Skip to main content
PMC home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 1999 Mar 16;96(6):3149–3153. doi: 10.1073/pnas.96.6.3149

The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A

James B Lawrence *, Claus Oxvig †, Michael T Overgaard †, Lars Sottrup-Jensen †, Gerald J Gleich ‡, Lara G Hays §, John R Yates III §, Cheryl A Conover *,
PMCID: PMC15910  PMID: 10077652

Abstract

Proteolytic cleavage of the six known insulin-like growth factor binding proteins (IGFBPs) is a powerful means of rapid structure and function modification of these important growth-regulatory proteins. Intact IGFBP-4 is a potent inhibitor of IGF action in vitro, and cleavage of IGFBP-4 has been shown to abolish its ability to inhibit IGF stimulatory effects in a variety of systems, suggesting that IGFBP-4 proteolysis acts as a positive regulator of IGF bioavailability. Here we report the isolation of an IGF-dependent IGFBP-4-specific protease from human fibroblast-conditioned media and its identification by mass spectrometry microsequencing as pregnancy-associated plasma protein-A (PAPP-A), a protein of unknown function found in high concentrations in the maternal circulation during pregnancy. Antibodies raised against PAPP-A both inhibited and immunodepleted IGFBP-4 protease activity in human fibroblast-conditioned media. Moreover, PAPP-A purified from pregnancy sera had IGF-dependent IGFBP-4 protease activity. PAPP-A mRNA was expressed by the human fibroblasts and osteoblasts, and PAPP-A protein was secreted into the culture medium. In conclusion, we have identified an IGF-dependent IGFBP protease and at the same time assigned a function to PAPP-A. This represents an unanticipated union of two areas of research that were not linked in any way before this report.

Insulin-like growth factors (IGFs) are essential polypeptides with potent anabolic and mitogenic actions both in vivo and in vitro (1, 2). IGF bioactivity is modulated by distinct high-affinity IGF binding proteins (IGFBPs), six of which have been identified to date (3, 4). Adding another layer of refinement to this growth regulatory system has been the recent discovery that IGFBPs can undergo limited proteolysis with consequent modification of IGFBP structure and function, and hence IGF action (5). Currently, there is active investigation into identifying, classifying, and ascertaining the biological significance of IGFBP proteases in a variety of biological systems.

We have characterized previously IGFBP-4-specific metalloprotease activity secreted by normal human fibroblasts in culture (6, 7). Incubation in human fibroblast-conditioned medium (HFCM) under cell-free conditions results in cleavage of IGFBP-4 (24 kDa nonreduced, 32 kDa reduced) in the midportion of the molecule producing distinct fragments of 18 and 14 kDa. The defining feature of this IGFBP-4 proteolytic reaction is its absolute dependence on IGFs for functional activity. Only very low concentrations of IGFs are needed, and, in general, IGF-II is more potent than IGF-I in activating proteolysis (6). Similar IGF-dependent IGFBP-4 proteolysis has been described in cultures of normal human osteoblasts (8, 9), vascular smooth muscle cells (10), endometrial stromal cells (11), decidual cells (12), and granulosa cells (13), as well as in ovarian follicular fluid (14), but in none of these systems was the responsible enzyme identified. This lack of identification has made delineation of a physiological role for IGFBP-4 proteolysis extremely difficult. Nonetheless, based on in vitro data showing increased bioavailability of IGFs with IGFBP-4 proteolysis in the various cell systems, it has been suggested that the IGF-dependent IGFBP-4 protease plays an important role in local proliferative responses such as wound healing, bone remodeling, and atherosclerotic plaque development, as well as in several aspects of human reproduction (5–15).

As a first step toward defining this IGF regulatory mechanism, we purified the IGF-dependent IGFBP-4 protease from HFCM. Herein we describe this enzyme purification and its identification as pregnancy-associated plasma protein-A (PAPP-A).

MATERIALS AND METHODS

HFCM.

Human fibroblasts from a normal male donor (GM03652, Coriell Institute, Camden, NJ), were grown to confluency in T75 flasks, washed twice with DMEM, and then changed to 15 ml of 50:50 Waymouth’s medium:DMEM containing 100 units/ml penicillin, 100 μg/ml streptomycin/4 mM glutamine/0.1% BSA for 6 hr at 37°C. The cells were again washed and changed to 10 ml of the serum-free medium and incubated for 72 hr at 37°C. This HFCM was placed in a sterile conical tube and centrifuged at 1,500 × g for 30 min at 4°C to remove cellular debris, decanted into another sterile conical tube, and stored at −30°C.

IGFBP-4 Protease Assay.

IGFBP-4 proteolysis was assayed, as described previously (6–8), by incubating sample overnight at 37°C with 2 mM CaCl2/50 mM Tris (pH 7.5)/10,000 cpm of [125I]IGFBP-4 in the absence and presence of 5 nM IGF-II in a total volume of 25 μl. Proteins were separated by reducing 15% SDS/PAGE and were visualized by autoradiography.

For some experiments, PAPP-A polyclonal antibody (16) or nonspecific rabbit IgG was added to the assay mixture (Fig. 2a) or was used in conjunction with protein G plus protein A-agarose (Oncogene Science) to immunoprecipitate IGFBP-4 protease activity before assay (Fig. 2b).

Figure 2.

Figure 2

Open in a new tab

Confirmation of IGF-dependent IGFBP-4 protease as PAPP-A. (a) Inhibition of IGFBP-4 protease activity in HFCM by PAPP-A/proMBP antibodies. HFCM was assayed for IGFBP-4 protease activity, as described in Fig. 1 and in Materials and Methods, in the presence of indicated titer of either PAPP-A/proMBP polyclonal antibody (anti-PAPP-A) or nonspecific rabbit IgG. (b) Immunodepletion of IGFBP-4 protease activity from HFCM by PAPP-A antibodies. HFCM was precleared with a 1:50 titer of nonspecific rabbit IgG. The precleared fractions were immunodepleted with indicated titer of PAPP-A/proMBP polyclonal antibody and assayed for IGFBP-4 protease activity. (c) IGF-dependent IGFBP-4 protease activity of purified PAPP-A. PAPP-A/proMBP (0.5 μg) purified from pregnancy sera was assayed for IGFBP-4 protease activity in the absence (-) and presence (+) of 5 nM IGF-II.

Purification.

HFCM (800 ml) in six 130- to 150-ml batches, were passed over a 25-ml bed volume of iminodiacetic acid immobilized to Sepharose 6B (Sigma) loaded with Zn+2 and equilibrated with 50 mM Tris/50 mM NaCl, pH 7.4. Bound proteins were eluted sequentially with 50 ml each of a 0.5 pH unit stepwise decreasing gradient. The pH 5.0 fraction was immediately adjusted to pH 7.4 and passed over a 1-ml bed volume wheat germ agglutinin column equilibrated with 20 mM Tris/100 mM NaCl (pH 7.5) at 4°C. Bound proteins were eluted with 15 ml each of 50 mM Tris/100 mM NaCl (pH 7.5) alone and then with 25 mM and 500 mM N-acetylglucosamine. Five-ml fractions were collected and assayed for IGFBP-4 protease activity. Fractions from the six chromatography runs were pooled, concentrated by ultrafiltration, and electrophoresed through a 5% acrylamide SDS/PAGE gel under reducing conditions. The SDS/PAGE gel was silver stained, and four bands at 400, 230, 200, and 175 kDa that correlated to IGFBP-4 protease activity were excised for mass spectrometric analysis.

Mass Spectrometry Sequencing.

The excised bands were subjected to in-gel trypsin digestion (17). Soluble fragments were recovered from the digestion, then resolved and characterized by microcolumn high-performance liquid chromatography (18) and automated tandem mass spectrometry. Microelectrospray columns were constructed from 360-μm o.d. × 100-μm i.d. fused silica capillary with the column tip tapered to a 5- to 10-μm opening (19). The microcolumns were packed with POROS 10 R2 (PerSeptive Biosystems, Framingham, MA) to a length of 10–15 cm. The mobile phase used for gradient elution consisted of (i) 0.5% acetic acid and (ii) acetonitrile/water 80:20 vol/vol containing 0.5% acetic acid. The gradient was linear from 0 to 40% over 30 min and from 40 to 60% over 5 min. The mass spectrometer used was a Finnigan-MAT (San Jose, CA) LCQ equipped with a microelectrospray ionization source. Tandem mass spectra were acquired during the entire gradient run automatically as previously described (20). The protein sequence and nucleotide sequence databases were searched directly with tandem mass spectra by using the computer program sequest (Finnigan, San Jose, CA) (21). Each sequence returned by sequest was verified by inspecting the fit of the amino acid sequence to the corresponding tandem mass spectrum.

Northern Blot Analysis.

Following previously published protocols (22), poly(A)-tailed mRNA (10 μg) was electrophoresed through a 1.5% agarose gel and transferred to nylon membrane (Hybond-N, Amersham Pharmacia). The membrane was prehybridized for 6 hr and then hybridized overnight at 43°C with 107 cpm of 32P-labeled PAPP-A cDNA probe corresponding to nucleotide 10–2365 (GenBank accession no. X68280). The membrane was washed three times, dried, and exposed to film.

Western Blot Analysis.

Samples were run on 8% Tris/tricine gels and blotted onto polyvinylidene difluoride membrane (Millipore). The membrane was blocked in 2% Tween 20 and washed with 50 mM Tris/500 mM NaCl/0.1% Tween 20/1% fetal bovine serum, pH 9.0. The blots were then incubated with monoclonal PAPP-A antibodies 234–2 and 234–5 (23). The secondary antibodies were peroxidase-conjugated anti(mouse-IgG) P260 (Dako). Blots were developed by using enhanced chemiluminescence (ECL, Amersham).

ELISA.

PAPP-A antigen was measured by using a standard sandwich ELISA. The capture antibody was polyclonal anti-PAPP-A/proMBP, and detection was done with anti-PAPP-A monoclonal antibodies 234–2 and 234–5 (23) followed by peroxidase-conjugated anti(mouse-IgG) P260. Highly purified PAPP-A/proMBP, prepared as described previously (16), was used for calibration.

RESULTS AND DISCUSSION

Purification of the IGF-dependent IGFBP-4 protease was monitored with a specific bioassay, i.e., cell-free degradation, of [125I]IGFBP-4 into 18- and 14-kDa radiolabeled fragments in the presence, but not the absence, of added IGF-II (6, 7). This approach had the clear advantage of ensuring that we were purifying our specific enzyme of interest. We obtained highly purified IGF-dependent IGFBP-4 protease from 800 ml of HFCM by a combination of zinc chelate and lectin affinity chromatography (Fig. 1). The final fraction containing the IGF-dependent IGFBP-4 protease activity was further analyzed by SDS/PAGE, which revealed four high molecular mass bands at 400, 230, 200, and 175 kDa. The proteins in these bands were identified by tandem mass spectrometry microsequencing (Table 1). All of the peptides represented known proteins or proteins deduced from known cDNA sequences. An extensive literature search involving a comparison of the characteristics of the IGF-dependent IGFBP-4 protease from HFCM and the proteins identified by tandem mass spectrometry revealed only one match. This candidate protein was PAPP-A. PAPP-A is one of four proteins originally isolated in 1974 from normal human pregnancy serum (24). Although knowledge of PAPP-A’s biological function was lacking, particular interest in PAPP-A had developed with its suggested clinical utility as an index of placental function and a first-trimester screen for Down’s Syndrome (25, 26). In addition, placental PAPP-A “knock-out” in humans appears to be associated with Cornelia de Lange syndrome, a condition involving incomplete fetal development and subsequent deformities (27). PAPP-A and the IGF-dependent IGFBP-4 protease from HFCM are similar in that they are both high molecular weight glycosylated proteins that bind Zn2+ (7, 16). Furthermore, amino acid sequence derived from cloned cDNA encoding PAPP-A reveals a specific Zn2+ binding motif (HEXXHXXGXXH) and a Met-turn further carboxyl terminal found only in the metzincin family of metalloproteases (28–31). In pregnancy, circulating PAPP-A is complexed to the proform of eosinophil major basic protein (proMBP) in a disulfide-bridged 2:2 heterotetramer denoted PAPP-A/proMBP (32). In nonreducing SDS/PAGE, PAPP-A/proMBP migrates with a molecular mass of ≈500 kDa. In reducing SDS/PAGE, the molecular mass of PAPP-A monomer is ≈200 kDa (32). Like PAPP-A, proMBP is synthesized in the placenta during pregnancy (33) and its role in pregnancy is unknown.

Figure 1.

Figure 1

Open in a new tab

Purification of the IGF-dependent IGFBP-4 protease from HFCM. (a) Metal-chelating affinity chromatography. Bound proteins were eluted with a descending stepwise pH gradient, and 50 μl aliquots of each of these fractions were dialyzed and assayed for IGFBP-4 protease activity. IGF-dependent IGFBP-4 protease activity is defined as the loss of intact 24-kDa [125I]IGFBP-4 and the appearance of 18- and 14-kDa radiolabeled fragments (denoted by arrows) in the presence (+), but not in the absence (-), of 5 nM IGF-II. HFCM, starting material; FT, flowthrough. (b) Wheat-germ agglutinin chromatography. Metal-chelating affinity chromatography fraction eluting at pH 5.0 was adjusted to pH 7.4 and immediately passed over a wheat-germ agglutinin column. Non- and weakly bound proteins came out in the flowthrough (FT) or were eluted with three washes of 0 mM and three washes of 25 mM N-acetylglucosamine. IGF-dependent IGFBP-4 protease activity was eluted with three washes of 500 mM N-acetylglucosamine.

Table 1.

Protein identification by means of mass spectrometry

Protein molecular mass, kDa Protein identity Peptides identified
175 Human α-2-macroglogulin 854–863 QTVSWAVTPK
935–945 LPPNVVEESAR
1264–1273 YGAATFTRTG
Human thrombospondin 1 61–83 IEDANLIPPVPDDKFQDLVDAVR
217–228 FVFGTTPEDILR
289–299 TIVTTLQDSIR
1086–1099 NALWHTGNTPGQVR
1148–1164 LGLFVFSQEMVFFSDLK
Human collagen 1063–1084 SGDRGETGPAGPAGPVGPVGAR
Bovine α-1-antichymotrypsin isoform pHHK1 37–58 GTSVDGHSLASSNTDFAFSLYK
321–329 DILSQLGIK
331–346 IFTSDADFSGITDDHK
Human soares testis NHT cDNA clone 727252 5′ 46–57 KLEEGTKQQWQG
Sommer Pristionchus Pristionchus pacificus 32–48 EVCNKLIPDSIGKDIEK
Human ribonuclease 6 124–139 GCLCLALLCLGGADKR
Caenorhabditis elegans cDNA clone yk182d6 68–75 FLIQVIHK
200 Human pregnancy-associated plasma protein-A 110–116 ADLELPR
133–143 SPAVITGLYDK
190–209 SYLPGQWVYLAATYDGQFMK
373–387 EQVDFQHHQLAEAFK
1071–1087 TISYPYSQLAQTTFWLR
1180–1195 SFDNFDPVTLSSCQRG
Rat hemiferrin 74–91 DQTVIQNTDGNNNEAWAK
Bovine transferrin 230–243 HSTVFDNLPNPEDR
230 Human laminin γ-1 chain 1122–1135 NTIEETGNLAEQAR
1407–1423 EAQQALGSAAADATEAK
Human laminin β-1 chain 1509–1530 NFLTQDSADLDSIEAVANEVLK
1597–1611 EALEEAEKAQVAAEK
400 Human collagen α-1 1–89 TLIPSTFFR
117–123 VIYLGLR
190–214 SSQALAFESSAGIFMGNAGATGLER
215–229 FTGSLQQLTVHPDPR
1243–1251 AAGLLSTYR
1344–1360 TADTAVTGLASPLSTGK
Open in a new tab

Peptides from bovine serum albumin were identified in each of the four analyzed bands and have been omitted for brevity. 

Identification of the IGF-dependent IGFBP-4 protease as PAPP-A was verified with further biochemical analyses. Anti-PAPP-A/proMBP polyclonal IgG, but not nonspecific rabbit IgG, inhibited IGF-dependent IGFBP-4 protease activity in HFCM in a dose-dependent manner (Fig. 2a). At a 1:50 titer of PAPP-A antibody, IGFBP-4 proteolysis was completely inhibited; a 1:500 titer inhibited 84% of the protease activity, whereas a 1:5,000 titer inhibited 15% of IGF-induced IGFBP-4 protease activity in these cell-free assays. In other experiments, we used anti-PAPP-A/proMBP polyclonal IgGs to immunodeplete specifically and completely IGFBP-4 protease activity from the medium (Fig. 2b). Moreover, PAPP-A/proMBP that had been purified from serum of pregnant women (16) exhibited IGF-dependent IGFBP-4 protease activity (Fig. 2c). In cell-free assay, PAPP-A/proMBP alone had no effect on [125I]IGFBP-4, but addition of IGF-II initiated proteolysis into radiolabeled fragments of 18 and 14 kDa, identical to what is seen with HFCM. Similar results were obtained with four different preparations of highly purified PAPP-A/proMBP (data not shown).

PAPP-A is most highly expressed in the syncytiotrophoblast of the placenta, which is the main source of circulating PAPP-A in pregnancy (33). However, PAPP-A has been detected in serum from nonpregnant as well as pregnant women and in preovulatory follicular fluid and has been immunolocalized to secretory endometrium, vascular endothelium, and actively proliferating fetal and adult tissues (25, 34, 35). These results were all based on polyclonal antisera that react with both PAPP-A and proMBP (32) and to an unknown extent with other proteins. In Fig. 3a, we demonstrate unequivocally PAPP-A expression in cultured human fibroblasts and bone cells by Northern blot analysis. Human fibroblasts expressed PAPP-A transcripts at ≈13 and 8.5 kilobases, similar to that detected by Northern blot analysis of placental mRNA (28, 29). The results from Northern blotting also show PAPP-A mRNA expression by normal human osteoblasts from adult and fetal sources and by osteoprogenitor cells, but not by several osteosarcoma cell lines. These findings are in agreement with those of previous studies showing IGF-dependent IGFBP-4 protease activity in medium conditioned by normal human osteoblasts but not by transformed osteoblastic cells (36). Not only is PAPP-A mRNA expressed by human fibroblasts, but PAPP-A protein is, in fact, secreted by cultured human fibroblasts. By ELISA, HFCM was found to contain 0.18 ± 0.02 μg/ml PAPP-A (n = 4). For comparison, term pregnancy serum contains ≈25 μg/ml PAPP-A. Western blotting analysis showed that HFCM reacted with monoclonal antibodies specific for PAPP-A (Fig. 3b). In nonreducing SDS/PAGE, PAPP-A from HFCM migrates with a molecular mass of ≈400 kDa, somewhat faster than PAPP-A/proMBP isolated from pregnancy serum but slower than PAPP-A monomer seen at ≈200 kDa in reducing SDS/PAGE. Expression of proMBP by fibroblasts was not detected by ELISA or Northern blotting (data not shown). Thus, PAPP-A in HFCM may form homodimers, although this remains to be determined.

Figure 3.

Figure 3

Open in a new tab

Expression and secretion of PAPP-A. (a) Northern analysis. Poly-A tailed mRNAs (10 μg) from cultured human cells were probed with 32P-labeled PAPP-A cDNA. The lanes are: a, adult fibroblasts; b, adult osteoblasts; c, fetal osteoblasts; d, marrow stromal cells; e, MG63 osteosarcoma cells; f, U2 osteosarcoma cells; g, TE85 osteosarcoma cells. (b) Western immunoblot. a, HFCM (10 μl of 30× concentrate corresponding to 0.05 μg PAPP-A by ELISA) and b, PAPP-A/proMBP purified from pregnancy serum (0.05 μg) were separated by SDS/PAGE under both nonreducing and reducing conditions, transferred to membrane, and immunoblotted with PAPP-A-specific monoclonal antibodies.

The IGF-dependent IGFBP-4 protease/PAPP-A is a new member of the metzincin family of metalloproteases (30, 31), and it is clear that this enzyme is unique among other multifunctional proteases capable of degrading IGFBP-4 (37, 38). The identification of PAPP-A as the IGF-dependent IGFBP-4 protease has immediate ramifications for placental function and fetal development. In addition, the identification of the IGF-dependent IGFBP-4 protease as PAPP-A and the availability of pure protein and associated molecular tools now allows determination of the mechanism underlying its IGF dependence and the biological role of localized IGF-dependent IGFBP-4 proteolysis in such diverse systems as wound healing, bone remodeling, atherosclerosis, and follicular development.

Acknowledgments

The authors acknowledge and thank Laurie Bale, Jay Clarkson, and Tufia Haddad for their excellent technical assistance.

This work was supported in part by the National Institutes of Heath (DK07352 to J.B.L., GM52095 to J.R.Y., and AI09728 to G.J.G.), the National Science Foundation (BIR9214821 to J.R.Y.), the Danish National Science Research Council (to C.O., M.T.O., and L.S.-J.), and the Mayo Foundation.

