Joitrnal of Neurochemistry Raven Press, Ltd., New York 0 1992 International Society for Neurochemistry
Biochemical Characterization of Recombinant Human Nerve Growth Factor Charles H. Schmelzer, Louis E. Burton, Wai-Pan Chan, Evelyn Martin, Cori Gorman, Eleanor Canova-Davis, Victor T. Ling, Mary B. Sliwkowski, Glynis McCray, Jonathan A. Briggs, Tue H. Nguyen, and Gian Polastri Genentech, Inc., South Sun Francisco, California, U.S.A.
Abstract: Recombinant human nerve growth factor (rhNGF) was expressed and secreted by Chinese hamster ovary cells and purified to homogeneity using ion-exchange and reversed-phase (RP) chromatography. The isolated product was shown to be consistent with a 120-amino-acid residue polypeptide chain by amino acid composition, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), RP-HPLC, and mass spectrometry and with an N-terminal sequence consistent with that expected from the cDNA for human nerve growth factor. By size-exclusion chromatography, rhNGF behaves like a noncovalent dimer. Limited enzymatic digests of the 120-residue monomer produced additional species of 118 (trypsin, removal of the C-terminal Arg"9-Ala'Zo sequence) and 1 17 (trypsin plus
carboxypeptidase B, removal of the C-terminal Arg"'Arg"9-Ala'20 sequence) residues. Each of these species was isolated by high-performance ion-exchange chromatography and characterized by amino acid and N-terminal sequence analyses, SDS-PAGE. RP-HPLC, and mass spectrometry. All three species were present in the digests as both homodimeric and heterodimeric combinations and found to be equipotent in both the chick dorsal root ganglion cell survival and rat pheochromocytoma neurite extension assays. Key Words: Nerve growth factor-Purification-Bioequivalence-Dimers. Schmelzer C. H. et al. Biochemical characterization of recombinant human nerve growth factor. J. Neurochem. 59, 1675-1683 (1992).
Nerve growth factor (NGF) is a protein required for the growth and survival of sympathetic and sensory neurons during development and in mature animals (Thoenen and Barde, 1980; Yankner and Shooter, 1982; L ~ i - M ~ f i t d d1987). ~ ~ i ,LT~ltilt . ~ ~ f i f lthe y , major source of this protein for biochemical and biological study has been the mouse submaxillary gland (Cohen, 1960; Varon et al., 1967a), although other tissue sources have been examined, most notably snake venoms (Hogue-Angeletti and Bradshaw, 1979)and prostate gland and its secretions (Harper et al., 1979, 1982; Harper and Thoenen, 1980; Chapman et al., 1981). The mouse submaxillary gland source yields two primary types of the active factor, dependent on
the approach to purification: the 7 s NGF complex (Varon et al., 1967a,b, 1968; Smith et al., 1968) and 2.5S, or P-NGF (Bocchini and Angeletti, 1969). pNGF, the subunit of the 7 s NGF complex exhibiting NGF activity, is a noncovalent dimer of two 13kDa polypeptides, the monomer of which occurs in one of three commonly described forms: 118 residues, or 6-NGF; 1 17 residues, or des Arg'18 P-NGF; and 1 10 residues, or des octa (or des 1-8) P-NGF (Moore et al., 1974; Mobley et al., 1976). Each of these forms has been well characterized, and their presence in mouse submaxillary gland preparations is also dependent on the mode of isolation (for review, see Longo et al., 1989).
Received December 3 1, 1991; revised manuscript received April 6, 1992; accepted April 14, 1992. Address correspondence and reprint requests to Dr. C. H. Schmelzer at Genentech, Inc., 460 Point San Bruno Boulevard, South San Francisco, CA 94080, U.S.A. Abbreviations used: CF, culture fluid; CHO, Chinese hamster ovary; DEFF, diethylaminoethane-Sepharose fast flow; DMEM, Dulbecco's modified Eagle's medium; ELISA, enzyme-linked immunosorbent assay; HPIEC, high-performanceion-exchange chromatography; MOPSO, 3-(N-morpholino)-2-hydroxypropanesul-
fonic acid; NGF, nerve growth factor; rhNGF, recombinant human nerve growth factor; RP, reversed-phase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SEC, size-exclusion chromatography; SSFF. sulfopropyl-Sepharosefast flow; TFA, trifluoroacetic acid. Unless otherwise indicated, the abbreviation for the form of the NGF molecule represents the homodimeric form. In the case of heterodimers, the different monomers in the dimeric complex are separated by a colon (:). For example, 120:1 18 represents the heterodimer between the two rhNGF monomers of 120 and 1 18 residues.
-
1675
C. H. SCHMELZER ET AL.
1676
The isolation of naturally occurring human NGF has been reported (Goldstein et al., 1978). However, no analytical data regarding sequence or amino acid composition of such preparations have been forthcoming. The cloning of human NGF has been reported (Ullrich et al., 1983), and several groups have successfully used recombinant technology to produce active recombinant human NGF (rhNGF) (Bruce and Heinrich, 1989; Kanaya et al., 1989; Iwane et al., 1990;Barnettetal., 1991;Buxseretal., 1991). Eachof these systems used eukaryotic organisms or yeast and generally used the expression of the pro form of the NGF molecule, which is subsequently processed to the mature molecule on secretion. In this report, we describe the expression, isolation, and characterization of mature rhNGF and several enzymatically modified versions of rhNGF, some of which mimic the structure exhibited by the wellknown murine molecule. We report for the first time that different dimeric forms of rhNGF have been purified and biologically evaluated in two commonly used assays to measure NGF potency. For the purposes of this report, abbreviations have been set to distinguish the variety of forms of rhNGF presented here. MATERIALS AND METHODS Transfection and expression of NGF in Chinese hamster ovary (CHO) cells The cDNA for human NGF (IJllrich et al., 1983) was inserted into a mammalian expression vector containing the SV40 enhancer and promoter and a 5’ intron (Huang and Gorman, 1990). At the 3‘ end of the NGF cDNA this vector contained the SV40 late polyadenylation signal. The backbone of this vector is pUC 118. The human NGF cDNA and a SV40-dhfr expression plasmid were cotransfected (Graham and van der Eb, 1973) into CHO dhfr- cells (Urlaub and Chasin, 1980). Two days posttransfection, cells were split into selective media: glycine-, hypoxanthine-, and thymidine-free Ham’s F12 Dulbecco’s modified Eagle’s medium (DMEM) ( I :1) and 7% dialyzed fetal bovine serum (GIBCO). Within 2 weeks individual clones were isolated and assayed for production of rhNGF by enzyme-linked immunosorbent assay (ELISA). The two-site ELISA used polyclonal antibodies raised to rhNGF in rabbits as both the capture and detection antibody. There is 95% pure (Fig. 3) and consisted of the mature form of rhNGF as a
J. Neurochem., Vol. 59, NO.5, 1992
C. H . SCHMELZER ET AL.
1680
TABLE 2. Amino acid compositions of rhNGF Residue Asx Thr Ser Glx Pro GlY Ala CYS Val Met Ile Leu
+ Cys-SH
TYr Phe His LYS Trp
Theoretical”
120:1206
120:118‘
118:118d
118:117‘
117:117*
13 10 II 6 3 7 7 6 13 2 6 3 2 7 4 9 3 8
13.2 f 0.2 9.5 f 0.1 9.6 f 0.1 6.3 f 0.0 3.8 f 0.3 7.0 f 0.1 6.9 f 0.1 4.2 f 0.5 13.0 f 0.2 1.9 f 0.0 5.8 f 0.1 3.2 % 0.1 2.1 f 0.1 7.1 f 0.2 3.9 f 0.0 9.0 f 0.1 ND 7.8 f 0.0
13.2 f 0.0 9.4 f 0.0 9.6 f 0.1 6.3 f 0.1 4.0 f 0.4 7.1 f 0.0 6.5 f 0.0 4.4 f 0.4 12.8 f 0.1 1.9 f 0.0 5.8 f 0.1 3.1 f 0.0 2.1 f 0.1 7.0 f 0.0 3.9 f 0.0 9.0 f 0. I ND 7.4 f 0.0
13.0 k 0.1 9.4 f 0.1 9.5 f 0.1 6.3 f 0.0 4.2 f 0.2 7.0 f 0.1 6.0 f 0.1 3.8 f 0.3 12.7 f 0.1 1.9 f 0.0 5.7 f 0.0 3.1 f 0.0 2.1 f 0.0 7.0 f 0. I 3.9 f 0.0 9.0 f 0.0 ND 6.8 f 0.1
12.9 f 0.0 9.3 f 0.1 9.7 f 0.0 6.1 f 0.0 3.1 f O . l 7.3 0.0 6.0 f 0.1 4.8 k 0.1 12.8 f 0.0 1.8 f 0.1 5.8 f 0.0 3.0 f 0.0 1.8 f 0.1 6.9 f 0.0 4.0 f 0.1 9.0 k 0.1 ND 6.3 f 0.0
12.9 k 0.0 9.4 0.1 9.5 f 0.1 6.3 f 0.0 4.6 f 0.1 7.1 & 0.1 5.9 f 0. I 3.5 f 0.2 12.7 f 0.1 1.8 f O . 1 5.7 f 0.0 3.0 f 0. I 2.1 f 0.0 7.1 f 0.0 4.0 f 0.0 9.0 f 0. I ND 5.8 f 0.1
*
*
ND, not determined. Theoretical residues per monomer of 120-residue rhNGF. 120:120 rhNGF molecule (Fig. 4, peak 3). 120:1 18 rhNGF molecule (Fig. 4, peak 2). I18:118 rhNGF molecule (Fig. 4, peak I). 1 18: 1 17 rhNGF molecule. 1 17: I 17 rhNGF molecule.
*
120-residue homodimer [Tables 1 and 2 ( 120: 120)]. This isolation method used two common ion-exchange techniques (DEFF and SSFF) to reduce the amount of host contaminating proteins to an acceptable level before RP-HPLC. Molecular sieving of this material reveals a noncovalent dimeric form under nondenaturing conditions. The production of rhNGF in CHO cells resulted in the mature product predicted by Ullrich et al. ( 1983). The N-terminal sequence data indicate a single product produced by this method from the prepro form. Work affecting the expression of recombinant murine NGF using its prepro form (Edwards et al., 1988) in mouse L929 fibroblasts was minimally successful in producing active NGF. When trypsin or the mouse 7 s NGF y-serine protease was added to process the putative precursor form secreted by those cells, increased TABLE 3. Mass spectral analysis of rhNGF fractions Sample in Fig. 5A
Theoretical masses
Observed masses
Curve A ( I l7:117) CurveB(118:117)
13,104.9 13,261.1 13,104.9 13,261.1 13,488.4 13,261.1 13,488.4
13,105.4 13,260.8 13,104.5 I3,26 1.2 13,488.7 13,261.5 13,488.8
Curve C (1 18: 118) Curve D ( 120:1 18) Curve E ( 120:120)
Theoretical masses are average masses.
J. Neurochem.. Vol. 59, No. 5, 1992
levels of NGF activity resulted. Proteolytic modifications of mature NGF, such as des octa NGF (murine NGF lacking the first eight N-terminal amino acids), seen in mouse submaxillary gland and saliva (Moore et al., 1974; Mobley et al., 1976; Burton et al., 1978), were not observed for these preparations of rhNGF in CHO cells. One of the enzymes reportedly responsible for posttranslational processing of murine NGF is the y-protease. The best-described activity associated with yprotease is the precursor processing of pro-murine NGF resulting in an enzyme-product complexation as part of the 7s NGF complex (Jongstra-Bilen et al., 1989).Such a complex or its potential has been shown only for murine sources (Darling and Fahnestock, 1988;Schwarz et al., 1989), and the in vivo processing status of the human molecule remains unknown. In light of this, we endeavored to demonstrate that regardless of the processing status of natural human NGF, some of the proteolytic modifications described were not crucial to the expression of biological activities of human NGF. Most studies regarding the isolation of various proteolytic species of the 0 form of murine NGF have resulted in mixtures of heterodimers and homodimers generally containing more than one of any of these species (Moore et al., 1974; Mobley et al., 1976). We have performed controlled enzymatic cleavages of rhNGF, followed by the isolation of resulting dimeric species, which mimic some of those observed for the murine molecule, to examine their relation-
1681
RECOM3INANT HUMAN NERVE GROWTH FACTOR A
100,
I
I
I
100
1 02
10'
I
20
40
60
80
103
104
NGF (pglml)
I
-
Time (rnin)
.001
.01
.1
1
lo
100
1000
NGF (ng/ml)
/
0
/
10
,
I
l
30
20
l
40
50
,
'
60
Time (rnin)
Analytic_.l HPLC of purifiedrhNGF. A Analyticc IPIEC of FI dimers. Individual samdes were iniected onto a Polvsulrh foethyl A cation-exchange (4.6-X 200-mm) column as descAbed in Materials and Methods: curve A, rhNGF 117:117; curve B, rhNGF 118:117; curve C, rhNGF 118:118; curve D, rhNGF 120:118; and curve E, rhNGF 120:120.B: Analytical C, RP-HPLC analysis of various rhNGF species. Individual samples were injected onto an analytical YMC C, RP-HPLC column as described in Materials and Methods. Curves A-E are as described in A.
