help button home button Endocrine Society JCEM JCEM Call for Nominations for EIC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yavelow, J.
Right arrow Articles by Finlay, T. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yavelow, J.
Right arrow Articles by Finlay, T. H.
The Journal of Clinical Endocrinology & Metabolism Vol. 82, No. 3 745-752
Copyright © 1997 by The Endocrine Society

Endocrinological Oncology

{alpha}1-Antitrypsin Blocks the Release of Transforming Growth Factor-{alpha} from MCF-7 Human Breast Cancer Cells1

Jonathan Yavelow, Anna Tuccillo2, Susan S. Kadner, Joseph Katz and Thomas H. Finlay

Department of Biology, Rider University (J.Y., A.T.), Lawrenceville, New Jersey 08648; and the Department of Obstetrics and Gynecology (S.S.K., J.K., T.H.F.) and the Kaplan Cancer Center (T.H.F.), New York University Medical Center, New York, New York 10016

Address all correspondence and requests for reprints to: Thomas H. Finlay, Ph.D., New York University Medical Center, 550 First Avenue, New York, New York 10016.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Human breast cancer cells synthesize and release a variety of growth-modulating substances in response to estrogen stimulation, and it is generally accepted that the growth-promoting effects of estrogens are due at least in part to this autocrine/paracrine mechanism. Several of these growth-modulating substances, including transforming growth factor-{alpha} (TGF{alpha}) and its analogs, have been shown to require pericellular proteolysis for activation or release. Recently, we reported that MCF-7 human breast cancer cells are able to synthesize {alpha}1-antitrypsin ({alpha}1-AT), the major elastase inhibitor in human serum, and that there is a negative correlation between anchorage-independent growth of MCF-7 cells in soft agar and synthesis of {alpha}1-AT. The studies we present here were undertaken to gain an understanding of the mechanisms responsible for this observation. We show that release of TGF{alpha} from its membrane-bound precursor on MCF-7 cells is blocked by {alpha}1-AT whether the cells were maintained in the presence or absence of estradiol and that there is a clear dose-response relationship between the {alpha}1-AT concentration and both the release of TGF{alpha} and growth in soft agar. Consistent with this, TGF{alpha} release was increased in the presence of antibody to {alpha}1-AT. In contrast, TGF{alpha} release and growth in soft agar were not blocked by peptide inhibitors specific for trypsin- or chymotrypsin-like enzymes. The {alpha}1-AT concentration required for a half-maximal effect is lower for inhibition of TGF{alpha} release than it is for inhibition of colony formation (0.4 vs.1.5 µmol/L). However, both values are in the range of concentrations one might expect at the cell surface in vivo. A new MCF-7 cell subline producing 10-fold higher levels of {alpha}1-AT than its parent cell line was constructed by stable transfection of MCF-7 ML cells (a subline producing low levels of {alpha}1-AT) with an {alpha}1-AT complementary DNA. Growth in soft agar and release of TGF{alpha} were significantly decreased in cells transfected with the {alpha}1-AT complementary DNA compared to those in cells transfected with vector alone, although, TGF{alpha} expression was the same. The above observations support a model for growth regulation in human breast ductal epithelial cells in which growth factor activation and release are dependent on the coordinate action of proteases and protease inhibitors. This model would predict that {alpha}1-AT can act as a tumor suppressor in inhibiting the growth of breast cancer cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
TRANSFORMING growth factor-{alpha} (TGF{alpha}), a peptide structurally and functionally related to epidermal growth factor (EGF), interacts with the EGF receptor and elicits a mitogenic response in a variety of cell types (1). Proteolytic cleavage is required to split out the soluble, 50-amino acid form of TGF{alpha} from its membrane-bound precursor (pro-TGF{alpha}) on the cell surface (2), and this process may constitute an important regulatory step in the release of soluble TGF{alpha}. The observation that cleavage occurs on cells growing in serum-free medium indicates that the protease(s) is synthesized by the cell itself. Pro-TGF{alpha} has been shown to interact with EGF/TGF{alpha} receptors on the surface of adjacent cells (3). This juxtacrine action may play a role in cell-cell adhesion and mitogenesis, and cleavage may be an obligate step in a process generating two active forms of TGF{alpha}, each with different properties. The amino acid residues at the two cleavage sites in pro-TGF{alpha} suggest that both reactions should be accomplished by a yet to be identified elastase-like enzyme(s) (4). Because of this elastase-like specificity, the TGF{alpha} protease would not appear to be a Kex2- or furin-like enzyme (5, 6) or hepsin (7). One would also expect it to be neutralized by {alpha}1-antitrypsin ({alpha}1-AT also known as {alpha}1-proteinase inhibitor), the major human extracellular elastase inhibitor (8), or other extracellular elastase inhibitors, such as the secretory leukocyte protease inhibitor (SLPI) (9). Chinese hamster ovary (CHO) cells, which normally do not express pro-TGF{alpha}, can release soluble TGF{alpha} after transfection with a TGF{alpha} complementary DNA (cDNA) (10); however, the CHO cell protease responsible for this hydrolysis does not appear to be sensitive to {alpha}1-AT. This does not rule out such a role for an {alpha}1-AT-sensitive protease in other cell types. For example, cell surface elastase-like enzymes sensitive to {alpha}1-AT have been isolated from transformed rat liver epithelial cells and Schwann cells, both of which express TGF{alpha} (11).

The expression of TGF{alpha} occurs in normal breast tissue, breast tumors, and breast cancer cells in culture, and TGF{alpha} has been proposed to act as a major autocrine mediator of estrogen-stimulated growth in estrogen-dependent breast cancer cells (12, 13). TGF{alpha} is expressed by MCF-7 cells and stimulates proliferation and anchorage-independent growth in an autocrine/paracrine manner (13, 14). Concurrent staining of TGF{alpha} and the EGF/TGF{alpha} receptor appears to be characteristic of a clinically aggressive subset of breast carcinomas (15).

