Aug 12, 1986 - Interferon A Using Fast Atom Bombardment Mass Spectrometry*. (Received for ... rated by HPLC on a reversed-phase column into several.
Vol. 262, No. 8, Issue of March 15, pp. 3541-3547 1987 Printed in L~.s.A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1987 by The American Society of Biological Chemists,lnc.
Chemical Characterization of Recombinant Human Leukocyte Interferon A Using Fast Atom Bombardment Mass Spectrometry* (Received for publication, August 12, 1986)
Toshifumi TakaoS, Makoto KobayashiO, Osamu Nishimurag, and Yasutsugu ShimonishiS1 From the $Institute for Protein Research, Osaka University, Suita, Osaka 565, Japan and $BiotechnologyLaboratories, Central Research Division, Takeda Chemical Industries, Ltd., Yodogawa-ku, Osaka 532, Japan
Proteolytic digests of biologicallyactive fractions of fractions with similar magnitudes of biological activity’ (15). recombinant human leukocyte interferonA expressed These findings suggest that structurally similar proteins are in large quantities in Escherichia coli were analyzed formed during the production of recombinant proteins in large by fast atombombardmentmassspectrometryand quantities in E. coli. high-performance liquid chromatography. The values In the present study, we examined the amino acid sequence observed in the mass spectra of digests of the major of rIFN-aA produced in large quantities in E. coli by FAB fraction of recombinant human leukocyte interferon A mass and HPLC analyses of the proteolytic digests of rIFNwith trypsin and Staphylococcus aureus protease V8 aA. The results provided evidence that the major product in accounted for93%of the amino acid sequences of hu- the preparation of this protein had the primary structure man leukocyte interferonA predicted from the nucleo-predicted from the gene encoding the protein and the same tide sequence of the gene encoding the protein, indi- disulfide linkages as described previously (16) but that minor cating that the major fraction of recombinant human leukocyte interferon A was expressed with the same components weremodified at the amino group of the Namino acid sequence as that translated from the nu- terminal amino acid residue or the thiol groups of Cys-1 and/ or Cys-98. This is the first example of detailed analysis of the cleotidesequenceofthegeneencodingtheprotein. primary structure of a recombinant protein produced in large Mass spectrometry of proteolytic digests two of minor fractions of recombinant human leukocyte interferon quantities in E. coli. A and mass and amino acid analyses of their highMATERIALS AND METHODS3 performance liquid chromatography fractions showed that the amino group of the N-terminal amino acid RESULTS AND DISCUSSION residue of interferon was in part acetylated, and the Cys-1 and Cys-98 residues were oxidized to cysteic One major and two minor fractions with IFN-aA activity acid or linked to glutathione. These findings suggest were isolated from E. coli with the IFN-aA gene by HPLC that amino acid residues in recombinant proteins pre-(named Mf-1, Mf-2, and Ms in the order of their elution on pared in large quantities in E. coli are modified post- HPLC; see “Materials and Methods” and RRf. 15), as shown transiationaIIy. in Fig. 1.To examine the primary structures of these proteins, we first measured their trypticdigests by FABmass spectrometry and compared them. Fig. 2 shows the FAB mass spectra in the range from 300 to 3600 amu of the tryptic digest of MfInterferons are a family of proteins characterized by their 1 before and after digestion with Staphylococcus aureus proability to induce antiviral activity in target cells (1-3). They tease V8. The observed mass values of the digests were corinhibit cell proliferation and modulate the immune response related with the theoretical mass values calculated from the (4); hence, they may have clinical value intreating viral amino acid sequence predicted from the nucleotide sequence infections and malignancies. Recently, the cloning and of the gene encoding human leukocyte interferon A ( 3 ,as expression of various human interferon genes in Escherichia summarized in Tables 1 and 2. The mass spectra accounted coli (5-11) has made it possible to produce the proteins in for all the IFN-nA sequences except the regions from 122 large quantities and to use them for clinical investigations. (tyrosine) to 125 (arginine), 132 (glutamic acid) to 133 (lyBiosynthetic human IFN-aA’ exhibits antiviral and antipro- sine), and 163 (serine) to 165 (glutamic acid) (Fig. 3). The liferative activities comparable to those of partially purified last two ofthese three regions couldnot be identified because natural IFN-(YA(12). However, purified rIFN-aA consists of their molecular weights could not be detected under the oligomers and two monomeric components that have similar, present conditions for mass measurement. The peptide from but not identical, mobilities on sodium dodecyl sulfate-poly- position 122 to 125 was not observed in the spectra in Fig. 2 acrylamide gel electrophoresis under unreduced conditions (13, 14). In addition, a preparation of rIFN-aA can be sepaNakagawa, S., Honda, S., Sugino, H., Kusumoto, S., Sasaoki K., rated by HPLC on a reversed-phase column into several Nishi, K., and Kakinuma, A. (1986) Biochim. Biophys. Acta, submitted for publication. Portions of this paper (including “Materials and Methods,” Figs. the payment of page charges. This article must therefore be hereby 4 and 6-8, and Tables 3 and 4) are presented in miniprint at theend marked “aduertisement” in accordance with 18 U.S.C. Section 1734 of this paper. Miniprint is easily read with the aid of a standard solely to indicate this fact. magnifying glass. Full size photocopies are available from the Journal V To whom correspondence should be addressed. of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. The abbreviations used IFN-&A, human leukocyte interferon A Request Document No. 86M-2782, cite the authors, and include a rIFN-aA, recombinant human leukocyte interferon A; FAB, fast atom check or money order for $3.60 per set of photocopies. Full size bombardment; HPLC, high-performance liquid chromatography; photocopies are also included in the microfilm edition of the Journal amu, atomic mass unit. that is available from Waverly Press.
* The costs of publication of this article were defrayed in part by
3541
Recombinant Human Leukocyte Interferon A
3542
for some unknown reason, but it was identified by HPLC as described below. The observed mass value 2244.6 corresponding to thatof T’3-T12 or T3-T’12 with signals at m/z = 910.2, 1038.4,1209.3, and 1337.4, which corresponded to T13, T3, T’12,and Tlz, respectively, as the fragments of T’3-T12and T3-T’12in the tryptic digest (upper spectrum in Fig. 2) indi-
cated that Cys-29 in TI3 or T3 was linked to Cys-138 in T12 or T’12by a disulfide bond in Mf-1 (16), because a peptide containing a cystine residue gaveion signals of a peptide linked by a disulfide bond and its fragments cleaved at the disulfide linkage on FAB mass spectra, as reported in our previous paper (17). This conclusion was supported by the observation of signals at m/z = 910.4,1038.4,1762.7, and 1890.7, corresponding to T’3, T3,T’3-T’12(V)and T,-T’,,(V) or T’3-T12(V),respectively, in the tryptic and S. aureus protease V8 digest (lower spectrum in Fig. 2). Furthermore, the observation of signals at m/z = 1313.5, 1645.6, 2386.3, and 2957.1, corresponding to T1, T7(VZ), T1-T7(V3)and Tl-T7(V2), respectively, in the tryptic and S. aureus protease V8 digests (lower spectrum in Fig. 2) provided evidence that Cys-1 in T1 was linked to Cys-98 in T7-V2 or T7-V3 by a disulfide bond in Mf-1. These results indicate that Mf-1 is a protein with the same amino acid sequence as thatpredicted from the nucleotide sequence of the clonedcDNA (7) and with the same disulfide linkages as reported previously (16). To obtain information on the structuraldifferences between Mf-1 on the one hand and Mf-2 and Ms on the other, we measured the FAB mass spectra of enzymatic digests of Mf2 and Ms. The spectra of Mf-2 and Ms were similar to those of Mf-1 but differed from those of Mf-1 in the range from 1300 to 1700 amu (Fig. 4). The signal of T1was very weakin the mass spectrum of the digest of Mf-2 with trypsin and S. aurew protease V8 (upper spectrum in Fig. 41, unlike in the spectrum of the digest of Mf-1 in which this signal was distinct (Fig. 2), and signals at m/z = 1355.3 and 1371.3 were newly observed. Furthermore, signals of Tl-T7(V2)and T1-T7(V3), observed in the spectrum of the digest of Mf-1, could not be detected, although the signal of T7(V2) was observed (Table 2). Moreover, the signals of T1, Tl-T7(VZ), and T1-T7(VA could not be seen in the mass spectrum of the digest of Ms
: 10
15 20 25 30 35 Retention time b i n )
40
FIG. 1. HPLC profile on a reversed-phase column of the Sephadex G-50 eluate of rIFN-nA cited from Ref. 15.
