| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
School of Animal and Microbial Sciences, University of Reading (R.J.W., C.F.K., P.J.L.), Whiteknights, Reading, United Kingdom RG6 6AJ; the Department of Rheumatology, Royal Berkshire and Battle Hospitals, National Health Service Trusts (J.D.), Reading, United Kingdom RG3 1AG; and the Institute of Food Research (I.G.S.), Reading, United Kingdom RG6 6BZ
Address all correspondence and requests for reprints to: Dr. R. J. Woods, School of Animal and Microbial Sciences, University of Reading, Whiteknights, Reading, United Kingdom RG6 6AJ. E-mail: r.j.woods{at}reading.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this report we show, using a new immunoradiometric assay (IRMA), that in addition to CRF, CRF-BP is present in the synovial fluid of patients with rheumatoid arthritis. We have previously reported that CRF-BP is elevated in the plasma of patients with arthritis (8) and also that it is partially cleaved (9). We have now found that cleaved CRF-BP is present in some synovial fluid samples, and cleavage of CRF-BP in vivo may constitute a pathological or physiological response to inflammation. The properties of this binding protein have previously been explored by means of experiments on human recombinant CRF-BP (9, 10), and we have recently observed that highly purified preparations are susceptible to spontaneous proteolytic cleavage yielding N- and C-terminal fragments (11). The size of the resulting C-terminal fragment is remarkably similar to that which we have now found in synovial fluid. The cleavage site in the recombinant protein, occurring between amino acid residues serine 234 and alanine 235, is unique and does not resemble any that are recognized by the common proteases. This cleavage has been attributed to autocatalytic proteolysis, and it may therefore reflect a physiological property of the native protein. Because of the significance of CRF in the inception of inflammation, the ligand-binding properties of its cleaved binding protein were considered for investigation, but as there is insufficient native binding protein in synovial fluid with which to experiment, the properties of cleaved recombinant CRF-BP were investigated instead.
Although the cleavage site of native CRF-BP has not been determined, it was assumed that the properties of both cleaved forms are similar. Cleaved recombinant CRF-BP also helped to further our understanding of both ligand binding and the subsequent dimerization of the ligand/CRF-BP complex reported previously (10).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human CRF (hCRF) and CRF-BP-(25–45) were supplied by Dr. Jean Rivier (Clayton Laboratories for Peptide Biology, The Salk Institute, La Jolla, CA).
hCRF-BP-(298–322) in the form of a synthetic peptide as well as rabbit antiserum to CRF-BP-(298–322) were donated by Prof. T. Suda (University of Hirosaki Medical School, Hirosaki, Japan).
Collection of samples
Blood was collected in ethylenediamine tetraacetate from 24 normal controls and 10 patients with classical rheumatoid arthritis assessed by the American Rheumatism Association criteria (12). The plasma was separated and stored at -20 C. Synovial fluid was collected from 27 rheumatoid arthritis patients and frozen in 30-mL aliquots.
The blood and synovial fluid samples were collected as part of a routine investigation in patients with rheumatoid arthritis. The synovial fluid was aspirated from effused rheumatoid knees to alleviate discomfort. All patients gave informed and written consent, agreeing to have their blood and synovial fluid samples used for research purposes. All patients were taking nonsteroidal antiinflammatory drugs and disease-modifying drugs such as penicillamine, gold, or salazopyrine. They were deemed to have active rheumatoid arthritis by physicians’ global assessment. The age range of the 11 males and 16 females was 30–60 yr; their mean hemoglobin level was 10.2 g/L (range, 8.8–11.9), and their C-reactive mean protein level was 19.4 mg/L (range, 9–54 mg/L).
Reagents
All reagents and chemicals, except for antisera and where otherwise specified, were obtained from Sigma Chemical Co. (Poole, UK).