1
2
3
4
5
6
7
FIG. 6. SDS-PAGE of rhNGF species. Samples were prepared and subjected to SDS-PAGE on 15% gels containing -4 M urea as described in Materials and Methods: lane 1, molecular-mass standards; lane 2, rhNGF 117:117; lane 3, rhNGF 118:117; lane 4, rhNGF 118:118; lane 5, rhNGF 118:120; lane 6, rhNGF 120:120; and lane 7, a mixture of rhNGF 120:120 and rhNGF 117:l 17.
FIG. 7. Biological characterization of rhNGF species. A PC12 assay of rhNGF species. Individual rhNGF species were analyzed in the PC12 assay (single determination): 120:l 20 (0),120:118 (O), 118:118 (O),118:117 (m), and 117:117 (A). 6: Chick dorsal root ganglion (DRG) survival assay of rhNGF species. Individual rhNGF species were tested in the chick DRG survival assay (n = 2 determinations):120:l 20 (0),120:118 (O), 118:118 (O),118:117 (m), and 117:117 (A).
ships with regard to specific activity. As described, the tryptic cleavage of 120:120, when separated on HPIEC (see Fig. 4), resulted in a mixture of dimeric forms composed primarily of 120:120, 120:1 18, and 1 18:1 18 dimers. Subsequent limited cleavage of the 118:118 pool by carboxypeptidase B resulted in two additional dimer forms: 1 18: 1 17 and 1 17:1 17. Each of these forms has been shown to be distinguishable by HPIEC (Fig. 5A). Biological assessment of the potency of each of the dimeric species described above shows that all are equally potent in either maintaining neuronal survival in primary culture in the chick dorsal root ganglion assay (Fig. 7B) or causing neurite extension in the rat PC 12 assay (Fig. 7A). The ED,, and the maximal effect within each assay for each of the dimeric species are reached at essentially the same concentrations. Murine 118-residue NGF prepared in our hands is equipotent with the rhNGF species described herein in the respective bioassays (L. E. Burton and C. H. Schmelzer, unpublished data). Analytical RP-HPLC (Fig. 5B) of the separated species noted in Fig. 5A indicates the gross homogeneity of each of the preparative HPIEC peaks (Fig. 4)as well
J. Neurochem., Vol. 59, No. 5, 1992
C. H. SCHMELZER ET AL.
1682
as showing, in the case of heterodimers such as 120:1 18, that this mode of chromatography reveals the monomeric forms of NGF present in each heterodimer. This observation is in contrast to that made by Barnett et al. (1991). These authors observed that monomers of putative 120-residue rhNGF in fact bound more tightly than their apparent dimer to an RP-HPLC system comparable to that used here. Although the data presented here indicate only very small elution condition differences between minor modified forms of rhNGF, we have also observed (data not shown) species of rhNGF that elute in similar positions with our system, i.e., with tighter resin association. We have noted that the contents of this (these) species increase with prolonged exposure to acid (TFA or HCl) or other denaturants (urea or guanidine-HCl) and that the rate of appearance is time and temperature dependent (data not shown). We suggest that this form(s) of rhNGF may constitute an additional structural change that is reversible under neutralized or native conditions (data not shown). Studies examining the character of this “form” of rhNGF are currently underway. To summarize, we have demonstrated that rhNGF can be produced in CHO cells by recombinant techniques, purified to homogeneity, and retain full biological activity, as judged by the chick dorsal root ganglion assay and the rat PC12 neurite extension assay. We have also shown definitively that rhNGF of 120 residues is equivalent to the 1 1%residue murine molecule in potency based on mass in these bioassays. Furthermore, direct comparison of characterized proteolytic modifications of the mature 120-residuespecies, i.e., 118 and 117 residues, definitively show no effect on the biological potency of the molecule. These data indicate that the rhNGF produced in CHO cells is a candidate source for material used for study of efficacy in humans. Acknowledgment: The authors are indebted to Reed Hams for N-terminal sequence analyses, Karen Wagner for amino acid compositional analyses, Dr. Eva Szonyi for dorsal root ganglion assays, William Mulhern for SEC, and Aldona T. Gorrell, Cathleen Yedinak, and Hannah Beasley for support in purification work.
REFERENCES Barnett J., Chow J., Nguyen B., Eggers D., Osen E., Jarnagin K., Saldou N., Straub K., Gu L., Chaing H.-S., Fausnaugh J., Townsend R. R., Lile J., Collins F., and Chan H. (1991) Physicochemical characterization of recombinant human nerve growth factor produced in insect cells with a baculovirus vector. J. Neurochem. 57, 1052-1061. Bennett G. L., Burton L. E., Chan W.-P., and Wong W. L. T. (1990) Sensitive enzyme immunoassay for recombinant human NGF. J. Cell. Biochem. 14F, 87. Bocchini V. and Angeletti P. U. (1969) The nerve growth factor: purification as a 30,000-molecular-weight protein. Proc. Natl. Acad. Sci. USA 64, 787-794.