5
t
i FIG. 2. Positive FAB mass spectra 2500 z000 1500 of the3500 tryptic3000 digests of Mf-1 before (upper) and after (lower) digestion with S. aurew protease VS. T, and Tm(V,) show tryptic peptides and tryptic and S. aureus protease V8 peptides, respectively. Tm-T,,or T,,,-T,(Vd denotes that T, and T,or T,(V,) are linked by a disulfide bond, respectively.
3580
3000
1000
2500
500
2000508 1500
1000
n/z
Recombinant Human Leukocyte Interferon A TABLE1 MOSSvalues and locations of trypticpeptides Mf-1 Peptide
bserved
mass value
TI Tz T I 2
TB T I 3
Mf-2
of rIFN-aA
Ms
Observed Observed mass value mass value
Theoretical Sequence massvalue
1313.2 1232.41232.41232.41232.713-22 1076.3 1076.4 1076.7 1076.6 14-22 1038.41038.41038.21038.623-31 910.2 910.3 910.3
1313.6 1-12
3543
Mf-1 in their sequences from position 1 to 12, as described above, we attempted to isolate the peptides with the sequence from position 1 to 12 from tryptic digests ofthree the components of rIFN-aA by HPLC. The HPLC profiles of tryptic digests of Mf-1, Mf-2, and Ms are shown in Fig. 5. In this experiment, we used different chromatographic conditions from those in a previous experiment' in which it had been difficult to differentiate the peptide maps of the digests of three components of rIFN-aA. All peak fractions were sepaand rated analyzed by amino acid analysis and FAB mass
spectrometry. Peptides containing the sequence from position 1to 12 were 23 31 recovered in peak fractions 1 and 2 in the tryptic digest of T3-T'12 1351144 Mf-1 shown in Fig. 5A. Amino acid analysis and FAB mass 24 31 T13-T'12 2116.92116.52117.1 1351144 spectrometry showed that peak fraction 1contained a peptide 2226.0 32-49 with the sequences from position 1 to 12 and from position -D T 4 2225.4 T'4 1954.4 1954.6 1955.3 1954.9 34-49 84 to 112 linked by a disulfide bond, while peak fraction 2 T 5 2459.72459.92459.52459.350-70 contained a peptide with the sequences from position 1to 12 T8 1030.41030.4KK30.41030.6113-121 and from position 71 to 112 linked by a disulfide bond (Table T'B902.1902.3902.2 3 and Figs. 6 and 7). In addition, a peptide with the sequence T I 0 750.2750.41337.4 750.4 1337.4b 1337.46 1337.7 134-144 from position 122 to 125, which could not be detected in the TIZ T'IZ 1209.31209.31209.51209.6135-144 FAB mass spectra of Mf-1 Fig. in 2, was found in the peak Tn 619.2 619.3 619.1 619.3 145-149 fraction marked by an asterisk in Fig. 5A. TI, 1481-31481.51481.21481.8150-162 In the tryptic digest of Mf-2, peaks 1and 2 were only small, Signals were not observed. unlike in HPLC the of Mf-1 in which they were distinct, and Weak but distinct signals were observed. peaks and 34 were newly found (Fig. 5B). Peak fractions 3 and 4 had similar amino acid compositions to those of peak TABLE 2 fractions 1 and 2, respectively (Table 3). Therefore, peaks 3 MOSSvalues and locations of the tryptic and s. aurem Protease v8 and 4 were considered to contain peptides with the sequences digests of rtFN-aA from position 1 to 12 and from position 84 to 112 or 71 to Mf-1 Mf-2 Ms 112, respectively, linked by a disulfide bond. However, the Peptide ___ ____ Theoretical Observed Observed Observed maasvalue Sequence FAB mass spectra of peak fractions 3 and4after digestion mass value mass value mass value with S. aureus protease