J. Biochem. 98, 483-492 (1985)
Isolation of Human, Swine, and Rat Prepepsinogens and Calf Preprochymosin, and Determination of the Primary Structures of Their N112-Terminal Signal Sequences1
Yoshikazu ICHIHARA,2 Kazuhiro SOGAWA,3 and Kenji TAKAHASH12
Department of Biochemistry, Primate Research Institute, Kyoto University, Inuyama, Aichi 484
Received for publication, April 1, 1985
The total RNAs were extracted from human, swine, rat, and calf gastric mucosae, and translated in vitro in the presence of radiolabeled amino acids using a wheat germ cell-free system. Upon sodium dodecyl sulfate (SDS)-polyacrylamide gel elec trophoresis of the translation products, a protein band with a molecular weight of about 43,000 was obtained in each case as one of the major products. These prod ucts could be specifically immunoprecipitated with a corresponding anti-pepsinogen or anti-chymosin antiserum. Radiosequence analysis of these translation products purified by SDS-polyacrylamide gel electrophoresis showed that each of them is a precursor form, i.e., prepepsinogen or preprochymosin, having an amino-terminal extension peptide (signal sequence) comprising 15 (human and swine) or 16 (rat and calf) amino acid residues. The primary structures of these signal sequences were determined to be as follows:
These signal sequences share common characteristics with those of other pre-secretory proteins, i.e., the presence of positive charges in the NH2-terminal region, hydrophobic amino acid clusters in the interior part, and amino acids with short side chains at the site of cleavage by the signal peptidase.
1 This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan and grants from the Ishida Foundation and The Naito Foundation. 2 Present address: Department of Biophysics and Biochemistry, Faculty of Science, The University of Tokyo , Hongo, Bunkyo-ku, Tokyo 113. 3 Present address: Department of tsiocnemistry, Lancer Institute, Japanese t-ounaation for Lancer xesearcn, Kami-Ikebukuro, Toshima-ku, Tokyo 170. Abbreviation: SDS, sodium dodecyl sulfate.
Vol. 98, No. 2, 1985 483 484 Y. ICHIHARA, K. SOGAWA, and K. TAKAHASHI
In the biosynthesis of eukaryotic secretory pro Ci/mmol), L-[2,6-3H]tyrosine (60 Ci/mmol) and a teins, the nascent proteins usually contain an NH. 14C-methylated protein mixture (myosin, phospho terminal extension peptide, the so-called signal se rylase-b, bovine serum albumin, ovalbumin, car quence, composed of 15 to 30 amino acid residues bonic anhydrase, and lysozyme) (Amersham); (1). This signal sequence is thought to be essen [14C]iodoacetamide (24.4 mCi/mmol), L-[2,3-3H] tial for the vectorial transport of the nascent poly arginine (20 Ci/mmol), L-[2,3-3H]tryptophan (17.4 peptide chain across the membranes of the rough Ci/mmol), EN3HANCE and Aquasol-2 (New Eng endoplasmic retieulum into the cisternal space (2). land Nuclear); Protein A-Sepharose CL-4B (Phar Pepsinogen is the zymogen of pepsin, a major macia, Sweden); trifluoroacetic acid, phenyliso gastric aspartic protease, and is known to be auto thiocyanate and N-, N'-dimethyl allylamine (Wako catalytically converted to pepsin under acidic con Pure Chemicals Ind., Tokyo); Freund adjuvant ditions. It can be classified into three major (Difco Laboratories, Detroit); and sperm whale groups, i.e., pepsinogen A, pepsinogen C (or pro myoglobin, porcine pepsinogen, and calf chymosin gastricsin) and prochymosin (or neonatal pepsino (Sigma). All other chemicals were of reagent gen). Since pepsinogen is a typical secretory pro grade and were used without further purification. tein, it is also expected to be synthesized as a Wheat germ was kindly provided by Nisshin Mill precursor form having an NH,-terminal signal Corporation, Nagoya. Rabbit anti-Wistar-Ima peptide. Indeed, we previously reported the oc michi rat pepsinogen antiserum (4) and anti-human currence of rat prepepsinogen, the biosynthetic peptinogen A antiserum (5) were kindly supplied precursor of rat pepsinogen, and the partial amino by Drs. C. Furihata and K. Miki, respectively. acid sequence of its signal peptide (3). Specimens of human stomachs obtained at surgery In order to characterize further the biosyn from gastric ulcer or gastric cancer patients were thetic precursors of the pepsinogen family, we have kindly provided by Dr. M. Ukai. Fresh swine isolated prepepsinogens and preprochymosin from and calf stomachs were obtained from a local translation mixtures of the total RNAs from hu slaughterhouse. Wistar-Imamichi male rats were man, swine, rat, and calf gastric mucosae in a purchased from the Imamichi Institute for Animal wheat germ cell-free system, and determined the Reproduction (Saitama. Janan). NH2-terminal amino acid sequences of these pre Purification of Pepsinogen and Prochymosin proteins by radiosequence analysis. The results Rat pepsinogen was purified by the method de established the complete amino acid sequences scribed by Muto and Tani (6). Human pepsinogen of the 15-residue signal sequences of human and and calf prochymosin were purified by a procedure swine prepepsinogens and the 16-residue signal similar to that used previously (7), except that sequences of rat prepepsinogen and calf prepro DEAE-Sepharose CL-6B was used at the step of chymosin. the 2nd DE-32 cellulose column chromatography. Swine pepsinogen (Sigma) was used without fur MATERIALS AND METHODS ther purification. The potential proteolytic activ ities of pepsinogen and prochymosin were deter Materials-The following materials were pur mined by the method described previously (8). chased from the respective sources indicated. L The purity of the enzymes was examined by poly [2,3-3H]Alanine (36 Ci/mmol), L-[2,3-3H]aspartic acrylamide disc gel electrophoresis according to acid (9.2 Ci/mmol), L-[35S]cysteine(100 Ci/mmol), Ornstein (9) and Davis (10). To prepare [14C] L-[G-3H]glutamic acid (29 Ci/mmol), L-[3,4-3H(N)] carboxamidomethylated pepsinogens and prochy glutamine (46.8 Ci/mmol), [2-3H]glycine (30 Ci/ mosin, the proteins were treated with [14C]iodo mmol), L-[2,5-3H]histidine (53 Ci/mmol), L-[4,5 acetamide essentially according to Waxdal et al. (11) 3H]isoleucine (81 Ci/mmol) , L-[4,5 -3H]leucine(130 except for the use of 8 M urea instead of 6 M guani Ci/mmol), L-[4,5-3H]lysine-HCl (40 Ci/mmol), L dine HCI. [methyl3H]methionine (15 Ci/mmol), L-[3bS]meth Preparation of Antiserum-Anti-swine pep ionine (1,120 Ci/mmol), L-[2,4,6-3H]phenylalanine sinogen antiserum was prepared as described by (77 Ci/mmol), L-[5-3H]proline (15.3 Ci/mmol), L Kushner et al. (12). Anti-chymosin antiserum was [3-3H]serine (28 Ci/mmol), L-[3,4(n)3H]valine (37 prepared in a similar manner.
J. Biochenl. SIGNAL SEOUENCFS OF PRFPFPS1NO(FNS ANfl PRFPROCHYM(OS1N ARS
Extraction of Total Gastric RNA-The total tate was washed 3 times with 500 Id of the above
RNA was extracted from gastric mucosae by the buffer. The precipitate was mixed with 50ƒÊ of phenol-SDS method (13) with slight modifications. 0.0625 M Tris-HCl buffer, pH 6.8, containing 2 % Gastric mucosae were rapidly frozen with solid SDS, 20% glycerin, and 5 % 2-mercaptoethanol, carbon dioxide immediately after excision and and then boiled for 5 min. The supernatant was stored at 80"C until use. The frozen tissue was subjected to SDS-polyacrylamide gel electropho cracked into pieces and to these were added 10 resis. vol. of 50 mm Tris-HCl buffer, pH 7.5, containing Electrophoretic Analysis of Cell-Free Transla
1% SDS, 5 mm EDTA, and 25 mm NaCl, and 10 tion Products-Cell-free translation products were vol. of phenol saturated with the above buffer. analyzed by SDS-polyracrylamide slab gel electro The tissue was then homogenized 4 times, for 10 s phoresis as previously described (3) according to each, with a Waring blender at the maximum speed. Laemmli (17) using gels containing 15 % acrylamide Each mixture was stirred for 20 min and then 10 (running gel). Fluorographic exposure of the vol. of chloroform was added. After stirring for a gels was carried out as described by Bonner and further 20 min the mixture was centrifuged at 2,500 Laskey (18). x g for 10 min and the aqueous phase was saved. Isolation of Cell-Free Translation Products
The procedure was repeated at least three more from Polyacrylamide Gels-After the electropho times with 10 vol. of phenol : chloroform (1 : 1) resis, the gels were slightly stained by incubation
saturated with the homogenizing buffer until no for 5 min at room temperature with 50°o acetic
protein was observed at the interface. RNA in acid containing 0.125% Coomassie brilliant blue the aqueous phase was precipitated by adding 10 M and 25% ethanol, and destained with 7.5% acetic
LiCI (final concentration, 2 M), and keeping over acid containing 5 % methanol. The gels were cut night at -20•Ž. After centrifugation at 10,000 into slices of 1 mm thickness with a razor blade. x g for 20 min, the RNA pellet was dissolved in A part of each slice was dissolved at 50•Ž in 0.1
water and precipitated twice in the same manner ml of 31 % hydrogen peroxide solution in a scintil as above. Approximately 15 to 60 A260 units of lation vial. One ml of an Aquasol-2 scintillation
RNA were obtained from 1 g of gastric mucosa. cocktail was added to each vial and the radio The RNA concentration was estimated assuming activity was measured with an Aloka scintillation that I mg of RNA equals 25 A260 units. counter LSC-671. The gel slices containing de
Cell-Free Protein Synthesis-Wheat germ ex sired products were submitted to electroelution by tract was prepared according to the method of the method of Allington et al. (19) with some Roberts and Paterson (14) with slight modifications modifications. The eluted translation products
(15). A single radiolabeled amino acid was used were dialyzed against distilled water and then in each translation experiment. One ml of each lyophilized. In the case of translation products
translation mixture corresponding to 100 ƒÊCi of labeled with [35S]eysteine, the samples were car the labeled amino acid was used for the sequence boxamidomethylated by the above method (11) determination. after lyophilization. The carboxamidomethylated Immunoprecipitation-Nascent forms of pep samples were dialyzed against distilled water and
sinogens or prochymosin synthesized in the cell then lyophilized for amino acid sequence deter free system were immunoprecipitated using rabbit mination.
antiserum and immobilized Staphylococcus aureus Amino Acid Sequence Determination-Each sample was mixed with 6 mg of sperm whale apo protein A (16). About 10ƒÊl of 5-fold diluted anti pepsinogen or anti-chyrnosin rabbit serum was myoglobin as carrier and subjected to NH2-termi added to 50ƒÊ1 of the cell-free translation system. nal sequence determination by the manual Edman Each mixture was incubated at 25•Ž for 30 min, method (20) with slight modifications (21). The
and then mixed with 20ƒÊl of a suspension of total radioactivity of the phenylthiohydantoin de Protein A-Sepharose CL-4B suspended in an equal rivative fraction was measured with the scintillation
volume of 10 mm Tris-HCl buffer, pH 7.5. After counter using Aquasol-2. Since only one radio incubation for 30 min at 25•Ž, the suspension was labeled amino acid was used at a time, the position
centrifuged at 1,700 •~ g for 5 min. The precipi of each amino acid was assigned simply by count
Vol. 98, No. 2, 1985 486 Y. ICHIHARA, K. SOGAWA, and K. TAKAHASHI
ing the radioactivity. prochymosin (Fig. 1, C and D, lanes 2 and 3). These results indicated that rat pepsinogen and RESULTS calf prochymosin were synthesized as larger molec Identification of Prepepsinogens and Prepro ular weight precursors, namely, prepepsinogen and chymosin by SDS-Polvacrylamide Gel Electropho preprochymosin, respectively. On the other hand, resis-When the total RNA extracted from the by means of SDS-polyacrylamide gel electropho gastric mucosa of each animal was translated in resis alone, we could not positively identify the the wheat germ cell-free system, and the radio precursor forms of the human and swine pepsino labeled translation products were examined by gens (Fig. 1, A and B, lanes 2 and 3), since in each SDS-polyacrylamide gel electrophoresis, the results case the translation product and the authentic shown in Fig. 1 (A-D, lanes 1) were obtained. In pepsinogen were practically indistinguishable as each case, a clear band was obtained at a position to mobility. corresponding to a molecular weight of about NH2-Terminal Amino Acid Sequence of Pre 43,000, and only the radiolabeled protein in this pepsinogens and Preprochymosin-The radiolabeled band was immunoprecipitated with the correspond translation products of about a molecular weight ing anti-pepsinogen or anti-chymosin antiserum of 43,000 purified by preparative SDS-polyacryl (Fig. 1, A-D, lanes 2). Furthermore, the posi amide gel electrophoresis (data not shown) were tion of this band was shown to correspond to a subjected to manual Edman degradation and their molecular weight slightly higher than those of the NH,-terminal amino acid sequences were deter corresponding carboxamidomethylated authentic mined. zymogens in the case of rat pepsinogen and calf Figure 2 shows the results of these radiose
Fig. 1. Analysis by SDS-polyacrylamide gel electrophoresis of cell-free translation
products directed by total RNA from human, swine, rat, and calf gastric mucosae. The total RNA was translated in the cell-free translation system (50ƒÊ1) of wheat germ
at 20•Žfor 90 min using [35S]methionine. The radiolabeled cell-free translation
products were immunoprecipitated using specific rabbit antiserum against the cor responding pepsinogen or chymosin and Protein A Sepharose CL-4B. The immuno
precipitate was analyzed by SDS-polyacrylamide gel (15%) electrophoresis followed by fluorography. A, human; B, swine; C, rat; D, calf. Lane 1, the translation
products of total RNA (20ƒÊg); lane 2, immunoprecipitates of the translation products using rabbit anti-pepsinogen or anti-chymosin antiserum; lane 3, 14C-labeled authentic
pepsinogen or prochymosin from the same animal. The molecular weight standards were: phosphorylase-b (92,500), bovine serum albumin (69,000), ovalbumin (46,000),
carbonic anhydrase (30,000) and lysozyme (14,300).
J. Biochem. SIGNAL SEQUENCES OF PREPEPSINOGENS AND PREPROCHYMOSIN 487
Fig. 2. NH,-terminal amino acid sequence analysis of prepepsinogens and preprochymosin by Edman degradation. Prepepsinogens and preprochymosin were separately labeled with various radioactive amino acids in the wheat germ cell-free translation system (1 ml), using one radioactive amino acid at a time. Each radiolabeled prepep sinogen or preprochymosin sample was mixed with 6 mg of sperm whale apomyoglobin as carrier and subjected to (continued)
Vol. 98, No. 2, 1985 488 Y. ICHIHARA, K. SOGAWA, and K. TAKAHASHI
quence analyses, and Fig. 3 the NH,-terminal corresponded to Ile-13 (data not shown). amino acid sequences deduced from these results. In the case of the calf radiolabeled product, In the case of the human radiolabeled product, we we found alanine at step 17, glutamic acid at step found lysine at steps 19, 25, and 26, valine at step 18, arginine at step 21, leucine at steps 24 and 20, and leucine at steps 22 and 28, corresponding 30, and lysine at steps 26 and 28, which corre to Lys-4, 10, and 11, Val-5, and Leu-7 and 15, sponded to Ala-1, Glu-2, Arg-5, Leu-8 and 14, and respectively, of human pepsinogen (Figs. 2A and Lys-10 and 12 of prochymosin, respectively (Figs. 3). Therefore, it became clear that the human 2D and 3). Thus the radiolabeled product was radiolabeled product was a prepepsinogen contain deduced to be preprochymosin containing a signal ing a signal sequence of 15 amino acid residues sequence of 16 amino acid residues as in the case at its NH2-terminus. of rat prepepsinogen at its NH,-terminus. When In the case of the swine radiolabeled product, the translation was carried out using radiolabeled the leucine at step 16, valine at steps 17, 19, and 22, arginine, a high level of radioactivity, presumably lysine at steps 18, 24, and 25, and serine at step 26 due to contamination, was seen mainly at the first corresponded to Leu-1, Val-2, 4, and 7, Lys-3, 9, step of the radiosequencing (Fig. 2D). This ac and 10, and Ser-11, respectively, of swine pepsino tivity could be decreased by repeating three times gen (Figs. 2B and 3). Thus, the swine radio the precipitation of the translation products with labeled product was also shown to be a prepep ethanol before polyacrylamide gel electrophoresis, sinogen containing a 15-residue signal sequence at and thus the presence of arginine at step 2 became its NH2-terminus. clear (Fig. 2D, insert). At step 2 a smaller amount As for the rat prepepsinogen, we previously of lysine was also detected. This is probably due reported its occurrence and NH,-terminal partial to contamination by bovine prepepsinogen which amino acid sequence (3). In the present study, the presumably has an NH,-terminal Met-Lys-Trp complete amino acid sequence of its signal peptide sequence like other prepepsinogens. Indeed, when (16 amino acid residues) was determined to be as preprochymosin prepared in the presence of radio shown in Fig. 3. The serine at steps 17 and 28, labeled lysine or tryptophan was purified using leucine at steps 18, 19, and 23, valine at step 21, anti-chymosin antiserum and Protein A-Sepharose proline at step 22, lysine at steps 25 and 27, and to remove contaminating prepepsinogen and radio methionine at step 26 of the prepepsinogen coin sequenced, no significant radioactivity was detected cided with Ser-1 and 12, Leu-2, 3, and 7, Val-5, at steps 2 and 3 (data not shown). Pro-6, Lys-9 and 11, and Met-10 of the pepsino gen, respectively (Figs. 2C and 3). Furthermore, the isoleucine at step 29 of the prepepsinogen
Fig. 2 (continued) manual Edman degradation. The resulting phenylthiohydantoin fraction was mixed with Aquasol-2 to measure the radioactivity with an Aloka scintillation spectrometer LSC-671. Data are shown only for those amino acids which were positively identified in the sequences. The positions of each amino acid residue identified are numbered. A: Human prepepsinogen; [3H]methionine (25,000), [3H]lysine (10,700), [3H]tryptophan (240,000), [3H]leucine (84,000), [3H]glycine (56,600), [3H]valine (7,100), [3H]alanine (15,000), [3H]serine (11,600), [3H]glutamic acid (110,700) and [35S]cysteine(53,400). B: Swine prepepsinogen; [3H]methionine (233,000), [3H]lysine (12,800), [3H]tryptophan (6,600), [3Hlleucine (87,700), [3H]serine (73,100), [3H]valine (69,400), [3H]glutamic acid (9,100) and [31S]cysteine (69,400). C: Rat prepepsinogen; [34S]methionine (460,000), [3H]lysine (50,600), [3H]tryptophan (69,400), [3H]valine (65,000), [3H]alanine (55,400), [3Hlleucine (287,000), [36S]cysteine (36,000), [3H]proline (34,900), [3H]glutamic acid (11,600) and [3H]serine (81,000). D: Calf preprochymosin; [3H]methionine (150,000), [3H]arginine (260,000 and 15,000 (insert)), [36S]cysteine (10,800), [3H]leucine (78,500), [3H]valine (17,000), [3H]alanine (18,900), [3H]phenyl alanine (14,000), [3H]serine (42,500), [3H]glutamine (27,900), [3H]glycine (7,700), [3H]glutamic acid (6,600) and [3H]lysine (6,400). The insert in the case of [3H]arginine-labeled calf preprochymosin shows the result obtained when the translation products were precipitated three times with ethanol before polyacrylamide gel electrophoresis to decrease the background. Each value in parentheses denotes the total radioactivity (cpm) in each sample analyzed.
J. Biochem. Fig. 3. The NH2-terminal amino acid sequences of prepepsinogens and preprochymosin. The NH2-terminal sequence of each mature zymogen is shown under the corresponding precursor sequence. Common residues between each precursor and its mature form are enclosed.
Vol. 98, No. 2, 1985 490 Y. ICHIHARA, K. SOGAWA, and K. TAKAHASHI
sinogen) groups, respectively. The primary struc
DISCUSSION tures of the signal sequences appear to reflect this difference. All these signal sequences, however, As described in the "RESULTS" section, human, share similar characteristics with those of other swine and rat pepsinogens, and calf prochymosin preproteins. The NH2-terminus is methionine, have been shown to be synthesized initially as followed by a basic residue which would help corresponding prepepsinogens and preprochymo anchor the signal peptide on the surface of the sin, respectively, each having a signal sequence of membranes. Most of the interior of the signal 15 or 16 residues at the NH2-terminus. On SDS sequence is occupied by hydrophobic residues, and polyacrylamide gel electrophoresis, rat prepepsino charged amino acid residues are lacking except for gen and calf preprochymosin appeared to have glutamic acid at position 2. Position 1 is slightly higher molecular weights than the corre occupied by an amino acid residue with a short sponding mature zymogens. This is consistent side chain, such as cysteine, alanine and glycine, with the above results. On the other hand, hu which is involved in the specific cleavage by the man and swine prepepsinogens were indistinguish signal peptidase. We previously deduced the pri able in mobility from the corresponding pepsino mary structure of the signal sequence of human gens. The reason for this discrepancy is not clear prepepsinogen A by elucidating the nucleotide at present. The number of residues in the signal sequence of its genomic DNA (25). The present sequences varies significantly among various pre results are in full accord with the results of DNA proteins, and the 15-residue sequence is, to our sequencing. More recently, the primary structure knowledge, the shortest one among the signal of the signal sequence of calf preprochymosin was sequences hitherto known. Besides human and independently elucidated from the results of nu swine prepepsinogens,pre-ƒÀ-casein (22), pretryp cleotide sequence studies on its complementary sinogen (23), prechymotrypsinogen (24) and pre DNA (26, 27). Our present results coincide com amylase (24) have also been reported to have a pletely with this. 15-residue signal sequence. At least a stretch of Table I shows the secondary structure predic 15 residues may be necessary for a preprotein tc TABLE I. Predictions of secondary structures of the pass through the membranes of the rough endo signal sequences of prepepsinogens and preprochymosin. plasmic reticulum. Indeed, the length of an ex tended 15-residue sequence roughly coincides with
the width (ca. 60 A) of the membranes. The amino acid sequences of the four signal sequences are similar to one another as shown in Fig. 4. Especially the signal sequences of human and swine prepepsinogens are alike; this is con
sistent with the fact that both zymogens belong to '-Predictions for the a -helix (
the pepsinogen A group. On the other hand, rat (
Fig. 4. Comparison of the sequences of prepepsinogens and preprochymosin. The amino acid sequences are aligned so as to maximize the homology. The amino acid residues common to at least three species are enclosed.
J. Biochem. SIGNAL SEQUENCES OF PREPEPSINOGENS AND PREPROCHYMOSIN 491
TABLE II. Amino acid sequence homology among the 2. Blobel, G. & Dobberstein, B. (1975) J. Cell Biol. domains of human and swine prepepsinogens and calf 67,835-851 preprochymosin. Sequence data for zymogens: human 3. Ichihara, Y., Sogawa, K., & Takahashi, K. (1982) (25), swine (29, 30) and calf (31, 32). J. Biochem. 92, 603-606 4. Furihata, C., Saito, D., Fujiki, H., Kanai, Y., Matsushima, T., & Sugimura, T. (1980) Eur. J. Biochem. 105, 43-50 5. Ichinose, M., Miki, K., Furihata, C., Kageyama, T., Hayashi, R., Niwa, H., Oka, H., Matsushima, T., & Takahashi, K. (1982) Clin. Chin. Acta 126, 183 191 6. Muto, N. & Tani, S. (1979) J. Biochem. 85, 1143 tions for these signal sequences according to the 1149 method of Chou and Fasman (28). The results 7. Kageyama, T. & Takahashi, K. (1980) J. Biochem. show that the central parts of these signal sequences 88, 571-582 exhibit a significantly higher potential for ƒÀ-struc 8. Kageyama, T. & Takahashi, K. (1980) J. Biochem. ture formation (
Vol. 98, No. 2, 1985 492 Y. ICHIHARA, K. SOGAWA, and K. TAKAHASHI
26. Harris, T.J.R., Lowe, P.A., Lyons, A., Thomas, J.P. (1973) Proc. Natl. Acad. Sci. U.S. 70, 3437 P.G., Eaton, M.A.W., Millican, T.A., Patel, T.P., 3439 Bose, C.C., Carey, N.H., & Doel, M.T. (1982) 30. Foltmann, B. & Pedersen, V.B. (1977) in Acid Pro Nucl. Acids Res. 10, 2177-2187 teases: Structure, Function, and Biology (Tang, J., 27. Moir, D., Mao, J., Schumm, J.W., Vovis, G.F., ed.) pp. 3-22, Plenum Press, New York Alford, B.L., & Taunton-Rigby, A. (1982) Gene 19, 31. Foltmann, B., Pedersen, V.B., Kauffman, D., & 127-138 Wybrandt, G. (1979) J. Biol. Chem. 254, 8447-8456 28. Chou, P.Y. & Fasman, G.D. (1978) Annu. Rev. 32. Foltmann, B., Pedersen, V.B., Jacobsen, H., Kauff Biochem. 47, 251-276 man, D., & Wybrandt, G. (1977) Proc. Natl. Acad. 29. Tang, J., Sepulveda, P., Marciniszyn, J., Jr., Chen, Sci. U.S. 74, 2321-2324 K.C.S., Huang, W.-Y., Tao, N., Liu, D., & Lanier,
J. Biochem.