J. Neurochem., Vol. 59, No. 5, 1992
Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72,248-254. Bruce G. and Heinrich G. (1989) Production and characterization of biologically active recombinant human nerve growth factor. Neurobiol. Aging 10, 89-94. Button L. E., Wilson W. H., and Shooter E. M. (1978) Nerve growth factor in mouse saliva. J. Biol. Chem. 253,7807-7812. Buxser S.. Vroegop S.. Decker D., Hinzmann J., Poorman R.. Thomsen D. R., Stier M., Abraham I., Greenberg B. D., Hatzenbuhler N. T., Shea M., Curry K. A., and Tomich C.4. C. (1991) Single-step purification and biological activity of human nerve growth factor produced from insect cells. J. Neurochem. 56, 1012-1018. Chapman C. A., Banks B. E. C., Vernon C. A., and Walker J. M. (1 98 1) The isolation and characterization of nerve growth factor from the prostate gland of the guinea-pig. Eur. J. Biochem. 115,347-35 1. Cohen S. (1960) Purification of a nerve growth promoting protein from the mouse salivary gland and its neurocytotoxic antiserum. Proc. Natl. Acad. Sci. USA 46, 302-3 1 1. Darling T. L. and Fahnestock M. (1988) The high molecular weight nerve growth factor complex from Mastomys natalensis differs from the murine nerve growth factor complex. Biochemistry 27.6686-6692. Davies A. M. ( 1989) Neurotrophic factor bioassay usingdissociated neurons, in IBRO Handbook Series: Methods in the Neurosciences, Vol. 12: Nerve Growth Factors (Rush R. A,, ed), pp. 95-109. Wiley and Sons, New York. Edwards R. H.. Selby M. J., Garcia P. D., and Rutter W. J. (1988) Processing of the native nerve growth factor precursor to form biologically active nerve growth factor. J. Biol. Chem. 263, 6810-6815. Goldstein L. D., Reynolds C. P., and Perez-Polo J. R. (1978) Isolation of human nerve growth factor from placental tissue. Neurochern. Res. 3, 175-183. Graham F. and van der Eb A. (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52,456457. Greene L. A. (1977) A quantitative bioassay for nervegrowth factor activity employing a clonal pheochromocytoma cell line. Brain Res. 133, 350-353. Harper G. P. and Thoenen H. (1980) The distribution of nerve growth factor in the male sex organs of mammals. J. Neurochem. 34,893-903. Harper G. P., Barde Y.-A,, Burnstock G.. Carstairs J. R., Dennison M. E., Suda K., and Vernon C. A. (1979)Guineapigprostateis a rich source of nerve growth factor. Nature 279, 160-162. HarperG. P.,Glanville R. W.,andThoenenH.(1982)Thepurification of nerve growth factor from bovine seminal plasma. J. Biol. Chem. 257,8541-8548. Hogue-Angeletti R. A. and Bradshaw R. A. (1979) Nerve growth factor in snake venoms, in Handbook ofExperimenta1 Pharmacolog.v, Vol. 52 (Lee C.-Y., ed), pp. 276-294. Springer-Verlag, Berlin. Huang M. and Gorman C. (1990) Intervening sequence increases efficiency of RNA 3’ processing and accumulation of cytoplasmic RNA. Nucleic Acids Res. 18,937-947. Iwane M., Kitamura Y., Kaisho Y., Yoshimura K., Shintani A,, Sasada R., Nakagawa S., Kawahara K., Nakahama K., and Kakinuma A. ( 1990) Production, purification and characterization of biologically active recombinant human nerve growth factor. Biochem. Biophys. Res. Commun. 171, 116122. Jongstra-Bilen J., Coblentz L., and Shooter E. M. (1989) The in vitro processing of the NGF precursors by the gamma-subunit ofthe 7s-NGF complex. Mol. Brain Res. 5, 159-169. Kanaya E., Higashizaki T., Ozawa F., Hirai K., Nishizawa M., Tokunaga M., Tsukui H., Hatanaka H., and Hishinuma F. ( 1 989) Synthesis and secretion of nerve growth factor by Saccharomyces cerevisiae. Gene 83, 65-74.
RECOMBINANT HUMAN NERVE GROWTH FACTOR Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685. Levi-Montalcini R. (1987) The nerve growth factor 35 years later. Science 237, 1154-1 162. Longo F. M., Woo J. E., and Mobley W. C. (1989) Purification of nerve growth factor, in IBRO Handbook Series: Methods in the Neurosciences, Vol. 12: Nerve Growth Factors (Rush R. A., ed), pp. 3-30. Wiley and Sons, New York. Mobley W. C., Schenker A., and Shooter E. M. (1976) Characterization and isolation of proteolytically modified nerve growth factor. Biochemistry 15, 5543-5552. Moore J. B. Jr., Mobley W. C., and Shooter E. M. (1974) Proteolytic modification of the beta nerve growth factor protein. Biochemistry 13, 833-839. Morrissey J. H. (1981) Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Anal. Biochem. 117,307-3 10. Murphy R. A,, Chlumecky V., Smillie L. B., Carpenter M., Nattriss M., Anderson J. K., Rhodes J. A.. Barker P. A., Siminoski K., Campenot R. B., and Haskins J. (1989) Isolation and characterization of a glycosylated form of beta nerve growth factor in mouse submandibular glands. J. Biol. Chem. 264, 1250212509. Petrides P. E. and Shooter E. M. ( I 986) Rapid isolation of the 7Snerve growth factor complex and its subunits from murine submaxillary glands and saliva. J. Neurochem. 46, 121-725. Schwarz M. A,, Fisher D., Bradshaw R. A., and Isackson P. J. (1989) Isolation and sequence of a cDNA clone of 0-nerve growth factor from the guinea pig prostate gland. J. Neurochem. 52, 1203-1209.
1683
Smith A. P., Varon S., and Shooter E. M. (1968) Multiple forms of the nerve growth factor protein and its subunits. Biochemistry 7,3259-3268. Thoenen H. and Barde Y.-A. (1980) Physiology of nerve growth factor. Physiol. Rev. 60, 1284-1335. Ullrich A., Gray A., Berman C., and Dull T. J. (1983) Human p nerve growth factor gene sequence highly homologous to that of mouse. Nature 303,821-825. Urlaub G. and Chasin A. J. (1980) Isolation of Chinese hamster cell mutants deficient in dihydrofolate reductase activity. Proc. Natl. Acad. Sci. USA 77, 4216-4220. Varon S., Nomura J., and Shooter E. M. (1967a) The isolation of the mouse nerve growth factor protein in high molecular weight form. Biochemistry 6, 2202-2209. Varon S., Nomura J., and Shooter E. M. (19676) Subunit structure of a high molecular-weight form of the nerve growth factor from mouse submaxillary gland. Proc. Natl. Acad. Sci. USA 57, 1782-1789. Varon S., Nomura J., and Shooter E. M. (1968) Reversible dissociation of the mouse nerve growth factor protein into its different subunits. Biochemistry 7, 1296-1303. Wise R. J., Barr P. J., Wong P. A., Keifer M. C., Brake A. J., and Kaufman R. J. (1990) Expression of a human proprotein processing enzyme: correct cleavage of the von Willebrand factor at a paired basic amino acid site. Proc. Natl. Acad. Sci. USA 87, 9378-9382. Yankner B. A. and Shooter E. M. (1982) The biology and mechanism of action of nerve growth factor. Annu. Rev. Biochm. 51, 845-868.
J. Neurochem.. Vol. 59, No. 5, 1992
Biochemical Characterization of Recombinant Human Nerve Growth Factor Charles H. Schmelzer, Louis E. Burton, Wai-Pan Chan, Evelyn Martin, Cori Gorman, Eleanor Canova-Davis, Victor T. Ling, Mary B. Sliwkowski, Glynis McCray, Jonathan A. Briggs, Tue H. Nguyen, and Gian Polastri Genentech, Inc., South Sun Francisco, California, U.S.A.
Abstract: Recombinant human nerve growth factor (rhNGF) was expressed and secreted by Chinese hamster ovary cells and purified to homogeneity using ion-exchange and reversed-phase (RP) chromatography. The isolated product was shown to be consistent with a 120-amino-acid residue polypeptide chain by amino acid composition, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), RP-HPLC, and mass spectrometry and with an N-terminal sequence consistent with that expected from the cDNA for human nerve growth factor. By size-exclusion chromatography, rhNGF behaves like a noncovalent dimer. Limited enzymatic digests of the 120-residue monomer produced additional species of 118 (trypsin, removal of the C-terminal Arg"9-Ala'Zo sequence) and 1 17 (trypsin plus
carboxypeptidase B, removal of the C-terminal Arg"'Arg"9-Ala'20 sequence) residues. Each of these species was isolated by high-performance ion-exchange chromatography and characterized by amino acid and N-terminal sequence analyses, SDS-PAGE. RP-HPLC, and mass spectrometry. All three species were present in the digests as both homodimeric and heterodimeric combinations and found to be equipotent in both the chick dorsal root ganglion cell survival and rat pheochromocytoma neurite extension assays. Key Words: Nerve growth factor-Purification-Bioequivalence-Dimers. Schmelzer C. H. et al. Biochemical characterization of recombinant human nerve growth factor. J. Neurochem. 59, 1675-1683 (1992).
Nerve growth factor (NGF) is a protein required for the growth and survival of sympathetic and sensory neurons during development and in mature animals (Thoenen and Barde, 1980; Yankner and Shooter, 1982; L ~ i - M ~ f i t d d1987). ~ ~ i ,LT~ltilt . ~ ~ f i f lthe y , major source of this protein for biochemical and biological study has been the mouse submaxillary gland (Cohen, 1960; Varon et al., 1967a), although other tissue sources have been examined, most notably snake venoms (Hogue-Angeletti and Bradshaw, 1979)and prostate gland and its secretions (Harper et al., 1979, 1982; Harper and Thoenen, 1980; Chapman et al., 1981). The mouse submaxillary gland source yields two primary types of the active factor, dependent on
the approach to purification: the 7 s NGF complex (Varon et al., 1967a,b, 1968; Smith et al., 1968) and 2.5S, or P-NGF (Bocchini and Angeletti, 1969). pNGF, the subunit of the 7 s NGF complex exhibiting NGF activity, is a noncovalent dimer of two 13kDa polypeptides, the monomer of which occurs in one of three commonly described forms: 118 residues, or 6-NGF; 1 17 residues, or des Arg'18 P-NGF; and 1 10 residues, or des octa (or des 1-8) P-NGF (Moore et al., 1974; Mobley et al., 1976). Each of these forms has been well characterized, and their presence in mouse submaxillary gland preparations is also dependent on the mode of isolation (for review, see Longo et al., 1989).
Received December 3 1, 1991; revised manuscript received April 6, 1992; accepted April 14, 1992. Address correspondence and reprint requests to Dr. C. H. Schmelzer at Genentech, Inc., 460 Point San Bruno Boulevard, South San Francisco, CA 94080, U.S.A. Abbreviations used: CF, culture fluid; CHO, Chinese hamster ovary; DEFF, diethylaminoethane-Sepharose fast flow; DMEM, Dulbecco's modified Eagle's medium; ELISA, enzyme-linked immunosorbent assay; HPIEC, high-performanceion-exchange chromatography; MOPSO, 3-(N-morpholino)-2-hydroxypropanesul-
fonic acid; NGF, nerve growth factor; rhNGF, recombinant human nerve growth factor; RP, reversed-phase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; SEC, size-exclusion chromatography; SSFF. sulfopropyl-Sepharosefast flow; TFA, trifluoroacetic acid. Unless otherwise indicated, the abbreviation for the form of the NGF molecule represents the homodimeric form. In the case of heterodimers, the different monomers in the dimeric complex are separated by a colon (:). For example, 120:1 18 represents the heterodimer between the two rhNGF monomers of 120 and 1 18 residues.
-
1675
C. H. SCHMELZER ET AL.
1676
The isolation of naturally occurring human NGF has been reported (Goldstein et al., 1978). However, no analytical data regarding sequence or amino acid composition of such preparations have been forthcoming. The cloning of human NGF has been reported (Ullrich et al., 1983), and several groups have successfully used recombinant technology to produce active recombinant human NGF (rhNGF) (Bruce and Heinrich, 1989; Kanaya et al., 1989; Iwane et al., 1990;Barnettetal., 1991;Buxseretal., 1991). Eachof these systems used eukaryotic organisms or yeast and generally used the expression of the pro form of the NGF molecule, which is subsequently processed to the mature molecule on secretion. In this report, we describe the expression, isolation, and characterization of mature rhNGF and several enzymatically modified versions of rhNGF, some of which mimic the structure exhibited by the wellknown murine molecule. We report for the first time that different dimeric forms of rhNGF have been purified and biologically evaluated in two commonly used assays to measure NGF potency. For the purposes of this report, abbreviations have been set to distinguish the variety of forms of rhNGF presented here. MATERIALS AND METHODS Transfection and expression of NGF in Chinese hamster ovary (CHO) cells The cDNA for human NGF (IJllrich et al., 1983) was inserted into a mammalian expression vector containing the SV40 enhancer and promoter and a 5’ intron (Huang and Gorman, 1990). At the 3‘ end of the NGF cDNA this vector contained the SV40 late polyadenylation signal. The backbone of this vector is pUC 118. The human NGF cDNA and a SV40-dhfr expression plasmid were cotransfected (Graham and van der Eb, 1973) into CHO dhfr- cells (Urlaub and Chasin, 1980). Two days posttransfection, cells were split into selective media: glycine-, hypoxanthine-, and thymidine-free Ham’s F12 Dulbecco’s modified Eagle’s medium (DMEM) ( I :1) and 7% dialyzed fetal bovine serum (GIBCO). Within 2 weeks individual clones were isolated and assayed for production of rhNGF by enzyme-linked immunosorbent assay (ELISA). The two-site ELISA used polyclonal antibodies raised to rhNGF in rabbits as both the capture and detection antibody. There is 95% pure (Fig. 3) and consisted of the mature form of rhNGF as a
J. Neurochem., Vol. 59, NO.5, 1992
C. H . SCHMELZER ET AL.
1680
TABLE 2. Amino acid compositions of rhNGF Residue Asx Thr Ser Glx Pro GlY Ala CYS Val Met Ile Leu
+ Cys-SH
TYr Phe His LYS Trp
Theoretical”
120:1206
120:118‘
118:118d
118:117‘
117:117*
13 10 II 6 3 7 7 6 13 2 6 3 2 7 4 9 3 8
13.2 f 0.2 9.5 f 0.1 9.6 f 0.1 6.3 f 0.0 3.8 f 0.3 7.0 f 0.1 6.9 f 0.1 4.2 f 0.5 13.0 f 0.2 1.9 f 0.0 5.8 f 0.1 3.2 % 0.1 2.1 f 0.1 7.1 f 0.2 3.9 f 0.0 9.0 f 0.1 ND 7.8 f 0.0
13.2 f 0.0 9.4 f 0.0 9.6 f 0.1 6.3 f 0.1 4.0 f 0.4 7.1 f 0.0 6.5 f 0.0 4.4 f 0.4 12.8 f 0.1 1.9 f 0.0 5.8 f 0.1 3.1 f 0.0 2.1 f 0.1 7.0 f 0.0 3.9 f 0.0 9.0 f 0. I ND 7.4 f 0.0
13.0 k 0.1 9.4 f 0.1 9.5 f 0.1 6.3 f 0.0 4.2 f 0.2 7.0 f 0.1 6.0 f 0.1 3.8 f 0.3 12.7 f 0.1 1.9 f 0.0 5.7 f 0.0 3.1 f 0.0 2.1 f 0.0 7.0 f 0. I 3.9 f 0.0 9.0 f 0.0 ND 6.8 f 0.1
12.9 f 0.0 9.3 f 0.1 9.7 f 0.0 6.1 f 0.0 3.1 f O . l 7.3 0.0 6.0 f 0.1 4.8 k 0.1 12.8 f 0.0 1.8 f 0.1 5.8 f 0.0 3.0 f 0.0 1.8 f 0.1 6.9 f 0.0 4.0 f 0.1 9.0 k 0.1 ND 6.3 f 0.0
12.9 k 0.0 9.4 0.1 9.5 f 0.1 6.3 f 0.0 4.6 f 0.1 7.1 & 0.1 5.9 f 0. I 3.5 f 0.2 12.7 f 0.1 1.8 f O . 1 5.7 f 0.0 3.0 f 0. I 2.1 f 0.0 7.1 f 0.0 4.0 f 0.0 9.0 f 0. I ND 5.8 f 0.1
*
*
ND, not determined. Theoretical residues per monomer of 120-residue rhNGF. 120:120 rhNGF molecule (Fig. 4, peak 3). 120:1 18 rhNGF molecule (Fig. 4, peak 2). I18:118 rhNGF molecule (Fig. 4, peak I). 1 18: 1 17 rhNGF molecule. 1 17: I 17 rhNGF molecule.
*
120-residue homodimer [Tables 1 and 2 ( 120: 120)]. This isolation method used two common ion-exchange techniques (DEFF and SSFF) to reduce the amount of host contaminating proteins to an acceptable level before RP-HPLC. Molecular sieving of this material reveals a noncovalent dimeric form under nondenaturing conditions. The production of rhNGF in CHO cells resulted in the mature product predicted by Ullrich et al. ( 1983). The N-terminal sequence data indicate a single product produced by this method from the prepro form. Work affecting the expression of recombinant murine NGF using its prepro form (Edwards et al., 1988) in mouse L929 fibroblasts was minimally successful in producing active NGF. When trypsin or the mouse 7 s NGF y-serine protease was added to process the putative precursor form secreted by those cells, increased TABLE 3. Mass spectral analysis of rhNGF fractions Sample in Fig. 5A
Theoretical masses
Observed masses
Curve A ( I l7:117) CurveB(118:117)
13,104.9 13,261.1 13,104.9 13,261.1 13,488.4 13,261.1 13,488.4
13,105.4 13,260.8 13,104.5 I3,26 1.2 13,488.7 13,261.5 13,488.8
Curve C (1 18: 118) Curve D ( 120:1 18) Curve E ( 120:120)
Theoretical masses are average masses.
J. Neurochem.. Vol. 59, No. 5, 1992
levels of NGF activity resulted. Proteolytic modifications of mature NGF, such as des octa NGF (murine NGF lacking the first eight N-terminal amino acids), seen in mouse submaxillary gland and saliva (Moore et al., 1974; Mobley et al., 1976; Burton et al., 1978), were not observed for these preparations of rhNGF in CHO cells. One of the enzymes reportedly responsible for posttranslational processing of murine NGF is the y-protease. The best-described activity associated with yprotease is the precursor processing of pro-murine NGF resulting in an enzyme-product complexation as part of the 7s NGF complex (Jongstra-Bilen et al., 1989).Such a complex or its potential has been shown only for murine sources (Darling and Fahnestock, 1988;Schwarz et al., 1989), and the in vivo processing status of the human molecule remains unknown. In light of this, we endeavored to demonstrate that regardless of the processing status of natural human NGF, some of the proteolytic modifications described were not crucial to the expression of biological activities of human NGF. Most studies regarding the isolation of various proteolytic species of the 0 form of murine NGF have resulted in mixtures of heterodimers and homodimers generally containing more than one of any of these species (Moore et al., 1974; Mobley et al., 1976). We have performed controlled enzymatic cleavages of rhNGF, followed by the isolation of resulting dimeric species, which mimic some of those observed for the murine molecule, to examine their relation-
1681
RECOM3INANT HUMAN NERVE GROWTH FACTOR A
100,
I
I
I
100
1 02
10'
I
20
40
60
80
103
104
NGF (pglml)
I
-
Time (rnin)
.001
.01
.1
1
lo
100
1000
NGF (ng/ml)
/
0
/
10
,
I
l
30
20
l
40
50
,
'
60
Time (rnin)
Analytic_.l HPLC of purifiedrhNGF. A Analyticc IPIEC of FI dimers. Individual samdes were iniected onto a Polvsulrh foethyl A cation-exchange (4.6-X 200-mm) column as descAbed in Materials and Methods: curve A, rhNGF 117:117; curve B, rhNGF 118:117; curve C, rhNGF 118:118; curve D, rhNGF 120:118; and curve E, rhNGF 120:120.B: Analytical C, RP-HPLC analysis of various rhNGF species. Individual samples were injected onto an analytical YMC C, RP-HPLC column as described in Materials and Methods. Curves A-E are as described in A.
1
2
3
4
5
6
7
FIG. 6. SDS-PAGE of rhNGF species. Samples were prepared and subjected to SDS-PAGE on 15% gels containing -4 M urea as described in Materials and Methods: lane 1, molecular-mass standards; lane 2, rhNGF 117:117; lane 3, rhNGF 118:117; lane 4, rhNGF 118:118; lane 5, rhNGF 118:120; lane 6, rhNGF 120:120; and lane 7, a mixture of rhNGF 120:120 and rhNGF 117:l 17.
FIG. 7. Biological characterization of rhNGF species. A PC12 assay of rhNGF species. Individual rhNGF species were analyzed in the PC12 assay (single determination): 120:l 20 (0),120:118 (O), 118:118 (O),118:117 (m), and 117:117 (A). 6: Chick dorsal root ganglion (DRG) survival assay of rhNGF species. Individual rhNGF species were tested in the chick DRG survival assay (n = 2 determinations):120:l 20 (0),120:118 (O), 118:118 (O),118:117 (m), and 117:117 (A).
ships with regard to specific activity. As described, the tryptic cleavage of 120:120, when separated on HPIEC (see Fig. 4), resulted in a mixture of dimeric forms composed primarily of 120:120, 120:1 18, and 1 18:1 18 dimers. Subsequent limited cleavage of the 118:118 pool by carboxypeptidase B resulted in two additional dimer forms: 1 18: 1 17 and 1 17:1 17. Each of these forms has been shown to be distinguishable by HPIEC (Fig. 5A). Biological assessment of the potency of each of the dimeric species described above shows that all are equally potent in either maintaining neuronal survival in primary culture in the chick dorsal root ganglion assay (Fig. 7B) or causing neurite extension in the rat PC 12 assay (Fig. 7A). The ED,, and the maximal effect within each assay for each of the dimeric species are reached at essentially the same concentrations. Murine 118-residue NGF prepared in our hands is equipotent with the rhNGF species described herein in the respective bioassays (L. E. Burton and C. H. Schmelzer, unpublished data). Analytical RP-HPLC (Fig. 5B) of the separated species noted in Fig. 5A indicates the gross homogeneity of each of the preparative HPIEC peaks (Fig. 4)as well
J. Neurochem., Vol. 59, No. 5, 1992
C. H. SCHMELZER ET AL.
1682
as showing, in the case of heterodimers such as 120:1 18, that this mode of chromatography reveals the monomeric forms of NGF present in each heterodimer. This observation is in contrast to that made by Barnett et al. (1991). These authors observed that monomers of putative 120-residue rhNGF in fact bound more tightly than their apparent dimer to an RP-HPLC system comparable to that used here. Although the data presented here indicate only very small elution condition differences between minor modified forms of rhNGF, we have also observed (data not shown) species of rhNGF that elute in similar positions with our system, i.e., with tighter resin association. We have noted that the contents of this (these) species increase with prolonged exposure to acid (TFA or HCl) or other denaturants (urea or guanidine-HCl) and that the rate of appearance is time and temperature dependent (data not shown). We suggest that this form(s) of rhNGF may constitute an additional structural change that is reversible under neutralized or native conditions (data not shown). Studies examining the character of this “form” of rhNGF are currently underway. To summarize, we have demonstrated that rhNGF can be produced in CHO cells by recombinant techniques, purified to homogeneity, and retain full biological activity, as judged by the chick dorsal root ganglion assay and the rat PC12 neurite extension assay. We have also shown definitively that rhNGF of 120 residues is equivalent to the 1 1%residue murine molecule in potency based on mass in these bioassays. Furthermore, direct comparison of characterized proteolytic modifications of the mature 120-residuespecies, i.e., 118 and 117 residues, definitively show no effect on the biological potency of the molecule. These data indicate that the rhNGF produced in CHO cells is a candidate source for material used for study of efficacy in humans. Acknowledgment: The authors are indebted to Reed Hams for N-terminal sequence analyses, Karen Wagner for amino acid compositional analyses, Dr. Eva Szonyi for dorsal root ganglion assays, William Mulhern for SEC, and Aldona T. Gorrell, Cathleen Yedinak, and Hannah Beasley for support in purification work.
REFERENCES Barnett J., Chow J., Nguyen B., Eggers D., Osen E., Jarnagin K., Saldou N., Straub K., Gu L., Chaing H.-S., Fausnaugh J., Townsend R. R., Lile J., Collins F., and Chan H. (1991) Physicochemical characterization of recombinant human nerve growth factor produced in insect cells with a baculovirus vector. J. Neurochem. 57, 1052-1061. Bennett G. L., Burton L. E., Chan W.-P., and Wong W. L. T. (1990) Sensitive enzyme immunoassay for recombinant human NGF. J. Cell. Biochem. 14F, 87. Bocchini V. and Angeletti P. U. (1969) The nerve growth factor: purification as a 30,000-molecular-weight protein. Proc. Natl. Acad. Sci. USA 64, 787-794.
J. Neurochem., Vol. 59, No. 5, 1992
Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72,248-254. Bruce G. and Heinrich G. (1989) Production and characterization of biologically active recombinant human nerve growth factor. Neurobiol. Aging 10, 89-94. Button L. E., Wilson W. H., and Shooter E. M. (1978) Nerve growth factor in mouse saliva. J. Biol. Chem. 253,7807-7812. Buxser S.. Vroegop S.. Decker D., Hinzmann J., Poorman R.. Thomsen D. R., Stier M., Abraham I., Greenberg B. D., Hatzenbuhler N. T., Shea M., Curry K. A., and Tomich C.4. C. (1991) Single-step purification and biological activity of human nerve growth factor produced from insect cells. J. Neurochem. 56, 1012-1018. Chapman C. A., Banks B. E. C., Vernon C. A., and Walker J. M. (1 98 1) The isolation and characterization of nerve growth factor from the prostate gland of the guinea-pig. Eur. J. Biochem. 115,347-35 1. Cohen S. (1960) Purification of a nerve growth promoting protein from the mouse salivary gland and its neurocytotoxic antiserum. Proc. Natl. Acad. Sci. USA 46, 302-3 1 1. Darling T. L. and Fahnestock M. (1988) The high molecular weight nerve growth factor complex from Mastomys natalensis differs from the murine nerve growth factor complex. Biochemistry 27.6686-6692. Davies A. M. ( 1989) Neurotrophic factor bioassay usingdissociated neurons, in IBRO Handbook Series: Methods in the Neurosciences, Vol. 12: Nerve Growth Factors (Rush R. A,, ed), pp. 95-109. Wiley and Sons, New York. Edwards R. H.. Selby M. J., Garcia P. D., and Rutter W. J. (1988) Processing of the native nerve growth factor precursor to form biologically active nerve growth factor. J. Biol. Chem. 263, 6810-6815. Goldstein L. D., Reynolds C. P., and Perez-Polo J. R. (1978) Isolation of human nerve growth factor from placental tissue. Neurochern. Res. 3, 175-183. Graham F. and van der Eb A. (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52,456457. Greene L. A. (1977) A quantitative bioassay for nervegrowth factor activity employing a clonal pheochromocytoma cell line. Brain Res. 133, 350-353. Harper G. P. and Thoenen H. (1980) The distribution of nerve growth factor in the male sex organs of mammals. J. Neurochem. 34,893-903. Harper G. P., Barde Y.-A,, Burnstock G.. Carstairs J. R., Dennison M. E., Suda K., and Vernon C. A. (1979)Guineapigprostateis a rich source of nerve growth factor. Nature 279, 160-162. HarperG. P.,Glanville R. W.,andThoenenH.(1982)Thepurification of nerve growth factor from bovine seminal plasma. J. Biol. Chem. 257,8541-8548. Hogue-Angeletti R. A. and Bradshaw R. A. (1979) Nerve growth factor in snake venoms, in Handbook ofExperimenta1 Pharmacolog.v, Vol. 52 (Lee C.-Y., ed), pp. 276-294. Springer-Verlag, Berlin. Huang M. and Gorman C. (1990) Intervening sequence increases efficiency of RNA 3’ processing and accumulation of cytoplasmic RNA. Nucleic Acids Res. 18,937-947. Iwane M., Kitamura Y., Kaisho Y., Yoshimura K., Shintani A,, Sasada R., Nakagawa S., Kawahara K., Nakahama K., and Kakinuma A. ( 1990) Production, purification and characterization of biologically active recombinant human nerve growth factor. Biochem. Biophys. Res. Commun. 171, 116122. Jongstra-Bilen J., Coblentz L., and Shooter E. M. (1989) The in vitro processing of the NGF precursors by the gamma-subunit ofthe 7s-NGF complex. Mol. Brain Res. 5, 159-169. Kanaya E., Higashizaki T., Ozawa F., Hirai K., Nishizawa M., Tokunaga M., Tsukui H., Hatanaka H., and Hishinuma F. ( 1 989) Synthesis and secretion of nerve growth factor by Saccharomyces cerevisiae. Gene 83, 65-74.
RECOMBINANT HUMAN NERVE GROWTH FACTOR Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685. Levi-Montalcini R. (1987) The nerve growth factor 35 years later. Science 237, 1154-1 162. Longo F. M., Woo J. E., and Mobley W. C. (1989) Purification of nerve growth factor, in IBRO Handbook Series: Methods in the Neurosciences, Vol. 12: Nerve Growth Factors (Rush R. A., ed), pp. 3-30. Wiley and Sons, New York. Mobley W. C., Schenker A., and Shooter E. M. (1976) Characterization and isolation of proteolytically modified nerve growth factor. Biochemistry 15, 5543-5552. Moore J. B. Jr., Mobley W. C., and Shooter E. M. (1974) Proteolytic modification of the beta nerve growth factor protein. Biochemistry 13, 833-839. Morrissey J. H. (1981) Silver stain for proteins in polyacrylamide gels: a modified procedure with enhanced uniform sensitivity. Anal. Biochem. 117,307-3 10. Murphy R. A,, Chlumecky V., Smillie L. B., Carpenter M., Nattriss M., Anderson J. K., Rhodes J. A.. Barker P. A., Siminoski K., Campenot R. B., and Haskins J. (1989) Isolation and characterization of a glycosylated form of beta nerve growth factor in mouse submandibular glands. J. Biol. Chem. 264, 1250212509. Petrides P. E. and Shooter E. M. ( I 986) Rapid isolation of the 7Snerve growth factor complex and its subunits from murine submaxillary glands and saliva. J. Neurochem. 46, 121-725. Schwarz M. A,, Fisher D., Bradshaw R. A., and Isackson P. J. (1989) Isolation and sequence of a cDNA clone of 0-nerve growth factor from the guinea pig prostate gland. J. Neurochem. 52, 1203-1209.
1683
Smith A. P., Varon S., and Shooter E. M. (1968) Multiple forms of the nerve growth factor protein and its subunits. Biochemistry 7,3259-3268. Thoenen H. and Barde Y.-A. (1980) Physiology of nerve growth factor. Physiol. Rev. 60, 1284-1335. Ullrich A., Gray A., Berman C., and Dull T. J. (1983) Human p nerve growth factor gene sequence highly homologous to that of mouse. Nature 303,821-825. Urlaub G. and Chasin A. J. (1980) Isolation of Chinese hamster cell mutants deficient in dihydrofolate reductase activity. Proc. Natl. Acad. Sci. USA 77, 4216-4220. Varon S., Nomura J., and Shooter E. M. (1967a) The isolation of the mouse nerve growth factor protein in high molecular weight form. Biochemistry 6, 2202-2209. Varon S., Nomura J., and Shooter E. M. (19676) Subunit structure of a high molecular-weight form of the nerve growth factor from mouse submaxillary gland. Proc. Natl. Acad. Sci. USA 57, 1782-1789. Varon S., Nomura J., and Shooter E. M. (1968) Reversible dissociation of the mouse nerve growth factor protein into its different subunits. Biochemistry 7, 1296-1303. Wise R. J., Barr P. J., Wong P. A., Keifer M. C., Brake A. J., and Kaufman R. J. (1990) Expression of a human proprotein processing enzyme: correct cleavage of the von Willebrand factor at a paired basic amino acid site. Proc. Natl. Acad. Sci. USA 87, 9378-9382. Yankner B. A. and Shooter E. M. (1982) The biology and mechanism of action of nerve growth factor. Annu. Rev. Biochm. 51, 845-868.
J. Neurochem.. Vol. 59, No. 5, 1992
