A splicing factor phosphorylated by protein kinase A is increased in HL60 cells treated with retinoic acid

Retinoic acid (RA) induces the differentiation of human promyelocytic leukemia HL60 cells into granulocytic cells and inhibits proliferation. Certain of actions of RA are mediated by RA nuclear receptors that regulate gene expression. However, it is also known that direct protein modification by RA (retinoylation) can occur. One such retinoylated protein in HL60 cells is a regulatory subunit of protein kinase A (PKA), which is increased in the nucleus following RA treatment and which then increases phosphorylation of other nuclear proteins. However, a complete understanding of which nuclear proteins are phosphorylated is lacking. In the current study, we employed mass spectrometry to identify one of the PKA-phosphorylated proteins as a serine/arginine-rich splicing factor 1 (SF2, SRSF1). We found that RA treatment increased the level of PKA-phosphorylated SF2 but decreased the level of SF2. While SF2 regulates myelogenous cell leukemia-1 (Mcl-1, anti-apoptotic factor), RA treatment reduced the level of Mcl-1L (full-length Mcl-1 long) and increased the level of Mcl-1S (Mcl-1 short; a short splicing variant of the Mcl-1). Furthermore, treatment with a PKA inhibitor reversed these effects on Mcl-1 and inhibited RA-induced cell differentiation. In contrast, treatment with a Mcl-1L inhibitor enhanced RA-induced cell differentiation. These results indicate that RA activates PKA in the nucleus, increases phosphorylation of SF2, raises levels of Mcl-1S and lowers levels of Mcl-1L, resulting in the induction of differentiation. RA-modified PKA may play an important role in inducing cell differentiation and suppressing cell proliferation.     

Protein kinase A activation by retinoic acid in the nuclei of HL60 cells

BackgroundActivation of protein kinase A (PKA) occurs during retinoic acid (RA)-induced granulocytic differentiation of human promyelocytic leukemia HL60 cells. It is known that the RIIα regulatory subunit of PKA, is modified by RA (retinoylated) in the early stages of differentiation. We have investigated the effects of RA on PKA during cell differentiation in order to understand the potential significance of this process in the retinoylation of RIIα subunits.MethodsImmunoblotting, immunoprecipitation, confocal microscopy, PCR, and PKA activity assays were employed for characterizing the effects of RA on PKA.ResultsWe found that RA induces intracellular mobility of RIIα and the activation of PKA in HL60 cells. Increases in RIIα levels were observed in RA-treated HL60 cells. RA treatment altered intracellular localization of the PKA subunits, RIIα and Cα, and increased their protein levels in the nuclei as detected by both immunoblotting and immunostaining analyses. Coincident with the increase in nuclear Cα, RA-treated HL60 cells showed increases in both the protein phosphorylation activity of PKA and the levels of phosphorylated proteins in nuclear fractions as compared to control cells. In addition, RIIα protein was stabilized in RA-treated HL60 cells as compared to control cells.ConclusionsThese results suggest that RA stabilizes RIIα protein and activates PKA in the nucleus, with a resultant increase in the phosphorylation of nuclear proteins.General significanceOur evidence suggests that retinoylation of PKA might contribute to its stabilization and activation and that this could potentially participate in RA's ability to induce granulocytic differentiation of HL60 cells.     

Control of retinoic acid receptor expression in mouse melanoma cells by cyclic AMP

Retinoic acid receptor (RAR) α and γ mRNAs were constitutively expressed in B16 melanoma cells with or without retinoic acid (RA) treatment. RARβ mRNA, however, was significantly expressed only after exposure to RA. Induction of RARβ by RA occurred within 1 h and was not inhibited by cycloheximide (i.e., did not require new protein synthesis). All three RAR mRNA levels were dramatically decreased with 8-bromo-cyclic AMP treatment and could not be rescued by addition of RA. Analysis of RARγ revealed that this decrease occurred within 1 h of exposure to 8-bromo-cyclic AMP and was not blocked by simultaneous treatment with cycloheximide. The stability of RARγ mRNA was not altered by cyclic AMP treatment. Nuclear extracts from 8-bromo-cyclic AMP-treated cells showed a large decrease in protein binding to a retinoic acid response element (RARE) oligonucleotide compared to control cells. This correlated with a marked reduction of RA-stimulated RARE-reporter gene activity in transfected cells which were treated with cyclic AMP. Pretreatment of B16 cells with cyclic AMP prior to RA addition dramatically reduced induction of PKCα, an early marker of RA-induced cell differentiation. Thus, cyclic AMP can antagonize the action of RA most likely via its ability to inhibit RAR expression. © 1996 Wiley-Liss, Inc.     

Transglutaminase activity increases in HL60 cells induced to differentiate with retinoic acid and TPA but not with DMSO

HL60 cells induced to differentiate into myeloid cells by retinoic acid exhibited a 300-fold increase in transglutaminase (TGase) activity which peaked on day 5. HL60 cells induced to differentiate into monocytes by a phorbol ester tetradecanoylphorbol-12-myristate-13-acetate (TPA) had a greater than 840-fold increase in TGase activity on day 7. In contrast, cells induced to differentiate along the myeloid pathway by dimethyl sulfoxide (DMSO) exhibited no increase in TGase activity. Elevation of TGase activity appears to be characteristic of monocyte differentiation and retinoic acid-induced myeloid differentiation but not of myeloid differentiation in response to DMSO.     

Calcium-activated, phospholipid-dependent protein kinase activity and protein phosphorylation in HL60 cells induced to differentiate by retinoic acid

Treatment of human promyelocytic (HL60) cells with retinoic acid for at least 48 h causes differentiation to more mature myeloid forms. Prior to commitment of cells to the myeloid pathway there is a marked increase in cytosolic calcium-activated, phospholipid-dependent protein kinase activity. This increase does not result from an intracellular redistribution of the enzyme. Concomitant with the increased enzyme activity there is enhanced phospholipid-dependent phosphorylation of proteins of 29, 49, 52, 58, 68, 69, 120, 170, 200 and 245 kDa.     

Effects of tunicamycin on the differentiation of F9 cells induced by either retinoic acid or retinoic acid and dibutyryl cyclic AMP

Tunicamycin (0.5 micrograms/ml) inhibited differentiation of F9 cells treated either with retinoic acid or with retinoic acid and dibutyryl cyclic AMP, as monitored by the activity of alkaline phosphatase and expression of cytokeratins. On the other hand, the pattern of the polysaccharide chain synthesis changed drastically with the treatment irrespective of the presence of tunicamycin. Therefore, phenotypes induced with retinoic acid are dissociated into two categories, one that is directly induced by the drug and the other that is induced indirectly by a mechanism in which glycoproteins play a role.     

Proteins in human myeloid leukemia cell line HL60 reacting with retinoic acid monoclonal antibodies

The vitamin A derivative, retinoic acid (RA) has various biological effects in mammalian cells and tissues. It is well known that RA induces differentiation of leukemia cells and inhibits cell growth. There are two pathways for RA action; one via RA nuclear receptors (RARs), and one via acylation of proteins by RA (retinoylation). However, an understanding of which actions of RA occur via RARs and which occur via retinoylation is lacking. Thus, we undertook the examination of HL60 proteins using anti-RA monoclonal antibodies (ARMAs). These ARMAs showed specific binding to proteins in a saturable manner depending on protein and antibody concentration. Proteins eluted by Mono Q anion exchange chromatography and separated using two-dimensional polyacrylamide gel electrophoresis were detected by ARMAs. One of these ARMA-bound proteins in HL60 cells was identified as alpha-actinin. These results indicate that retinoylated proteins in HL60 cells can be recognized by ARMAs and that alpha-actinin modified by RA may play a significant role in RA-induced differentiation, including the promotion of cytomorphology changes.     

In vivo and in vitro induction of 'tissue' transglutaminase in rat hepatocytes by retinoic acid.

Tissue transglutaminase (tTG) expression was found to be induced in rat liver following in vivo retinoic acid (RA) treatment (Piacentini et al. (1988) Biochem. J. 253, 33-38). Here we show that the increased enzyme expression in rat liver is at least partially the result of the action of RA in parenchymal cells. In fact, (a) when hepatocytes are isolated from RA-treated animals their transglutaminase protein content is much higher than in similarly isolated control cells; (b) higher tTG protein level is also found by immunoelectronmicroscopy in the hepatocytes of the RA-treated rats as compared with the very low amount detected in the controls; (c) RA induces tTG in hepatocytes under culture conditions as well. One of the functions of tTG is to form a protein polymer in dying apoptotic cells by epsilon(gamma-glutamyl)lysine and, specifically gamma-glutamylpolyamine cross-links (Fesus et al. (1989) FEBS Lett. 245, 150-154). Noteworthy, after in vivo and in vitro RA-treatment we could not determine any increase (there was even a slight decrease) in the number of the cross-linked apoptotic envelopes. In keeping with this is the significant reduction of protein bound gamma-glutamylpolyamine detected in hepatocytes exposed to RA in culture. These findings suggest that the RA-induced tTG in parenchimal cells is an inactive form.     

Correlation between transglutaminase activity and polyamine levels in human neuroblastoma cells. Effect of retinoic acid and alpha-difluoromethylornithine

The human neuroblastoma cell line SK-N-BE can be induced to differentiate by retinoic acid (RA) or by alpha-difluoromethylornithine (DFMO). The former inducer produces neurite outgrowth, 60% reduction of growth rate, overexpression of neural antigens, and enhanced gamma-aminobutyric acid (GABA) and acetylcholinesterase levels. In contrast, DFMO causes cell body elongation, complete growth inhibition, and higher binding of antibodies directed against neuroectodermal antigens. Polyamine metabolism is also differently affected by the two agents. In particular a large spermine catabolism is induced by RA, while DFMO treatment leads to a small increase in the level of this compound. The neural differentiation induced by RA is accompanied by a marked increase in transglutaminase activity and its induction is paralleled by a transient increase of putrescine and spermidine. The putrescine and spermidine depletion determined by DFMO is accompanied instead by a large inhibition of transglutaminase activity. The inhibiting effect of DFMO treatment on transglutaminase is reversed by the addition of 1 mM putrescine to the culture medium. In the presence of both RA and DFMO a mixed morphological and biochemical pattern is observed. The possibility that the expression of transglutaminase associated to cellular differentiation may be modulated by the level of its substrates is also discussed.     

Rapid induction of alkaline phosphatase activity by retinoic acid

Retinoic acid (vitamin A acid) increased alkaline phosphatase activity in cultured cells derived from both normal rat prostate and the Dunning R-3327 transplantable prostatic adenocarcinoma. Retinoic acid was found to be 3–4-fold more effective as an inducer of enzyme activity than retinol or retinal. In one rapidly-growing cell line (UMS-1541Q) which has a barely-detectable level of enzyme activity in the uninduced state, increased activity could be detected as early as 3–4 hours after the addition of 10μM retinoic acid. This increase was totally blocked by actinomycin D and cycloheximide. The demonstrated rapid inducibility of alkaline phosphatase activity provides a specific marker for the action of retinoic acid at the molecular level.     

Cyclic AMP potentiates the retinoic acid-induced expression of tissue transglutaminase in peritoneal macrophages

Fresh serum and retinoids induce the expression of tissue transglutaminase in cultured mouse resident peritoneal macrophages. Analogues of cyclic AMP, such as dibutyryl cyclic AMP, and agents that increase intracellular cyclic AMP levels enhance the induction. Dibutyryl cyclic AMP alone has little effect on transglutaminase expression, but it increases the sensitivity of macrophages to low concentrations of either serum or retinoic acid. Dibutyryl cyclic AMP potentiates the transglutaminase-inducing activity of both free retinoic acid and retinoic acid bound to the serum retinol-binding protein. Pretreating macrophages with dibutyryl cyclic AMP or retinoic acid does not prime the cells to respond to the other agent; instead, both agents must be present simultaneously to obtain the synergistic induction of transglutaminase. Our studies suggest that the modulation of intracellular cyclic AMP levels may have pronounced effects on retinoic acid-induced gene expression in myeloid cells.     

Physiological cyclic AMP stimulation and oncogene mRNA levels in HL-60 cells

Altered gene expression of the proto-oncogenes c-fos and c-myc is associated with differentiation of the human promyelocytic leukemia (HL-60) cells. We studied changes of cyclic AMP levels and c-fos and c-myc mRNA levels after stimulation with theophylline and theophylline plus isoproterenol. Reduced c-fos and c-myc mRNA levels were detected, but the reduction could not, however, be related to the observed alternations in cyclic AMP concentrations. Expression of c-jun and c-Ha-ras was not affected under these conditions.     

Dual regulation of arachidonic acid release by P2U purinergic receptors in dibutyryl cyclic AMP-differentiated HL60 cells.

ATP promoted biphasic effects on both basal and fMLP-stimulated arachidonic acid (AA) release in neutrophil-like HL60 cells: stimulation in the micromolar range (EC50 = 3.2 +/- 0.9 microM) and inhibition at higher concentrations (EC50 = 90 +/- 11 microM). ATP also inhibited UTP- and platelet activating factor-stimulated AA release. Only stimulatory effects of ATP on basal or fMLP-stimulated phospholipase C were observed. The inhibitory effect of ATP on AA release was not due to reacylation of released AA, chelation of extracellular Ca2+, cell permeabilization, or changes in the rise of [Ca2+]i induced by agonist. The inhibition was rapid, being detected within 5-15 s. The inhibitory effect of ATP on fMLP-stimulated AA release could be desensitized by pretreatment of the cells with 2 mM ATP, but not 20 microM ATP, the concentration that resulted in maximal release of AA and inositol phosphates. The inhibition by ATP was neither dependent on generation of adenosine by ATP hydrolysis nor the result of direct interaction of ATP with P1 purinergic receptors. Among other nucleotides tested (CTP, GTP, ITP, TTP, XTP, adenosine 5'-(beta,gamma-methylene)triphosphate (AMP-PCP), adenyl-5'-yl imidodiphosphate (AMP-P(NH)P), ADP, adenosine 5'-O-(3-thiotriphosphate) (ATP gamma S), and UTP), only UTP and ATP gamma S displayed biphasic effects with potencies and efficacies almost identical to those of ATP. The other nucleotides only exhibited stimulatory effects (EC50 = 60-300 microM). The results are consistent with a model of dual regulation of AA release by two distinct subtypes of P2U receptors in HL60 cells.     

Inhibition of chondrogenesis by retinoic acid in limb mesenchymal cells in vitro: Effects on PGE2 and cyclic AMP concentrations

Effects of retinoic acid (RA) on prostaglandin E₂ (PGE₂) and cyclic AMP (cAMP) concentrations were investigated in high density, micromass cultures of mesenchymal cells derived from chick limb buds. Exposure of cells during the initial 24 h of culture to RA concentrations between 0.05–1.0 μg/ml inhibited chondrogenesis in a dose-dependent manner with 1.0 μg/ml totally inhibiting cartilage formation. Concentrations of PGE₂ and cAMP increased during the prechondrogenic period in control cells in a closely related way and remained elevated throughout the six-day period examined. Addition of RA (0.05 and 0.5 μg/ml) did not significantly alter cAMP concentrations at any time point, but significantly elevated PGE₂ levels relative to control cells in six-day cultures in a concentration-dependent manner. Addition of dibutyryl cAMP enhanced chondrogenesis in control cells between days 3 and 4, but failed to alter the inhibitory effect of RA on chondrogenesis. The results indicate that while PGE₂ and cAMP are important signals in cartilage differentiation, the inhibitory effects of RA on this process are mediated through some other mechanism.