{alpha}1-AT is a broad spectrum inhibitor of serine proteases, including trypsin-, chymotrypsin-, and elastase-like enzymes (8). Its major physiological role is the inhibition of leukocyte elastases released at sites of inflammation. {alpha}1-AT is present at significant levels in blood and at lower levels in other extracellular fluids, including breast milk (16). Until recently, expression of {alpha}1-AT was thought to be restricted to hepatocytes and, to a lesser degree, to monocytes (17). However, we have shown that MCF-7 human breast cancer cells are also able to synthesize and secrete {alpha}1-AT and the closely related serine protease inhibitor (i.e. serpin) {alpha}1-antichymotrypsin ({alpha}1-ACHY) (18). Using a series of MCF-7 cell variants expressing different levels of {alpha}1-AT, we found a negative correlation between the synthesis of {alpha}1-AT and anchorage-independent growth of MCF-7 cells in soft agar (19, 20). We also demonstrated the expression of {alpha}1-AT and {alpha}1-ACHY messenger ribonucleic acid (mRNA) and protein in epithelioid trophoblast cells, which resemble cancer cells in their invasive behavior (21). The function of the {alpha}1-AT and {alpha}1-ACHY released by epithelial cells is unclear, although, they may play a role in the regulation of growth processes.

In this communication we describe in vitro studies that provide a possible mechanism for the reduced anchor-dependent growth of MCF-7 cells expressing high levels of {alpha}1-AT. We show that the release of TGF{alpha} from its membrane-bound precursor on MCF-7 cells is stimulated by estradiol (E2) and blocked by {alpha}1-AT, and that there is a clear dose-response relationship between {alpha}1-AT, whether added to the medium or expressed by the tumor cell itself, and the release of TGF{alpha} and growth in soft agar.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials

Geneticin (G418), ultrapure agarose, FBS, and other cell culture materials were obtained from Life Technologies (Gaithersburg, MD). Bowman-Birk inhibitor (BBI), steroid hormones, and immobilized protein A were purchased from Sigma Chemical Co. (St. Louis, MO). O-Tetradecanoyl-phorbol-13-acetate (TPA) was purchased from LC Laboratories (Woburn, MA). {alpha}1-AT was purchased from Athens Research and Technology (Athens, GA). The phAT85 plasmid containing a full-length human {alpha}1-AT cDNA in pBR322 was obtained from the American Type Culture Collection (Rockville, MD). Reagents for PAGE were obtained from Bio-Rad (Richmond, CA), and GeneScreen membrane was obtained from DuPont (Wilmington, DE). Rabbit antibodies to human {alpha}1-AT were obtained from Accurate Chemical & Scientific (Westbury, NY). Human TGF{alpha} and the human TGF{alpha} enzyme-linked immunosorbent assay (ELISA) kit were purchased from Oncogene Sciences (Cambridge, MA). Sheep polyclonal antibody to TGF{alpha} was purchased from R&D Systems (Minneapolis, MN). T7 RNA polymerase and the rabbit reticulocyte lysate cell-free translation system were obtained from Promega (Madison, WI). All other materials were of high purity from commercial sources.

Maintenance of cell cultures

MCF-7 cell sublines, obtained as previously described (18, 19), were maintained in DMEM supplemented with 10% heat-inactivated FBS, 2 mmol/L glutamine, and insulin (6 ng/mL). Cultures were maintained in humid air containing 5% CO2. Medium was changed every 2–3 days.

To observe the effects of protease inhibitors in our system (whether added or synthesized endogenously), the FBS used in the culture medium was depleted of its trypsin inhibitory capacity (TIC) by titration with trypsin covalently linked to Affigel-10 as previously described (18). This procedure removes more than 95% of the TIC in FBS as determined by a two-stage assay in which serum samples are incubated with a known amount of trypsin for 2 min at room temperature, after which residual trypsin activity is determined by hydrolysis of the chromogenic substrate Bz-Phe-Val-Arg-p-nitroanilide.

Growth of MCF-7 cells in soft agar

Soft agar transformation assays of MCF-7 cells were carried out essentially as described previously (19). Cells, maintained for 24–48 h in RPMI 1640 medium containing 10% charcoal-treated FBS, were plated in RPMI 1640 medium containing 0.3% agar, 10% charcoal-treated and TIC-depleted FBS, penicillin, streptomycin, and glutamine with and without E2 over a bottom layer of 0.5% agar using 104 cells/well in 30-mm tissue culture dishes and incubated at 37 C in an atmosphere containing 5% CO2. After 21 days, the plates were stained for 24 h with 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride, and the number of colonies per well (aggregates of more than 40 cells, >100 µm) was counted manually.

Immunological determination of TGF{alpha} by ELISA

Because of the low level of TGF{alpha} released from MCF-7 cells, it was necessary to concentrate conditioned medium before ELISA assay. Conditioned supernatants from MCF-7 cells in a single T-25 flask were concentrated on Sep-Pak C18 cartridges equilibrated with 10% acetonitrile-0.1% trifluoroacetic acid (TFA) eluting with 40% acetonitrile-0.1% TFA. Samples were taken to dryness, resuspended in 50 µL phosphate-buffered saline (PBS), and loaded into ELISA wells. TGF{alpha} levels were determined using a sandwich TGF{alpha} ELISA (Oncogene Science, Cambridge, MA) according to the manufacturer’s instructions. The ELISA is sensitive to less than 25 pg TGF{alpha}.

Immunoprecipitation

Aliquots of spent medium were brought to 2% in SDS and heated for 2 min at 95 C. Nonspecific rabbit ({alpha}1-AT) or goat (TGF{alpha}) IgG (10 µg/mL spent medium) was added, and the reaction was incubated for 60 min at 4 C. Protein A-agarose ({alpha}1-AT) or protein G-agarose (TGF{alpha}), as a 25% suspension in buffer A (50 mmol/L Tris-HCl, 150 mmol/L NaCl, and 6 mmol/L ethylenediamine tetraacetate, pH 7.4, containing 2.5% Triton X-100), was added, and the incubation was continued for 60 min longer. The protein A (or G)-IgG complexes were removed by centrifugation, and 4 vol buffer A were added followed by either a rabbit antibody to {alpha}1-AT or a goat antibody to TGF{alpha}. The incubation was continued overnight at 4 C. Immune complexes were removed with immobilized protein A (or G) as described above. After washing six times with buffer A, the labeled inhibitors were released from immobilized protein A by heating to 95 C in sample buffer containing SDS and subjected to PAGE in the presence of SDS under reducing conditions. Prestained molecular mass markers were included on all gels. After electrophoresis, labeled proteins on dried gels were visualized by autoradiography.