Footnotes

Abbreviations: IGF, insulin-like growth factor; IGFBP, insulin-like growth factor binding protein; HFCM, human fibroblast-conditioned medium; PAPP-A, pregnancy-associated plasma protein-A; proMBP, proform of major basic protein.

References

  • 1.Daughaday W H, Rotwein P. Endocr Rev. 1989;10:68–90. doi: 10.1210/edrv-10-1-68. [DOI] [PubMed] [Google Scholar]
  • 2.D’Ercole A J. Growth Dis. 1996;25:573–590. doi: 10.1016/s0889-8529(05)70341-8. [DOI] [PubMed] [Google Scholar]
  • 3.Shimisaki S, Ling N. Prog Growth Factor Res. 1991;3:243–266. doi: 10.1016/0955-2235(91)90003-m. [DOI] [PubMed] [Google Scholar]
  • 4.Jones J I, Clemmons D R. Endocr Rev. 1995;16:3–34. doi: 10.1210/edrv-16-1-3. [DOI] [PubMed] [Google Scholar]
  • 5.Fowlkes J L. Trends Endocrinol Metab. 1997;8:299–306. doi: 10.1016/s1043-2760(97)00112-4. [DOI] [PubMed] [Google Scholar]
  • 6.Conover C A, Kiefer M C, Zapf J. J Clin Invest. 1993;91:1129–1137. doi: 10.1172/JCI116272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Lawrence, J. B., Bale, L. K., Haddad, T. C., Clarkson, J. T. & Conover, C. A. Growth Horm. IGF Res., in press. [DOI] [PubMed]
  • 8.Durham S K, Kiefer M C, Riggs B L, Conover C A. J Bone Miner Res. 1994;9:111–117. doi: 10.1002/jbmr.5650090115. [DOI] [PubMed] [Google Scholar]
  • 9.Kanzaki S, Hilliker S, Baylink D J, Mohan S. Endocrinology. 1994;134:383–392. doi: 10.1210/endo.134.1.7506211. [DOI] [PubMed] [Google Scholar]
  • 10.Parker A, Gockerman A, Busby W H, Clemmons D R. Endocrinology. 1995;136:2470–2476. doi: 10.1210/endo.136.6.7538463. [DOI] [PubMed] [Google Scholar]
  • 11.Irwin J C, Dsupin B A, Giudice L C. J Clin Endocrinol Metab. 1995;80:619–626. doi: 10.1210/jcem.80.2.7531715. [DOI] [PubMed] [Google Scholar]
  • 12.Myers S E, Cheung P T, Handwerger S, Chernausek S K. Endocrinology. 1993;133:1525–1531. doi: 10.1210/endo.133.4.7691578. [DOI] [PubMed] [Google Scholar]
  • 13.Iwashita M, Kudo Y, Takeda Y. Mol Hum Reprod. 1998;4:401–405. doi: 10.1093/molehr/4.4.401. [DOI] [PubMed] [Google Scholar]
  • 14.Chandrasekher Y A, Van Dessel H J H M, Fauser B C J M, Giudice L C. J Clin Endocrinol Metab. 1995;80:2734–2739. doi: 10.1210/jcem.80.9.7545699. [DOI] [PubMed] [Google Scholar]
  • 15.Conover C A. Prog Growth Factor Res. 1995;6:301–309. doi: 10.1016/0955-2235(95)00032-1. [DOI] [PubMed] [Google Scholar]
  • 16.Oxvig C, Sand O, Kristensen T, Kristensen L, Sottrup-Jensen L. Biochim Biophys Acta. 1994;1201:415–423. doi: 10.1016/0304-4165(94)90071-x. [DOI] [PubMed] [Google Scholar]
  • 17.Sheuchenko A, Wilm M, Mann M. J Protein Chem. 1997;16:481–490. doi: 10.1023/a:1026361427575. [DOI] [PubMed] [Google Scholar]
  • 18.Yates J R, III, Eng J K, McCormack A L, Schieltz D. Anal Chem. 1995;67:1426–1436. doi: 10.1021/ac00104a020. [DOI] [PubMed] [Google Scholar]
  • 19.Gatlin T, Kleemann G, Hays L, Yates J. Anal Biochem. 1998;263:93–101. doi: 10.1006/abio.1998.2809. [DOI] [PubMed] [Google Scholar]
  • 20.Link A J, Hays L, Carmack E, Yates J R., III Electrophoresis. 1997;18:1314–1334. doi: 10.1002/elps.1150180808. [DOI] [PubMed] [Google Scholar]
  • 21.Eng J K, McCormack A L, Yates J R., III Am Soc Mass Spec. 1994;5:976–989. doi: 10.1016/1044-0305(94)80016-2. [DOI] [PubMed] [Google Scholar]
  • 22.Conover C A, Clarkson J T, Bale L K. Endocrinology. 1995;136:1403–1410. doi: 10.1210/endo.136.4.7534698. [DOI] [PubMed] [Google Scholar]
  • 23.Qin Q-P, Christiansen M, Oxvig C, Pettersson K, Sottrup-Jensen L, Koch C, Norgaard-Pedersen B. Clin Chem. 1997;43:2323–2332. [PubMed] [Google Scholar]
  • 24.Lin T M, Halbert S P, Kiefer D J, Spellacy W N, Gall S. Am J Obstet Gynecol. 1974;118:223–236. doi: 10.1016/0002-9378(74)90553-5. [DOI] [PubMed] [Google Scholar]
  • 25.Stabile I, Grudzinskas J G, Chard T. Obstet Gynecol Surv. 1988;43:73–82. [PubMed] [Google Scholar]
  • 26.Wald N J, George L, Smith D, Densem J W, Petterson K. Br J Obstet Gynaecol. 1996;103:407–412. doi: 10.1111/j.1471-0528.1996.tb09765.x. [DOI] [PubMed] [Google Scholar]
  • 27.Westergaard J G, Chemnitz J, Teisner B, Poulsen H K, Ipsen L, Beck B, Grudzinskas J G. Prenatal Diagn. 1983;3:225–232. doi: 10.1002/pd.1970030307. [DOI] [PubMed] [Google Scholar]
  • 28.Kristensen T, Oxvig C, Sand O, Hundahl Moller N P, Sottrup-Jensen L. Biochemistry. 1994;33:1592–1598. doi: 10.1021/bi00172a040. [DOI] [PubMed] [Google Scholar]
  • 29.Haaning J, Oxvig C, Overgaard M T, Ebbesen P, Kristensen T, Sottrup-Jensen L. Eur J Biochem. 1996;37:159–163. doi: 10.1111/j.1432-1033.1996.0159n.x. [DOI] [PubMed] [Google Scholar]
  • 30.Hooper N M. FEBS Lett. 1994;343:1–6. doi: 10.1016/0014-5793(94)01079-x. [DOI] [PubMed] [Google Scholar]
  • 31.Bond J S, Beynon R T. Protein Sci. 1995;4:1247–1261. doi: 10.1002/pro.5560040701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Oxvig C, Sand O, Kristensen T, Gleich G J, Sottrup-Jensen L. J Biol Chem. 1993;268:12243–12246. [PubMed] [Google Scholar]
  • 33.Bonno M, Oxvig C, Kephart G M, Wagner J M, Kristensen T, Sottrup-Jensen L, Gleich G J. Lab Invest. 1994;71:560–566. [PubMed] [Google Scholar]
  • 34.Schindler A-M, Bischof P. Gynecol Obstet Invest. 1984;18:88–94. doi: 10.1159/000299055. [DOI] [PubMed] [Google Scholar]
  • 35.Sjoberg J, Wahlstrom T, Seppaa M, Rutanen E M, Loistinen R, Koskimies A I, Tenhunen A, Sinosich M J, Grudzinskas J G. Fertil Steril. 1984;41:551–557. doi: 10.1016/s0015-0282(16)47776-9. [DOI] [PubMed] [Google Scholar]
  • 36.Durham S K, Riggs B L, Harris S A, Conover C A. Endocrinology. 1995;136:1374–1380. doi: 10.1210/endo.136.4.7534697. [DOI] [PubMed] [Google Scholar]
  • 37.Conover C A, Perry J E, Tindall D J. J Clin Endocrinol Metab. 1995;80:987–993. doi: 10.1210/jcem.80.3.7533776. [DOI] [PubMed] [Google Scholar]
  • 38.Remacle-Bonnet M M, Garrouste F L, Pommier G J. Int J Cancer. 1997;72:835–843. doi: 10.1002/(sici)1097-0215(19970904)72:5<835::aid-ijc21>3.0.co;2-6. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES