Regulation of Transferrin Receptor Cycling by Protein Kinase C Is Independent of Receptor Phosphorylation at Serine 24 in Swiss 3T3 Fibroblasts*

Treatment of Swiss 3T3 fibroblasts with tumor-pro- moting phorbol diester or with platelet-derived growth factor caused the phosphorylation of the transferrin receptor by protein kinase C (Ca2+/phospholipid-de-pendent enzyme) at serine 24 and increased the cell surface expression of the transferrin receptor. The hypothesis that the regulation of transferrin receptor cycling by protein kinase C is causally related to the phosphorylation of the receptor at serine 24 was crit- ically tested. Site-directed mutagenesis of the human transferrin receptor cDNA was used to substitute serine 24 with threonine or alanine residues in order to create phosphorylation defective receptors. Wild- type and mutated transferrin receptors were expressed in Swiss 3T3 fibroblasts using the retrovirus vector pZipNeoSV (X). These receptors were functionally ac-tive and caused the receptor-mediated endocytosis of diferric transferrin. Incubation of the fibroblasts with phorbol diester caused the phosphorylation of the wild-type (Ser-24) human transferrin receptor, but this treatment did not result in the phosphorylation of the mutated (Ala-24 and Thr-24) receptors. The cycling of the phosphorylation defective receptors was regulated by phorbol diester and platelet-derived growth factor in a manner similar to that observed for the wild-type receptor. We conclude that the regulation of transfer- rin receptor cycling by protein kinase C is independent of receptor phosphorylation at serine 24 in Swiss 3T3 fibroblasts. single We conclude that these clones contain single intact and These clones were designated Ala-24 to Thr-24 were then washed and incubated with fluorescein conjugated goat anti-mouse IgG antibody (1:300) for 40 min. After this incubation the cells were washed and examined by fluorescence microscopy. In other experi- ments the monoclonal antibody R17217 was employed and fluores-cein-conjugated goat anti-rat IgG antibody was used to visualize the murine transferrin receptors. (pH 3.0) to remove the surface-bound transferrin. Intra- cellular 1251-transferrin was measured as the cell-associated radioactivity following washing at pH 3.0. Cell surface '*sI-transferrin was measured as the intracellular radioactivity subtracted from the total cell-associated radioactivity. The results represent the mean of triplicate determinations. Similar results were obtained in three separate exwriments. murine transferrin receptor. Similar results were obtained in three separate experiments.

Treatment of Swiss 3T3 fibroblasts with tumor-promoting phorbol diester or with platelet-derived growth factor caused the phosphorylation of the transferrin receptor by protein kinase C (Ca2+/phospholipid-dependent enzyme) at serine 24 and increased the cell surface expression of the transferrin receptor. The hypothesis that the regulation of transferrin receptor cycling by protein kinase C is causally related to the phosphorylation of the receptor at serine 24 was critically tested. Site-directed mutagenesis of the human transferrin receptor cDNA was used to substitute serine 24 with threonine or alanine residues in order to create phosphorylation defective receptors. Wildtype and mutated transferrin receptors were expressed in Swiss 3T3 fibroblasts using the retrovirus vector pZipNeoSV (X). These receptors were functionally active and caused the receptor-mediated endocytosis of diferric transferrin. Incubation of the fibroblasts with phorbol diester caused the phosphorylation of the wildtype (Ser-24) human transferrin receptor, but this treatment did not result in the phosphorylation of the mutated (Ala-24 and Thr-24) receptors. The cycling of the phosphorylation defective receptors was regulated by phorbol diester and platelet-derived growth factor in a manner similar to that observed for the wild-type receptor. We conclude that the regulation of transferrin receptor cycling by protein kinase C is independent of receptor phosphorylation at serine 24 in Swiss 3T3 fibroblasts.
Transferrin is a serum protein that binds iron and is an essential requirement for the growth of cultured cells (1). T h e uptake of iron into cells is mediated by specific cell surface receptors for diferric transferrin that cycle between the plasma membrane and endosomal membranes (reviewed in Ref. 2). Activation of the Ca2+/phospholipid-dependent protein kinase (protein kinase C)' by treatment of cells with phorbol diester (3-7) or with exogenous diacylglycerols (8) acutely regulates this process. Regulation of transferrin receptor cycling is also observed in cells treated with growth factors * This work was supported, in part, by Grants GM37845, AM30898, and CA39240. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The mechanism by which protein kinase C regulates the cycling of the transferrin receptor is not understood. May et al. (4,15,16) have proposed that phosphorylation of the transferrin receptor by protein kinase C regulates cycling. The purpose of the experiments described here was to critically test this hypothesis. The approach that we took was to examine the regulation of the cycling of transferrin receptors that are defective as protein kinase C substrates. The site of transferrin receptor phosphorylation by protein kinase C has been identified as serine 24 (17). By site-directed mutagenesis we substituted serine 24 with a threonine and an alanine residue and expressed these mutant receptors in Swiss 3T3 cells using a retrovirus vector. The cycling of the phosphorylation defective receptors was regulated by PMA in a manner similar to that observed for the wild-type receptor. We conclude that the regulation of transferrin receptor cycling by protein kinase C is independent of receptor phosphorylation at serine 24 in Swiss 3T3 fibroblasts. States Biochemical Corp., respectively. Dideoxynucleotides were from Boehringer Mannheim. Phorbol diesters and protein A-Sepharose CL-4B were from Sigma. The synthetic peptide Lys-Arg-Thr-Leu-Arg-Arg was obtained from Peninsula Laboratories (Belmont, CA). Epidermal growth factor was purified (18,19) and iodinated (20) as described. Platelet-derived growth factor (porcine) was from Bioprocessing Ltd. G418 was obtained from GIBCO. Transferrin was from Behring Diagnostics and was further purified by gel filtration chromatography before use. Fluorescein-conjugated antibodies were from Cappel. Diferric transferrin, [59Fe]diferric transferrin, and diferric lZ5I-transferrin were prepared as described previously (9).

Materials
Cell Culture-Swiss 3T3 cells were obtained from the American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium supplemented with 5% calf serum. Psi-2 cells were obtained from Dr. R. Mulligan (Whitehead Institute). WI-38 human fetal lung fibroblasts were obtained from the American Type Culture Collection and maintained in modified Eagle's medium supplemented with 5% fetal calf serum. Hybridoma cells OKT9, R17217, and R17208 were obtained from the American Type Culture Collection.
Site-directed Mutagenesis-The full-length 4.9-kb cDNA for the Dr. F. H. Ruddle (Yale University) (21). A BamHI-Hind111 0.9-kb human transferrin receptor in the plasmid pcDTRl was provided by fragment isolated from pcDTRl (which contains the 5'-untranslated region and the coding region for the amino-terminal cytoplasmic domain) was subcloned into M13mp8. Site-directed mutagenesis of serine 24 (&) was carried out according to Zoller and Smith (22) using 17-mer oligonucleotides coding for threonine (5"AGGGTGA-ACCGGGTATA-3') or alanine (5'-AGWGAACCGGGTm-3'). Mutants were selected by washing at increasing temperatures in 6 X 16041 Transferrin Receptor Phosphorylation by Protein Kinase C N:tCl/Tris, pH 7.5 (23), and confirmed by sequencing the entire 0.9k'l fragment with [35S]dATP and ddNTPs (24). The native BamHI-HmdIII 0.9-kb fragment as well as the alanine and threonine mutants at amino acid 24 were isolated and ligated to a HindIII-BglII 1.9-kb fragment of pcDTRl that contains the coding region for the carboxylterminal domain of the transferrin receptor. After digestion with BamHI and BgtII, a 2.8-kb cDNA containing the coding region and portions of the untranslated regions was isolated by agarose gel electrophoresis and cloned into the BamHI site of the retrovirus vector pZipNeoSV(X) (25) in both orientations using standard techniques (26). The plasmids obtained (pZTRSer-24, pZTRAla-24, and pZTRThr-24) were used for the expression of wild-type (Ser-24) and mutated human transferrin receptors in murine cells. Expression of the Transferrin Receptor cDNA in Swiss 3T3 Celk-The plasmid constructs pZTRSer-24, pZTRAla-24, and pZTRThr-24 were transfected into Psi-2 cells using the Capol method (25). Stable colonies resistant to G418 (1 mg/ml) were isolated. The viral titer in the tissue culture supernatant was measured for each clone by investigating the colony formation by Swiss 3T3 cells in G418 (1 mg/ml) after incubation with serial dilutions of supernatant treated with 8 pg/ml Polybrene (25). Clones with a titer of >lo5 cfu/ml were selected and used to produce virus stocks for the infection of Swiss 3T3 cells. Swiss 3T3 cells were infected with the recombinant retrovirus as described (25), and stable colonies resistant to G418 (1 mg/ml) were isolated.
The expression of human transferrin receptors at the cell surface was investigated using the antibody OKT9. Swiss 3T3 clones were seeded in 16-mm wells and grown to confluence. The cell monolayers were washed three times and incubated in 120 mM NaCl, 6 mM KCl, 1.2 mM MgCl,, 1 mM CaC12, 25 mM Hepes (pH 7.4), 0.2% bovine serum albumin for 30 min at 37 "C. The cells were cooled to 0 "C and incubated with 10 pg/ml OKT9 for 2 h. The cells were washed and subsequently incubated with ['*51]sheep anti-mouse IgG (500,000 cpm) for 2 h at 0 "C. The monolayers were then washed, solubilized with 1 M NaOH, and the radioactivity associated with the cells was measured with a gamma counter. Nonspecific binding of the sheep anti-mouse Ig was estimated in incubations without the monoclonal antibody OKT9.
Southern blot analysis was used to investigate the proviral integration in the fibroblast clones. The probe used was a BamHI-Hind111 1.4-kb fragment containing the aminoglycoside 3'-phosphotransferase (NeoR) gene isolated from pNeo (Pharmacia). Sac1 digestion of genomic DNA isolated from all clones gave a single 7.0-kb band of equal intensity corresponding to the expected length of the retrovirus containing the 2.8-kb transferrin receptor cDNA. Hind111 digestion gave a common band at 2.45-kb and a single band at 4.2 kb (clone 4.20 Ser-24), 5.0 kb (clone 5.01 Ala-24), and 6.4 kb (clone 3.05 Thr-24) (data not shown). We conclude that these fibroblast clones contain a single intact integrated provirus and were chosen for further study. These clones were designated Ser-24 (wild type), Ala-24 (serine to alanine mutation), and Thr-24 (serine to threonine mutation).
Immunoprecipitation of Transferrin Receptors-Cells were prepared by incubation for 24 h in (a) phosphate-free Dulbecco's modified Eagle's medium supplemented with 3 mCi/ml [32P]phosphate and 0.5% calf serum or ( b ) methionine-free Dulbecco's modified Eagle's medium supplemented with 10 p~ [%]methionine (100 pci/ ml) and 0.5% calf serum. The cells were then solubilized with 1% Nonidet P-40,25 mM Hepes (pH 7.8), 1 M NaCl, 50 mM NaF, 100 p M Na3V05, 5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 pg/ ml leupeptin. After centrifugation at 100,000 X g for 30 min at 4 "C, the extracts were incubated at 22 "C with 20 pl of packed protein A-Sepharose CL-4B coated with goat anti-mouse IgG and the monoclonal antibody OKT9 which is specific for the human transferrin receptor (27). In other experiments the monoclonal antibodies R17217 (rat IgG) and R17208 (rat IgM) were used to immunoprecipitate murine transferrin receptors (28, 29). For these experiments a goat anti-rat IgG or goat anti-rat IgM antibody was used. After 60 min the protein A-Sepharose CL-4B was washed with lysis buffer and finally washed with 0.2% Nonidet P-40, 25 mM Hepes (pH 7.8). The immunoprecipitates were reduced with dithiothreitol and then used for polyacrylamide gel electrophoresis in the presence of 0.1% sodium dodecyl sulfate. Gels with samples isolated from [32P]phosphatelabeled cells were analyzed by autoradiography using Kodak X-OMAT AR film and a Dupont Lightning Plus enhancing screen. Samples containing ["Slmethionine were analyzed by fluorography using En3Hance (Du Pont-New England Nuclear).
Phosphopeptide Mapping-Transferrin receptors were isolated from [32P]phosphate-labeled cells by immunoprecipitation and poly-acrylamide gel electrophoresis. The receptors were digested with trypsin and the [32P]phosphopeptides obtained were resolved by high pressure liquid chromatography as described (17). A Vydac CIS column equilibrated with 0.1% trifluoroacetic acid was employed. After injection of the sample, the column was washed for 5 min, and the peptides were eluted with a linear gradient of acetonitrile (0-60%) over 60 min. The flow rate was 1 ml/min. Fractions eluted from the column were collected, and peptides were detected by Cerenkov counting.
Measurement of Protein Kinase C Function and Actiuity-The specific binding of 5 nM [3H]PDBU to cell monolayers was measured as described (30). The regulation of the EGF receptor was investigated by measuring the specific binding of '"I-EGF to cell monolayers as described (30). Protein kinase C activity was measured using a synthetic peptide substrate assay?
Analysis of Transferrin Binding and Uptake-The binding of diferric lZ5I-transferrin to cell surface receptors was performed as described previously (9). The uptake of [59Fe]diferric transferrin was measured at 37 "C (9). The rate of release of ['251]apotransferrin from cell monolayers was measured as described by Wiley and Kaplan (32).
Indirect Immunofluorescence-Cells were rapidly transferred from medium at 37 "C to 3% (w/v) formaldehyde, 120 mM NaCl, 6 mM KCl, 1 mM MgC12,l mM CaC12, 25 mM Hepes (pH 7.4) at 22 "C. After 20 min the cells were washed and then incubated with 50 mM NH&l in Hepes-buffered saline for 10 min. Subsequently, the cells were incubated for 3 min with 0.2% Triton X-100 in Hepes-buffered saline to expose intracellular antigenic sites. The cells were washed and then incubated with 20 pg/ml anti-human transferrin receptor monoclonal antibody OKT9 for 40 min at 22 "C. The cells were then washed and incubated with fluorescein conjugated goat anti-mouse IgG antibody (1:300) for 40 min. After this incubation the cells were washed and examined by fluorescence microscopy. In other experiments the monoclonal antibody R17217 was employed and fluorescein-conjugated goat anti-rat IgG antibody was used to visualize the murine transferrin receptors.

Expression of the Human Transferrin Receptor in Swiss 3T3
Fibroblasts-Expression of the human transferrin receptor in Swiss 3T3 fibroblast clones was investigated by immunoprecipitation of cells metabolically labeled with [%]methionine using a monoclonal antibody specific for the human transferrin receptor (OKT9). No expression of human transferrin receptors ( M r = 94,000) was observed in control Swiss 3T3 cells, but specifically immunoprecipitated human transferrin receptors were observed in fibroblast clones infected with retrovirus ( Fig. 1). Expression of the murine transferrin receptor ( M , = 100,000) was examined by immunoprecipitation with the rat monoclonal antibodies R17217 and R17208 which bind to the murine transferrin receptor. Expression of the murine transferrin receptor was observed in control Swiss 3T3 fibroblasts and in fibroblast clones. However, the human transferrin receptor was expressed at a higher level than the murine transferrin receptor (Fig. 1) and resulted in an increased level of diferric transferrin binding to the fibroblast clones ( Table I).
The transferrin receptor is a disulfide-linked dimer (33). Therefore, in fibroblast clones expressing both murine and human transferrin receptors it is possible that a mixture of murine and human homodimers and heterodimers are present. If heterodimers are expressed in the fibroblast clones, it would be expected that the human and murine transferrin receptors would be coimmunoprecipitated by monoclonal antibodies. In experiments with the monoclonal antibody OKT9, only human transferrin receptors were isolated from fibroblast clones (Fig. 1). However, when the rat monoclonal antibodies R17217 and R17208 (which bind to the murine transferrin receptor) were used an equal number of human ( M , = 94,000) and murine (Mr = 100,000) transferrin receptors were immuno-* R. J. Davis and Czech, M. P. (1987) J. Biol TABLE I Regulntion of transferrin receptor distribution by phorbol diesters The distribution of transferrin receptors between the cell surface and intracellular compartments was investigated in control Swiss 3T3 fibroblasts and in fibroblast clones expressing human transferrin receptors. The fibroblast clones were incubated with 20 pg/ml of the IgM monoclonal antibody R17208 for 12 h prior to the experiment in order to down-regulate the murine transferrin receptors. Confluent monolayers of the cells were incubated at 37 "C for 2 h with 300 nM diferric "'1-transferrin. The cells were then treated with and without 10 nM PMA for 5 min. Subsequently, the cells were rapidly washed with cold medium and the radioactivity associated with the cells was measured. Diferric '*'I-transferrin bound to cell surface receptors was identified by washing the cells (3 min, 0 "C) with 120 mM NaCI, 50 mM glycine (pH 3.0) to remove the surface-bound transferrin. Intracellular 1251-transferrin was measured as the cell-associated radioactivity following washing at pH 3.0. Cell surface '*sI-transferrin was measured as the intracellular radioactivity subtracted from the total cell-associated radioactivity. The results represent the mean of triplicate determinations. Similar results were obtained in three separate exwriments.  (Fig. 1). As the human receptor is expressed a t a significantly higher level than the murine receptor, we conclude that the anti-murine transferrin receptor antibodies R17217 and R17208 immunoprecipitate heterodimeric transferrin receptors composed of murine and human disulfide- Although human homodimeric receptors are over-expressed in the fibroblast clones, the presence of murine-human heterodimeric transferrin receptors represents a significant problem for the interpretation of the effect of specific mutations in the cDNA for the human receptor on the cycling process (Fig. 1). A strategy to decrease the expression of heterodimeric receptors was therefore employed in order to characterize the fibroblast clones obtained. The rat monoclonal antibody R17208 is an IgM which specifically inhibits the binding of diferric transferrin to the murine transferrin receptor (29). Incubation of cultured cells with anti-transferrin receptor antibodies results in the down-regulation and degradation of the transferrin receptor (35). The antibody R17208 was therefore used to down-regulate the murine transferrin receptors in fibroblast clones by incubation of the cells for 12 h with 20 pg/ml of R17208. Fig. 2 demonstrates that this treatment results in the loss of murine transferrin receptors and consequently of heterodimeric receptors. The only receptor form observed in the down-regulated fibroblasts was the human homodimer, although this was expressed a t a reduced level compared with untreated cells (Fig. 2). Similar results were obtained in experiments using clones expressing the Ser-24, Thr-24, and Ala-24 human transferrin receptors (data not shown). We conclude that the fibroblasts incubated with antibody R17208 are suitable for the analysis of the properties of the expressed human receptor and were used in further experiments.

Regulation of Transferrin Receptor Phosphorylation by Phorbol Diester and Growth Factors-The effect
of growth factors on the phosphorylation state of the human (Ser-24) transferrin receptor expressed in Swiss 3T3 fibroblasts was examined. Treatment of the fibroblasts with PMA or PDGF caused a marked increase in the phosphorylation state of the human transferrin receptor (Fig. 3). A small increase in phosphorylation was observed when the fibroblasts were treated with EGF (Fig. 3). Diferric transferrin was found to have no effect on the phosphorylation state of the human transferrin receptor (data not shown). Phosphopeptide mapping of the human transferrin receptor indicated that the effect of PMA,  PDGF, and EGF was to increase the level of two tryptic phosphopeptides (data not shown). These phosphopeptides have been previously demonstrated to be the result of incomplete trypsin digestion of the transferrin receptor, and the phosphorylated residue has been identified as serine 24 (17).
Mutation of the human transferrin receptor by substitution of alanine for serine at residue 24 resulted in the loss of the PMA-stimulated phosphorylation of the transferrin receptor as expected (Fig. 4). Similarly, substitution of threonine for serine at residue 24 also resulted in the loss of the PMAstimulated phosphorylation of the transferrin receptor (Fig.  4). The differences in the phosphorylation of the transferrin receptor between fibroblast clones (Fig. 4) could be due to either the effect of specific mutations introduced into the receptor or to a defect in the response of the clones to phorbol diester. T o resolve these possibilities the activity of protein kinase C in the fibroblast clones was examined. A similar level of expression of phorbol diester receptors and protein kinase C activity was observed in control fibroblasts and clones expressing human transferrin receptors (Table 11). In order to investigate the functional activity of protein kinase C, the transmodulation of the EGF receptor caused by PMA (reviewed in Ref. 36) was examined. A similar inhibition of the binding of '2sI-EGF to cell surface receptors was observed in control fibroblasts and the isolated clones (Table 11). We conclude that the lack of PMA-stimulated phosphorylation of the transferrin receptor in the Thr-24 and Ala-24 clones is a result of the receptor mutations and not because of a defect in the response of the clones to phorbol diester.
Regulation of the Cell Surface Expression of Transferrin Receptors by Phorbol Diester-Treatment of Swiss 3T3 fibroblasts with PMA causes a rapid increase in the cell surface expression of the transferrin receptor (Fig. 5 ) . Clones of Swiss 3T3 fibroblasts expressing human transferrin receptors (Ser-24, Thr-24, and Ala-24) exhibited a similar time course and dose response of PMA action (Fig. 5 ) . The increase in the cell surface expression of transferrin receptors is accompanied by a decrease in the number of intracellular transferrin receptors (Table I). We conclude that the action of PMA is to cause a redistribution of intracellular transferrin receptors to the cell surface that is independent of serine 24 phosphorylation. The mechanism by which PMA causes this rapid redistribution of receptors was examined by investigating the rate of exocytosis of the transferrin receptor, which can be measured as the rate of release of ['2sI]apotransferrin from cells. Treatment with PMA caused an increase in the first-order rate constant for ['2sI]apotransferrin release from 0.10 min" to 0.19 min" (Fig.  6). Similar effects of PMA were observed on the cycling of the wild-type human transferrin receptor and receptors in which serine 24 was substituted with threonine or alanine residues (Table I, Figs. 5, and 6).

TABLE I1 Characterization of protein kinuse C function and activity in Swiss
3T3 fibroblast clones Control Swiss 3T3 fibroblasts and fibroblast clones expressing human transferrin receptors (4.20 Ser-24, 5.01 Ala-24 and 3.05 Thr-24) were investigated for the presence of functional protein kinase C activity. First, the specific binding of 5 nM ['H]4,9-phorbol-12,9-, 130dibutyrate (PDBU) to cell monolayers was measured. Second, the Ca2+ and phospholipid-dependent phosphorylation of the synthetic peptide Lys-Arg-Thr-Leu-Arg-Arg by cell extracts was examined. Third, the transmodulation of the EGF receptor in cells treated with 10 nM PMA was investigated. The effect of PMA to regulate the EGF receptor is expressed as the specific binding of 200 p~ '''I-EGF to cells treated with 10 nM PMA for 30 min at 37 "C compared with the binding observed to cells treated without PMA. The results are presented as the mean of observations made in two separate experiments.
["HIPDBU Regulation of Iron Accumulation by Phorbol Diester-The functional significance of the regulation of the cycling of the transferrin receptor in Swiss 3T3 fibroblasts was examined by investigating the rate of accumulation of radioactivity by cells incubated with 100 nM [59Fe]diferric transferrin. Treatment of control Swiss 3T3 fibroblasts with PMA increased the rate of accumulation of [59Fe]diferric transferrin (Fig. 7). The redistribution of transferrin receptors caused by PMA is therefore associated with an increase in the rate of uptake of iron by the fibroblasts. Incubation of the fibroblasts with 20 ,ug/ml of antibody R17208 for 12 h caused a marked inhibition of iron accumulation. This result is consistent with the effect of R17208 to inhibit diferric transferrin binding to the murine transferrin receptor (29) and to cause down-regulation of the murine transferrin receptor (Fig. 2). Fibroblast clones expressing the wild-type (Fig. 7B) and mutated (Fig. 7, C and   D ) human transferrin receptor accumulated [59Fe]diferric transferrin a t a greater rate than control Swiss 3T3 cells. Incubation of these clones with the antibody R17208 for 12 h did not result in a significant decrease in the rate of uptake of [59Fe]diferric transferrin (Fig. 7). Treatment with PMA caused an increase in the rate of uptake of [59Fe]diferric transferrin in all cases. We conclude that the uptake of diferric transferrin mediated by the wild-type and mutated forms of the human transferrin receptor is regulated similarly by tumor-promoting phorbol diesters.
Effect of Growth Factors on the Cell Surface Expression of Transferrin Receptors-PDGF and EGF cause an increase in the cell surface expression of transferrin receptors in fibroblasts (9,32). PDGF caused a marked increase in the cell surface expression of murine and human transferrin receptors in Swiss 3T3 fibroblasts (Fig. 8). EGF also increased the cell surface expression of transferrin receptors (Fig. 8), but this effect was small compared to that reported for BALB/c 3T3 fibroblasts (9). Similar results were obtained with clones expressing Ala-24 and Thr-24 human transferrin receptors (Fig. 8).
Indirect Immunofluorescence Analysis of Transferrin Receptor Expression-The distribution of transferrin receptors was examined with the antibody R17217 in control Swiss 3T3 fibroblasts treated with Triton X-100 to expose intracellular receptors (Fig. 9). Most of the murine transferrin receptors were observed to be in an intracellular juxtanuclear location. Cell surface receptors were frequently concentrated at the spreading margins of the cells. Similar results were obtained when the distribution of the wild-type human transferrin receptor (Ser-24), and the Ala-24 or Thr-24 mutant receptors were examined using the antibody OKT9 (Fig. 9).

DISCUSSION
Expression of the Human Transferrin Receptor cDNA-The strategy we have used to express the human transferrin re-ceptor cDNA in cultured cells was to employ the retrovirus vector pZipNeoSV (X) described by Cepko et al. (25). Transfection of Psi-2 cells with plasmid constructs containing the transferrin receptor cDNA yielded stable clones resistant to G418 that provided high (>lo5 cfu/ml) titers of recombinant retrovirus. Infection of Swiss 3T3 fibroblasts with the retrovirus allowed the isolation of stable clones that express the human transferrin receptor. As the transferrin receptor is a disulfide-linked dimer, a problem encountered was the biosynthesis of heterodimeric receptors composed of murine and human monomers (Figs. 1 and 2). In order to investigate the cycling of the human homodimeric receptors, it was necessary to down-regulate the murine transferrin receptors by incubating the cells with the rat IgM monoclonal antibody R17208 which binds to the murine transferrin receptor and inhibits the binding of diferric transferrin (Fig. 2). The cells obtained by this treatment expressed human homodimeric transferrin receptors (Fig. 2). The monoclonal antibody OKT9 which binds specifically to the human transferrin receptor was used to characterize the expressed human homodimeric receptors.
Regulation of the Cycling of Phosphorylation Defective Transferrin Receptors-Serine 24 is the major site on the human transferrin receptor that is phosphorylated by protein kinase C (17). Replacement of the serine residue at position 24 with alanine resulted in the loss of PMA-stimulated phosphorylation of the transferrin receptor as expected (Fig. 4). A similar result was obtained when the serine was replaced by a threonine residue (Fig. 4). This result was unexpected because the threonine substitution represents a relatively conservative mutation. Recently, House et al. (37) have investigated the substrate specificity of protein kinase C using synthetic peptides corresponding to the local phosphorylation site sequence of glycogen synthase. It was observed that the substitution of threonine for a serine residue at the phosphorylation site caused a marked decrease in the apparent K,,, for phosphorylation. These results suggest that the kinetics of substrate phosphorylation by protein kinase C can be strongly sequence dependent.