Analysis of synovial fluid samples by SDS-PAGE and immunoblotting
Samples of synovial fluid from three arthritic patients were examined by SDS-PAGE. Using a procedure previously described (13), antibody to intact recombinant CRF-BP was extracted from antiserum raised in rabbits by affinity chromatography with a column containing a peptide corresponding to the last 25 amino acid residues of the protein (298–322) coupled to cyanogen bromide-activated Sepharose 4B. The antibody was then coupled to the solid phase using the same procedure. Columns containing 0.25 mL immobilized antibody directed against C-terminal epitopes were used to extract native CRF-BP from synovial fluid. Fluid samples were diluted 4-fold and clarified by centrifugation. Each was passed through a column, which was then washed with 10 mL 0.1 mol/L ammonium bicarbonate solution. Binding protein was eluted in 4 mL 0.05 mol/L triethylamine in aqueous solution containing 20% acetonitrile. Mannitol (2 mg in 20 µL water) was added to each eluate, and all were then lyophilized before resuspension in sample treatment buffer used for SDS-PAGE. As a control, a synovial fluid sample containing 4 µg recombinant CRF-BP was processed in the same way. Subsequent to SDS-PAGE (14), the proteins were transferred from the gel to a polyvinylidene difluoride membrane by electroblot for 1 h at a constant voltage (10 V). The membrane was washed in phosphate-buffered saline (PBS), then soaked overnight in blocking solution (PBS, 1% casein, and 1% Polypep) at 4 C. It was then washed in PBS and incubated for 2 h at 20 C with primary sheep antiserum raised against intact recombinant CRF-BP at a 1:1000 dilution in blocking solution. The membrane was washed in PBS before incubation with secondary antiserum raised in rabbits against the Fc region of sheep immunoglobulin G and conjugated to horseradish peroxidase at a 1:1000 dilution in blocking solution for 2 h at 20 C. The membrane was washed again with PBS, and then the blot was developed with the Supersignal chemiluminescent substrate from Pierce Chemical Co. (Rockford, IL) and Warriner (Chester, UK) and exposed to autoradiograph film.
Production of recombinant hCRF-BP
Chinese hamster ovary cells were transfected with a complementary DNA sequence for CRF-BP isolated from a human liver library (15). Intact CRF-BP was then isolated from the culture medium by affinity chromatography (10, 13). All preparations were checked for integrity by SDS-PAGE both before and after storage at -20 C using a polyacrylamide concentration of 16%. Experiments involving cleaved CRF-BP were conducted with protein that had fragmented spontaneously, yielding the glycosylated, N-terminal fragment [CRF-BP-(25–234)] with an apparent mass of 27 kDa and a C-terminal [CRF-BP-(235–322)] of 9.6 kDa. Upon separation by SDS-PAGE and staining with Coomassie blue dye, no intact protein was observed in cleaved preparations.
Analysis of recombinant CRF-BP by tryptophan excitation fluorescence
Samples (0.5 mL) of intact and cleaved recombinant CRF-BP at concentrations exceeding 150 µg/mL were analyzed at 22 C using a Perkin Elmer Corp. LS 50B scanning emission fluorometer (Palo Alto, CA). The excitation wavelength was 295 nm, and the emission was measured between 300 and 400 nm using a 3-nm slit width. Sample size was 0.5 mL.
Column chromatography
This was carried out on a 90 x 1-cm bed of Sephacryl S200 developed at 20 C with the buffer described above at a flow rate of 3 mL/h, collecting fractions every 20 min. Relative elution volumes represented by Kav were calculated using the formula: Kav = (Ve - Vo)/Vt - Vo), where Ve is the elution volume, Vo is the void volume, and Vt is the total volume of the column.
BSA eluted with a Kav of 0.28, and human GH with a Kav of 0.58. CRF-BP eluted with a Kav of 0.45 consistent with that of a 37-kDa protein. Intact and cleaved CRF-BP were loaded onto the column in 0.5 mL sheep serum containing 1.5 µg CRF-BP (40.5 pmol/L). Where appropriate, CRF-BP was preincubated at 20 C for 30 min with a 1.7-fold molar ratio of hCRF (395 ng) or with 8 mol/L urea, which was added in solid form.
IRMAs
A fresh preparation of recombinant CRF-BP that was subsequently confirmed by SDS-PAGE to be intact was diluted 30-fold in sheep serum to produce an assay standard containing 1.8 mg/L. Aliquots of 250 µL were kept at -20 C and further diluted in sheep serum to produce concentrations of 200, 100, 50, 25, 12. 5, 6.25, and 3.125 µg/L. An assay reagent containing a 100-fold dilution of rabbit antiserum raised against intact CRF-BP was prepared in 0.05 mol/L sodium phosphate buffer, pH 7.4, containing 5 g/L protease-free BSA (BSA buffer), 100 mL/liter normal sheep serum, 2.0 g/L ethylenediamine tetraacetate, 1 g/L sodium azide, and 2.0 mL/L of the nonionic detergent, Igepal CA-630. It was able to bind 50% of a radioiodinated CRF-BP tracer at a dilution of approximately 1:20,000.
Antibody from sheep immunized with recombinant CRF-BP was affinity
purified using a synthetic peptide comprising the first 21 amino acid
residues of the CRF-BP attached to solid phase. The procedure was that
reported previously for the preparation of other antipeptide antibodies
(9). A sample comprising 18 µg IgG in 0.1 mol/L Tris buffer at pH 7.4
was radioiodinated by the Iodogen procedure (16). Bound and free iodine
were separated by gel filtration chromatography using the method
described above. An aliquot of the radiolabeled IgG was added to the
assay reagent to obtain an activity of 90,000 cpm/100 µL. The
protocol for the IRMA required mixing 50 µL unextracted plasma
samples or standard with 100 µL of the above assay reagent followed
by incubation at room temperature for 16 h. Bound and free
radiolabel were then separated by addition of 200 µL sheep antiserum
raised against the Fc region of rabbit IgG and diluted 10-fold in BSA
buffer (13). After 30 min, 2 mL 150 mmol/L sodium chloride containing
0.2% Igepal CA-630 were added to each tube, and samples and standards
were subjected to centrifugation at 4 C for 30 min. Supernatants were
aspirated, radioactivity in the remaining pellets was estimated, and
analyte concentrations were automatically calculated by a computer
connected to the output of a 
-ray spectrometer. Assays were
conducted to measure the recoveries of recombinant CRF-BP added to
samples of plasma from control and subjects with arthritis.
Concentrations of analyte were measured with and without the presence
of an additional 1.5 nmol/L CRF-BP in nine arthritic and five control
subjects. The recovery from controls was 93% (SEM = 5),
and that from arthritic subjects was 108% (SEM = 9). The
IRMA was further validated by preparing serial dilutions of samples and
comparing the assay results for parallelicity with the standard curve.
Those for arthritis and controls could be superimposed. The results of
determining quality control samples on a weekly basis were also found
not to vary. The interassay coefficient of variance was less than
5%.
Synovial fluid samples were clarified by centrifugation and diluted 10-fold before assay. Data for the comparison of CRF-BP levels in control and arthritic subjects was analyzed by one-way ANOVA, and the significance determined by nonpooled Student’s t test.
The CRF two-site IRMA design was essentially the same as that described previously (5). This IRMA used an antibody raised in rabbit to human CRF-(1–20) and a radiolabeled antibody that had been raised in sheep to hCRF-(21–41). The assay conditions and separation of bound and free radiolabel were as described for the CRF-BP IRMA. The interassay coefficient of variance for the lowest CRF standard (39 ng/L) was less than 5%.
RIAs
CRF-BP RIA was conducted as described previously (10). All standards were made from stocks of intact CRF-BP in normal sheep serum. Two rabbit anti-CRF-BP antisera were used in all the RIAs described below. One, raised against intact CRF-BP, bound to CRF-BP-(25–234), and the second, raised against a synthetic peptide comprising the last 25 amino acid residues of the protein, was used to measure CRF-BP-(235–322).
Ligand IRMA (LIRMA)
LIRMA was carried out as previously described (9) using the assay reagent and standards described for the above IRMA, except that radioiodinated hCRF was substituted for the radiolabeled anti-CRF-BP-(25–234) antibody. This assay measured functional CRF-BP by its capacity to bind radioiodinated ligand. The same rabbit antiserum, raised against whole molecule CRF-BP and recognizing both CRF-BP-(25–234) and CRF-BP-(235–322) that was used for the CRF-BP IRMA described above, was also used as a link antibody for the LIRMA.
Samples and a range of standards (50 µL) having the same range of concentrations as those for the IRMA were prepared in duplicate in BSA buffer. Radioiodinated CRF in sheep serum (50 µL) with a radioactivity of 40,000 cpm was added to each tube. After 30 min, 100 µL IRMA reagent, containing the link antibody at a 100-fold dilution, were added. Bound and free radioactivity were separated after further incubation for 1 h at room temperature by the same procedure as for the CRF-BP IRMA.
Ligand binding and Scatchard analysis
hCRF was radioiodinated at pH 8.5 by the Iodogen procedure (16) and separated from free radioiodine. A short silica column coated with C4 was used to remove iodinated peptide from the iodination mixture and eluted stepwise with 1.0-mL aliquots of water and methanol mixtures containing 0.1% trifluoroacetic acid. Radioiodinated CRF was eluted in 70% methanol. It was assumed that both histidine residues were disubstituted, and the concentration of radiolabeled peptide was estimated accordingly from the specific activity of the 125I radioisotope (Amersham International). Cleaved and intact CRF-BP (100 µL containing 21.6 µg, 584 pmol) were incubated simultaneously with 50 µL radiolabeled CRF (165 pmol/L) for 48 h at 20 C in the presence of unlabeled CRF in a range of concentrations prepared as double dilutions, from 6.25–0.049 nmol/L. Rabbit antiserum (50 µL) recognizing both CRF-BP-(25–234) and CRF-BP-(235–322) was added to a final concentration of 1:400. Bound and free radiolabel were separated by addition of 200 µL sheep antirabbit Ig diluted 10-fold in BSA buffer followed after 30 min by 2 mL 150 mmol/L sodium chloride containing 0.2% Igepal CA-630 as described for the CRF-BP IRMA. The ratio of bound to free radioligand was plotted, by the method of least squares, against the molar concentration of displaced radiolabeled ligand.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As can be seen from Fig. 1
, immunoblotting with a sheep antiserum raised to intact recombinant
binding protein shows the presence of intact native binding protein in
the synovial fluid samples. The blot also reveals the presence of a
C-terminal fragment in two of the three synovial samples tested. These
appear as distinct bands corresponding to a molecular mass of
approximately 10 kDa. This is consistent with the presence of a
C-terminal fragment of CRF-BP in synovial fluid, similar to that
observed previously in cleaved samples of purified recombinant CRF-BP
(11). For comparison, an immunoblot of cleaved recombinant CRF-BP made
with the same sheep antiserum showing the position of the
CRF-BP-(235–322) is included. In the third sample of synovial fluid,
it is possible that the N-terminal, 27-kDa fragment of CRF-BP has
copurified in association with the C-terminal fragment.
|
Mean concentrations of CRF-BP in both plasma and synovial fluid from patients suffering from arthritis and in plasma from control subjects did not differ significantly according to whether they were taken from males or females. The results were therefore combined. The mean plasma concentration of CRF-BP in 24 control subjects was 1.38 nmol/L (SD = 0.35 nmol/L), and that in 10 arthritic patients was 2.89 nmol/L (SD = 0.84 nmol/L). CRF-BP levels in plasma from arthritic subjects were elevated significantly with respect to control values (P < 0.0001). Synovial fluids samples from 27 arthritic subjects contained 0.51 nmol/L (SD = 0.24 nmol/L). The mean concentration of CRF measured by the CRF IRMA in the synovial fluid samples was 6.31 pmol/L (SD = 0.63 pmol/L).
Analysis of recombinant CRF-BP by tryptophan excitation fluorescence
The two spectra shown in Fig. 2 are
those for intact and cleaved recombinant CRF-BP. The emission peak of
the intact protein was at 310 nm, whereas that for cleaved protein was
of lower energy at 350 nm.
|
Without preincubation with hCRF the elution peak of
CRF-BP-(25–234) was symmetrical (Kav = 0.46), as shown in Fig. 3. CRF-BP-(235–322) eluted as a diffuse
peak, and it appeared to be in part physically associated with
CRF-BP-(25–234). Preincubation of a 1.7-fold molar excess of CRF
caused CRF-BP-(25–234) to undergo partial dimerization. The results
are presented in Fig. 4. Most of the
protein eluted in a volume greater (Kav = 0.38) than that
previously described for a dimeric complex of intact CRF-BP with CRF
(Kav = 0.32) (10). In addition, a "shoulder" was observed to
elute in the same position as the CRF-BP-(25–234) peak depicted in
Fig. 3. After preincubation with CRF, CRF-BP-(235–322) was associated
more closely with CRF-BP-(25–234).
|
|
After incubation with 8 mol/L urea, cleaved recombinant binding
protein was separated into N- and C-terminal fragments by gel
filtration. These were measured by RIAs for CRF-BP-(25–234) and
CRF-BP-(235–322). Estimates of the intact protein made using
antibodies to the CRF-BP-(25–234) and to the CRF-BP-(235–322) were in
good agreement (data not shown). Figure 5
shows that CRF-BP-(235–322) eluted as a Gaussian peak and was not
associated with the CRF-BP-(25–234). The Kav of CRF-BP-(25–234) was
0.45, whereas that for CRF-BP-(235–322) was 0.63.
|
. Radioiodinated CRF is
bound by fractions containing CRF-BP-(25–234), but not
CRF-BP-(235–322). Identical results (not shown) were obtained with
radioiodinated urocortin. Chromatography of cleaved CRF-BP after incubation with CRF and 8 mol/L urea
Cleaved binding protein was subjected to gel filtration after
incubation with a 1.7-fold molar excess of CRF and 8 mol/L urea. It is
clear from Fig. 6 that there was
sufficient time after loading samples for separation of urea and
protein to occur and for CRF to bind to all the binding protein.
CRF-BP-(25–234) dimerized with a peak of Kav 0.38, which, as no
C-terminal fragment was present, must have consisted only of
CRF-BP-(25–234) and CRF. Some monomeric CRF-BP was also apparent at
its trailing edge. CRF, measured by CRF IRMA, was bound to both
monomeric and dimeric CRF-BP-(25–234). A peak of unbound CRF was also
observed to elute in the later fractions (Kav = 0.62).
|
Scatchard analysis of the binding of CRF to identical molar
amounts of intact and cleaved CRF-BP (Fig. 7) revealed that affinity for ligand was
unaffected by cleavage. The dissociation constants
(Kd) for intact and cleaved protein were 113 and
137 pmol/L, respectively. The maximum ligand binding in these
experiments was 51 pmol/L for intact and 28 pmol/L for cleaved protein.
This represents a fall in activity of 45% for this cleaved
preparation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The tryptophan fluorescence spectrum of intact binding protein has a
remarkably high peak energy emission at 310 nm comparable only to that
of azurin (308 nm), in which the tryptophan residue is located in a
hydrophobic pocket in the protein (18). After cleavage there is a
considerable reduction in the energy of this peak associated with
exposure of the sole tryptophan residue to solvent molecules, denoting
conformational change in the CRF-BP-(25–234) fragment (19). This
degree of unfolding does not affect affinity for ligand, although it
may account for impaired dimerization. Its behavior when exposed to
ligand was investigated by gel filtration chromatography. It is evident
from Fig. 3 that after cleavage the resulting fragments remain in close
association, although their mutual affinity must be low for they
separated gradually during chromatography even under nondenaturing
conditions. Addition of a molar excess of CRF increases the retention
of CRF-BP-(235–322) by the CRF-BP-(25–234), but in molar terms there
is much less CRF-BP-(235–322) than CRF-BP-(25–234) in the dimerized
component. Unlike intact binding protein, cleaved CRF-BP is capable
only of partial dimerization by CRF, although at least some of the
remainder that does not dimerize can still bind ligand. It is clear
from experiments in which fragments are separated by exposure to 8
mol/L urea that CRF-BP-(25–234) alone is capable of binding CRF and
that CRF-BP-(235–322) is not. This was confirmed both by observations
made after chromatography with radioiodinated ligand and by measurement
of eluted CRF. It would appear also that CRF-BP-(25–234), after
treatment with urea, can refold without cooperation from
CRF-BP-(235–322) to regain its ligand-binding activity.
A discrepancy is apparent in Fig. 5; estimates of CRF-BP-(25–234) in
chromatographic fractions made by LIRMA were lower than those made by
RIA. Furthermore, peak values for CRF-BP-(25–234) estimated by RIA and
binding of radiolabeled ligand, do not coincide. Both indicate the
presence of inactive protein. The difference revealed by Scatchard
analysis between the values for maximum binding of CRF by equimolar
preparations of cleaved and intact CRF-BP confirmed that approximately
half of the cleaved protein did not bind ligand.
Scatchard analysis also gave Kd values for both cleaved and intact CRF-BP that were in close agreement with the published value (15). That monophasic binding curves were observed suggests that dimerization did not increase affinity for ligand. It is unlikely, therefore, that CRF participates in the binding sites involved in dimer formation. Loss of the C-terminal fragment (30% of the protein) does not affect ligand binding, and results of Scatchard analysis show the affinity of CRF for intact and cleaved protein to be the same.
The results obtained with recombinant CRF-BP may allow tentative conclusions to be drawn concerning the activity of native CRF-BP in synovial fluid. As in pregnancy, secretion of even large amounts of CRF is not itself associated with increased cytokine activity, but CRF production is an early event in the cellular inflammatory response. Endogenous CRF has a proinflammatory effect in the joints of animals with experimental arthritis (2) and is capable of producing an acute inflammatory response by degranulation of mast cells (3). Immunoreactive CRF (20) and urocortin (21) are also reported to be present in human peripheral blood leukocytes, and CRF has been demonstrated to stimulate proliferation of human lymphocytes (4) as well as to enhance their production of interleukin-1 and interleukin-2 (22). It was therefore relevant to ask whether cleavage of CRF-BP might lead to the release of CRF-like activity and thereby exacerbate the cycle of inflammation. Crofford et al. (23) reported a mean CRF concentration, measured by RIA, of 28 pmol/L in synovial fluid taken from patients with rheumatoid arthritis. This may be an overestimate, as their assay may have detected immunoreactive CRF fragments as well as the intact peptide. By means of a CRF IRMA, which is specific for intact CRF, we arrived at a lower estimate of 6.3 pmol/L. We also report that CRF-BP is present in an 80-fold molar excess, at approximately 500 pmol/L. If these values are representative and if all of the protein was functionally intact, virtually all of the CRF would be protein bound and therefore physiologically inactive (6, 7). Evidence from immunoblotting indicates that in two synovial samples, substantial cleavage has occurred, but it is not reasonable to suppose that the level of free CRF in synovial fluid would rise as a consequence. A physiological role for cleavage is not, therefore, immediately apparent.
The factors that promote cleavage of native or recombinant CRF-BP are not known, but circumstantial evidence indicates that cleavage in vivo may be both specific and of some functional importance, rather than resulting from nonspecific activity of proteolytic enzymes accumulating in arthritic joints. Synovial fluid from rheumatoid arthritis patients contains insulin-like growth factor-binding protein-3 that is intact, whereas that from healthy controls is predominantly cleaved (24). Cleavage of hormone-binding proteins as a mechanism for raising the level of free hormone has been recognized for the insulin-like growth factor-binding proteins (25), but it is not solely restricted to this role. Cleavage can also result in the generation of a new bioactivity. For example, cleavage of the 22-kDa hormone PRL yields an N-terminal fragment of 16 kDa that is a potent inhibitor of angiogenesis (26), and autoproteolytic cleavage of the p53 tumor suppressor protein predisposes cells to either apoptosis or cell growth arrest depending upon whether cleavage is N- or C-terminally directed (27). It remains to be seen whether fragments derived from CRF-BP have novel activities in their own right.
|
Footnotes |
|---|
Received January 6, 1999.
Revised April 20, 1999.
Accepted May 4, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,6-diphenyl glycouril (Iodogen). Anal
Biochem. 117:136–146.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  P. Florio, R. Romero, T. Chaiworapongsa, J. P. Kusanovic, M. Torricelli, P. J. Lowry, and F. Petraglia Amniotic Fluid and Umbilical Cord Plasma Corticotropin-Releasing Factor (CRF), CRF-Binding Protein, Adrenocorticotropin, and Cortisol Concentrations in Intraamniotic Infection and Inflammation at Term J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3604 - 3609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. O. Huising, J. M. Vaughan, S. H. Shah, K. L. Grillot, C. J. Donaldson, J. Rivier, G. Flik, and W. W. Vale Residues of Corticotropin Releasing Factor-binding Protein (CRF-BP) That Selectively Abrogate Binding to CRF but Not to Urocortin 1 J. Biol. Chem., April 4, 2008; 283(14): 8902 - 8912. [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 |
) RIAs.
).
