| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
National Cooperative Program for Infertility Research, Reproductive Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Dr. Patrick M. Sluss, Reproductive Endocrine Unit Assay Laboratory, Edwards 003, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: sluss.patrick{at}mgh.harvard.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Synthetic peptide mimotopes identified within a 63-residue N-terminal domain two discontinuous sequences capable of binding labeled activin A. The first is located in a region (amino acids 3–26) of follistatin, a site previously identified by directed mutagenesis as important for activin binding. The second epitope, predicted to be located between amino acids 46 and 59, is newly identified. Although the sequences 3–26 and 46–59 code for activin binding, native follistatin only binds activin if disulfide bonding is intact. Furthermore, pyridylethylation of Cys residues followed by N-terminal sequencing and amino acid analysis revealed that all of the Cys residues in follistatin are involved in disulfide bonds and lack reactive free sulfhydryl groups.
Specific ligands were used to probe the structural effects of activin binding on the other domains of the full-length molecule, comprised largely of the three 10-Cys follistatin module domains. No effects on ligand binding to follistatin-like module I or II were observed after the binding of activin A to native protein. In contrast, activin binding diminished recognition of domain III and enhanced that of the C domain by their respective monoclonal antibody probes, indicating an alteration of the antigenic structures of these regions. Thus, subsequent to activin binding, interactions are likely to occur between regions of follistatin located in different domains and separated by considerable lengths of linear sequence. Such interactions could have important functional significance with respect to the structural heterogeneity of naturally occurring follistatins.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite the critical importance of follistatin’s neutralization of activin, the structural basis of activin binding to follistatin is poorly understood. Follistatin is a monomeric glycoprotein translated from messenger RNAs generated by alternative splicing of transcripts of a single gene (16, 17). Thus, isoforms of follistatin derive from transcription of five exons (follistatin-288 isoforms) or six exons (follistatin-315 isoforms). Posttranslational glycosylation and/or proteolytic processing of these two core proteins can generate additional follistatin isoforms. However, all follistatin isoforms have similar activin-binding characteristics (6, 18).
Perhaps the most remarkable structural feature of follistatin is the
large number of cysteine (Cys) residues (n = 36) contained in this
small, 32,000–35,000 Mr protein. Also striking
is the presence of distinct structural domains, evident from the exon
organization and primary sequence of follistatin (16, 17), which are
characteristic of mosaic proteins derived from exon shuffling during
evolution. After removal of a 29-amino acid leader sequence, the mature
protein contains 5 domains (Fig. 1
): an
N-terminal domain of 63 amino acids (a.a.), three 73- to 75-amino acid
repeats of what has become known as the follistatin module (19, 20, 21),
and, in follistatins derived from the splice variant in which all 6
exons are transcribed (follistatin-315), a highly acidic C-terminal
domain of 27 a.a. encoded by exon 6. The 10-Cys follistatin module
is a motif found in a number of primarily extracellular matrix proteins
involved in cellular differentiation and growth (20, 22, 23, 24, 25).
|
The current studies were undertaken to identify specific epitopes directly involved with activin binding and to determine whether activin binding to intact follistatin altered domain-specific antigenic epitopes located outside the N domain. We designed synthetic peptide mimotopes based on the 288-amino acid sequence common to all isoforms of follistatin as direct probes (31) to identify binding epitopes for activin A. Heparan sulfate or monoclonal antibodies were then used as site-specific probes to reveal structural changes elsewhere in the follistatin molecule after activin binding.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peptide synthesis and sequencing
Figure 1
shows the peptides synthesized and the positions of the
follistatin sequences they represent within the native protein.