Isolation of mRNA and Northern blot analysis

Total cellular RNA was isolated from MCF-7 cells by guanidine isothiocyanate extraction and centrifugation through a CsCl gradient (22). The plasmid containing the {alpha}1-AT cDNA insert was labeled with [32P]deoxy-CTP using random hexamers as primers (Multiprime DNA Labeling System, Amersham, Arlington Heights, IL). Unincorporated nucleotides were removed by gel exclusion chromatography (Push Column, Stratagene, La Jolla, CA). Total RNA samples (~20 µg) were electrophoresed on 1.5% agarose/formaldehyde gels and transferred to GeneScreen nylon membrane. Blots were hybridized under conditions of high stringency (50% formamide, 42 C, for 12–18 h) and washed under these same conditions.

Construction of hyperexpressing {alpha}1-AT MCF-7 cell sublines

Using the MCF-7 ML variant as a parent, new sublines expressing large amounts of {alpha}1-AT were constructed by transfection with a full-length {alpha}1-AT cDNA. The 1.4-kilobase (kb) EcoRI fragment containing the complete coding sequence for {alpha}1-AT was excised from pBR322 and blunt-ended using the Klenow fragment of Escherichia coli DNA polymerase I. After the addition of HindIII linkers, the fragment was subcloned into the eukaryotic expression vector pRc/CMV (Invitrogen, San Diego, CA). This vector, which has a gene for neomycin acetyl transferase that confers resistance to the aminoglycoside antibiotic G418 and a cytomegalovirus promoter in front of a polylinker site followed by a bovine GH polyadenylation signal, is capable of directing high levels of protein expression in eukaryotic cells. The orientation of the insert with respect to the cytomegalovirus promoter was determined by restriction mapping of the subclones, and plasmid containing {alpha}1-AT in the sense orientation was amplified for transfection.

To test whether the construct was capable of directing the expression of active {alpha}1-AT, linearized plasmid was transcribed in vitro using T7 RNA polymerase. The RNA generated was translated in a rabbit reticulocyte lysate system in the presence of [35S]methionine, and the translation products were subjected to SDS-PAGE (Fig. 1Go). The translated RNA yielded a labeled protein with a molecular mass of approximately 50 kDa (lower arrow). This protein was active {alpha}1-AT, as it was able to form a SDS-stable complex with trypsin (upper arrow).



View larger version (99K):
[in this window]
[in a new window]
  		Figure 1. SDS-PAGE of {alpha}1-AT prepared by in vitro transcription/translation. The full-length {alpha}1-AT cDNA was transcribed in vitro, and the resulting RNA was translated in the presence of [35S]methionine with a rabbit reticulocyte lysate cell-free translation system. Aliquots (2 µL) of the reaction mixture were incubated alone or with trypsin for 3 min at room temperature in a total volume of 10 µL. Reactions were subjected to SDS-PAGE on 12% gels and visualized by autoradiography. Lane 1, {alpha}1-AT; lane 2, {alpha}1-AT plus 1 µg/mL trypsin (1 min); lane 3, {alpha}1-AT plus 2 µg/mL trypsin (1 min). The numbers on the left indicate the positions of standard molecular mass markers. The upper arrow on the right, at about 72 kDa, indicates the position of the {alpha}1-AT-trypsin complex. The lower arrow indicates the position of {alpha}1-AT.

 
MCF-7 ML cells were transfected with the {alpha}1-AT-containing expression vector or with vector alone using the calcium phosphate coprecipitation method (23). Posttransfection, cells were plated in medium containing 350 µg/mL active G418. Individual resistant colonies were isolated and expanded. After expansion, cells were maintained in medium containing 185 µg/mL active G418. Five colonies were further characterized from the cells transfected with the {alpha}1-AT-containing expression vector (Fig. 2Go, A and B, lanes 1–5). As no significant differences was detected in their growth properties, one single clone (lane 2), designated MCF-7 AT, was used in all subsequent experiments. Similarly, a single clone of the MCF-7 ML cells transfected with vector alone (Fig. 2Go, A and B, lane 6), designated MCF-7 PRC, was used in all subsequent experiments. Expression of {alpha}1-AT mRNA in MCF-7 AT cells was dramatically enhanced compared to that in MCF-7 PRC cells, as determined by Northern blotting (Fig. 2AGo). The approximately 2.0-kb transcript (upper arrow) represents the {alpha}1-AT messenger from the transfected cDNA (1.4 kb and polyadenylated tail) and is slightly larger than the native {alpha}1-AT transcript. The band at approximately 1.8 kb present in all transfected clones (lower arrow) represents the transcript from the neomycin-resistant gene. The autoradiogram was deliberately overexposed to show the position of the {alpha}1-AT mRNA in the parental cell line (lane 8) and the large difference in the level of expression between transfected and parental sublines. Expression of {alpha}1-AT protein after metabolic labeling with [35S]methionine was consistent with the overexpression of {alpha}1-AT mRNA in the {alpha}1-AT-transfected cells (Fig. 2BGo). The difference in the apparent molecular mass of the {alpha}1-AT in Figs. 1Go and 2BGo results from the glyosylation of the protein made in MCF-7 cells. No obvious morphological differences among the MCF-7 ML, AT, and PRC sublines were apparent when these cells were grown on plastic. Steady state TGF{alpha} mRNA levels were roughly comparable in each of the three sublines, with highest levels in the MCF-7 AT subline (data not shown).



View larger version (103K):
[in this window]
[in a new window]
  		Figure 2. Production of {alpha}1-AT mRNA and protein by MCF-7 ML cells transfected with an {alpha}1-AT cDNA. A, Northern blot analysis. Eighteen micrograms of total RNA from ML cell clones were electrophoresed on a 1.5% agarose/formaldehyde gel, transferred to nylon membrane, and hybridized with a 32P-labeled plasmid vector containing an {alpha}1-AT (and neomycin resistance gene) cDNA. Lanes 1–5, RNA from clones transfected with {alpha}1-AT-containing vector; lanes 6 and 7, RNA from clones transfected with vector alone; lane 8, RNA from the ML parent cell line. The upper arrow indicates the position of {alpha}1-AT mRNA; the lower arrow shows the position of neomycin resistance gene mRNA. B, SDS-PAGE of {alpha}1-AT after immunoprecipitation from conditioned medium from the same cells metabolically labeled with [35S]methionine.

 

   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Dose dependence for the inhibition of anchorage-independent growth of MCF-7 cells by {alpha}1-AT

The ability of cells to grow under anchorage-independent conditions (i.e. to form colonies in soft agar) has been correlated with their tumorigenicity or state of malignant transformation (24). In an earlier study we found an apparent negative correlation between anchorage-independent growth of MCF-7 cells and the presence of {alpha}1-AT whether added exogenously or expressed by the cell itself (19). We have examined this phenomenon more closely and show that when MCF-7 ML cells (a subline synthesizing low levels of {alpha}1-AT) are grown in soft agar in {alpha}1-AT-free medium (TIC depleted), addition of {alpha}1-AT significantly reduces colony formation in a dose-dependent fashion in both the presence and absence of E2 (Fig. 3Go). The {alpha}1-AT concentration required for half-maximal effect was approximately 1.5 µmol/L. The plant trypsin/chymotrypsin inhibitors, soybean trypsin inhibitor and BBI, neither of which have activity against elastase-like enzymes, had no effect on anchorage-independent growth even at a level of 100 µg/mL (4.5 and 12.5 µmol/L respectively; data not shown).



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. Effect of {alpha}1-AT concentration on anchorage-independent growth of MCF-7 ML cells in soft agar. MCF-7 ML (104) cells were seeded in soft agar in medium containing 10% TIC-depleted and charcoal/dextran-treated FBS with and without {alpha}1-AT or E2 (as indicated). The number of colonies per plate (>100 µm) was counted after 21 days. Control experiments had no {alpha}1-AT.

 
{alpha}1-AT blocks release of soluble TGF{alpha} from MCF-7 cells

TGF{alpha} is expressed by MCF-7 cells and stimulates proliferation and anchorage-independent growth in an autocrine/paracrine manner (13, 14). To determine whether there was a relationship between TGF{alpha} and the inhibition of anchorage-independent growth by {alpha}1-AT, we tested {alpha}1-AT for its effect on TGF{alpha} release (Fig. 4Go and Table 1Go). Under the conditions of the experiments described here, MCF-7 ML cells show a time-dependent increase in TGF{alpha} released from 30 min to 4 h, which then remains constant to 24 h (data not shown). The amount of TGF{alpha} released by MCF-7 ML cells is 50–60 pg/106 cells·4 h (Fig. 4Go). After a 24-h exposure to E2 (10-7 mol/L), the amount of TGF{alpha} released increased 4-fold to 200 pg/106 cells·4 h. Contrary to its effect in CHO cells (10), TPA (50 ng/mL) had little or no effect on TGF{alpha} release by MCF-7 cells unless the cells had been previously exposed to E2. Treatment with both TPA and E2 resulted in an additional 50% increase in TGF{alpha} release above that caused by E2 alone. Significantly, the presence of {alpha}1-AT during the incubation inhibited TGF{alpha} release in a dose-dependent manner, with the concentration for half-maximal effect being approximately 0.5 µmol/L. BBI (3.1 µmol/L) had no significant effect on TGF{alpha} release under these conditions (Table 1Go). Consistent with the inhibition of TGF{alpha} release by {alpha}1-AT, a polyclonal rabbit antibody able to neutralize {alpha}1-AT significantly stimulated TGF{alpha} release (Table 2Go).



View larger version (71K):
[in this window]
[in a new window]
  		Figure 4. {alpha}1-AT blocks TGF{alpha} release from MCF-7 cells, as measured by ELISA. MCF-7 ML cells (2 x 106 cells/T-25 flask) were incubated in phenol red-free DMEM containing 10% dextran/charcoal-stripped FBS for 48 h. E2 (10-7 mol/L) was added (as indicated), and the incubation was continued for an additional 24 h. Cells were washed twice with PBS and incubated for an additional 4 h in serum-free DMEM containing E2, TPA (50 ng/mL), and various levels of {alpha}1-AT or BSA (as indicated). Conditioned supernatants were concentrated on Sep-Pak C18 cartridges equilibrated with 10% acetonitrile-0.1% TFA eluting with 40% acetonitrile-0.1% TFA. Samples were taken to dryness, resuspended in 50 µL PBS, and loaded into ELISA wells. TGF{alpha} levels were determined using a sandwich TGF{alpha} ELISA (Oncogene Science) sensitive to less than 25 pg TGF{alpha}. Data represent the mean ± SD of triplicate determinations.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Effect of protease inhibitors on TGF{alpha} release from MCF-7 cells

 

View this table:
[in this window]
[in a new window]
  		Table 2. Antibody to {alpha}1-antitrypsin stimulates TGF{alpha} release from MCF-7 cells

 
The ELISA data were confirmed in an experiment in which TGF{alpha} release was measured by immunoprecipitation and SDS-PAGE after metabolic labeling of MCF-7 ML cells with [35S]cysteine (Fig. 5Go). The antibody used in this experiment was raised to the 6-kDa TGF{alpha} peptide. The only TGF{alpha} species released from MCF-7 cells under these conditions appears to be the mature, 50-amino acid, 6-kDa form. Inclusion of unlabeled TGF{alpha} during the immunoprecipitation blocked the precipitation of an 35S-labeled peptide migrating with authentic TGF{alpha} at approximately 6 kDa (Fig. 5Go, lane 4). TGF{alpha} levels were increased approximately 4-fold in medium from cells treated with E2 and TPA (Fig. 4Go, column 3, and Fig 5Go, lane 3). In addition, TPA stimulation of TGF{alpha} release appears to be dependent on pretreatment of the cells with E2. Whether this represents increased expression of pro-TGF{alpha} or protease, stimulation of protease activity, or a combination of these remains to be determined. Consistent with the ELISA results (Fig. 4Go and Table 1Go), TGF{alpha} release was reduced when {alpha}1-AT was present during the incubation period (Fig. 5Go, lane 5).



View larger version (81K):
[in this window]
[in a new window]
  		Figure 5. Effects of E2, TPA, and {alpha}1-AT on the release of TGF{alpha} from MCF-7 cells. MCF-7 cells were maintained in medium containing charcoal-stripped 10% FBS for at least 2 days before the start of the experiment. Cells, at approximately 80% confluence, were incubated with or without 10-8 mol/L E2 for 24 h. Medium was then changed to serum-free medium with or without E2 and containing [35S]cysteine (250 µCi/mL). After an 18-h pulse, medium was removed, and cells were chased with serum-free, [35S]cysteine-free medium containing the additions indicated for 6 h. [35S]TGF{alpha} was immunoprecipitated from spent medium with a sheep polyclonal antibody to human TGF{alpha}. Samples were subjected to discontinuous SDS-PAGE (10% spacer and 16.5% separating gels) following the procedure of Schagger and von Jagow (42). After electrophoresis, the gels were dried and visualized by autoradiography. Lane 1, No additions; lane 2, no E2, TPA (50 ng/mL) present during the second incubation; lane 3, E2 present during the entire incubation, TPA present during the second incubation; lane 4, E2 present during the entire incubation, TPA present during the second incubation, TGF{alpha} present during immunoprecipitation; lane 5, E2 present during the entire incubation, TPA and {alpha}1-AT (100 µg/mL) present during the second incubation. -TGF indicates the migration of authentic TGF{alpha}, visualized with Coomassie blue.

 
Anchorage-independent growth and TGF{alpha} release are reduced in {alpha}1-AT-hyperproducing MCF-7 cells

The above results suggested a negative correlation between the expression of {alpha}1-AT and anchorage-independent growth of MCF-7 cells. However, because of the possibility that this growth inhibition may have occurred by an {alpha}1-AT-independent mechanism, we reexamined this in an {alpha}1-AT-hyperexpressing MCF-7 subline (MCF-7 AT) constructed by transfection of MCF-7 ML cells with an expression vector containing a full-length {alpha}1-AT cDNA (Fig. 2Go). A control cell line (MCF-7 PRC) was transfected with vector alone. Table 3Go clearly shows an inverse relationship between expression of {alpha}1-AT, and TGF{alpha} release and colony formation in soft agar. Colony formation by the MCF-7 AT subline compared to that of the MCF-7 PRC control was reduced by more than 90% in E2-free medium and by more than 80% after a 24-h incubation with 10-7 mol/L E2. Comparable results were obtained for TGF{alpha} release.


View this table:
[in this window]
[in a new window]
  		Table 3. Effect of endogenous synthesis of {alpha}1-AT by MCF-7 cells on colony formation in soft agar and release of TGF{alpha}

 
MCF-7 cells have a cell surface protease able to bind {alpha}1-AT

If a physiological response depends on the presence of a protease inhibitor, it follows that a protease must also be involved. To test whether MCF-7 cells express a cell surface protease able to form a stable complex with {alpha}1-AT, cells were incubated with highly labeled {alpha}1-AT, prepared by in vitro transcription of an {alpha}1-AT cDNA followed by in vitro translation of the transcribed messenger in the presence of [35S]methionine (Fig. 6Go). Although this {alpha}1-AT contains a leader sequence and is not glycosylated, it is able to neutralize serine proteases and forms an SDS-PAGE-stable 73-kDa complex with trypsin (Fig. 1Go). MCF-7 (ML) cells were incubated with [35S]{alpha}1-AT (~0.75–1.5 µg/mL) in serum-free medium for 60 min at 37 C. At the end of the incubation, cells and media were subjected to SDS-PAGE, and the gel was examined by autoradiography (Fig. 6AGo). Under these conditions, essentially all of the [35S]{alpha}1-AT associated with the MCF-7 cells was either converted to an apparent 60-kDa {alpha}1-AT-protease complex or else degraded. Although the 72-kDa {alpha}1-AT- trypsin complex is not as prominent as that in Fig. 1Go, the 46-kDa {alpha}1-AT cleavage product (lanes 2 and 5) characteristic of neutralization of trypsin by {alpha}1-AT is clearly evident. The apparently low molecular mass of the cell surface protease-{alpha}1-AT complex is disturbing, but could be explained by partial degradation of the complex or if the protease had two or more chains linked by disulfide bonds. The absence of any indication of complex formation in medium from cells incubated with [35S]{alpha}1-AT (Fig. 6AGo, lane 4) strongly suggests that the {alpha}1-AT target protease is membrane bound. To clear up possible ambiguities resulting from the presence of radiolabeled impurities in the [35S]{alpha}1-AT preparation, cells were incubated with [35S]{alpha}1-AT in the presence of various modulators, and the {alpha}1-AT-cell surface protease complex was isolated by immunoprecipitation before SDS-PAGE (Fig. 6BGo). Binding to MCF-7 cells was completely inhibited by 10-3 mol/L diisopropyl flurophosphate (DIFP) and 50 µmol/L 3,4-dichloroisocoumarin, confirming that the pericellular protease is a serine protease. The serum-enzyme complex receptor agonist peptide 105Y (25) had no effect, suggesting that the {alpha}1-AT-pericellular protease complex is not internalized after binding to a putative serum-enzyme complex receptor.



View larger version (38K):
[in this window]
[in a new window]
  		Figure 6. {alpha}1-AT binds to an MCF-7 cell membrane protease. In vitro translated [35S]{alpha}1-AT (150–300 ng/well) was incubated for 60 min with confluent MCF-7 ML cells in serum-free DMEM in 24-well plates (total volume, 200 µL). A, Medium and cell lysate were subjected to SDS-PAGE on 12% gels and visualized by autoradiography. Lane 1, [35S]{alpha}1-AT alone; lane 2, [35S]{alpha}1-AT incubated for 1 min with 4 µg/mL trypsin; lane 3, cell lysate; lane 4, conditioned medium; lane 5, conditioned medium incubated for 1 min with 4 µg/mL trypsin. B, After incubation with the additions as indicated, cells were washed with PBS, lysed, and heated for 2 min at 95 C. [35S]{alpha}1-AT and complexes were immunoprecipitated with a polyclonal antibody to {alpha}1-AT and subjected to SDS-PAGE as described above. Lane 1, No additions; lane 2, 1 mmol/L DIFP; lane 3, 50 µmol/L 3,4-dichloroisocoumarin; lane 4, 1 µmol/L peptide 105Y.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this communication we show that release of TGF{alpha} from its membrane-bound precursor on MCF-7 cells is blocked by {alpha}1-AT, which, despite its name, is primarily an inhibitor of elastase-like enzymes (8). This is in agreement with the observation that TGF{alpha} release can be achieved in vitro by the action of elastases (11). The effect of {alpha}1-AT is particularly pronounced in MCF-7 ML cells, a subline expressing low levels of {alpha}1-AT. We also demonstrate a clear dose-response relationship between the concentration of {alpha}1-AT and the release of TGF{alpha} and anchorage-independent growth in soft agar. The {alpha}1-AT concentration necessary for a half-maximal effect is lower for inhibition of TGF{alpha} release than it is for inhibition of colony formation (0.4 vs. 1.5 µmol/L). However, both values are in the range of concentrations one might expect at the cell surface in vivo (27). BBI, an 8-kDa inhibitor of trypsin- and chymotrypsin-like serine proteases with anticarcinogenic activity in other systems (26), did not inhibit TGF{alpha} release, nor did it reduce colony formation in soft agar. Consistent with the effect of {alpha}1-AT on the release of TGF{alpha}, we have identified a cell surface serine protease on MCF-7 cells that can be neutralized by {alpha}1-AT. The fate of the putative {alpha}1-AT-membrane protease complex remains to be elucidated. Whether this protease is responsible for TGF{alpha} release has not yet been established.

A polyclonal antibody to {alpha}1-AT significantly increased the release of TGF{alpha}. However, as this antibody neutralized both human and bovine {alpha}1-AT, it was not clear whether the increase in TGF{alpha} release resulted from the inhibition of {alpha}1-AT synthesized endogenously or {alpha}1-AT from the medium adsorbed to the cell surface. To distinguish between these two possibilities, a new MCF-7 cell subline producing more than 10-fold higher levels of {alpha}1-AT than its parent cell line was constructed by stable transfection of MCF-7 ML with an {alpha}1-AT cDNA. Growth in soft agar and release of TGF{alpha} were significantly decreased in cells transfected with the {alpha}1-AT cDNA compared to those in cells transfected with vector alone.

The role of {alpha}1-AT as an acute phase reactant inhibitor of neutrophil elastase is well documented (8). Our results suggest that a second function of endogenous serpins is the regulation of autocrine growth factor activity. Zou et al. (28) have identified a gene for a protein, Maspin, that is expressed by normal mammary epithelial cells and appears to function as a true tumor suppressor. The deduced Maspin amino acid sequence shows a high degree of homology with that of the serpins such as {alpha}1-AT. However, recent evidence suggests that Maspin does not undergo the stress to relaxed transition typical of the interaction of serpin protease inhibitors with proteases, and Maspin may serve as a protease substrate rather than as an inhibitor (29). Our previous observation that the nonserpin elastase inhibitor SLPI is effective in blocking colony formation in soft agar by MCF-7 cells (19) suggests that in this instance {alpha}1-AT is acting as a protease inhibitor.

Recently, it has been suggested that TGF{alpha} acts in a juxtacrine fashion in human tumor cells, expressing both TGF{alpha} and EGF receptor (30). However, if a juxtacrine mode of action was the major mechanism triggering cell proliferation, then agents blocking TGF{alpha} release would be expected to have little effect. Our data clearly demonstrate that MCF-7 cells transfected with {alpha}1-AT cDNA fail to grow in agar compared to controls, suggesting that the release of either TGF{alpha} or a yet to be identified growth factor is important in these cells.

About a third of all cases of advanced breast carcinoma are responsive to estrogens, and recent epidemiological studies and studies using human breast cancer cells in culture strongly suggest a correlation between estrogens and the pathogenesis of breast cancer (13). The mechanisms by which this occurs are not entirely clear. Certainly, estrogens have a direct effect on cell growth, but in addition, they can stimulate the expression and release of a variety of polypeptide growth factors, and it is highly likely that the tumorigenic effects of estrogens are due at least in part to the autocrine/paracrine action of these factors (31, 32). Several of these polypeptides, including he EGF and its analogs, heregulin and TGF{alpha} (2), insulin-like growth factors I and II (33), mast cell growth factor (34), and tumor necrosis factor-{alpha} (35) have been shown to require pericellular proteolysis for activation or release. To achieve homeostasis, levels of the growth-modulating proteases must also be regulated. However, the mechanism involved in the release process is not understood. Our data suggest that release is regulated through the action of locally synthesized protease inhibitors. This would predict that an imbalance in the ratio between local levels of particular proteases and protease inhibitors would be responsible for increases in tumorigenic potential.