V8 both gave intense signals at m/z 1313.6 1-12 = 1355.4, 1371.2, 1645.2, 2998.4, and 3014.5, although peaks TI 1313.4" 1313.5 b 1 12 and 3 4 did not give their molecular ion signals because they TI-T7(VZ) 2957.1 -b 2956'5 97r112 were too big todetect under the present experimental condi1 12 Tl-TV(V3)2386.3 -' -' 2386.1 971107 tions. The values of 1355.4 and 1371.2 were consistent with T? 1232.5 1232.3 1232.3 1232.7 13-22 those observed in the spectrum of the tryptic and S. aureus T ' 2 1076.51076.51076.51076.614-22 protease V8 digests of Mf-2 (Fig. 4) and were 42 and 58 amu, T3 1038.4 1038.2 1038.2 1038.6 23-31 respectively, more than that calculated from the sequence 1 TrT'12W 23 31 to 12 in Table 2. The value of 1645.2 corresponded thatto of 135T141 the sequence from 97 to 112, but the values 2998.4 and 3014.5 or 1890.9 1891.1 1890.9 1890.7 or 24 31 were 42 and 58 amu, respectively, greater than those of the T'3-TdV) 1341141 sequences from 1 to 12 and from 97 to 112. These results T I 3 g10.4 910-3 24-31 suggestthe that mass values 1355.4 and 1371.2 resulted from 24 31 T'3-T'12(V) 1762.7 1763.0 1763.0 1762.8 1351141 the sequence from position 1to 12 andthat Cys-1 was linked TdVA 1008.41008.41008.11008.550-58 to Cys-98 by a disulfide bond. On the other hand, a previous TdV'J 808.4 808.4 808.1 808.5 52-58 experiment' showed that the recoveries of N-terminal amino TdV2) 1469.51469.9146g651469.859-70 acid residues were rather lower on Edman degradation of 1990.9 1991.0 lg90.' TdVJ* 71-87 reduced and carboxymethylated Mf-2 than of similarly treated 1875.1 1876.1" 1876.1" 1875.9 72-87 Tdv'1)* Mf-1. These results suggest that the N-terminal amino group 79-87 1129.5 112g,3 112g.1 112g.6 TdV2) of Mf-2 is partially blocked by substituents with mass values 88-96 T4VJ 1135.6 1135.1 1135.2 1135.5 T4V2) 1645.6 1646.0 1646.0 1645.9 97-112 of 43 and 59. The mass value 43 corresponds to an acetyl 589.3 T7(V4) 108-112 589.3 589.1 589.1 group. T't3 902.3 902.1 902.1 902'5 '13-120 TO obtain further information on the structure of the 773.5 114-120 173.2 T'JV) 773.4 773.2 750.5 126-131 substituents, we reduced and carboxymethylated peaks 3and TIO 750.1 750.1 750.4 TI3 619.3 619.1 619.1 619.3 145-149 4 and isolated the peptides with the amino acid composition TIO) 1125.41125.31125.41125.5150-153 from position 1to 12. The FAB mass spectra of these peptides Weak but distinct signals were observed. were 42 and 58 amu greater than the mass value calculated Signals were not observed. from the sequence 1 to 12, supporting the above result that the mass values of the sequence from position 1 to 12 were (lower spectrum in Fig. 4). These observations suggest that 42 and 58 amu more in Mf-2 than in Mf-1. Peak fractions 1 there are some modifications of the amino acid residues in and 2 from Mf-1 were allowed to react with acetic acid N-hythe sequence from position 1to 12 of Mf-2 and Ms, and also, droxysuccinimide ester, and after subsequent reduction and that in Ms, Cys-1 is not linked to Cys-98. carboxymethylation, the peptide with the amino acid compoSince Mf-2 and Ms were considered to be different from sition of position 1-12 was isolated. This peptide had the T'3-T12 or
or
2244.6 2245.5 2245.1 2245.2
Q
~~~~~~~
___^
3544
Recombinant Human Leukocyte Interferon A 1
‘P
2?
3?
40
50
7?
6?
80
CDLP9THSLGSRRTLMLLAQMRKISLFSCLKDRHDFGFPflEEFGNOFOKAETIPVLHEMI~~IFNLFSTKDSSAA!~DETLLDKFY (2225.4) (1038.4) (1232.4) (1313.2)
T4----”------5-
+4;+I-p3+l----q---4
( 2459.7)
(1954.4) (910.2) (1076.3)
FIG. 3. Amino acidsequence of IFN-aA predicted from the nucleotide sequence of the gene encoding theprotein (7). 2’; and T;(V,)show tryptic peptides and tryptic and 5’. aureus protease V8 peptides, respectively, observed in the mass spectra in Fig. 2, and numbers inparentheses are observed mass values. Signals of peptides marked by an asterisk (*) were observed in a different spectrum from that in Fig. 2.
1;o
170
15.0
1qo
TELYQQLNDLEACVIOGVGVTETPLMKEDSILAVRKYFflRITLYLKEKKYSPCAWEVVRAEIMRSFSLSTNLQESLRSKE “ g C
ICTD-
-TpPTu-t---TM+
(1481.3) (1337.4) (1030.4) (619.2) (750.4)
+%(1209.3)
-+-yy)”S(y”~---r (1125.4) (619.3)
(750.4)
+ %
(902.2) r T B t T M ( V r) -- T? n 7
(902.3) (1645.6)(1135.5)
“I
same retention time on HPLC as the peptide isolated from peaks 3 and 4 (Fig. 8) and had a mass value of 42 amu more than that calculated from the sequence from position 1 to 12, as described above. These results indicate that thedifference of 42 amu between peak fractions 1 and 2 and peak fractions 3 and 4 is due to an acetyl group, that is the amino group of the N-terminal cysteine residue of Mf-2 is in part acetylated. In this connection it isnoteworthy that theN-terminal amino group of eglin c produced in E. coli appears to be acetylated (18).The peptide with a mass value of 58 amu more than that of the sequence from 1 to 12 was considered to be modified like that with a mass value of 42 amu more than that of the sequence, but its structurecould not bedetermined. The tryptic digest of Ms did not show peaks corresponding to either peaks 1 and 2 or peaks 3 and 4 found in the digests of Mf-1 and Mf-2, respectively (Fig. 5C). Peptides with the amino acid composition of the sequence from 1 to 12 were found in the peaks marked 5 to 8. Peak 5 was deduced to consist mainly of the sequence from 1to 12 linked by glutathione through a disulfide bond, because it had the amino acid composition of the sequence from 1 to 12 with that of glutathione and showed signals at m/z = 1313.1 and 1618.0. The former valuewas identical with the theoretical mass value calculated from the sequence 1 to 12, and the latter value was identical with that calculated from the sequence from 1 to 12 with glutathione. Namely, Cys-1 was linked to the cysteine residue in glutathione by a disulfide bond. Peak fraction 6 had the amino acid composition of the sequence from 1to 12 but with cysteic acid instead of half-cystine. This result was supported by the finding that themass value 1361.0 of this peak fraction was identical with the theoretical mass value of the sequence from 1to 12, in which the Cys-1 residue was oxidizedto cysteic acid. Peak fractions7 and 8 had similar amino acid compositions to the sequence from 1to 12. However, their structures could not be elucidated because their FAB mass spectra showed various unidentified signals, suggesting that modifications had occurred in the amino acid residues in the sequence from position 1 to 12. These results indicate that Cys-1 is not involved in a disulfide linkage and,
therefore, raises the problem of whether Cys-98, the partner of Cys-1 in a disulfide linkage in IFN-aA, is concerned in a disulfide linkage in Ms. Peptidescontaining Cys-98 were found in peak fractions 9-13 in Fig. 5C. Amino acid and FAB mass analyses (Table 4) indicated the following. (i)Peak fractions 9 and 10 consisted of the sequence from 84 to 112. Cys-98 was oxidized to cysteic acid in both peak fractions, and Met-111 was changed to Met(0) in peak fraction 9. (ii) Peak fractions 11 and 12 consisted of sequence 84 to 112 linked to glutathione by a disulfide bond, and inpeak fraction 11, Met-111 was also oxidized to Met(0). (iii) Peak fraction 13 consisted of the sequence from position 71 to 112 linked by glutathione through a disulfide bond. Thus, it was concluded that Cys-98 is oxidized to cysteic acid or linked to glutathione in Ms. These results are compatible with a previous finding thata glycine residue, in addition to a glutamic acid residue, was found as the C-terminal amino acid residue of Ms. The evidence of a protein-glutathione conjugate was obtained by the finding that Ms was treated with performic acid to give a peptide, which was eluted at the same position on HPLC as glutathione treated with performic acid under similar conditions, with a composition of glutamic acid, glycine, and cysteic acid (data not shown). In our experience, this is the first example of recombinant proteins that are modified at the thiol groups of cysteine residues by glutathione, although the formation of human growth hormone-glutathione conjugate in E. coli has recently been suggested (19). This protein-glutathione conjugate may be an intermediate in the formation of a disulfide linkage in the folding process of proteins formed in the cytoplasm. A possibly existing enzyme system to form disulfide linkages in proteins might not function efficiently in such proteins produced in large quantities in E. coli. Moreover, the present finding is consistent with the report that IFN-aA has antiviral activity even when a disulfide bond is not formed between Cys-1 and Cys98 (20). The results described above clearly demonstrate that Mf-1 has thesame amino acid sequence as thatpredicted from the nucleotide sequence encoding the gene of IFN-aA. This Mf-1
Recombinant Human Leukocyte Interferon A A
3545
the present investigation indicates the possibility that biologically active but structurally modified proteins are present in the preparation of E. coli, although the observed structural modifications may have occurred during the biosynthesis or purification of the proteins. Therfore, these proteins should be characterized by analysis of their preparations, even if their primary structures are confirmed by sequencing cDNA. The present method is a suitable procedure for such analyses.
I '0
io io O I
IO !O
Acknowledgments-We (M. K. and 0. N.) are grateful to Dr. Yukio Sugino, Dr. Masao Nishikawa, and Dr. Atsushi Kakinuma for their encouragement and discussion throughout this work. We also thank Dr. Kiyoshi Nishi for supplying Mf-1, Mf-2, and Ms and Mrs. Shizue Nakagawa for amino acid analysis. Thanks are due to Dr. James R. Miller for reading the manuscript.
LO
REFERENCES E E 0
B
N N
.@
m
-
70 60 50 z 40 om 30
aJ V
m
n
L 0 In
n
m aJ
20 10
>
.r
c,
m
i I
7
aJ
C 170 ,60 ,50 140
-30 v
20
.lo
0
10
20 30 40 50 60 Retention time (min) FIG. 5. HPLC profiles on a reversed-phase column of the For chromatryptic digests of Mf-1 (A),Mf-2 (B),and Ms (0.
tographic conditions see text.
(15) is now being used as IFN-(YA inclinical studies. Mf-2 is in partmodified at the amino group of the N-terminal cysteine residue by an acetyl group, and Ms is modified at Cys-1 and Cys-98 to cysteic acid or linked to glutathione. Furthermore,
1. Isaacs, A., and Lindermann, J. (1957) Proc. R. Soc. Lond. B Biol. Sci. 147,258-267 2. Isaacs, A., Lindermann, J., and Valentine, R. C. (1957) Proc. R. Soc. Lond. B Bwl. Sci. 147,268-273 3. Nagano, Y., and Kojima, Y. (1958) C. R. Seances SOC.Biol. Fil. 162,1627-1629 4. Stewart, W. E. (1979) The Interferon System, Springer-Verlag New York Inc., New York 5. Nagata, S., Taira, H., Hall, A., Johnsrud, L., Streuli, M., Escodi, J., Boll, W., Cantell, K., and Weissmann, C. (1980) Nature 284,316-320 6. Taniguchi, T., Guarente, L., Robert, T. M., Kimelman, D., Douhan, J., 111, and Ptashne, M. (1980) P m .NatL Acad. Sci. U.S. A. 77,5230-5233 7. Goeddel,D. V., Yelverton, E., Ullrich, A., Heyneker, H. L., Miozzari, G., Holmes, W., Seeburg, P. H., Dull, T., May, L., Stebbing, N., Crea, R., Maeda, S., McCandliss, R., Sloma, A., Tabor, J. M., Gross, M., Familletti, P. C., and Pestka, S. (1980) Nature 287,411-416 8. Derynck, R., Remaut, E., Saman, E., Stanssens, P., De Clercq, E., Content, J., and Fiers, W. (1980) Nature 287,193-197 9. Maeda, S., McCandliss, R., Gross, M., Sloma, A., Familletti, P. C., Tabor, J. M., Evinger, M., Levy, W. P., and Pestka, S. (1980) Proc. Natl. Acad. Sci. U.S. A. 7 7 , 7010-7013 10. Goeddel, D. V., Sheppard, H. M., Yelverton, E., Leung, D., Crea, R.,Sloma, A., and Pestka, S. (1980) Nucleic Acids Res. 8,40574074 11. Yelverton, E.,Leung, D., Weck, P., Gray, P. W., and Goeddel, D. V. (1981) Nucleic Acids Res. 9 , 731-741 12. Staehelin, T., Hobbs, D. S., Kung, H., Lai, C.-Y., and Pestka, S. (1981) J. Bwl. Chem. 256,9750-9754 13. Pestka, S., Kedler, B., Tarnowski, D. K., and Tarnowski, S. J. (1983) Anal. Biochem. 132,32%333 14. Felix, A. M., Heimer, E. P., Lambros, T. J., Swistok, J., Tarnowski, S. J., and Wang, C.-T. (1985) J. Chromutogr. 3 2 7 , 359-368 15. Honda, S., Sugino, H., Nishi, K., and Kakinuma, A. (1986) J. Biotechnol., in press 16. Wetzel, R. (1981) Nature 289,606-607 17. Takao, T.,Yoshida, M., Hong, Y.-M., Aimoto, S., and Shimonishi, Y. (1984) Bwmed. Mass Spectrom. 11,549-556 18. Griitter, M. G., Miirki, W., and Walliser, H.-P. (1985) J. Biol. Chem. 260,11436-11437 19. Axelsson, K. (1985) Acta Chem. Scand. B39,69-77 20. Morehead, H., Johnston, P. D., and Wetzel, R. (1984) Biochemistry 23,2500-2507 21. Takao, T.,Watanabe, H., and Shimonishi, Y. (1985) Eur. J. Biochem. 146,503-508 22. Takao, T., Hitouji, T., Shimonishi, Y., Tanabe, T., Inouye, S., and Inouye, M. (1984) J. Bwl. Chem. 259,6105-6109
Continued on next page.
Recombinant Human Leukocyte Interferon A
3546
SUPPLEMENTARY MATERIAL TO CHEMICAL CHARACTERIZATION OF RECOMBINANTHUMANLEUKOCYTEINTERFERON A USING FASTATOMBOMBARDMENTMASSSPECTROMETRY Toshifumi Talcso. Makoto Kobayarhi , Ovmu Nirhimura and Yarutsugu 1129.2
Shimonishi MATERIALSANDMETHODS
2386.7 Recombinant Human LeukocvtelnterfemnA rlFN-aAwasobtainedfpm E. mlicelircontaininga plasmidwith human IFNd