Synthetic peptides were designed to 1) represent the entire N domain
(residues 1–63) for mapping the activin-binding site, 2) represent the
heparan sulfate-binding site (27), 3) represent the human-specific
antigenic epitope of human follistatin (a.a. 168–178), and 4)
represent intra-Cys sequences from each of the 10-Cys follistatin
module domains of the follistatin-288 core protein. Peptides were also
used to generate site-directed monoclonal antibodies for use as
structural probes of the C-termini of follistatin-288 and
follistatin-315. Progressive N-terminal truncations were introduced in
the two peptides identified as containing activin-binding epitopes.
All peptides were prepared at the Peptide Core Facilities of the Reproductive Endocrine Sciences Center at Massachusetts General Hospital by solid phase synthesis using F-moc chemistry. The peptides were purified by reverse phase HPLC, and mass spectroscopy was used to verify the sequences and confirm peptide homogeneity.
Evaluation of free sulfhydryl groups in peptide mimotopes
The presence of free sulfhydryl groups on Cys residues of the
peptide mimotopes was evaluated using Ellman’s reagent (32) purchased
from Pierce Chemical Co. (Rockford, IL). Peptides were
dissolved initially in 0.1 M acetic acid to 10 mg/ml and
then further diluted to 1 mg/ml (0.33–0.94 mM) with
Ellman’s reagent solution (0.1 M sodium phosphate, pH
8.0). Two hundred and fifty microliters of each of the peptides or
assay standard (cysteine hydrochloride monohydrate, 0.25–1.5
mM) were tested in duplicate following the protocol
recommended by the manufacturer.
Generation of monoclonal antibodies
The monoclonal antibody recognizing follistatin module domain II
was generated in mice immunized with recombinant human follistatin-288
as previously described (33, 34). Monoclonal antibodies against
follistatin module domain III and the C-terminal domain were generated
against synthetic peptides representing amino acid residues 274–287
and 300–315, respectively. Mice were immunized with protein or
synthetic peptide-BSA conjugates emulsified in Freund’s complete
adjuvant (Sigma, St Louis, MO). Booster immunizations were
composed of protein or peptide conjugated to BSA in Freund’s
incomplete adjuvant emulsifants. Both primary and booster immunizations
were administered sc. After achieving suitable serum titers, spleen
cells were harvested and fused with SP2/O myeloma cells obtained from
the American Type Culture Collection (Manassas, VA).
Hybridomas were selected by standard dilution cloning and enzyme
immunoassay methods. Monoclonal antibodies were purified using protein
G affinity chromatography from ascites fluid generated from
pristane-primed mice inoculated ip with 5 x
106 hybridoma cells. Harlan Laboratories, Inc.
(Indianapolis, IN), generated and harvested the ascites fluids.
The use of mice in these studies was performed in compliance with NIH guidelines. Experimental procedures were reviewed and approved by an internal review board for the use of animal resources. Animals were housed and cared for in American Association for Accreditation of Laboratory Animal Care-accredited facilities operated by the Massachusetts General Hospital or Harlan Laboratories.
Solid phase binding assays
These assays were used to identify and refine activin, heparin,
and antibody-binding epitopes. Follistatin peptides or follistatins
were coated onto Immulon-4 enzyme-linked immunosorbent assay (ELISA)
plates (Dynex Technologies, Chantilly, VA) by incubating 100 µl of
the appropriate solution at 4 C overnight. Follistatin peptides were
initially dissolved in 0.1 M acetic acid to 10 mg/ml and
then were diluted with 0.1 M sodium bicarbonate to 0.3, 1,
3, 10, 30, or 100 µg/ml. Native follistatin was coated in a similar
fashion using a 0.2 µg/ml solution. Nonspecific binding sites on
ligand-coated (peptide or follistatin) plates were blocked by
incubation with 200 µl of a blocking solution (PBS containing 0.05%
gelatin and 0.05% Tween-20) at room temperature for 2 h. Binding
of [125I]activin,
[3H]heparin, or monoclonal antifollistatin IgG
was determined by incubating these probes in a total volume of 200 µl
in ligand-coated or control wells at room temperature for 4 h with
gentle shaking. After incubation, unbound probes were decanted, and the
plates were washed three times with PBS-0.05% Tween-20 to reduce
nonspecific binding.
Bound [125I]activin was measured by direct
counting of the wells (inserted into 12 x 75-mm plastic tubes) in
a 
-counter. Tritiated heparin bound to peptide or follistatin was
measured by submerging washed wells in 10 ml liquid scintillation fluid
in a 20-ml glass scintillation vial and were counted in a ß-counter
(Beckman Coulter, Inc., Los Angeles, CA). Bound antibody
was measured with an ELISA reader (405 nM absorbance) after
sequentially incubating with an antimouse IgG-alkaline phosphatase
conjugate (Pierce Chemical Co.; 1:4000 dilution of
conjugated antibody) and then a p-nitrophenyl phosphate
substrate (Pierce Chemical Co.; 2.7 mM
solution).
[125I]Activin was prepared using lactoperoxidase and was purified by gel electrophoresis as previously described (35). The specific activity of the [125I]activin was approximately 30 µCi/µg based upon the method of Greenwood and Hunter (36).
Additionally, a solid phase sandwich-type assay was used to examine the effects of activin on follistatin module domains I and II. Follistatin-288 in the absence or presence (5 and 50 ng/ml) of activin was allowed to bind at room temperature for 12 h to the capture monoclonal antibody, 7FS30, coated on the bottom of ELISA wells. Unbound follistatin was removed by washing. Tritiated heparin, the detection probe for follistatin module I, was added and incubated at room temperature for an additional 3 h to allow its binding to follistatin (free or activin-bound) captured by the solid phase antibody. The wells were then washed to remove free tritiated heparin, and the amount of follistatin-bound tritiated heparin was measured by counting using liquid scintillation fluid and a ß-radiation counter. This basic sandwich-type assay format was also employed in a modified form using solid phase capture antibody covalently coupled to paramagnetic particles and a detection antibody conjugated to dimethylacrydinium ester. In this modification, bound reagents were separated from free using magnetic tube racks, and the chemiluminescent detection antibody was measured in a Ciba-Corning Magic Light Analyzer II (Chiron Corp., Walpole, MA).
Determination of follistatin peptide coating efficiency
The amount of peptide actually adsorbed to the microtiter wells
was measured to determine the mass of peptide mimotopes used in each
solid phase assay. Microtiter wells (10 wells/determination) were
coated with peptide (3.29–9.43 µM peptide; see Table 1) as described above for the preparation
of solid phase assay plates, then washed 3 times with PBS to remove
unbound peptides. Eighty microliters of 0.5 M NaOH were
added to each well, and the plates were incubated at room temperature
with constant shaking for 1 h. Solubilized peptides from 10 wells
were pooled, the pH was neutralized with concentrated HCl, and total
peptide was determined using a dye-ligand protein assay kit
(MicroBCA, Pierce Chemical Co.) and BSA as the
protein standard.
|
Pyridylethylation of follistatin
To probe for free sulfhydryl groups within follistatin, the
sulfhydryl-reactive reagent vinylpyridine was used for derivatization
of native follistatin along with a control preparation fully reduced by
mercaptoethanol (0.4%, 2 h, 37 C). The respective preparations
were incubated with vinylpyridine in 6 M guanidine/Tris-HCl
buffer (0.25 M, pH 8.5) for 2 h at 37 C and separated
from reagents using a Waters Corp. (Milford, MA) Sep-Pak
C18 cartridge. The stable pyridylethyl Cys
residues formed were quantitated by amino acid analysis (model 6300
analyzer, Beckman Coulter, Inc.) after total acid
hydrolysis (6 N HCl, 110 C, 24 h). N-Terminal sequence
analysis to characterize Cys at position 3 was performed on the
PE Applied Biosystems model 477A Microsequencer (Foster
City, CA).