Aribas and Massagué (36), using a transfected CHO cell model, have proposed that the release of TGF{alpha}, ß-amyloid precursor protein, and other membrane-bound proteins occurs by a common mechanism. They have provided genetic evidence for an unidentified regulatory molecule that could be a regulator of protease activity, such as a protease inhibitor. The one apparently common feature of the release process is the involvement of protein kinase C or calcium influx into the cytosol. A potential pro-TGF{alpha} cleavage enzyme, activated by phorbol esters, has been partially purified from CHO cell membranes (37). The release of TGF{alpha} by MCF-7 cells differs in one significant way from its release in CHO cells, namely inhibition of the process by {alpha}1-AT. Although the specific protease(s) responsible for TGF{alpha} release has not been identified, on the basis of the peptide bonds cleaved in pro-TGF{alpha} and the fact that release is blocked by DIFP, {alpha}1-AT, and SLPI, it appears to be an elastase-like serine protease (8). Consistent with our results, a cell surface elastase-like enzyme sensitive to {alpha}1-AT has been isolated from transformed rat liver epithelial cells able to express TGF{alpha} (11). The importance of elastase inhibitors as potential tumor suppressors is underscored by a recent report of elafin, a nonserpin elastase inhibitor, which is present in normal cells, but down-regulated in breast tumor cell lines (38).

Stromelysin-3 (ST-3), a novel metalloproteinase initially identified in the stroma surrounding invasive breast neoplastic cells, is expressed in many human carcinomas, with highest mRNA levels in tumors demonstrating a high degree of invasiveness (39). Recently, it has been shown that {alpha}1-AT is the major target of ST-3 (40, 41). Cleavage by ST-3 results in the total loss of {alpha}1-AT protease inhibitory activity. ST-3, by destroying {alpha}1-AT activity at the tumor cell-stromal interface, might be at least partially responsible for the regulation of TGF{alpha} release. It is conceivable that {alpha}1-AT and TGF{alpha} from the breast epithelial cell and ST-3 from the stromal cell form part of an autocrine loop that regulates normal cell growth and tumorigenesis by the cancer cell.


   Footnotes
 
1 This work was supported in part by grants from the Elena Calas Memorial Fund, the Children’s Brain Tumor Foundation, the NIH (Grant CA-67348), and the U.S. Army (Grant DAMD17–96-1–6238). The National Science Foundation is thanked for its support of the computing resources through Grant BIR-9318128. Back

2 Current address: Department of Biochemistry, University of Medicine and Dentistry, Johnson Medical School, Piscataway, New Jersey 08854. Back

Received September 11, 1996.

Revised November 12, 1996.

Accepted November 20, 1996.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Lee DC, Fenton SE, Berkowitz EA, Hissong MA. 1995 Transforming growth factor {alpha}: expression, regulation, and biological activities. Pharm Rev. 47:51–85.[Medline]
  2. Massagué J, Pandiella A. 1993 Membrane-anchored growth factors. Annu Rev Biochem. 62:515–541.[CrossRef][Medline]
  3. Teixido J, Wong ST, Lee DC, Massagué J. 1990 Generation of transforming growth factor-{alpha} from the cell surface by an O-glycosylation-independent multistep process. J Biol Chem. 265:6410–6415.[Abstract/Free Full Text]
  4. Anklesaria P, Texido J, Laiho M, Pierce JH, Greenberger JS, Massagué J. 1990 Cell-cell adhesion mediated by binding of membrane-anchored transforming growth factor {alpha} to epidermal growth factor receptors promotes cell proliferation. Proc Natl Acad Sci USA. 87:3289–3293.[Abstract/Free Full Text]
  5. Lindberg I. 1991 The new eukaryotic precursor processing proteinases. Mol Endocrinol. 5:1361–1365.[Medline]
  6. Ehlers MRW, Riordan JF. 1991 Membrane proteins with soluble counterparts: role of proteolysis in the release of transmembrane proteins. Biochemistry. 30:10065–10074.[CrossRef][Medline]
  7. Torres-Rosado A, O’Shea KS, Tsuji A, Arai T, Chou S-H, Kurachi K. 1993 Hepsin, a putative cell-surface serine protease is required for mammalian cell growth. Proc Natl Acad Sci USA. 90:7181–7185.[Abstract/Free Full Text]
  8. Potempa J, Korzus E, Travis J. 1994 The serpin superfamily of proteinase inhibitors: structure, function and regulation. J Biol Chem. 269:15957–15960.[Free Full Text]
  9. Rice WG, Weiss SJ. 1990 Regulation of proteolysis at the neutophil-substrate interface by secretory leukoprotease inhibitor. Science. 249:178–181.[Abstract/Free Full Text]
  10. Pandiella A, Bosenberg MW, Huang EJ, Besmer P, Massagué J. 1992 Cleavage of membrane-anchored growth factors involves distinct protease activities regulated through common mechanisms. J Biol Chem. 267:24028–24033.[Abstract/Free Full Text]
  11. Cappelluti E, Strom SC, Harris RB. 1993 Potential role of two novel elastase-like enzymes in processing pro-transforming growth factor-{alpha}. Biochemistry. 32:551–560.[CrossRef][Medline]
  12. Harris JR, Lippman ME, Veronesi UV, Willett W. 1992 Breast cancer. N Engl J Med. 327:319–328, 390–398, 473–480.[Medline]
  13. Bates SE, Davidson NE, Valverius EM. 1988 Expression of transforming growth factor alpha and its messenger ribonucleic acid in human breast cancer: its regulation by estrogen and its possible functional significance. Mol Endocrinol. 2:543–555.[Abstract]
  14. Manni A, Wright C, Buck H. 1991 Growth factor involvement in the multihormonal regulation of MCF-7 breast cancer growth in soft agar. Breast Cancer Res Treat. 20:43–52.[CrossRef][Medline]
  15. Castelliani R, Visscher DW, Wykes S, Sarkar FH, Crissman JD. 1994 Interaction of transforming growth factor-alpha and epidermal growth factor receptor in breast carcinoma. Cancer. 73:344–349.[CrossRef][Medline]
  16. McGilligan KM, Thomas DW, Eckert CD. 1987 Alpha-1-antitrypsin concentration in human milk. Pediatr Res. 22:268–270.[Medline]
  17. Perlmutter DH, Cole FS, Kilbridge P, Rossing TH, Colten HR. 1985 Expression of the {alpha}1-proteinase inhibitor gene in human monocytes and macrophages. Proc Natl Acad Sci USA. 82:795–799.[Abstract/Free Full Text]
  18. Tamir S, Kadner SS, Katz J, Finlay TH. 1990 Regulation of antitrypsin and antichymotrypsin synthesis by MCF-7 breast cancer cell lines. Endocrinology. 127:1319–1328.[Abstract]
  19. Finlay TH, Tamir S, Kadner SS, Yavelow J, Levitz M. 1993 {alpha}1-Antitrypsin and anchorage-independent growth of MCF-7 breast cancer cells. Endocrinology. 133:996–1002.[Abstract]
  20. Finlay TH, Kadner SS, Tamir S. 1993 Protease inhibitor synthesis by MCF-7 breast cancer cells. In: Troll W, Kennedy A, eds. Protease inhibitors as cancer chemopreventative agents. New York: Plenum Press; 141–159.
  21. Bergman D, Kadner SS, Cruz MR, et al. 1993 Synthesis of {alpha}1-antichymotrypsin and {alpha}1-antitrypsin by human trophoblast cells. Pediatr. Res. 34:312–317.
  22. Chirgwin JM, Pryzbyla AE, McDonald RJ, Rutter WJ. 1979 Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry. 18:5294–5299.[CrossRef][Medline]
  23. Sambrook J, Fritsch EF, Maniatis T. 1989 Molecular cloning–a laboratory manual, 2nd ed. Cold Spring Harbor: Cold Spring Harbor Laboratory.
  24. Freedman VH, Shin S. 1974 Cellular tumorigenicity in nude mice: correlation with cell growth in semi-solid media. Cell. 3:355–359.[CrossRef][Medline]
  25. Perlmutter DH, Joslin G, Nelson P, Schasteen C, Adams SP, Fallon RJ. 1990 Endocytosis and degradation of {alpha}1-antitrypsin-protease complexes is mediated by the serpin-enzyme complex (SEC) receptor. J Biol Chem. 265:16713–16716.[Abstract/Free Full Text]
  26. Billings PC, Habres JM. 1992 A growth-regulated protease activity that is inhibited by the anticarcinogenic Bowman-Birk protease inhibitor. Proc Natl Acad Sci USA. 89:3120–3124.[Abstract/Free Full Text]
  27. Crystal RG. 1990 {alpha}1-Antitrypsin deficiency, emphysema, and liver disease: genetic basis and strategies for therapy. J Clin Invest. 85:1343–1352.
  28. Zou Z, Anisowicz A, Hendrix MJC, et al. 1994 Maspin, a serpin with tumor-suppressing activity in human mammary epithelial cells. Science. 263:526–529.[Abstract/Free Full Text]
  29. Pemberton PA, Wong DT, Gibson HL, et al. 1995 The tumor suppressor maspin does not undergo the stressed to relaxed transition or inhibit trypsin-like serine proteases: evidence that maspin is not a protease inhibitor. J Biol Chem. 270:15832–15837.[Abstract/Free Full Text]
  30. Baselga J, Mendelsohn J, Kim Y-M, Pandiella A. 1996 Autocrine regulation of membrane transforming growth factor-{alpha} cleavage. J Biol Chem. 271:3279–3284.[Abstract/Free Full Text]
  31. Preston-Martin S, Pike MC, Ross RK, Jones PA, Henderson BE. 1990 Increased cell division as a cause of human cancer. Cancer Res. 50:7415–7421.[Abstract/Free Full Text]
  32. Lippman ME, Dickson RB. 1989 Growth control of normal and malignant breast epithelium. Proc R Soc Edinburgh. 95B:89–106.
  33. Lamson G, Giudice LC, Cohen P, et al. 1993 Proteolysis of IGFBP-3 may be a common regulatory mechanism of IGF action in vivo. Growth Regul. 3:91–95.[Medline]
  34. Majumdar MK, Feng L, Medlock E, Toksoz D, Wiliams DA. 1994 Identification and mutation of primary and secondary proteolytic cleavage sites in murine stem cell factor cDNA yields biologically active, cell-associated protein. J Biol Chem. 269:1237–1242.[Abstract/Free Full Text]
  35. Jue D-M, Sherry B, Luedke C, Manogue KR, Cerami A. 1990 Processing of newly synthesized cachetin/tumor necrosis factor in endotoxin-stimulated macrophages. Biochemistry. 29:8371–8377.[CrossRef][Medline]
  36. Arribas J, Massagué J. 1995 Transforming growth factor-{alpha} and ß-amyloid precursor protein share a secretory mechanism. J Cell Biol. 128:433–441.[Abstract/Free Full Text]
  37. Harano T, Mizuno K. 1994 Phorbol ester-induced activation of a membrane-bound candidate pro-transforming growth factor-{alpha} processing enzyme. J Biol Chem. 269:20305–20311.[Abstract/Free Full Text]
  38. Zhang M, Zou Z, Maass N, Sager R. 1995 Differential expression of elafin in normal mammary epithelial cells and carcinomas is regulated at the transcriptional level. Cancer Res. 55:2537–2541.[Abstract/Free Full Text]
  39. Basset P, Wolf C, Chambon P. 1993 Expression of the stromelysin-3 gene in fibroblastic cells of invasive carcinomas of the breast and other human tissues: a review. Breast Cancer Res Treat. 24:185–193.[CrossRef][Medline]
  40. Pei D, Majmudar G, Weiss SJ. 1994 Hydrolytic inactivation of a breast carcinoma cell-derived serpin by human stromelysin-3. J Biol Chem. 269:25849–25855.[Abstract/Free Full Text]
  41. Pei D, Weiss SJ. 1995 Furin-dependent intracellular activation of the human stromelysin-3 zymogen. Nature. 375:244–247.[CrossRef][Medline]
  42. Schagger H, von Jagow G. 1987 Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem. 111:368–379.



This article has been cited by other articles:


Home page
 Endocr Relat CancerHome page
C L Wilson, A H Sims, A Howell, C J Miller, and R B Clarke
Effects of oestrogen on gene expression in epithelium and stroma of normal human breast tissue.
Endocr. Relat. Cancer, June 1, 2006; 13(2): 617 - 628.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yavelow, J.
Right arrow Articles by Finlay, T. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yavelow, J.
Right arrow Articles by Finlay, T. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE