The Synthesis of Human Placental Lactogen by Ribosomes Derived

Proc. Nat. Acad. Sci. USA Vol. 71, No. 4, pp. 1322-1325, April 1974

The Synthesis of Human Placental Lactogen by Ribosomes Derived from Human Placenta (reproduction/protein synthesis/hormone/pregnancy) IRVING BOIME AND SOPHIE BOGUSLAWSKI Departments of Obstetrics and Gynecology and Pharmacology, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, Missouri 63110 Communicated by Oliver H. Lowry, December 7, 1973

ABSTRACT In a very active cell-free system containing tional Biochemicals. Rat liver tRNA was generously provided polysomes derived from human placenta and a cell-sap Dr. fraction prepared from ascites tumor cells, the syn- by Dolph Hatfield. thesis of the hormone human placental lactogen (HPL) Isolation of Placental Ribsomes. Ribosomes derived from was detected. The identification was based on the following: (a) The in vitro synthesized protein labeled with first and third trimester placentas were prepared as described P5Simethionine migrated at the same rate as authentic previously (4). The tissue, which was washed free of blood, HPL on sodium dodecyl sulfate-polyacrylamide gels and was pressed through a 1.5-mm grid to remove connective (b) tryptic fingerprint analysis of the labeled protein tissue and vasculature. The preparations were further homog- yielded peptides having the same mobilities as seen with enized on a 1:1 v/w basis in a buffer containing 30 mM the same analysis of purified HPL. The amount of HPL synthesized in a cell-free system Tris - HCl, (pH 7.5), 120 mM KCl, 7 mM 2-mercaptoethanol, containing polysomes derived from term placenta was 5 mM magnesium acetate, and 0.5 mM EDTA. Homogeniza- about 10% of the total proteins synthesized aind in a com- tion was carried out in the cold for about 3 min with motor- parable system containing first trimester ribosomes the driven Teflon and glass homogenizers (Thomas Co., Philadel- level of synthesis was about 5%. These data suggest the potential for quantitating the HPL mRNA activity as a phia, Pa.). The homogenate was then centrifuged at 8500 X function of the period of gestation and for isolating the g for 10 min at 4°. The supernatant fluid was brought to 1% mRNA itself. deoxycholate concentration with a 10% solution. This suspension was then layered on a discontinuous sucrose gradient It is well established that the human placenta synthesizes at composed of 4 ml each of 40 and 45% sucrose solutions pre- least two protein hormones, human chorionic gonadotrophin pared in the homogenizing buffer. The gradients were placed (HCG) and human placental lactogen (HPL). The levels of in a Beckman type 60 titanium rotor and centrifuged at these hormones peak at different periods of gestation; HCG 200,000 X gfor3hr at 4'. reaches its peak level in serum and urine at 10-12 weeks, The top layers were then aspirated and the tubes containing whereas, the highest titers of HPL are attained at term. In the pellets were gently rinsed with homogenizing buffer and fact, HPL seems to represent a major protein synthesized in the pellets were resuspended in this buffer with a small hand the placenta at terrh (1-3). Therefore, it is apparent that the homogenizer. (Occasionally, some white, fluffy material synthesis and/or secretion of these two hormones is differen- collected around the ribosome pellet; much of this was re- tially coupled to gestation. moved with a stirring rod.) In order to investigate the factors involved in regulating the synthesis of the placental protein hormones, it would be Preparation of Ribosome-Free Supernates (Cell-Sap) from helpful to examine their production from placental polysomes Placenta and Ascites Tumor Cells. The preparation of the cell- in vitro. Such a cell-free system would also be useful for study- sap fraction from the placenta was carried out as described ing the biosynthesis of placental peptide hormones as a func- above except for the following: (a) there was no EDTA in tion of gestation. Furthermore, the possession of polysomes the homogenizing buffer and (b) the post-mitochondrial synthesizing these proteins is a prerequisite for isolating supernatant fluid was not treated with deoxycholate, and it mRNA's specific for HPL and HCG. was centrifuged at 250,000 X g for 2 hr in the absence of We have previously shown that ribosomes with relatively of sucrose solutions. The cell-sap fraction so obtained was high endogenous activity can be prepared from first and dialyzed overnight against the homogenizing buffer without third trimester placentas (4). In the present work, it was ob- EDTA and was stored in 200- to 300-Al aliquots in liquid served that in cell-free systems containing either first or third nitrogen. The ribosomal and cell-sap fractions derived from trimester polysomes, the placental hormone HPL was syn- Krebs II ascites tumor cells, except for the omission of the thesized and represented a substantial portion of total pro- preincubation step, were prepared as described elsewhere (5). tein synthesized. Assay for Protein Synthesis. Protein synthesis was assayed MATERIALS AND METHODS in 0.06 ml reaction mixtures composed of 30 mM Trist [I5S]Methionine was obtained from Amersham Searle. Human HCl (pH 7.5), 3.3 mM magnesium acetate, 70 mM KCl, 7 placental lactogen (95% pure) was purchased from Nutri- mM 2-mercaptoethanol, 1 mM ATP, 0.1 mM GTP, 0.6 mM CTP, 10 mM creatine phosphate, 0.16 mg/ml of creatine Abbreviations: HPL, human placental lactogen; HCG, human kinase, 40 AM each of 19 nonradioactive amino acids and chorionic gonadotrophin. 0.5 MM [35S]rnethionine (specific activity 200--300 Ci/mmol). 1322 Downloaded by guest on September 27, 2021 Proc. Nat. Acad. Sci. USA 71 (1974) Cell-Free Synthesis of Human Placental Lactogen 1323 TABLE 1. The incorporation of [tIS]methionine into proteins synthesized by placenta ribosomes Molecular weight [35S]Met incorp Ribosomes Cell-sap (cpm/0.06 ml) None Ascites 9,500 Placenta (F.T.) Ascites 625,000 Placenta (T.T.) Ascites 675,000 Placenta (T.T.) Placenta (F.T.) 545,000 Placenta (T.T.) None 10,800

The assays were performed as described in Materials and Meth- 47,500 ods. Where indicated, each reaction mixture contained the cell-sap equivalent of 80 and 130 ,ug of protein for ascites and placenta, respectively, and 50 ,g of ribosomal RNA. The placentas were from either the first trimester (F.T.) or third trimester (T. T.) In addition, 3 Ag of rat-liver tRNA was added to all reactions. The amount of ribosomes and cell-sap added will be noted in the appropriate experiment. Incubation was at 330 for 90 21,600- min. The reactions were stopped by the addition of either 0.2 ml of 0.1 N KOH or 25 gg of pancreatic ribonuclease. Incubation was continued for 20 min and 1 ml of 10% cold CC13COOH was then added. The mixture was cooled at 00 for 5 min and the precipitate was collected on a 0.45-jum pore size Millipore filter, washed three times with 3 ml each of 5% A P CC13COOH, dried and counted in a Packard liquid scintilla- tion counter. FIG. 1. Autoradiograph of [35S]methionine labeled-proteins Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis. synthesized in a cell-free system containing ribosomes from third trimester placenta (P) or Krebs II ascites tumor cells (A). The A five-fold scaled up reaction mixture containing [35S ]methio- cell-sap used was prepared from ascites tumor cells. Sodium do- nine was treated with 0.2 ml of 0.1 N KOH per 0.06 ml as de- decyl sulfate-polyacrylamide gradient (7-28%) gel electrophoresis scribed above. Following incubation, the mixture was ad- was carried out as described in Materials and Methods. Approx- justed to 10% in CClaCOOH, allowed to incubate for 10 min imately 100,000 cpm of CCl3COOH-precipitable material was ap- at 40, then centrifuged at 4000 rpm in a Sorvall SS-34 rotor plied to each lane. The molecular weight standards indicated for 10 min. The precipitate was washed once with cold 5% correspond to heavy chain of IgG (47,500) and HPL (21,600; 95% CC13COOH and twice with acetone to remove residual CC13- pure, Nutritional Biochemicals). COOH. The samples of in vitro products were prepared for analysis was cut out with scissors and placed in a centrifuge tube by dissolving them in 0.02 M Tris * HCl (pH 6.8) containing with 5-10 ml of H20. The suspension was incubated first for 1% sodium dodecyl sulfate, 1% 2-mercaptoethanol, 10% 2 hr at 370 and then overnight at 4°. The mixture was centri- glycerol, and 0.001% bromphenol blue, followed by heating at fuged at 10,000 X g for 15 min and the supernatant fluid was 100° for 1 min. The proteins were electrophoresed at 200 V decanted and saved. This fraction was lyophilized and the for 4-5 hr in a slab containing a linear 7-28% gradient of residue was taken up with 1 ml of water. Seven milligrams of polyacrylamide and subsequently stained and destained ac- purified unlabeled HPL were added; this mixture was dena- cording to procedures previously described (5). The dried tured, treated with trypsin, and then the digests were chroma- slabs were then exposed to x-ray film (Kodak RPR-54) for tographed and electrophoresed to yield tryptic maps as de- 1-3 days. scribed previously (5). The maps were exposed to x-ray film Tryptic Peptide Analysis. Ten to 20-fold scaled up reaction and stained with ninhydrin to localize the unlabeled peptides. mixtures were incubated for 90 min after which 25 jig of ribonuclease per 0.06 ml was added. The mixtures were then RESULTS incubated and processed for acrylamide gel electrophoresis as The high endogenous protein synthetic activity of placenta described above. A sample containing 2 to 4 X 106 cpm was ribosomes is shown by the incorporation of [35S]methionine distributed to nine 0.6-cm slots extending across the top of an (Table 1). As found previously (4), the cell-sap fraction from SDS-polyacrylamide slab gel. In the two lanes near the ex- ascites tumor cells is somewhat more active than homologous tremities of the gel 10 jig of purified HPL was applied. Im- cell-sap. It can also be seen that the endogenous activities mediately after electrophoresis the two end strips containing of the polysomes from first trimester and term are comparable. the HPL standard and one adjacent lane containing an aliquot To investigate the nature of the proteins synthesized, ribo- of the labeled mixture were sliced away from the main portion somes derived from third trimester tissue were incubated with of the gel and stained. The remaining untreated gel was cell-sap from ascites cells. The [3S]methionine labeled prod- immediately dried and then an autoradiograph was obtained ucts were analyzed by sodium dodecyl sulfate-acrylamide gel following a 10-hr exposure. The band corresponding to HPL electrophoresis (Fig. 1). There were apparently several pro- Downloaded by guest on September 27, 2021 1324 Biochemistry: Boime and Boguslawski Proc. Nat. Acad. Sci. USA 71 (1974)

Opp,or VW._ ._-W Molecular weight

..;

.j ......

..f _h _ Al.. : 47,500 - _A.*_ .. .:'.' : 4 .-.. - '. __

21,600-

i :

-w - - I..~~~~Ae 2~~~~~~~~~~~~~~~~~~~~~~~~~~ -T.T. F.T. -F.T. 5 --T.T. FIG. 3. Autoradiograph of [35S]methionine-labeled proteins e4} synthesized in the cell-free system containing ribosomes from first w trimester (F.T.) or third trimester (T.T.) placenta. Approxi- B mately 100,000 cpm of CCl3COOH-precipitable material was ap- ~* plied to lanes denoted F.T. and T.T. The third lane contained 50,000 cpm each of the F.T. and T.T. samples. FIG. 2. Two-dimensional tryptic fingerprint analyses of 7 mg of unlabeled carrier HPL and of a mixture of the labeled protein acid sequence of HPL (6, 7), tryptic hydrolysis should theoret- that electrophoresed at the same rate as HPL. The equivalent of ically yield 21 peptides. The ninhydrin-stained map displays about 200,000 cpm was mixed with the carrier. Panel A is the 19 major peptides and about 5 minor ones (Fig. 2B). autoradiograph of panel B, which has been stained with ninhy- As can be seen in Fig. 2A, there are more than six [13S]- drin. The ninhydrin-positive peptides of HPL which show the methionine-containing peptides present on the autoradio- same mobility as the labeled peptides are denoted by the dotted rings. graph. However, four labeled peptides have identical mobilities with peptides of purified HPL. These are denoted in teins synthesized in the cell-free system, many of which were Fig. 2B by the dotted lines surrounding the corresponding difficult to resolve from the background. Some of the same ninhydrin positive peptides. This pattern was consistently bands appeared when proteins synthesized by ribosomes from observed on maps of this protein in each of three independent ascites cells were also examined. However, at least one major experiments. distinct protein was synthesized only by the placental ribo- Peptide number 2 in Fig. 2 does not correspond precisely somes. This protein did not appear when pancreatic ribonu- to a ninhydrin positive peptide although it was always present clease was used to inhibit protein synthesis and migrated at on maps of the labeled protein. Peptide numbers 1 and 3 to 5 the same rate as purified HPL. apparently overlap with ninhydrin-stained peptides; how- More direct identification was obtained as follows: The ever, this correspondence is not reproducible and thus may band containing the protein was eluted from a preparative reflect nonspecific tryptic cleavages. The other labeled pep- gel, mixed with purified unlabeled HPL, digested with trypsin, tides do not have ninhydrin counterparts and must reflect and the resulting peptides analyzed by two dimensional peptides derived from other proteins eluted from the gel. - chromatography and electrophoresis. The fingerprints were The peptide spots were cut out of the map and their radio- subjected to autoradiography and then sprayed with ninhy- activity determined. The four labeled coincident peptides drin in order to localize the peptides derived from the purified constituted 50-60% of the total radioactivity among the carrier. The exposed x-ray film was then compared with the labeled areas detected by the autoradiograph. (The radio- ninhydrin-stained fingerprint. HPL contains six methionine activity at the origin was not included in this calculation.) residues (6, 7) each of which is distributed in a single tryptic The intensity of the labeling of the peptides probably re- peptide. Two of these methionine containing peptides contain flects their position in the HPL molecule since the amount of 21 and 29 amino acids and it is probable they would not mi- reinitiation in this system is not great (unpublished obser- grate in the solvent systems employed. Based on the amino- vation). Therefore, peptides nearest the carboxyl end of the Downloaded by guest on September 27, 2021 Proc. Nat. Acad. Sci. USA 71 (1974) Cell-Free Synthesis of Human Placental Lactogen 1325

protein will have the greatest radioactivity. This assumes large molecular-weight proteins were precipitated with HPL that the yield of each tryptic peptide is the same. Preliminary antibody. While the predominant protein synthesized in the sequence analyses performed in collaboration with Dr. Ruth cell-free system is HPL, it is possible that a precursor is Hogue Angeletti suggest that the least radioactive peptide of generated which is rapidly cleaved to HPL. Furthermore, a the four coincident ones seen in Fig. 2A is the peptide closest protein with a small variation in molecular weight might not to the amino terminus. be detected on polyacrylamide gels. Perhaps a more defini- The ninhydrin-stained fingerprint in Fig. 2B represents tive answer regarding this point can be obtained by iso- carrier HPL plus a much smaller mass of protein eluted from lating the HPL mRNA and translating it in a nonplacental the gel. It is unlikely that the eluted protein was sufficient to cell-free system. This approach might be fruitful since the give rise to visible spots (which might account for the coin- cleavage activity for some precursors may reside in the micro- cident peptides). This conclusion is supported by a finger- somal fraction of the cell (12). It was demonstrated that the print of the carrier by itself; this was indistinguishable from mRNA encoding for the light chain derived from a myeloma that in Fig. 2B. As a further control, labeled proteins syn- tumor was translated in a heterologous cell-free system and thesized by ascites polysomes were eluted from the same an apparent precursor of this protein was detected (11, 12). region of a preparative polyacrylamide gel as the labeled Since polysomes from placenta can synthesize a significant protein of Fig. 2 and fingerprinted. There was no corre- amount of a specific human placental hormone, HPL, they spondence between the labeled areas and the ninhydrin presumably contain correspondingly high levels of the specific positive peptides. These data strongly suggest that the band mRNA, which can be isolated and used as a reagent to study migrating at the same rate as HPL on the polyacrylamide gel the regulation of the biosynthesis of this peptide hormone. was in fact HPL. of the proteins synthesized by We are grateful to Dr. Dolph Hatfield for his generous gift of Polyacrylamide gel analysis tRNA and to Dr. Ruth Hogue Angeletti for performing some se- first trimester ribosomes also revealed a band migrating as quence analyses on the tryptic peptides and to Charles Lawrence HPL (Fig. 3). Trypsin treatment of this protein yielded a for his helpful discussion. The authors would also like to thank radioactive fingerprint similar to that obtained with term Kathy Neely for her excellent assistance in preparing this manu- ribosomes. script. This work was aided in part by a grant from the Popula- For quantitating levels of the protein in the HPL region, the tion Council (#M73, 135), and NIH Grant #Am-16865. band was cut out from the dried gel and the radioactivity de- 1. Szabo, A. & Grimaldi, R. D. (1970) Advan. Metab. Disord. 4, termined. In the case of third trimester ribosomes the band 185-228. contained about 10% of the total amount of radioactivity 2. Kaplan, S. L., Gurpidie, E., Sciarra, J. J. & Grumbach, M. M. J. Clin. Endocrinol. Metab. 28, 1450-1460. in case (1968) applied to the gel; the of first trimester ribosomes the 3. Grumbach, M. M., Kaplan, S. L., Sciarra, J. J. & Burr, I. M. figure was about 5%. These values somewhat overestimate (1968) Ann. N. Y. Acad. Sci. 148, 501-531. the HPL synthesized since, while the band is very discrete, a 4. Boime, I., Corash, L. & Gross, E. (1974) Pediat. Res., in certain amount of other labeled protein comigrates in this press. region, as shown above. 5. Boime, I. & Leder, P. (1972) Arch. Biochem. Biophys. 153, 706-713. DISCUSSION 6. Sherwood, L. M., Handwerger, S., McLaurin, W. l). & Lan- ner, M. (1971) Nature New Biol. 233, 59-69. The (overestimated) value for the percentage of HPL syn- 7. Li, C. H., Dixon, J. S. & Chung, D. (1973) Arch. Biochem. thesized in the cell-free system is in satisfactory agreement Biophys. 153, 95-110. with the data obtained by Friesen, et al. (8) who showed that 8. Friesen, H., Belanger, C., Guyda, H. & Hwang, P. (1972) in Lactogenic Hormones, Ciba Foundation (Churchill Living- in term placenta slices, HPL constitutes 2-4% of the total ston, London), pp. 83-110. protein synthesized. The results are also consistent with in 9. Spellacy, W. N. (1972) in Lactogenic Hormones, Ciba Founda- vivo findings which indicate that in the third trimester of tion (Churchill Livingston, London), pp. 223-239. pregnancy the placenta secretes larger quantities of the protein 10. Friesen, H. G., Suwa, S. & Pare, P. (1969) Recent Progr. into maternal serum than in the first trimester (3, 9). Hormone Res. 25, 161-205. 11. Swan, D., Aviv, H. & Leder, P. (1972) Proc. Nat. Acad. Sci. It has been proposed that HPL might be derived from a USA 69, 1967-1971. precursor protein (8, 10). This was based on the finding that 12. Milstein, C., Brownlee, G. G., Harrison, T. M. & Mattbews, in a mixture of labeled proteins from term placenta slices, some M. B. (1972) Nature New Biol. 239, 117-120. Downloaded by guest on September 27, 2021