Ligand blotting
SDS-PAGE and LIGAND blot analysis were performed as previously
described (37). X-Ray film (X-OMAT, Eastman Kodak Co.,
Rochester, NY) was exposed to the dried nitrocellulose blots for 3–7
days at -80 C.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Nonspecific
binding by each of the ligands was monitored by inclusion of two types
of controls in each binding experiment. Control 1 represented binding
to microtiter wells containing no solid phase test peptide, but
otherwise prepared identically to wells coated with synthetic peptide
mimotopes. Control 2 represents binding of ligands to plates coated
with an irrelevant synthetic peptide (R-K-G-Y-S-R-K-G-F-D-C; from an
enzyme unrelated to follistatin) rather than synthetic fragments of
human follistatin. Activin bound to only two synthetic fragments,
follistatin-(1–26) and follistatin-(44–59), from the N-terminal
domain (a.a. 1–63). Under similar experimental conditions, no activin
binding was observed to synthetic peptides representing epitopes within
the follistatin module domains I, II, and III. Also shown in Fig. 2 is
the mapping of tritiated heparin (Fig. 2B) and monoclonal antibody
7FS30 (Fig. 2C) using this series of synthetic peptides. Heparin
binding mapped to separate epitopes within the N-terminal domain and
follistatin module domains I, respectively. Heparin binding to the a.a.
74–86 fragment replicates the previously identified heparin binding
site (27). The monoclonal antibody 7FS30 bound only to the amino acid
168–178 fragment within the follistatin module domain II.
|
Because relatively low affinity interactions (nonspecific
protein-protein or protein-matrix binding) can be detected using small
synthetic peptides as solid phase ligands, the ability of native
follistatins to block activin binding to solid phase peptide was used
to further define the specificity of activin binding to the two
putative follistatin mimotopes identified in Fig. 2.
Follistatin-activin complexes (0.36 nM activin/2.9
nM follistatin-288 or 0.36 nM activin/3.1
nM follistatin-315) were incubated with solid phase
follistatin-(1–26) (197 pmol/well) or follistatin-(44–59) (334
pmol/well) for 4 h; under these conditions the follistatin-activin
complexes do not dissociate (35). The binding of
[125I]activin that had been preincubated with
BSA only for 1 h at room temperature was tested using BSA-coated
wells (nonspecific control, bar 1) or using peptide mimotope-coated
wells (positive controls, bars 2 and 5). As shown in Fig. 3, the binding of
[125I]activin to the synthetic follistatin
epitopes was completely blocked after coupling with native
follistatins. Thus, the binding of activin to follistatin-(1–26) and
follistatin-(44–59) appeared to rely upon the same binding regions
involved in its binding to native follistatins.
|
and 3) indicates the presence of these peptides on the
microtiter plates. However, the absence of ligand binding can represent
either the inability of the peptide mimotope to bind ligand or the
inability of the peptide to adsorb to the microtiter plate. To
distinguish between these two possibilities and to allow the
interpretation of negative binding data, the coating efficiency for
each of the peptides tested was determined. Two methods were used to
measure solid phase peptides. An improved commercial dye-ligand protein
assay was used with BSA protein calibrators to quantitate peptide
removed from microtiter wells with 0.5 M NaOH. Response
curves generated from dilutions of the solubilized solid phase peptides
were linear and parallel to each other as well to as a BSA reference
curve, allowing quantitative comparisons (data not shown). The plating
efficiency (percentage of peptide concentration used to coat the
microtiter plate wells) calculated for each peptide is summarized in
Table 1. Plating efficiencies based on amino acid analysis of solid
phase peptides subjected to acid hydrolysis in the microtiter wells is
also summarized in Table 1. These results indicate that despite very
large differences (70- to >100-fol) with respect to ligand binding
(Fig. 2), all of the peptide mimotopes were present in closely
equivalent amounts (differences <2-fold).