Characterization of the wild-type (Ser-24) and the phosphorylation defective (Ala-24 and Thr-24) human transferrin receptors expressed in Swiss 3T3 fibroblasts demonstrated that they were functional and caused the receptor-mediated endocytosis of diferric transferrin (Fig. 7). Examination of the subcellular distribution of the phosphorylation defective receptors by indirect immunofluorescence microscopy indicated no significant differences with the distribution observed for the wild-type human receptor (Fig. 9). Furthermore, the cycling of the phosphorylation-defective receptors was regulated by PMA, PDGF, and EGF in a manner similar to that observed for the wild-type human receptor (Figs. 5-8). We conclude from these results that the regulation of transferrin receptor cycling by protein kinase C is independent of receptor phosphorylation a t serine 24 in Swiss 3T3 fibroblasts.
Regulation of Transferrin Receptor Cycling by Protein Kim e C-Treatment of cultured cells with tumor-promoting phorbol diesters that stimulate the activity of protein kinase C cause the phosphorylation of the transferrin receptor at serine 24 and the acute regulation of transferrin receptor cycling. In Swiss 3T3 fibroblasts (Fig. 5) and murine peritoneal macrophages (7), PMA causes a rapid increase in the expression of transferrin receptors at the cell surface. A mechanism by which PMA regulates the cell surface transferrin receptor expression in Swiss 3T3 fibroblasts is a marked increase in the rate of transferrin receptor exocytosis (Fig. 6). A functional consequence of this is an increase in receptormediated endocytosis of diferric transferrin resulting in a stimulation of iron accumulation by the fibroblasts (Fig. 7). Increased endocytosis of transferrin receptors has also been reported in experiments with HepG2 hepatoma cells (6), HL60 promyelocytic leukemia cells (3,4, 15), and K562 erythroleukemia cells (5) treated with phorbol diester. However, in contrast to the results presented here with Swiss 3T3 fibroblasts, a decrease in the cell surface transferrin receptor expression is observed (3-6, 15). This difference in the response of cultured cells to PMA treatment suggests that the cycling of the transferrin receptor is under complex regulatory control. Several steps in the cycling pathway may be regulated and in different cell types the primary site of regulation may not be identical. The morphological pathway by which the transferrin receptor cycles (reviewed in Ref. 16) comprises many steps that could potentially be regulated by protein kinase C. Data in the literature indicate that membrane flow through several of these morphological compartments can be acutely regulated in PMA-treated cells. For example PMA causes an increase in endocytosis as well as exocytosis in the toad urinary bladder (38). Furthermore, in macrophages PMA stimulates pinocytosis and redirects the flow of intracellular pinocytotic fluid (31). Future progress toward understanding how protein kinase C regulates the cycling of the transferrin receptor will require identification of the relevant morphological compartment(s) of cycling transferrin receptors and the identification of the biochemical step(s) regulated.
Conclusions-We have tested the hypothesis that phosphorylation of the transferrin receptor by protein kinase C regulates receptor cycling in Swiss 3T3 fibroblasts. The site of transferrin receptor phosphorylation by protein kinase C has been identified as serine 24 (17). Wild-type and phosphorylation defective transferrin receptors in which serine 24 was substituted with either an alanine or a threonine residue were expressed in Swiss 3T3 fibroblasts. The cycling of the phosphorylation-defective receptors was regulated by PMA, PDGF, and EGF in a manner similar to that observed for the wild-type receptor. We conclude that the regulation of transferrin receptor cycling by protein kinase C is independent of receptor phosphorylation a t serine 24 in Swiss 3T3 fibroblasts.