Molecular Genetics and Metabolism 68, 56 – 67 (1999) Article ID mgme.1999.2889, available online at http://www.idealibrary.com on
Determination of Queuosine Modification System Deficiencies in Cultured Human Cells Rana C. Morris, Marissa C. Galicia, Kari L. Clase, and Mark S. Elliott 1 Department of Chemistry and Biochemistry, Old Dominion University, Norfolk, Virginia 23529-0126 Received June 11, 1999, and in revised form June 24, 1999
Key Words: queuine; queuosine; adenocarcinomas; cell culture; tRNA modifications; assessment.
Queuosine-deficient tRNAs are often observed in neoplastic cells. In order to determine possible sites for malfunction of the multistep queuosine modification system, comprehensive studies were performed on two human neoplastic cell lines, the HxGC 3 colon adenocarcinoma and the MCF-7 breast adenocarcinoma, which are 100 and 50 – 60% queuosine deficient, respectively. These results were compared with data obtained from normal human fibroblast (HFF) cultures which maintain 100% queuosine-modified tRNA populations. Queuine uptake in all three cell types was similar and each demonstrated activation by protein kinase C (PKC). However, incorporation of queuine into tRNA by tRNA:guanine ribosyltransferase (TGRase; E.C. 2.4.2.24) and PKC-catalyzed activation of this enzyme occurred only in HFF and MCF-7 cells. The HxGC 3 cell line exhibited no TGRase activity as was expected. Treatment with 5-azacytidine (5-azaC) induced TGRase activity to a level 20% of that in HFF and MCF-7 cells; however, this 5-azaCinduced TGRase activity was not regulated by PKC. Salvage of the queuine base from tRNA degradation products has been shown in mammalian cells and was measured in the HFF cells. However, salvage activity in the MCF-7 cell line was deficient. Therefore, it was shown by direct measurements that the HxGC 3 cell line is completely lacking in queuosine-modified tRNA due to loss of functional TGRase, while the MCF-7 cell line has an inefficient queuine salvage mechanism resulting in a significant deficiency of queuosine-modified tRNA. These techniques can be applied to any cultured cell types to determine specific lesions of the queuosine modification system, which have been suggested to be associated with neoplastic progression. © 1999 Academic Press
Transfer RNA (tRNA) anticodons are posttranscriptionally modified to contain queuosine in the wobble position of tRNA asp, tRNA asn, tRNA his, and tRNA tyr (1). This is the case in normal differentiated human cells; however, neoplastic cells derived from biopsies of human tumors exhibit deficiencies in the level of queuosine-modified tRNAs (2). The degree of hypomodification has been shown to be related to the metastatic potential of colon (3), ovarian (4), brain (5), lung (6), leukemia, and lymphoma (7) tumor biopsies. The degree of hypomodification also varies in cultured cell lines (2,3), with deficiencies as much as 50 to 60% in the MCF-7 breast adenocarcinoma cultured cell line and up to 100% in the HxGC 3 colon adenocarcinoma cell line. While diagnostic procedures have been developed based on queuosine deficiencies, it is still not known if the breakdown of a single component or multiple steps of the queuosine modification system is the cause of the wide variability in deficiencies. The mammalian queuosine modification system is significantly different from the well-known prokaryotic system which involves multistep intracellular synthesis of queuosine directly on a tRNA scaffold (8). Mammalian cells obtain queuine from diet or from secretions of intestinal flora in intact organisms or from serum supplements to growth media in the case of cultured cells (9). The queuosine modification system in mammalian cells is composed of three steps: uptake of the queuine base into the cells by a queuine-specific membrane transporter, incorporation of the queuine base into the wobble position
1
To whom correspondence should be addressed. Fax: (757) 683-4628. E-mail:
[email protected]. 56 1096-7192/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
QUEUOSINE IN CULTURED CELLS
of tRNA by tRNA:guanine ribosyltransferase (TGRase; EC 2.4.2.24), and salvage of the queuine base from queuosine 59-monophosphate resulting from tRNA degradation for reuse by the TGRase enzyme. These three steps are all potential points of deficiency in the maintenance of cellular queuosinemodified tRNA levels. In addition, it was determined that protein kinase C and protein phosphatase I modulate the rates of both queuine uptake into the cell and its subsequent incorporation into the anticodon of tRNAs (10,11). Therefore, aberrant kinase and/or phosphatase activity may also affect queuosine-modified tRNA levels in the cell. PKC and protein phosphatase modulation of the queuine salvage mechanism does not occur (data not shown). The studies presented here attempt to identify the malfunctioning sites of the queuosine modification system in two