Refinement of N-terminal peptide binding sequences
Specific activin binding to two nonadjacent, linear sequences
(Figs. 2 and 3) indicated that there may be two distinct
activin-binding epitopes (a.a. 1–26 and a.a. 44–59) located within
the N domain. To define these activin-binding mimotopes more precisely,
a series of N-terminal truncated peptides was synthesized (Tables 2 and 3). Each series of peptides successively truncated at the N-terminus
was directly tested as solid phase ligands for their ability to bind
activin. As determined by direct measurement of solid phase peptides
(data shown in Fig. 4, insets), N-terminal truncation did not significantly alter
the ability of the peptide to stick to the microtiter wells. Figure 4A
summarizes the results of mapping the binding of activin to
follistatin-(1–26). Because follistatin-(1–26) and
follistatin-(3–26) are equipotent activin-binding peptides, amino acid
residues 1 and 2 of follistatin are not required for activin binding.
However, none of the peptides truncated beyond a.a. 3–26 were able to
bind activin, indicating that a.a. 3–5 of follistatin are essential
for the binding activity of follistatin-(3–26).
|
|
|
summarizes the mapping of the follistatin-(44–59) mimotope.
Removal of three a.a. from the N-terminus did not significantly affect
[125I]activin binding. However, the subsequent
deletion of a.a. 46–47 resulted in a complete loss of
[125I]activin binding, suggesting that those
two a.a. are critical to activin binding to this mimotope. As this
potential epitope had not been previously identified, an additional
control was prepared by synthesizing a full-length (a.a. 44–59)
peptide with a scrambled sequence of the identical amino acid
composition (Table 3). This peptide (scrambled amino acid 44–59
sequence; Fig. 4B) was inactive, supporting the specificity of the
native sequence for activin.
Activin binding to chemically modified native follistatins
The binding of activin to synthetic fragments of follistatin
indicates that at least two separate regions are involved in the
binding reaction. However, binding to synthetic peptide mimotopes
provides little insight into the conformational requirements, if any,
of the native protein to allow these epitopes to interact in the
binding of activin to native follistatin. Hence, follistatin-288 was
unfolded by reduction of Cys residues to determine whether the linear
epitopes, follistatin-(3–26) and follistatin-(46–59), bind activin in
linearized native follistatin or constitute a conformation-dependent
binding site(s) within the N-terminal domain. Ligand blotting
([125I]activin) after SDS gel-electrophoresis
of reduced or unreduced follistatin-288 was used to assess the ability
of activin to bind epitopes within the unfolded protein. As shown in
Fig. 5, activin binding to the complete
288-amino acid follistatin depends on the disulfide bonding of the
protein. The lower limit of detection by ligand blotting was
approximately 1 ng/lane. No activin binding could be detected when 20
ng reduced follistatin were loaded, and the amount of activin binding
was near the limit of detection when the gels were deliberately
overloaded with 100 ng follistatin/lane. It can be estimated from these
considerations that reduction of the disulfide bonds resulted in at
least a 99% loss of activin binding.
|
). Its importance is also suggested
by loss of function mutants created by amino acid insertions between
residues 2 and 3 of follistatin (26). This Cys could be part of a
disulfide bond in the native protein, or it could be present in an
unbonded, reduced state. If Cys-3 is reduced in the native protein, its
free sulfhydryl group might participate in postbinding reactions
between activin and follistatin, explaining the essentially
irreversible behavior of the follistatin-activin complex (35). Thus,
N-terminal composition and sequence analyses were performed after
reacting follistatin-288 with 4-vinylpyridine to specifically
derivatize free sulfhydryls (38, 39).
As shown by the compositional analysis in Fig. 6, the pyridylethyl Cys residue eluting
at 51 min in the reduced vinylpyridine-treated control preparation was
absent from the derivatized native follistatin. To determine the state
of the sulfhydryls within the N-terminal 3–26 binding epitope
precisely, each preparation was subjected to Edman microsequence
analysis through 15 cycles. Phenylthiohydantoins representing
pyridylethyl Cys, eluting between Tyr and Pro by on-line HPLC, were
observed at positions 3 and 13 of the reduced, pyridylethylated control
sample. No pyridylethyl Cys peak was found at either position in the
pyridylethylated native preparation (Fig. 7). Hence, Cys residues in native
follistatin, including those at the active N-terminus, are present in
the disulfide-bonded form and appear to be unavailable for covalent
cross-linking with activin ligand.
|
|
Tritiated heparin was used as a probe for the heparan sulfate-binding
site located in domain I (Fig. 2B). Domain II was probed using a
monoclonal antibody, 7FS30, specific to the 168–178 sequence (Fig. 2C). Activin-induced changes in follistatin module domains I and II
were probed simultaneously using a sandwich-type immunoassay composed
of solid phase antibody bound to plastic plates to capture follistatin
and the tritiated heparin to detect the antibody-bound protein. The
activin concentrations tested ranged from physiological (40) to
slightly greater than a 2-fold molar excess relative to the
concentration of follistatin. A full range of follistatin doses was
examined so as to detect either competitive- or noncompetitive-type
effects on probe binding. However, as shown in Fig. 8, the response curves (which require
binding of both probes to follistatin) were superimposable. Thus,
activin binding did not impair follistatin recognition by specific
probes to either domain I or II.
|
, activin binding to follistatin
resulted in a marked decrease in recognition by the monoclonal antibody
directed against the domain III epitope. Assay conditions were
optimized to provide sensitive detection of follistatin (0.3
nM) and a reasonable dose-response working range (10-fold).
These assay conditions allowed the testing of excess activin
concentrations ranging from about 2-fold (7 nM activin/3
nM follistatin) to about 20-fold (7 nM
activin/0.3 nM follistatin). Because the effects of activin
binding could have been due to alteration of follistatin conformation
or steric hindrance of probe binding, detailed dose-response curves for
follistatin were examined.
|
, activin
binding enhanced antibody binding to the C domain epitope. This
detailed analysis of follistatin response curves generated under
optimal assay conditions indicated that enhanced binding could be
detected even at a 1:1 molar ratio of activin to follistatin. This
observation suggests conformational changes associated with activin
binding, resulting in a more favorable C domain antigenic epitope,
either through higher affinity recognition or an enriched ensemble of
favorable conformers among the population of follistatin-315
molecules.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first activin-binding epitope predicted by mimotope binding data is located in the same region of follistatin previously identified by site-directed mutagenesis as important for activin binding based upon insertion of a.a. between residues 2 and 3 (26). The second epitope, predicted to be located between a.a. 46 and 59, is newly identified in the current work. Each of these activin-binding epitopes are completely contained within inter-Cys sequences of the N-terminal domain common to follistatin-288 and -315.
Synthetic peptide mimotopes, being relatively small and unconstrained
by tertiary or quaternary structure, are able to adopt multiple
conformations (41, 42). Thus, the current data demonstrate that
properly folded native follistatin can display two activin-binding
epitopes in the N-terminal domain. Because reduced native follistatin
has no activin-binding activity (see below), it appears that the
predicted activin-binding epitopes are not folded properly when the
disulfide bonding of the intact protein is disrupted. Alternatively,
the epitopes in the reduced protein may express the conformation
necessary for activin binding, but a such a low affinity that it can
only be detected in in vitro settings where excess peptide
is available or when brought together by the disulfide-mediated folding
of the native protein to produce a high affinity site of juxtaposed
epitopes. The small amount of activin binding seen to reduced
follistatin when overloaded gels were examined in Fig. 6 might support
this alternative, but that small amount of activin binding could also
reflect residual unreduced follistatin on the blot.
Currently, no conformational or three-dimentional studies of follistatin have been reported. Thus, we cannot be certain that the activin-binding epitopes predicted by the peptide mimotopes are folded into a complete activin-binding site by disulfide bonding of the native protein. It is possible that additional activin-binding epitopes exist within other domains of follistatin. These may be topographical sites composed of individual residues from several regions or linear sequences not represented by the probes used in this study.
One of the predicted activin-binding epitopes, 3–26, contains a Cys residue (a.a. 3). Thus, two alternative mechanisms could explain the loss of activin binding after the reduction of native follistatin. First, this loss could be due to conformational consequences of breaking disulfide bonds in a highly folded protein. Alternatively, if Cys-3 is not disulfide bonded in the native protein, reducing conditions could prevent the free sulfhydryl group from forming a disulfide bond with activin. The current studies rule out the second alternative, because we determined that all of the Cys residues, including Cys-3, are disulfide bonded in native follistatin-288. These observations together with the identification of two activin-binding sequences within the N-terminal domains suggest that the disulfide bond-dependent folding of the N-terminal domain assembles the activin-binding site.
The function of the three 10-Cys follistatin module domains remains unknown. No association has been reported between activin binding and the presence of similar follistatin module domains in numerous extracellular proteins (20, 22, 23, 24, 25). Although our studies were not designed to screen the entire follistatin protein sequence for activin binding, none of the synthetic peptides representing sequences outside the N-terminal domain was able to bind activin directly. Thus, rather than contributing directly to activin binding, the role of the 10-Cys follistatin module domains may be 1) strictly structural, providing the scaffolding necessary for the proper orientation of the activin-binding site or for disulfide bond-dependent assembly of the activin-binding site; 2) functional in an indirect manner, influencing activin binding by allosteric effects or by defining the orientation or alignment of those regions directly contacting activin; or 3) functional in a direct, albeit currently undefined, manner to stabilize the bound activin-follistatin complex, which is extremely stable under physiological conditions (35).
Insight into interactions between the N-terminal domain and other domains is provided by our findings using heparin and monoclonal antibody probes in conjunction with activin binding. An important structural role for follistatin module domain I in maintaining a properly assembled activin-binding site in the N-terminal domain was suggested by reports (28) that heparin binding, localized to this region (27), reduces the affinity of follistatin for activin. However, we observed that 1) activin cannot bind to the heparin-binding site located in domain I [follistatin 77–89)]; and 2) activin and heparin bind simultaneously despite overlapping epitopes in the first 26 a.a. of the N-terminal domain. In fact, binding of ligand probes for domains I (heparin) and II (antibody 7FS30) to native follistatin in the presence of activin suggests minimal interaction between the N-terminal domain and domains I and II.
In contrast, the effects of bound activin on antibody binding by
segments of domains III and C strongly suggest interactions between
these domains and the N-terminal domain. This observation may reflect
steric effects of the bound activin molecule or conformational changes
introduced by the binding process that alter the presentation of
antigenic epitopes on follistatin domain III and the C-terminal domain.
Steric effects of activin binding, such as epitope masking, could
explain the effects observed in Fig. 9, but not those seen in Fig. 10.
In this context it is of interest to note the presence of a
thioredoxin/protein disulfide isomerase consensus sequence (43, 44) in
both the N-terminal (a.a. 59–63) and domain III (a.a. 237–241). These
Cys pairs, although fully oxidized in native follistatin, could result
in a reshuffling of the disulfide-bonding structure (45) upon activin
binding, in turn mediating conformational changes within the
follistatin protein. The role, if any, of these Cys pairs during the
activin-induced conformational changes implied by our data is an
important subject for future investigation.
Our studies provide new insights into the structure-function relationships of follistatin. However, an understanding of the complex nature of the activin-binding site is still far from complete. With this goal in mind, the linear activin-binding epitopes we identified using synthetic peptides represent precise targets for mutagenesis to generate follistatins with altered activin-binding properties. Recognition of these sequences will also aid in the interpretation of effects of mutations made to other regions of follistatin. Finally, synthetic peptides incorporating these linear activin-binding epitopes are candidates for the development of follistatin mimetics for modulating follistatin and/or activin biological activities.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received February 15, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chain by immunochemistry of synthetic
peptides. J Protein Chem 4:171–184[CrossRef]
This article has been cited by other articles:
 |
|
 |
|
  M. I. Rosenberg, S. A. Georges, A. Asawachaicharn, E. Analau, and S. J. Tapscott MyoD inhibits Fstl1 and Utrn expression by inducing transcription of miR-206 J. Cell Biol., October 9, 2006; 175(1): 77 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. F. Rocnik, P. Liu, K. Sato, K. Walsh, and C. Vaziri The Novel SPARC Family Member SMOC-2 Potentiates Angiogenic Growth Factor Activity J. Biol. Chem., August 11, 2006; 281(32): 22855 - 22864. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Harrison, K. L. Chan, and D. M. Robertson Activin-A Binds Follistatin and Type II Receptors through Overlapping Binding Sites: Generation of Mutants with Isolated Binding Activities Endocrinology, June 1, 2006; 147(6): 2744 - 2753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Glister, N. P Groome, and P. G Knight Bovine follicle development is associated with divergent changes in activin-A, inhibin-A and follistatin and the relative abundance of different follistatin isoforms in follicular fluid J. Endocrinol., February 1, 2006; 188(2): 215 - 225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. R. Kumar Too Many Follistatins: Racing Inside and Getting Out of the Cell Endocrinology, December 1, 2005; 146(12): 5048 - 5051. [Full Text] [PDF]
|
|
|
|
|
|
Y. Sidis, A. L. Schneyer, and H. T. Keutmann Heparin and Activin-Binding Determinants in Follistatin and FSTL3 Endocrinology, January 1, 2005; 146(1): 130 - 136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Schneyer, Q. Wang, Y. Sidis, and P. M. Sluss Differential Distribution of Follistatin Isoforms: Application of a New FS315-Specific Immunoassay J. Clin. Endocrinol. Metab., October 1, 2004; 89(10): 5067 - 5075. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. T. Keutmann, A. L. Schneyer, and Y. Sidis The Role of Follistatin Domains in Follistatin Biological Action Mol. Endocrinol., January 1, 2004; 18(1): 228 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Innis and M. Hyvonen Crystal Structures of the Heparan Sulfate-binding Domain of Follistatin: INSIGHTS INTO LIGAND BINDING J. Biol. Chem., October 10, 2003; 278(41): 39969 - 39977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Hill, Y. Qiu, R. M. Hewick, and N. M. Wolfman Regulation of Myostatin in Vivo by Growth and Differentiation Factor-Associated Serum Protein-1: A Novel Protein with Protease Inhibitor and Follistatin Domains Mol. Endocrinol., June 1, 2003; 17(6): 1144 - 1154. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Welt, Y. Sidis, H. Keutmann, and A. Schneyer Activins, Inhibins, and Follistatins: From Endocrinology to Signaling. A Paradigm for the New Millennium Experimental Biology and Medicine, October 1, 2002; 227(9): 724 - 752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sidis, D. V. Tortoriello, W. E. Holmes, Y. Pan, H. T. Keutmann, and A. L. Schneyer Follistatin-Related Protein and Follistatin Differentially Neutralize Endogenous vs. Exogenous Activin Endocrinology, May 1, 2002; 143(5): 1613 - 1624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. V. Tortoriello, Y. Sidis, D. A. Holtzman, W. E. Holmes, and A. L. Schneyer Human Follistatin-Related Protein: A Structural Homologue of Follistatin with Nuclear Localization Endocrinology, August 1, 2001; 142(8): 3426 - 3434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sidis, A. L. Schneyer, P. M. Sluss, L. N. Johnson, and H. T. Keutmann Follistatin: Essential Role for the N-terminal Domain in Activin Binding and Neutralization J. Biol. Chem., May 18, 2001; 276(21): 17718 - 17726. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
