stimulated ß-casein synthesis 8.7-fold over insulin alone (striped bars at 0.5 µg/ml vs. 0 µg/ml) and 83-fold over prolactin alone (striped vs. open bars at 0.5 µg/ml). As with total protein synthesis (Fig. 2A), the stimulation was less at 0.8 µg/ml than at 0.5 µg/ml prolactin (Fig. 2B).
To verify that these hormonal effects were due to changes in translational efficiency, it was necessary to show that mRNA levels did not change. We employed real-time PCR to quantitate levels of ß-casein mRNA relative to rRNA (Table 1). The concentration of ß-casein mRNA increased 3.2 x 106-fold when prolactin was included at 0.5 µg/ml during differentiation. When a lower concentration of prolactin was used (0.05 µg/ml), mRNA induction was correspondingly lower (1.5 x 106-fold). During the 14 h of hormone deprivation, ß-casein mRNA decayed to a relative value of 2.6 x 105. Most significantly for the question of translational control, adding back either insulin alone, prolactin alone, or insulin plus prolactin did not increase the level of ß-casein mRNA during the 30 min over which protein synthesis was measured. Comparable results were obtained when GAPDH mRNA rather than rRNA was used as a control RNA (data not shown). Thus, the increase in the rate of ß-casein synthesis observed upon hormone readdition in Figs. 2 and 3 is due to changes in the rate of mRNA translation, not mRNA levels.
|
The overall polysome distribution, as measured by OD254, indicated that the fraction of ribosomes in polysomes increased in cells incubated for 30 min with insulin (Fig. 4, B vs. A) and increased even more with insulin plus prolactin (Fig. 4, C vs. B). The polysomes were disaggregated when EDTA was included in the gradients, indicating that these profiles do not represent nonpolysomal ribonucleoprotein complexes (data not shown). ß-Casein mRNA was biphasically distributed when no hormones were added back to cells, with peaks in fractions 2 and 5 (Fig. 4D). By analysis of distances sedimented, we calculated (19, 20) that fraction 2 corresponds to the untranslated messenger ribonucleoprotein complex (mRNP) pool, and fraction 5 corresponds to low polysomes (5.1 ribosomes per mRNA). Insulin shifted most of the ß-casein mRNA out of mRNPs and into polysomes (Fig. 4E), and the average polysome size increased to 6.5 ribosomes per mRNA. Insulin plus prolactin caused an even more pronounced loss of ß-casein mRNA from mRNPs and a further shift to heavy polysomes (Fig. 4F), with a peak at 9.7 ribosomes per mRNA. Translation of GAPDH mRNA was also initiated more efficiently in cells treated with insulin compared with no hormone (Fig. 4, H vs. G), the peak shifting from 8.2 to 11.2 ribosomes per mRNA. However, the polysome distribution did not change when prolactin was included (Fig. 4, I vs. H). Overall, these data indicate that the combination of insulin plus prolactin increases the initiation rate of ß-casein mRNA relative to the elongation/termination rate but has no effect on GAPDH mRNA.
|
). Rapamycin inhibits the mammalian target of rapamycin (mTOR), an insulin-stimulated kinase that promotes initiation of highly cap-dependent mRNAs and 5'-TOP mRNAs (21). PD98059 inhibits MAPK (22), which is required for insulin-stimulated protein synthesis (23, 24). LY294002 is an inhibitor of phosphatidylinositol 3-kinase [PI3K (25)], which is upstream of the mTOR regulation of PHAS-I and p70S6K as well as the protein kinase B (PKB) regulation of glycogen synthase kinase-3 (GSK-3). LY294002 also inhibits mTOR directly (26, 27).
|
. However, in this experiment, the inhibitors were added 30 min before the simultaneous addition of hormones and [35S]Met. At 25 ng/ml, rapamycin reduced insulin-stimulated synthesis of total protein (Fig. 5A, solid triangles) as well as ß-casein (Fig. 5E) to the level seen with no hormone (open circles). This concentration is similar to that observed to inhibit insulin-stimulated protein synthesis in other systems (24). At higher concentrations of rapamycin, translation in the absence of hormones or in the presence of prolactin alone (solid circles) was inhibited further, presumably because mTOR has some basal activity in the absence of insulin. Most significantly, translation in the presence of insulin plus prolactin (open triangles) remained higher than with insulin alone at all rapamycin concentrations. This indicates that the pathway responsible for synergistic translational stimulation of total protein and ß-casein is not sensitive to rapamycin.
Insulin-stimulated synthesis of total proteins and ß-casein was also eliminated by 50 µM PD98059 (Fig. 5, panels B and F, solid triangles vs. open circles), which is similar to the concentration needed to inhibit MAPK in other systems (28), but the synergistic hormonal stimulation was maintained even up to 100 µM (open triangles vs. solid triangles). This indicates that the MAPK branch of the insulin pathway to translational stimulation is active in these cells but not responsible for the synergistic hormonal stimulation.
LY294002 at 25 µM produced the same effect on insulin-stimulated synthesis of total proteins and ß-casein as did rapamycin and PD98059 (Fig. 5, C and G). This LY294002 concentration is similar to that needed to inhibit PI3K in other systems (26), but the synergistic hormonal stimulation was maintained even up to 50 µM. This eliminates pathways downstream of PI3K as being responsible for this phenomenon.
The opposite behavior was seen with the chain terminator cordycepin (3'-deoxyadenosine), an inhibitor of polyadenylation (29, 30). There was no significant effect of cordycepin on synthesis of total proteins (Fig. 5D) or ß-casein (Fig. 5H) in the absence of hormone readdition (open circles), in the presence of insulin alone (solid triangles), or in the presence of prolactin alone (solid circles). However, cordycepin progressively inhibited synthesis of total proteins as well as ß-casein in the presence of insulin plus prolactin (open triangles). By 100 µg/ml, the rate was the same as with insulin alone. This concentration is similar to that observed to block cytoplasmic polyadenylation in other systems (31). These results suggest that the increase in translational efficiency of ß-casein mRNA is due to cytoplasmic polyadenylation.
Insulin Plus Prolactin Induces Cytoplasmic Poly(A) Elongation of ß-Casein, But Not GAPDH, mRNA
To test directly for a hormone-dependent change in the poly(A) length of ß-casein mRNA, we isolated total RNA from cells prepared as in Figs. 2–4 and examined mRNA length by northern blotting (Fig. 6A). The mobility of full-length ß-casein mRNA was the same in cells treated for 30 min with either no hormone, prolactin alone, or insulin alone, but it decreased slightly for cells treated with insulin plus prolactin (lanes 1–4). Because small differences in length are difficult to visualize with intact mRNAs, we examined only the 3'-portion of ß-casein mRNA by first cleaving specifically with ribonuclease H (RNase H) (32, 33) in the presence of an oligonucleotide complementary to nucleotide (nt) 454–473 (lanes 5–11). The decrease in mobility of the 3'-terminal RNA fragment for cells treated with insulin plus prolactin could be seen more clearly (lane 8 vs. lanes 5–7). The mobility shift was diminished by pretreatment of the cells with cordycepin in a dose-dependent manner (lanes 9–11). By 100 µg/ml of cordycepin, a concentration that eliminated synergistic stimulation of translation (Fig. 5, D and H), the mobility was the same as that from cells treated with no hormone (Fig. 6A, lane 10 vs. lane 5).
|
, lanes 12–15). This results in site-specific cleavage of ß-casein mRNA as well as removal of the poly(A) tract. The 3'-terminal mRNA fragments from cells treated under all four hormonal conditions had the same mobility. Thus, the reduced mobility observed in lane 8 is due to poly(A) length and not other structural changes (e.g. RNA editing, methylation, alternative splicing).
As a control, the poly(A) length of GAPDH mRNA was measured (Fig. 6B). No change in full-length GAPDH mRNA was detected with any of the hormonal treatments (lanes 1–4). Similarly, when only the 3'-terminal fragment was visualized, there was no change in length (lanes 5–8). Treatment with cordycepin did not change the length of the GAPDH mRNA fragment (lanes 9–11). Including oligo(dT) in the RNase H digestion resulted in a slight increase in mobility, reflecting the loss of a short poly(A) tract. Thus, the hormone-induced increase in poly(A) length observed with a typical milk protein mRNA, ß-casein, does not occur with a typical nonmilk protein mRNA, GAPDH.
The effects of various signal transduction inhibitors on hormone-stimulated polyadenylation were tested. Rapamycin and LY294002, which inhibit PHAS-I phosphorylation in CID 9 cells (10), were tested for an effect on the hormone-induced poly(A) lengthening of ß-casein mRNA (Fig. 6C). In addition, PD98059, which inhibits insulin-stimulated protein synthesis but not PHAS-I phosphorylation, was tested. None of these inhibitors affected poly(A) length under the four hormonal conditions tested. In particular, the poly(A) lengthening produced only in the presence of insulin and prolactin was not prevented (lanes 4, 8, 12, and 16).
The magnitude of poly(A) lengthening can be estimated from electrophoretic mobilities. The ß-casein mRNA fragment with no poly(A) (Fig. 6A, lanes 12–15) migrated at approximately 665 nt, similar to the 657 nt calculated from sequence information (34). The mRNA fragment from cells treated with no hormone, insulin alone, or prolactin alone (lanes 5, 6, and 7, respectively) was about 690 nt, whereas that from cells treated with insulin plus prolactin (lane 8) was approximately 869 nt. The kinetics of poly(A) lengthening of ß-casein mRNA after the addition of insulin and prolactin were investigated. As shown in Fig. 7, the poly(A) tract progressively increased from about 20 to approximately 200 nt over 30 min.
|
To detect polyadenylation changes in the exogenous PAC mRNA, RNase H was again employed, but this time with an oligonucleotide complementary to nt 1135–1154, immediately upstream of the ß-casein 3'-UTR. Northern blotting with a ß-casein mRNA probe detected the cleaved 3'-UTR (Fig. 8A). A reduction in mobility occurred in the presence of both insulin and prolactin (lane 4) but not with the other hormonal treatments (lanes 1–3). The mobility shift was prevented by pretreatment of the cells with 200 µg/ml cordycepin (lane 5). Controls with oligo(dT) in combination with the PAC-specific oligonucleotide indicated that the change in mobility was due to polyadenylation (lanes 6–9).
|
). In cells transfected with pPUR-3'ß, insulin alone produced a small but significant increase in PAC activity compared with no hormone (1.25-fold; shaded vs. open bars), whereas prolactin alone produced no change (cross-hatched vs. open bars). However, insulin plus prolactin caused a 2-fold increase in PAC activity (striped vs. open bars). In cells transfected with the empty vector pPUR, insulin alone increased PAC activity, but there was no further increase with insulin plus prolactin.
ß-Casein mRNA contains two sequences that resemble previously described mouse CPEs, at nt 977–982 and 1061–1068, as well as the AAUAAA hexanucleotide sequence at nt 1101–1106 (37). We wished to determine whether mutation of one or both of these putative CPE sequences resulted in loss of hormone-stimulated translation. The sequence of pPUR-3'ß was modified to generate an RNA in which the ß-casein 3'-UTR at nt 977–982 was changed from UUUUAU to UUUGGU (pPUR-3'ßM1). Similarly, the sequence was changed from UUUUUUUU to UUUGGUUU at nt 1061–1068, resulting in pPUR-3'ßM2. In cells transfected with pPUR-3'ßM1, the increase in PAC activity for insulin plus prolactin was the same as for insulin alone (Fig. 8), indicating that the CPE at nt 977–982 is responsible. Transfection with pPUR-3'ßM2, on the other hand, gave the same result as transfection with pPUR-3'ß, indicating that the putative CPE at nt 1061–1068 is not functional.
Insulin Plus Prolactin Increases PHAS-I Phosphorylation, but This Is Not Responsible for the Synergistic Stimulation of Milk Protein Synthesis
When CID 9 cells are differentiated in the presence of insulin alone for 5 d and then prolactin is added to the medium, phosphorylation of PHAS-I increases (10). This was offered as a possible mechanism for regulation of milk protein synthesis by lactogenic hormones. We therefore investigated whether PHAS-I phosphorylation might be responsible for the synergistic stimulation of translation. Insulin alone caused an increase in the more highly phosphorylated ß- and -forms of PHAS-I, and insulin plus prolactin converted most of the PHAS-I into the -form (data not shown). However, cordycepin did not affect the higher phosphorylation induced by insulin plus prolactin. Furthermore, agents that prevent PHAS-I phosphorylation did not block synergistic hormone-stimulated translation (Fig. 5) or poly(A) elongation (Fig. 6C). Thus, insulin plus prolactin causes two events: phosphorylation of PHAS-I and mRNA-specific poly(A) lengthening, but neither is a prerequisite for the other.
Insulin Plus Prolactin Causes Phosphorylation of CPEB but Does Not Change PAP Activity
In Xenopus oocytes, mRNA-specific polyadenylation is triggered by the phosphorylation of CPEB (38). We tested for a similar mechanism in CID 9 cells treated with insulin plus prolactin. Cells were prepared as described in Figs. 2–4 and then tested for CPEB phosphorylation after a 30-min exposure to each of the hormone combinations (Fig. 9A). Neither insulin alone nor prolactin alone caused a detectable change in the mobility of CPEB (lanes 2 and 3). However, the combination of the two hormones resulted in a portion of the CPEB shifting into the phosphorylated form (lane 4).
|
-casein mRNAs with the levels of PAP mRNA and proposed that the poly(A) length is determined by expression of PAP (7). We therefore tested whether the hormone-induced increase in ß-casein mRNA polyadenylation in CID 9 cells was due to a change in the activity of PAP. Cells were deprived of hormones overnight, and then various combinations of hormones were restored for 30 min. Cytoplasmic extracts were then assayed for PAP activity (Fig. 9B). The end-labeled poly(A) primer was elongated during a 10-min incubation with CID 9 cell extract (lanes 2–5) and elongated even more over 20 min (lane 6). Yeast poly(A) polymerase served as a positive control (lane 7). However, there was no difference in poly(A) polymerase activity among the four hormonal treatments (lanes 2–5). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-mediated routes, both of which are downstream of phosphatidylinositol-dependent kinase-1 and PI3K (39, 40). Insulin also decreases eIF2 phosphorylation in some cells (e.g. Ref. 41), which prevents its sequestration in an inactive complex with eIF2B. Another mechanism for insulin-stimulated initiation is via recruitment of mRNA to the 43S initiation complex. This involves phosphorylation of p70S6K and PHAS-I, both of which are downstream of mTOR, PKB, phosphatidylinositol-dependent kinase-1, and PI3K. Insulin-stimulated initiation of protein synthesis also requires MAPK (23, 24), possibly via Src homology domain protein tyrosine phosphatase 2 (SH-PTP2) (42) or p90S6K, the latter of which in turn regulates GSK-3 (43). Rapamycin blocks the insulin-stimulated pathway to PHAS-I phosphorylation but not the MAPK pathway (44, 45). The elongation phase of protein synthesis is also stimulated by insulin via two mechanisms: phosphorylation of eEF1 and inhibition of phosphorylation of eEF2 (46). The present study shows that protein synthesis in CID 9 cells is stimulated by insulin, and the signaling pathways involve mTOR, PI3K, and MAPK, based on the ability of rapamycin, LY294002, and PD98059, respectively, to inhibit insulin-stimulated translation (Fig. 5). Thus, many of the same pathways identified in other cells operate in CID 9 cells.
To our knowledge, the effect of prolactin on the translational machinery, on the other hand, has been the subject of only one prior investigation (10). When CID 9 cells were differentiated in the presence of insulin alone for 5 d, and then prolactin was added to the medium, the most highly phosphorylated -form of PHAS-I was enriched compared with the less phosphorylated - and ß-forms. Accumulation of the -form was abolished by mTOR and PI3K inhibitors but not MAPK inhibitors. Furthermore, addition of prolactin increased the rate of protein synthesis approximately 3-fold over a period of 24 h. This correlation suggested that selective translation of milk protein mRNAs may proceed via PHAS-I phosphorylation and the concomitant release of eIF4E.
The present study indicates that even though prolactin stimulates the phosphorylation of PHAS-I in the presence of insulin, this is not responsible for the synergistic stimulation of protein synthesis. Inhibitors of mTOR and PI3K, which prevent PHAS-I phosphorylation in CID 9 cells (10), also inhibit the stimulation of protein synthesis by insulin alone or insulin plus prolactin, but they do not eliminate the differential translational stimulation (Fig. 5) or poly(A) elongation (Fig. 6C) conferred by prolactin. On the other hand, cordycepin has no effect on PHAS-I phosphorylation in response to insulin or insulin plus prolactin (data not shown) but completely inhibits the synergistic stimulation of total protein and ß-casein synthesis (Fig. 5, D and H). Thus, the phosphorylation of PHAS-I is likely to contribute to insulin-stimulated translation (47, 48) but not synergistic hormone-stimulated translation.
Rather than PHAS-I, the increase in rate of ß-casein mRNA initiation is due to poly(A) lengthening because 1) cordycepin prevents the synergistic stimulation of translation but not the stimulation caused by insulin alone (Fig. 5, D and H); 2) the poly(A) length of ß-casein, but not GAPDH, mRNA increases with prolactin plus insulin but not with either hormone alone (Fig. 6); and 3) the dose response of cordycepin is the same for inhibition of synergistic translation and cytoplasmic polyadenylation of ß-casein mRNA (Figs. 5 and 6). Poly(A) lengthening increases the rate of translational initiation due to the binding of poly(A)-binding protein (PABP) to a specific site near the N terminus of eIF4G (49, 50). Poly(A) stabilizes the PABP-eIF4G-eIF4E complex, in effect causing the circularization of mRNAs by association of these proteins with the 5'-cap and 3'-poly(A) tract, which in turn leads to mRNA stabilization and enhanced translational reinitiation (51, 52). The translational stimulation conferred by the cap acts synergistically rather than additively with that conferred by the poly(A) tract (53, 54). In our system, we envision that the combination of insulin and prolactin causes two events that increase the rate of initiation: the release of eIF4E by phosphorylation of PHAS-I and the lengthening of the poly(A) of selected mRNAs. The shift of ß-casein mRNA to higher polysomes after insulin addition (Fig. 4) confirms that the stimulation is primarily through initiation rather than elongation. The observation that polysome size is further increased for ß-casein, but not GAPDH, mRNA in the presence of both hormones is consistent with the stimulatory effect of poly(A) lengthening on initiation rate (53, 54).
Meiotic maturation of stage VI Xenopus oocytes provides the best studied precedent for hormone-activated mRNA-specific increases in poly(A) length leading to a stimulation of translational initiation (reviewed in Ref. 35). During vertebrate oogenesis, gene expression is governed primarily by translational control (55, 56). The developing oocyte accumulates maternal components, including mRNAs, proteins, and ribosomes, that will be required during the rapid development that follows fertilization (57). The stored maternal mRNAs encode cell cycle-regulatory proteins such as c-Mos, Cdk2, and cyclins A, B1, and B2, which play a role in early embryonic development (38, 58, 59). These translationally dormant mRNAs reside in stable ribonucleoprotein complexes and contain short 3'-poly(A) tracts (60, 61). Upon progesterone stimulation, some of the stored maternal mRNAs are recruited to the protein synthetic machinery whereas many of the housekeeping mRNAs cease to be translated (62, 63). Progesterone initiates a series of events that result in the cytoplasmic addition of 100–300 A residues to these mRNAs (64). The activated aurora kinase Eg2 phosphorylates CPEB at Ser-174 (65, 66). Phosphorylated CPEB recruits the cleavage and polyadenylation specificity factor (CPSF) as well as PAP and PABP, which are necessary for polyadenylation (67). Cytoplasmic polyadenylation requires two cis-acting elements in the 3'-UTR: the CPE, which is bound by CPEB, and the hexanucleotide AAUAAA (35). Translation is repressed by binding of eIF4E to maskin, which interacts with nonphosphorylated CPEB (68, 67). Phosphorylation of CPEB also leads to dissociation of maskin from eIF4E, a step that is prevented by cordycepin (69). This is followed by the binding of PABP to the newly elongated poly(A) tract, the interaction of PABP with eIF4G, and the replacement of the inhibitory maskin-eIF4E complex with the active eIF4E-eIF4G complex that promotes cap-dependent translation.
Similar mechanisms for regulated polyadenylation appear to operate in the embryonic development of other organisms. In the surf clam Spisula solidissima, three abundant mRNAs encoding cyclin A, cyclin B, and ribonucleotide reductase are held in an inactive or masked state in the oocyte (70). The masked mRNAs are loaded onto polysomes after fertilization at a time when translationally regulated mRNAs undergo poly(A) extension (71, 72). These translationally up-regulated mRNAs contain several copies of CPE-like U4–6AA/U motifs in their 3'-UTRs (73). Deletion of the CPE of the ribonucleotide reductase mRNA prevents the binding of p82, the functional homolog of Xenopus CPEB (74). In Drosophila embryos, regulated polyadenylation is essential for correct embryonic patterning. The mRNAs for bicoid, Toll, and torso, which play important roles in body patterning, undergo poly(A) elongation and translational activation (75). Toll mRNA has a 192-nt element in the 3'-UTR that is necessary, but not sufficient, for polyadenylation (76). Cytoplasmic polyadenylation also takes place in maturing mouse oocytes. Several dormant maternal mRNAs possess short poly(A) tracts, but they are elongated to 100–150 nt after the induction of oocyte maturation, after which translation ensues (77). Mouse cyclin B1 mRNA contains functional CPEs that direct polyadenylation-stimulated translation in maturing oocytes. Mouse oocytes have factors homologous to those involved in Xenopus oocyte maturation, including CPEB, CPSF, maskin, and IAK1/Eg2 (78). The activities of both the aurora kinase and protein phosphatase PP1 are tightly regulated during prophase I progression. Mouse CPEB is dephosphorylated by PP1 at embryonic stage E18.5, thereby rendering it, and CPE-mediated mRNA translation, inactive until oocyte maturation (79). Active aurora kinase phosphorylates CPEB on Thr-171 (78).
Although cytoplasmic polyadenylation is a hallmark of early animal development, evidence for this mechanism for modulation of gene expression in late development or in adult tissues is much less complete (77). CPEB and the other polyadenylation translation factors are expressed in the mammalian brain, particularly the hippocampus (80, 81 81A ). Synaptic stimulation results in polyadenylation and translation of the CPE-containing -Ca2+/calmodulin-dependent protein kinase II mRNA, but not of the CPE-lacking neurofilament mRNA (80). Polyadenylation occurs at synapses, because glutamate or N-methyl-D-aspartate treatment of synaptosomes isolated from rat hippocampal neurons also stimulates -Ca2+/calmodulin-dependent protein kinase II mRNA polyadenylation (81, 31). As noted in the Introduction, casein mRNAs also undergo changes in poly(A) length as a function of physiological state. The poly(A) length increases during lactogenesis and decreases during weaning in mouse mammary gland (6). This is accompanied by corresponding changes in the level of PAP mRNA (7).
During Xenopus oocyte maturation, cdc2 kinase phosphorylates PAP at a number of sites (82, 83). As the enzyme become phosphorylated, it becomes progressively less active (84, 83). PAP alone does not recognize pre-mRNAs specifically but requires the AAUAAA element and CPSF, the latter of which binds PAP through its 160-kDa subunit (85). Even in the presence of CPSF, PAP activity remains weak, but its activity is stimulated by binding of PABP to the poly(A) tract (86). In Drosophila, a single type of PAP is responsible for regulation of cytoplasmic polyadenylation (87). An increase in the level of PAP in vivo affects cytoplasmic, but not nuclear, polyadenylation, leading to very long poly(A) tracts and embryonic lethality. In vertebrates, two PAP isoforms differing in their C termini have been described, PAP I (70 kDa) and PAP II (83 kDa) (88, 89). Spermatogenesis is regulated by a testis-specific, cytoplasmic poly(A) polymerase (90). Thus, it is clear that both the levels and intrinsic activity of PAP can affect cytoplasmic polyadenylation. In the present study, we tested directly whether the activity of PAP changes in response to short-term hormonal treatment in CID 9 but found this was not the case (Fig. 9B).
Experiments involving only the 3'-UTR of ß-casein mRNA indicated that this was sufficient for mRNA-specific cytoplasmic polyadenylation (Fig. 8A) and that only one of the two possible CPE-like sequences is involved (Fig. 8B). The active UUUUAU at nt 977–982 is similar to the maturation type of CPE (37). Also, CPEB was phosphorylated upon addition of insulin plus prolactin (Fig. 9A), providing evidence that polyadenylation of ß-casein mRNA is controlled by regulation of CPEB activity. However, at present, nothing is known of the signaling pathways leading from exposure of cells to insulin plus prolactin to CPEB phosphorylation. To our knowledge, this is the first reported instance in which two separate hormones are required to increase mRNA polyadenylation. Understanding the mechanism and pathways responsible may have important medical and agricultural implications.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-casein (93), mouse ß and -caseins (93), mouse WAP (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), rabbit GAPDH (Research Diagnostic, Inc., Flanders, NJ), mouse CPEB (79), and mouse actin (Santa Cruz Biotechnology, Inc.).
Culture and Differentiation of CID 9 Cells
Proliferating CID 9 cells were grown on 100-mm dishes in DMEM/F12 (1:1; Life Technologies, Gaithersburg, MD) containing 5% heat-inactivated fetal calf serum (Life Technologies), 50 µg/ml gentamycin (Life Technologies), and 5 µg/ml bovine insulin (Sigma) at 37 C in 5% CO2. To achieve differentiation, 0.6 x 106 cells were seeded on 35-mm dishes coated with 135 µl Matrigel (BD Biosciences, Bedford, MA) (d 0). The next day (d 1), the medium was changed to DMEM/F12 without serum but containing gentamycin, 5 µg/ml insulin, and 0.05 µg/ml ovine prolactin. In some experiments, prolactin was either omitted or used at other concentrations, as noted.
Withdrawal and Readdition of Hormones
In experiments testing the effects of hormone combinations, the medium of d 5 cells was changed to DMEM/F12 plus gentamycin but without hormones, and the cells were maintained for 14 h (unless noted otherwise in the figure legends). Medium with no hormones was replaced with medium containing various combinations of insulin and prolactin, and culture was continued for 30 min. When the effects of signal transduction inhibitors were tested, either rapamycin (Calbiochem, San Diego, CA; dissolved at 1 mg/ml in ethanol), PD98059 [Calbiochem; provided in dimethylsulfoxide (DMSO) at 5 mg/ml], LY294002 (Calbiochem; provided at 10 mM in DMSO), or cordycepin (Sigma; dissolved at 5 mg/ml in ethanol-water, 1:1) were added 30 min before hormone addition. In separate experiments, we showed that there were no changes in protein synthesis at the highest concentrations of either DMSO (0.5% vol/vol) or ethanol (2% vol/vol) alone.
Plasmid Construction and Transfection
The ß-casein 3'-UTR was amplified from endogenous ß-casein mRNA in CID 9 cells by PCR using as primers 5'-TATCTAGAAACTGACTGAAACTGGAAAT-3' and 5'-TAGGATCCTAATACGACTCACTATATACATTAAAAGTGAATGATCT3 '), the latter of which contains the T7 promoter sequence. The resulting DNA product was introduced between the XbaI and BamHI sites downstream of the PAC coding sequences of pPUR to produce the new plasmid pPUR-3'ß. Site-directed mutagenesis of pPUR-3'ß was performed by PCR using primers 5'-GACGGCGCCGCGGTGGCGGT-3', 5'-GAAATTCCTGTAATATA ACT-3', 5'-GTAAGTGTTCAATATGGAGTT-3', and 5'-GGGTGTTGGGCCCTTGTGCA-3' to produce pPUR-3'ßM1, and primers 5'-GACGGCGCCGCGGTGGCGGT-3', 5'-CAAATAGAAAATCCAATTTT-3', 5'-GTTTCTTTGAGAACCTATTTC-3', and 5'-GGGTGTTGGGCCCTTGTGCA-3' to produce pPUR-3'ßM2. The PCR-amplified DNA fragments were introduced between the KspI and ApaI sites of pPUR-3'ß. Proliferating CID 9 cells (0.6 x 105) were transfected with either pPUR, pPUR-3'ß, pPUR-3'ßM1, or pPUR-3'ßM2 using the TransFast Transfection reagent (Promega Corp., Madison, WI). The next day, cells were seeded on Matrigel and allowed to differentiate for 3 d in the presence of 5 µg/ml insulin and 0.05 µg/ml prolactin.
Metabolic Labeling with [35S]Met
Day 5 cells that had been deprived of hormones for 14 h were incubated with prewarmed methionine-free DMEM for 10 min. The medium was changed to 1 ml prewarmed methionine-free DMEM plus F12 (1:1) containing 15–50 µCi [35S]Met (ICN Biochemicals, Inc., Cleveland, OH) and the hormones indicated in the figure legends. Incubation was continued for 30 min. Metabolic labeling was stopped by three washes with ice-cold PBS containing 10 mM Met. The cells were lysed in 50 mM Tris-HCl, pH 7.5, 5 mM EDTA, 150 mM NaCl, 0.5% Nonidet P-40, 1.0 mM phenylmethylsulfonylfluoride (Sigma), and 1% aprotinin (Sigma). Total protein synthesis was measured in aliquots containing 50 µg of protein that were spotted on 24-mm GF/C disks (Whatman, Inc., Clifton, NJ), placed in 10% trichloroacetic acid, and heated to 90 C for 10 min, and the radioactivity determined by scintillation spectrometry. ß-Casein synthesis was measured by immunoprecipitation (see below). Acid-soluble radioactivity was measured to determine the uptake of [35S]Met into CID 9 cells. Protein extracts (50 µg) were precipitated on ice for 15 min with 10% trichloroacetic acid containing 40 µg/ml BSA as carrier. The radioactivity of the supernatant was measured after extraction with ethyl acetate.
Immunological Procedures
Cell lysates (250 µg of protein) were diluted to a total volume of 500 µl with the lysis buffer described above. The solution was preadsorbed by the addition of 25 µl protein A-agarose (Pierce Chemical Co., Rockford, IL; supplied as a 50% aqueous slurry) and 50 µl DMEM/F12 containing 10% fetal calf serum followed by incubation for 30 min at room temperature with rotation. After 5 min of centrifugation, the supernatant was transferred to another tube, specific antibodies were added, and the sample was incubated for 90 min at room temperature with rotation. Protein A-agarose (50 µl) was added, and the sample was kept on ice for 30 min. After centrifugation for 15 min, the pellet was washed twice in 50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 0.5% NP-40; 5% sucrose; and 5 mM EDTA; and once in 50 mM Tris-HCl, pH 7.4; 150 mM NaCl; and 5 mM EDTA. The bound proteins were eluted by boiling in sodium dodecyl sulfate sample buffer, resolved by SDS-PAGE on a 12% gel, and visualized by autoradiography. Western blotting (94) was performed after SDS-PAGE and transfer to Immobilon membranes (Millipore Corp., Bedford, MA). The primary antibodies are described above, and either peroxidase-labeled antimouse IgG or alkaline phosphatase-conjugated antirabbit IgG (Vector Laboratories, Inc., Burlingame, CA) were used as secondary antibodies. The specificity of these primary antibodies for their individual antigens was demonstrated in preliminary experiments.
Preparation of Polysomes
Isokinetic sucrose gradients were prepared using a device with a constant-volume mixing chamber (19, 20). CID 9 cells (0.6 x 106) were washed twice with RNase-free PBS containing 50 µg/ml cycloheximide and then lysed in 200 µl of 10 mM 3-(N-morpholino)propanesulfonic acid, pH 7.2; 250 mM NaCl; 2.5 mM Mg(OAc)2; 0.5% NP-40; 0.1 mM phenylmethylsulfonylfluoride; 200 µg/ml heparin; and 50 µg/ml cycloheximide. After centrifugation at 25,000 x g for 15 min, the supernatant was layered on a 12-ml 15–35% isokinetic sucrose gradient containing 25 mM HEPES, pH 7.0; 50 mM KCl; 2 mM Mg(OAc)2; 50 µg/ml cycloheximide; and 15 mM 2-mercaptoethanol; and centrifuged at 200,000 x g in a Beckman SW41Ti rotor for 1 h.
RNase H Digestion
Total RNA was isolated from CID 9 cells using Trizol (Life Technologies). Aliquots of 5 µg were incubated for 5 min at 65 C in a total volume of 10 µl with 10 pmol of specific oligodeoxynucleotide (5'-TTATGAGGCGGAGCACAGTT-3' for cleavage of ß-casein mRNA, 5'-AATGCCAAAGTTGTCATGGA-3' for GAPDH mRNA, and 5'-AGTCGGTGGGCCTCGGGGGC-3' for PAC mRNA), followed by cooling to room temperature over 10 min. In some experiments, 200 pmol dT16 were included. The oligodeoxynucleotide-annealed RNA was digested in a total volume of 20 µl with 2 U of RNase H (Promega Corp.) for 1 h at 37 C in 10 mM HEPES, pH 8.0; 100 mM KCl; 50 mM MgCl2; 0.5 mM dithiothreitol; and 25 µg/ml BSA followed by phenol extraction.
Northern Blotting
Northern blotting (94) was performed using riboprobes for the 3'-terminal portion of either ß-casein mRNA, made by in vitro transcription of Alw44 I-digested Clone 111, or GAPDH mRNA, made from DdeI-digested pTRI-GAPDH-mouse antisense control template. The riboprobe for detection of the ß-casein 3'-UTR expressed by plasmid pPUR-3'ß was made by in vitro transcription of XbaI-digested pPUR-3'ß. Plasmids were transcribed in a total volume of 20 µl in the presence of 5 µCi [-32P]GTP (ICN Biochemicals). The membrane was prehybridized for 2 h and then hybridized with the probe overnight at 65 C. A size marker of 665 nt was synthesized by in vitro transcription of EaeI-digested 6ßcM5'(+) (91). A size marker of 869 nt was synthesized by in vitro transcription of Alw44 I-digested 6ßcM5'(+) followed by digestion with RNase H in the presence of the ß-casein-specific oligodeoxynucleotide.
Real-Time PCR
Reverse transcription was performed on 40 ng RNA, isolated as described above, in a 20-µl reaction mixture containing 5.5 mM MgCl2, 500 µM of each deoxyribonucleoside triphosphate, 2.5 µM random hexamers, 0.2 U RNase Inhibitor (Applied Biosystems, Foster City, CA), and 0.8 U MultiScribe reverse transcriptase (Applied Biosystems). The reaction mixture was incubated at 25 C for 10 min, 48 C for 30 min, and 95 C for 5 min. Real-time PCR was performed on an ABI Prism 7700 (ABI Advanced Biotechnologies, Inc., Columbia, MD) in 50-µl reactions containing 5 µl of the reverse transcription reaction mixture (10 ng cDNA), 25 µl TaqMan PCR Universal Master Mix (Applied Biosystems), 400 nM primers, and 200 nM probe. The thermal cycle conditions consisted of 2 min at 50 C, 10 min at 95 C, and 40 cycles of 15 sec at 95 C, and 1 min at 60 C. For mouse ß-casein mRNA, the forward primer was 5'-GCAGGCAGAGGATGTGCTC-3', the reverse primer was 5'-GAGCATATGGAAAGGCCT-3', and the probe was 5'-FAM-AGGCTAAAGTTCACTCCAGCATCCAGTCACA-TAMRA-3'. For human 3-BSD mRNA, the forward primer was 5'-CGGCTAACGGGTGGAATCTG-3', the reverse primer was 5'-CCCCATAGATATACATGGGTCGTAAG-3', and the probe was 5'-FAM-ACGGCGGCACCCT-TAMRA-3'. Probes and primers for mouse rRNA, mouse GAPDH mRNA, and human GAPDH mRNA were purchased from Applied Biosystems.
Relative ß-casein and GAPDH mRNA levels were calculated as described in User Bulletin no. 2 for the ABI Prism 7700 Sequence Detection System. To control for variations in RNA yield in polysomal fractions, human RNAs were added to polysomal fractions before RNA isolation as internal standards. The levels of both mouse and human mRNAs were determined, and the latter were used to normalize the former. Each 250-µl polysomal fraction in which mouse ß-casein was to be quantitated received 1.25 µg total RNA from human K562 cells before extraction of RNA. Each 250-µl polysomal fraction in which mouse GAPDH was to be quantitated received 2 x 106 copies of 3-BSD mRNA, synthesized in vitro from XbaI-digested pGEM-3B-HSD1. The difference between the threshold cycles, CT, of the target mRNA and the internal standard was calculated (
CT). The CT of the first fraction (top of the polysome gradient) was subtracted from the CT of each other mRNA (CT). Relative mRNA levels were then calculated as 2CT. For measurement of ß-casein mRNA during differentiation of CID-9 cells, CT values for ß-casein mRNA and for rRNA were determined in the same PCR reaction mixture. The CT of rRNA was subtracted from the CT of ß-casein mRNA (CT) for each cell preparation. The CT for cells allowed to differentiate for 5 d on insulin alone was subtracted from the CT of all other cells to yield CT. A similar determination of changes in ß-casein mRNA levels during CID 9 cell differentiation was made using mouse GAPDH mRNA instead of rRNA for normalization.
Assay of PAP Activity
The reaction mixture (100 µl; Ref. 95) contained 50 mM Tris-HCl, pH 8.3; 7% glycerol; 0.3% polyvinyl alcohol; 0.05% acetylated BSA; 40 mM KCl; 0.5 mM MnCl2; 25 mM (NH4)2SO4; 0.25 mM dithiothreitol; 20 U RNasin (Promega); 250 µM ATP; and 1 ng poly(A) (Amersham Pharmacia Biotech, Arlington Heights, IL) labeled with [-32P]ATP by T4 polynucleotide kinase (Promega Corp.). CID 9 cells were lysed as described above and the lysates subjected to centrifugation in a Beckman Airfuge (Beckman Coulter, Inc., Fullerton, CA) with an A-100/18 rotor (34,000 x g). PAP reactions were initiated by addition of either 50 µg CID 9 cytoplasmic extract or 4 µl yeast PAP (500,000 U/ml, United States Biochemical Corp., Cleveland, OH). Reaction products were extracted with phenol/chloroform followed by precipitation with ethanol, using 1 µg of yeast RNA as carrier. The elongated poly(A) molecules were separated by TBE-PAGE on 6% gels containing 7 M urea (94).
Assay of PAC Activity
PAC activity was measured in an assay based on the differential distribution of acetyl-CoA and N-acetylpuromycin between organic and aqueous phases (96). CID 9 cells were lysed with 50 mM Tris-HCl, pH 8.5, 2 mM EDTA, 10% (vol/vol) glycerol, and 0.5% (vol/vol) NP-40. Reaction mixtures (50 µl) contained 0.1 M Tris-HCl, pH 8.0; 0.2 mM puromycin; and 1.0 mM [3H]acetyl-CoA (3.4 Ci/mmol). The reaction was initiated by adding cell extracts (50 µg of protein). Incubation was for 30 min at 30 C followed by addition of 200 µl of 5 M NaCl; 0.1 M borate, pH 9.0.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: 3-BSD, 3ß-Hydroxysteroid dehydrogenase; CPE, cytoplasmic polyadenylation element; CPEB, CPE binding protein; CPSF, polyadenylation specificity factor; DMSO, dimethylsulfoxide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GSK-3, glycogen synthase kinase-3; mTOR, mammalian target of rapamycin; nt, nucleotide; PAC, puromycin acetyltransferase; PABP, poly(A)-binding protein; PAP, poly(A) polymerase; PI3K, phosphatidylinositol 3-kinase; PKB, protein kinase B; RNase H, ribonuclease H; UTR, untranslated region; WAP, whey acidic protein.
Received for publication December 16, 2003. Accepted for publication March 31, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
for stimulation of protein synthesis by insulin. Mol Cell Biol 17:5184–5192[Abstract]
-subunit of eukaryotic initiation factor (eIF) 2B is insufficient for the activation of eIF2B by insulin. Biochem J 367:475–481[CrossRef][Medline]
and restores protein synthesis in vulnerable hippocampal neurons after transient brain ischemia. J Cereb Blood Flow Metab 19:1010–1019[CrossRef][Medline]
-CaMKII mRNA at synapses. Neuron 21:1129–1139[CrossRef][Medline]
CaMKII mRNA polyadenylation at synapses. EMBO J 21:2139–2148[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  C. Berlato and W. Doppler Selective Response to Insulin Versus Insulin-Like Growth Factor-I and -II and Up-Regulation of Insulin Receptor Splice Variant B in the Differentiated Mouse Mammary Epithelium Endocrinology, June 1, 2009; 150(6): 2924 - 2933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bier, I. Oviedo-Landaverde, J. Zhao, Y. Mamane, M. Kandouz, and G. Batist Connexin43 pseudogene in breast cancer cells offers a novel therapeutic target Mol. Cancer Ther., April 1, 2009; 8(4): 786 - 793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Cocolakis, M. Dai, L. Drevet, J. Ho, E. Haines, S. Ali, and J.-J. Lebrun Smad Signaling Antagonizes STAT5-mediated Gene Transcription and Mammary Epithelial Cell Differentiation J. Biol. Chem., January 18, 2008; 283(3): 1293 - 1307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Toerien and J. P. Cant Abundance and Phosphorylation State of Translation Initiation Factors in Mammary Glands of Lactating and Nonlactating Dairy Cows J Dairy Sci, June 1, 2007; 90(6): 2726 - 2734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. V. Loladze, M. A. Stull, A. M. Rowzee, J. DeMarco, J. H. Lantry III, C. J. Rosen, D. LeRoith, K.-U. Wagner, L. Hennighausen, and T. L. Wood Epithelial-Specific and Stage-Specific Functions of Insulin-Like Growth Factor-I during Postnatal Mammary Development Endocrinology, November 1, 2006; 147(11): 5412 - 5423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Yaras, M. Ugur, S. Ozdemir, H. Gurdal, N. Purali, A. Lacampagne, G. Vassort, and B. Turan Effects of Diabetes on Ryanodine Receptor Ca Release Channel (RyR2) and Ca2+ Homeostasis in Rat Heart Diabetes, November 1, 2005; 54(11): 3082 - 3088. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