human neoplastic cell lines, the human colon adenocarcinoma (HxGC 3), a TGRase-deficient control, and previously uncharacterized human breast adenocarcinoma (MCF-7) cell lines, in comparison to fully characterized queuosine-containing normal human fibroblast cultures as positive controls. Recent descriptions of the queuosine modification system in eukaryotes indicate that the loss of the queuosine modification might come from the malfunction of one or a number of different steps, such as the loss of functional incorporation enzyme, of activation of this enzyme or the queuine membrane transporter by protein kinase C, or the ability to salvage the queuine base from cytosolic tRNA turnover products. Using an indirect method involving RPC-5 liquid chromatography of HxGC 3 tRNA, this cell line was determined to be completely deficient in queuosine-modified tRNA (3). The authors concluded that this was due to a deficiency in the level of TGRase, the enzyme responsible for incorporating queuine into the anticodon of tRNA. TGRase activity was suggested to be marginally inducible in these cells with exposure to the transcriptional activator 5-azacytidine based on the appearance of queuosine modification in the tRNA samples as analyzed by RPC-5 chromatography. However, it was not specifically determined if the original deficiency of the queuosine modification was due to the loss of TGRase activity because of a decrease in the protein’s synthesis, because of the lack of protein kinase C modulation, or because of the lack of salvage of the queuine base to maintain substrate availability. Methods for the direct assay of queuosine modification system activities have been developed using a
57
radiolabeled analogue of the queuine base, rQT 3 (12). In this paper, we describe the results of direct analysis of queuine uptake, incorporation, salvage, and protein kinase/phosphatase modulation assays performed on cultured colon and breast adenocarcinoma cell lines. Human fibroblast cell cultures derived from neonatal foreskin (HFF) were assayed as a “normal” reference model system (10,11,13–15). The HxGC 3 cell line, known to be completely deficient in queuosine, was studied as a proposed TGRase-deficient control. The MCF-7 cell-line is 50 to 60% deficient in the queuosine modification, but has not yet been studied with regard to the lesion(s) in this cell-line’s queuosine modification system. Comparisons between the results of the fibroblast cultures and those of the neoplastic cell lines were made to determine the sites and relative levels of deficiency within the multicomponent queuosine modification system. Utilization of this set of experimental techniques can allow for dissection of the queuosine modification system in any cultured cell type. This will allow for more fully characterized relationships between queuosine levels and diagnoses of cultured cells from tumor explants with regard to neoplastic tendency and metastatic potential. METHODS AND MATERIALS Materials The radiolabeled reduced analog of queuine, tritiated dihydroqueuine (rQT 3), used in these studies was a gift from Dr. Ronald W. Trewyn (Kansas State University). The modified base queuine was isolated from third trimester bovine amniotic fluid (16) and then subjected to reductive trititiation by Amersham Chemical Company. The addition of rQT 3 to cell cultures was employed to monitor cellular uptake of this base, its incorporation into tRNA, or salvage of the incorporated base during tRNA turnover. Two lots of lyophilized rQT 3 were dissolved in water to yield stock concentrations of 1 mM with specific activities of 3 mCi/pmol and stored at 220°C. Protein kinase and protein phosphatase modulators were received as lyophilized powders and stock solutions were made by dissolving each in acetone at 1000 times the working concentration. The protein kinase C activator 12-tetradecanoyl phorbol-13-acetate (TPA) (Sigma Chemical Company, St. Louis, MO) was used at a concentration of 20 nM. The PKC inhibitor staurosporine and the protein phosphatase
58
MORRIS ET AL.
inhibitor okadaic acid (Biomol Research Laboratories, Plymouth Meeting, PA) were used at 100 and 10 nM, respectively. The DNA methylase inhibitor, 5-azacytidine (5-azaC), employed in the generation of the 5-azaC-treated HxGC 3 cultures was purchased from Sigma Chemical Company and used at a working concentration of 5 mM. All culture work was performed in sterile polystyrene plasticware from Corning (Corning, NY).
ysis were conducted at 37°C for up to 6 h and terminated by a rinse of the cell monolayer four times with 5 mL of ice-cold phosphate-buffered saline. Then, the cells were lysed with 1.0 mL of 95% ethanol for 5 min. The lysate was aspirated and radioactivity determined by liquid scintillation. The amount of radioactivity in the cell lysate, normalized to the number of picomoles of rQT 3 per 10 5 cells, was representative of rQT 3 uptake into the cultured cells.
Cell Culture The HFF cells were established from circumcised neonatal foreskin by established methods (14). The HxGC 3 human colon adenocarcinoma cell line was obtained as a generous gift from Dr. Jon R. Katze (University of Tennessee, Memphis), and the MCF-7 human breast adenocarcinoma cell line was obtained from the American Type Culture Collection (Bethesda, MD). Cell cultures were established and maintained in either 10% neonatal calf serum-supplemented minimum essential media (HFF and HxGC 3 cell types) or 10% neonatal calf serum-supplemented Delbecco’s modified eagle media (MCF-7 cell line) as previously described (17). 5-AzaCtreated HxGC 3 cultures were exposed to 5 mM 5-azaC for 24 h and then incubated with minimal essential media containing 10% neonatal calf serum until confluence. Queuine-deficient cultures for the incorporation and salvage studies were generated by growing the cells in media supplemented with 10% charcoal-stripped queuine-free calf serum (14), for up to 3 weeks before the studies were initiated by treatment with rQT 3. rQT 3 Uptake Assays Cells were subcultured into 35-mm dishes at a density of 4 3 10 4 cells/mL in a final volume of 2 mL of media containing 10% serum. The cultures were incubated until they were a confluent monolayer with a final density of approximately 10 5 cells/cm 2. When required, 5-azaC treatment of HxGC 3 cultures was performed at this point. At confluence, the media was decanted from the cells and 1 mL of media supplemented with 10% calf serum and 100 nM rQT 3 (0.10 mCi) was added. In phosphorylation modulation studies, uptake of rQT 3 was compared between untreated control cultures and with cultures exposed to PKC and protein phosphatase-modulating agents at concentrations listed above with the modulating agents added to the culture at the same time as the rQT 3. Incubations for rQT 3 uptake anal-
rQT 3 Incorporation Assays Cells cultures were grown in media supplemented with 10% charcoal-stripped queuine-free calf serum for three passages before the start of the incorporation experiments to ensure that the cellular tRNAs were completely unmodified with respect to queuosine. Cells were then subcultured into 35-mm dishes at a density of 4 3 10 4 cells/cm 3 in a final volume of 2 mL of media containing 10% charcoalstripped neonatal calf serum. When required for induction of TGRase activity, 5-azaC treatment of HxGC 3 cultures at 5 mM for 24 h was performed at this point. At confluence, the medium was decanted from the cells and 1 mL of media supplemented with 10% charcoal-stripped neonatal calf serum and 100 nM rQT 3 (0.10 mCi) was added. Incorporation of rQT 3 into the acid-precipitable fraction (tRNA) of various control cultures and those treated with PKC and protein phosphatase-modulating agents were compared in phosphorylation modulation studies. PKC and protein phosphatase modulators were added to the cultures at concentrations listed above and at the same time as the rQT 3. Incubations for rQT 3 incorporation analyses were conducted at 37°C for timed intervals of up to 24 h. Incubations were terminated by a rinse of the cell monolayer four times with 5 mL of ice-cold phosphate-buffered saline, followed by disruption of the cells with 0.5 mL of lysis buffer (10 mM Tris (pH 7.5), 0.01% SDS, 0.01% Triton X-100) for 10 min at room temperature. The lysate was transferred to a small test tube, treated with 0.25 mL of ice-cold 30% trichloroacetic acid (TCA), and placed on ice for 10 min. The resulting precipitate was collected by vacuum filtration through GFA 2.4-cm glass fiber filter disks. Each disk was thoroughly rinsed with 40 mL of ice-cold 5% TCA and a final rinse of 5 mL ice-cold 95% ethanol. The filters were analyzed for the presence of bound radioactive rQT 3-modified tRNA by liquid scintillation. The amount of radioactivity on the fil-
QUEUOSINE IN CULTURED CELLS
ter disks, normalized to number of picomoles of rQT 3 per 10 5 cells, was reflective of rQT 3 incorporation into the acid-precipitable fraction (tRNA) from cultured cells. rQT 3 Salvage Fibroblasts and MCF-7 cells were subcultured into 25-mL culture flasks at a density of 1 3 10 4 cells/cm 3 in a final volume of 5 mL of media containing 10% calf serum. When the cells reached confluence, the medium was decanted and 5 mL of media supplemented with 10% charcoal-stripped queuinefree calf serum was added for 1 day, 1 week, or 3 weeks. Exposure of cells to media containing no queuine was used to induce the queuine salvage mechanism. After the pretreatment with charcoalstripped serum, 100 nM rQT 3 (0.10 mCi) was added to the cultures and incubated for 24 h to fully saturate the tRNAs with the radiolabeled queuine. Then, the cell monolayer was washed four times with sterile phosphate-buffered saline to remove unincorporated rQT 3 and incubated in either media containing 10% serum supplemented with 0.10 A 260 units unlabeled queuine (in order to measure the baseline tRNA turnover rate) or media supplemented with 10% charcoal-stripped serum (in order to measure the retention of the radiolabeled or salvage of rQT 3). Incubations for rQT 3 salvage analysis were conducted at 37°C at 24-h intervals for up to 4 days. Incubations were terminated by rinsing, lysing, and precipitating the acid-insoluble tRNAs as in the incorporation assays. Filter-bound tRNA was analyzed for radioactivity by liquid scintillation. The level of radioactivity on the filter disks was representative of the amount of remaining rQT 3 in the acid-precipitable fraction (tRNA) from cultured cells over time, and is a direct measure of cellular salvage capability. RESULTS Uptake of Queuine and Modulation by Phosphorylation The HxGC 3 and MCF-7 cell lines exhibit rQT 3 uptake rates comparable to those observed in normal HFF cells (Fig. 1). Both neoplastic cultures demonstrate a linear uptake of the radiolabeled queuine analogue through 60 min and a plateau at approximately 120 min. In addition, the maximal quantities of rQT 3 taken up by the three cell types are all very close in range (7.4 to 7.8 pmol/10 5 cells). The 5-azaC-
59
treated HxGC 3 cells were assayed for the ability to transport rQT 3 and were shown to exhibit the same uptake profile as the untreated HxGC 3 cell cultures. This culture set was studied to ensure that any difference of queuine incorporation into tRNA observed in this cell line was due solely to the incorporation step and not due to uptake insufficiencies. In the rQT 3 uptake studies, both neoplastic cell lines and the “normal” HFF cells exhibited sensitivity to modulators of PKC and protein phosphatase activity (Fig. 2). Treatment with the PKC activator, TPA, or treatment with the protein phosphatase inhibitor, okadaic acid, both resulted in increased cellular uptake of rQT 3 for all cells studied. The PKC inhibitor, staurosporine, resulted in a decrease in rQT 3 uptake for all cells studied. Pretreatment of HxGC 3 cells with 5-azaC did not effect uptake modulation by these agents. These results indicate that the queuine uptake mechanism is activated by phosphorylation in both HxGC 3 and MCF-7 cell lines, and to a level comparable with normal fibroblast cultures. Incorporation of Queuine into tRNA and Modulation by Phosphorylation Despite the proper functioning of the queuine uptake step, direct time-course incorporation assays show no incorporation of rQT 3 into HxGC 3 tRNA even after 24 h of exposure to this queuine analogue (Fig. 3). Pretreatment of these cells with the DNA methylase inhibitor and transcriptional activator, 5-azaC, does not effect the uptake of rQT 3. However, incorporation of rQT 3 into tRNA in 5-azaC-treated cells is distinguishable above background levels as early as 6 h after exposure to the rQT 3. The level of rQT 3-modified tRNA elevates to 4.5-fold background levels at 12 h and continues to rise through 24 h. This pattern is in contrast to the HFF cell line’s incorporation profile which is linear through 9 h and plateaus at 12 h. Although the level of queuosine modification in the HxGC 3 cultures is 5-fold lower than the incorporation levels in the HFF cells at 24 h, it remains induced significantly above that of background. In contrast to the HxGC 3 cell-line, MCF-7 cultures exhibited no deficiency in queuine incorporation activity. Incorporation of rQT 3 in MCF-7 cells is linear to 6 h and plateaus at 12 hours. This is very similar to the HFF culture incorporation assay. Maximal incorporation levels of rQT 3 in HFF and MCF-7 cells (2.48 pmol/10 5 cells) are also identical (Fig. 3).
60
MORRIS ET AL.
FIG. 1. Time-courses of queuine uptake in human foreskin fibroblast (HFF; ✚), colon adenocarcinoma (HxGC 3; ), 5-azacytidinetreated HxGC 3 (Œ), and breast adenocarcinoma (MCF-7; F) cell types. These cell types were grown to near confluence, then 100 nM rQT 3 (a tri-tritiated queuine analogue) was added at Time 0, and uptake assays were performed. The amount of radioactivity taken up into the cells was measured at 0, 0.5, 1, 1.5, 2, 2.5, 3, and 6 h. The values were normalized to the number of picomoles of rQT 3 per 10 5 cells. Deviation bars indicated represent the standard deviation with n 5 6.
Exposure to phosphorylation modulators produced similar results for the rQT 3 incorporation MCF-7 and HFF cell types, indicating that phosphorylation is important for maintaining elevated queuine incorporation rates in these cells (Fig. 4). However, in HxGC 3 cells even after 5-azaC induction of TGRase, the incorporation rate of rQT 3 is not effected by exposure to the protein kinase C or phosphatase modulators. Phosphorylation is not a factor in control of the inducible TGRase activity in the
HxGC 3 cell line. The mammalian TGRase enzyme is a heterodimer made up of a large regulatory subunit that is a target for protein kinase C-catalyzed phosphorylation and a smaller catalytic subunit (8). It is proposed that the absence of regulation by phosphorylation is due to the 5-azaC-induced synthesis of only the smaller catalytic subunit and not the larger regulatory subunit. This could explain the observation of a low baseline activity that is incapable of physiological activation.
QUEUOSINE IN CULTURED CELLS
61
FIG. 2. The effect of exposure to a protein kinase C activator (TPA), inhibitor (staurosporine), and phosphatase inhibitor (okadaic acid) as compared to untreated cultures on queuine uptake in human foreskin fibroblast (HFF; ✚), colon adenocarcinoma (HxGC 3; ), 5-azacytidine-treated HxGC 3 (Œ), and breast adenocarcinoma (MCF-7; F) cell types. These cell types were grown as under Materials and Methods. When the cells reached near confluence, 100 nM rQT 3 (a tri-tritiated queuine analogue) was added, and uptake assays were initiated. The uptake assays were stopped after 3 h and the cytosolic fraction was isolated and counted by liquid scintillation. The values were normalized to the number of picomoles of rQT 3 per 10 5 cells. Deviation bars indicate the standard deviation with n 5 6.
Queuine Salvage The initial rQT 3 incorporation values for HFF and MCF-7 cell cultures are comparatively close for the salvage studies (Fig. 5). The HFF cells are able to retain nearly 100% of incorporated rQT 3 for up to 4 days after pulse labeling with rQT 3 following either 1 day or 1 week treatment with charcoal-stripped queuine-free serum in growth media. Only a small decline in rQT 3 retention was observed after 3 weeks of exposure to the queuine-free serum indicating an active queuine salvage mechanism. In contrast to
queuine salvage in the “normal” cells, the MCF-7 cell line exhibits virtually no latent salvage capability. After only 1 day of queuine starvation, these cells show a similar pattern of rQT 3-modified tRNA retention between Q-chased and control cultures indicating normal tRNA turnover with no salvage of the queuine substrate from the tRNA degradation product queuosine 59-monophosphate. After 1 week of queuine starvation only a 50% salvage efficiency of rQT 3 is evident as compared with the 100% level of salvage induction in HFF cells. Salvage of queuine in MCF-7 cells was barely measurable above base-
62
MORRIS ET AL.
FIG. 3. Time-courses of queuine incorporation into tRNA of human foreskin fibroblast (HFF; ✚), colon adenocarcinoma (HxGC 3; ), 5-azacytidine-treated HxGC 3 (Œ), and breast adenocarcinoma (MCF-7; F) cell types. These cell types were grown to near confluence, then 100 nM rQT 3 (a tri-tritiated queuine analogue) was added at Time 0, and incorporation assays were performed. The amount of radioactivity inserted into tRNA was measured at 0, 3, 6, 9, 12, and 24 h. The values were normalized to the number of picomoles of rQT 3 per 10 5 cells. Deviation bars indicated represent the standard deviation with n 5 6.
line turnover after 3 weeks of queuine deprivation demonstrating MCF-7 cells extremely inefficient salvage mechanism. Finally, it is difficult to directly assess salvage in HxGC 3 cells due to their intrinsically low TGRase activity even when induced by 5-azaC. However, it was reported previously that the queuosine-deficient HxGC 3 cells maintain a poor queuine salvage efficiency of 18% (3), which is suggested to be due, in
part, to the lack of substrate for queuine salvage enzymes. DISCUSSION Comparisons of queuosine levels between normal human fibroblast cells and neoplastic HxGC 3 colon and MCF-7 breast adenocarcinoma cell lines indicate that the characteristic hypomodification ob-
QUEUOSINE IN CULTURED CELLS
63
FIG. 4. The effect of exposure to a protein kinase C activator (TPA), inhibitor (staurosporine), and phosphatase inhibitor (okadaic acid) as compared to untreated cultures on queuine incorporation into tRNA of human foreskin fibroblast (HFF; ✚), colon adenocarcinoma (HxGC 3; ), 5-azacytidine-treated HxGC 3 (Œ), and breast adenocarcinoma (MCF-7; F) cell types. These cell types were grown to near confluence, then 100 nM rQT 3 (a tri-tritiated queuine analogue) was added at Time 0, and incorporation assays were performed. The incorporation assays were stopped at 8 h and the acid-precipitable fraction of the cytosol was isolated and counted by liquid scintillation. The values were normalized to the number of picomoles of rQT 3 per 10 5 cells. Deviation bars indicate the standard deviation with n 5 6.
served in cancer cells occurs in these cultures at 100 and 50 to 60%, respectively. Studies were undertaken to address whether the malfunctions of the queuosine modification system in HxGC 3 and MCF-7 neoplastic cell lines were due to lack of adequate queuine uptake across the cell membrane, incorporation into tRNA (TGRase activity), and salvage capability or due to aberrant PKC-catalyzed modulation of uptake or incorporation (Fig. 6). There appears to be no abnormality in the uptake mechanisms of either of the two neoplastic cell lines which could create the queuosine deficiency ob-
served in the cultures when compared to normal HFF cells. However, results of in vivo incorporation assays of rQT 3 into tRNA of HxGC 3 support previous conclusions, based on indirect observations with RPC-5 chromatography, that there is no incorporation of queuine into tRNA in these cells due to the lack of cellular TGRase activity. A 4.5-fold increase in 5-azaC-induced TGRase activity in the formation of queuosine-modified tRNA was measured directly by rQT 3 incorporation into tRNA, rather than indirectly by RPC-5 liquid chromatography as previously reported (3). Furthermore, this activity was
64
MORRIS ET AL.
FIG. 5. Comparison of queuine salvage in “normal” human fibroblasts and mammary adenocarcinoma cell cultures. Cultures of human fibroblast (HFF; ✚) or breast adenocarcinoma (MCF-7; F) cells were grown to near confluence in normal media, fed with charcoal-stripped serum (lacking queuine) for 1 day, 1 week, or 3 weeks with regular media changes, and then submitted to a queuine salvage assay. A dose of 100 nM tri-tritiated analogue of queuine (rQT 3) was placed in the media 24 h before the start of the assay to saturate the tRNA with the radiolabeled base. At Time 0 extracellular rQT 3 was washed away, and then a media change with charcoal-stripped media was performed. For each cell type, two sets of parallel cultures were established. One set of cultures was exposed to 0.10 A 260 units of unlabeled queuine (solid line) to provide an indication of the tRNA turnover rate within those cells. No queuine was added to the other set of cultures (dashed line) which represents the ability of the cells to retain the radiolabel in the cell’s tRNA despite the degradation processes involved in normal tRNA turnover. The counts per minute values obtained via liquid scintillation were normalized to the number of pmol rQT 3 contained within the acid-precipitable fraction of the 10 5 cells that had been devoid of queuine for 1 day, 1 week, or 3 weeks. Deviation bars indicated represent the standard deviation with n 5 4.
not modulated by phosphorylation as is the case with “normal” TGRase enzymes. In addition, it is shown that malfunction of TGRase in the queuosine modification system is not the cause for the observed queuosine deficiency in MCF-7 cells, since the incorporation rate in this cell line is equivalent to that observed in the normal HFF control. Normal HFF cells contain both queuine uptake and incorporation mechanisms that are directly activated by protein kinase C-induced phosphoryla-
tion, whereas salvage is not effected by phosphorylation levels (10,11). In contrast, protein phosphatases reverse the activation of queuine uptake and incorporation by removing the protein kinase C-attached phosphate group, thereby inhibiting both systems. Accordingly, TPA and okadaic acid both stimulate queuine uptake and incorporation in these cells by activating PKC or inhibiting protein phosphatase, respectively. Staurosporine decreases both uptake and incorporation rates by inhibiting
QUEUOSINE IN CULTURED CELLS
65
FIG. 6. The eukaryotic queuosine modification system in human cells. This system involves an uptake step facilitated by a specific queuine transmembrane transporter, tRNA:guanine ribosyltransferase which catalyzes the incorporation reaction of queuine base into tRNA asn, tRNA asp, tRNA his, or tRNA tyr, and a salvage step to recycle free queuine base from queuosine-59-monophosphate, a tRNA degradation product, for use in the incorporation step. The rates of both the uptake and incorporation steps (*) are shown to be stimulated by direct phosphorylation by protein kinase C (PKC), while the salvage step does not appear to be effected by PKC or a PKC-regulated modulation system. The results shown here indicate a deficiency in the salvage step for the MCF-7 cell line and the complete loss of the functional incorporation step in the HxGC 3 cell-line.
PKC activity. The response of the queuine uptake mechanisms in both HxGC 3 and MCF-7 cell-lines when exposed to modulators of PKC and protein phosphatase was comparable to that of the normal HFF cultures, indicating there was no deficiency in either PKC or protein phosphatase in these cells. However, in studies on TGRase in 5-azaC-treated HxGC 3 cells, the induced queuine incorporation activity was not effected by exposure to PKC or phosphatase modulators. Therefore, phosphorylation does not enhance the basal activity of the induced TGRase in HxGC 3 cells. Since TGRase is known to
be a heterodimer, the 5-azaC induced expression of TGRase activity may be due to the transcriptional activation of only the gene for the small active subunit of the enzyme without expression of the larger regulatory subunit, which is the target for phosphorylation (11). The MCF-7 cell line, however, exhibited a response comparable to that of the normal HFF cell cultures in queuine incorporation upon exposure to phosphorylation modulators indicating that both the TGRase-specific protein kinase C system and the catalytic and regulatory subunits of TGRase are intact.
66
MORRIS ET AL.
Due to the importance of PKC in tumor biology and the existence of defects of this enzyme in several neoplastic systems (18), it is also likely that ineffective phosphorylation-based modulation of the uptake and incorporation systems in tumor cells might contribute to queuosine hypomodification levels. While it was reported that the HxGC 3 cell line maintains a level of only 18% salvage activity (3), it was not possible to measure salvage activity in these cells using the assay method employed in these studies due to the lack of the queuosine-59-monophosphate tRNA turnover product. However, queuine salvage studies demonstrated a distinct deficiency in the salvage ability of the MCF-7 cell line. In cells pretreated with queuine-free media, the salvage activity is induced to recycle queuine from the queuosine-59-monophosphate generated by tRNA turnover, as shown in the dramatic increase in the ability of HFF cells to maintain the queuosine modification despite prolonged queuine starvation past the typical half-life (24 to 48 h) of the tRNA population. The MCF-7 cells are shown to have no latent and only mildly inducible salvage activity with queuine starvation for 1 week. In fact, after 3 weeks of treatment with queuine-free media, the cells exhibit a loss of all salvage activity. If either a decrease in the availability of dietary queuine or an increase in cell division rate or the malfunction of the TGRase-specific protein kinase C system were to occur, these cells would not be able to sustain high levels of intracellular queuosine-modified tRNA by salvage of the queuine base alone as predicted by Katze et al. (19). The lack of an efficient salvage mechanism in the rapidly growing MCF-7 cell line could help to explain the relatively moderate level of queuosine deficiency (50 to 60%) exhibited as compared with the HxGC 3 cells (100%) which exhibit no queuine incorporation activity (Fig. 6). The results of this research suggest that the deficiencies of the queuosine-modified tRNA population found in cancer cells, such as the HxGC 3 and MCF-7 cells examined in this study, may arise from the malfunction of any of a number of steps. Although the queuine uptake system has yet to be shown deficient in a cell line, the incorporation and salvage mechanisms are now established as defective in two adenocarcinomas. The HxGC 3 cell line is an example of a cell type containing a transcriptional inability to produce the queuine incorporation enzyme that is essential to form queuosine-modified tRNA. In this cell line, the lone mechanism for creating this modification is absent, which explains the complete de-
ficiency of queuosine in HxGC 3 cells. The MCF-7 cell line is an example of the disruptive effect of an inefficient queuine salvage mechanism on queuosine-modification levels in tRNA. In these cells, the salvage mechanism is nonexistent and when induced only functions to 50% of the capacity of “normal” fibroblasts. Due to the inability to salvage an intracellular source of queuine substrate for the production of the modification, these cells are able to maintain only 50% of the substrate tRNA modified with queuosine even with exposure to high queuine levels in serum-supplemented media. We suspect the deficiency would be worse in vivo perhaps approaching 80 –90% queuosine deficiency. This is because the normal circulating level of queuine in human serum is approximately 1 nM (19) and not 100 nM as used in this study. At physiological concentration, queuine uptake and incorporation would be less efficient. In combination with reduced modification efficiency, lack of latent salvage activity, and rapid growth rates, MCF-7 cells in vivo would be almost completely deficient in queuosine. The use of this comprehensive approach for study of the queuosine modification system is likely to produce reliable screening data for defects in the queuosine modification of cell cultures of established cell lines and those from tumor explants. The results can be used to further correlate the relationship of queuosine deficiency to progression and metastatic potential of neoplastic disease. ACKNOWLEDGMENTS The authors thank Dr. Jon R. Katze (University of Tennessee, Memphis) for the gracious gift of the HxGC 3 cell line and Dr. Ronald W. Trewyn (University of Kansas) for the generous donation of the rQT 3. This research was funded by the National Cancer Institute branch of the National Institutes for Health (R29-CA45213).
REFERENCES 1.
Katze JR, Basile B, McCloskey JA. Queuine, a modified base incorporated post-transcriptionally into eukaryotic transfer RNA: Wide distribution in nature. Science 216:55–56, 1982.
2.
Dirheimer G, Baranowski W, Keith G. Variations in tRNA modifications, particularly of their queuine content in higher eukaryotes. Its relations to malignancy grading. Biochimie 77:99 –103, 1995.
3.
Gu¨ndu¨z U, Elliott MS, Seubert PH, Houghton JA, Houghton PJ, Trewyn RT, Katze JR. Absence of tRNA-Guanine transglycosylase in a human colon adenocarcinoma cell line. Biochim Biophys Acta 1139:229 –238, 1992.
4.
Baranowski W, Dirheimer G, Jakowicki JA, Keith G. Defi-
QUEUOSINE IN CULTURED CELLS
5. 6.
7.
8.
9.
10.
11.
ciency of queuine, a highly modified purine base, in transfer RNAs from primary and metastatic ovarian malignant tumors in women. Cancer Res 54:4468 – 4471, 1994. Aytac U, Gu¨ndu¨z U. Q-modification of tRNAs in human brain tumors. Cancer Biochem Biophys 14:93–98, 1996. Huang BS, Wu RT, Chien KY. Relationship of the queuine content of transfer ribonucleic acids to histopathological grading and survival in human lung cancer. Cancer Res 52:4696 – 4700, 1992. Emmerich B, Zubrod E, Weber H, Maubach PA, Kersten H, Kersten W. Relationship of queuine-lacking transfer RNAs to the grade of malignancy in human leukemias and lymphomas. Cancer Res 45:4308 – 4314, 1985. Slany RK, Kersten H. Genes, enzymes, and coenzymes of queuosine biosynthesis in procaryotes. Biochimie 76:1178 – 1182, 1994. Katze JR, Gu¨ndu¨z U, Smith DL, Cheng CS, McCloskey JA. Evidence that the nucleic acid base queuine is incorporated intact into tRNA by animal cells. Biochemistry 23:1171– 1176, 1984. Morris RC, Brooks BJ, Eriotou P, Kelly DF, Sagar S, Hart KL, Elliott MS. Activation of transfer RNA-guanine ribosyltransferase by protein kinase C. Nucleic Acids Res 23:2492– 2498, 1995. Morris RC, Brooks BJ, Hart KL, Elliott MS. Modulation of queuine uptake and incorporation into tRNA by protein
67
kinase C and protein phosphatase. Biochim Biophys Acta 1311:124 –132, 1996. 12.
Muralidhar G, Utz ED, Elliott MS, Katze JR, Trewyn RW. Identifying inhibitors of queuine modification of tRNA in cultured cells. Anal Biochem 171:346 –351, 1988.
13.
Elliott MS, Trewyn RW. Queuine hypomodification of tRNA induced by 7-methylguanine. Biochem Biophys Res Commun 104:326 –332, 1982.
14.
Elliott MS, Katze JR, Trewyn RW. Relationship between a tumor promoter-induced decrease in queuine modification of transfer RNA in normal human cells and the expression of an altered phenotype. Cancer Res 44:3215–3219, 1984.
15.
Elliott MS, Katze JR. Inhibition of queuine uptake in diploid human fibroblasts by phorbol-12,13-didecanoate. J Biol. Chem 261:13019 –13025, 1986.
16.
Katze JR. Q-factor: A serum component required for the appearance of nucleoside Q in tRNA in tissue culture. Biochem Biophys Res Commun 84:527–535, 1978.
17.
Gu¨ndu¨z U, Katze JR. Queuine salvage in mammalian cells. J Biol Chem 259:1110 –1113, 1984.
18.
Glazer RI. In Protein Kinase C (Kuo JF, Ed.). Oxford: Oxford University Press, pp 171–198, 1994.
19.
Katze JR, Beck WT, Cheng CS, McCloskey JA. Why is tumor tRNA hypomodified with respect to Q nucleoside? Recent Results Cancer Res 84:140 –159, 1983.