| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology and Cell Biology, Forschungszentrum Borstel, Borstel, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1
nM), but only IL-8 exhibits similar high affinity for CXCR-1, while the
other ELR-CXC chemokines bind with 20- to 100-fold lower affinity
(3, 4, 5, 6, 7). With respect to regulation of its activity, NAP-2 holds a
unique position among ELR-CXC chemokines. While the production and
hence the activity levels of all other members of this subfamily are
predominantly regulated by differential gene expression (1), the
unfolding of NAP-2 activity is controlled by proteolytic processing
(for review, see 8 . In fact, NAP-2 is stored in the 
-granules
of platelets as the 70-amino-acid-long C-terminal part of several
precursor molecules, the by far most abundant of these being the
connective tissue-activating peptide III (CTAP-III, 85 amino acids, see
Fig. 1
) (9). NAP-2 and its immunologically cross-reactive precursors
are comprised under the term ß-TG Ag. CTAP-III, which becomes
released from activated platelets in high amounts (mean serum
concentrations amounting to about 3 µM), neither binds to nor
activates PMN (10). Only site-specific N-terminal cleavage, which can
be catalyzed in vitro, e.g., by the serine protease cathepsin G,
results in the formation of the mature NAP-2 (11). This kind of
proteolytic conversion is also mediated by intact leukocytes such as
monocytes and PMN, but not lymphocytes (10). Interestingly, the most
efficient leukocyte populations appear to be the PMN, which were shown
to rapidly mediate the processing through a cathepsin G-like membrane
protease (10, 11). The circumstance, that PMN themselves, the target
cells for NAP-2, may generate the chemokine from inactive CTAP-III,
confers an indirect desensitizing capacity to CTAP-III: during the
initial phase of CTAP-III conversion, NAP-2 appears at low
concentrations that are insufficient to trigger degranulation.
Nevertheless, at these concentrations, NAP-2 selectively binds to and
down-regulates CXCR-2 from the PMN surface (4, 10, 12). Continuous
accumulation of NAP-2 during ongoing processing renders the cells
increasingly refractory toward stimulation by NAP-2 and other ELR-CXC
chemokines, especially to those preferentially utilizing the NAP-2 high
affinity receptor CXCR-2 (10). Thus, NAP-2 exhibits not only activating
properties for PMN, but in the course of its generation from CTAP-III,
also has a desensitizing impact on processing cells. This mechanism may
serve to limit deleterious effects of PMN during an inflammatory
reaction.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CTAP-III, NAP-2 (1–70), and NAP-2 (1–66) were purified as
reported previously (11, 13). Like NAP-2 (1–66), novel C-terminally
truncated ß-TG Ag isoforms were derived from culture supernatants of
platelet-containing, 4ß-PMA/Ca-ionophore-stimulated PBMC. This
material, concentrated by absorption to silica and prefractionated by
cation-exchange chromatography, was kindly provided by Dr. H. Mohr
(Blood Bank of German Red Cross, Springe, Germany). All isoforms were
purified in a three-step procedure, as described previously (11).
Briefly, total ß-TG Ag was isolated by immunoaffinity chromatography
on a fast protein liquid chromatograpy (FPLC) unit (Pharmacia/LKB,
Freiburg, Germany) using ß-TG Ag-specific mAb C-24 coupled to
CNBr-activated Sepharose. The eluate was adjusted to 10 mM phosphate,
pH 7, and further separated by FPLC on a cation-exchange column
(Mono S HR 10/10; Pharmacia/LKB) under a linear gradient of 0 to 0.5 M
NaCl in the same buffer. Eluting proteins as detected at 
= 214 nm
were pooled as indicated in Figure 2A. Pools were further
separated by reversed-phase HPLC using an analytical cyanopropyl column
(4.6 x 250 mm, 5-µm-wide pore; J. T. Baker,
Phillipsburg, NJ). Samples acidified with trifluoroacetic acid
(TFA) were directly loaded and eluted at 0.5 ml/min with a gradient of
0 to 35% 1-propanol in 0.1% TFA. Eluting protein peaks were collected
according to absorbance at 214 nm, lyophilized, and resuspended in
0.1% TFA, and finally protein concentration was determined using the
microbicinchoninic acid method, with BSA serving as standard.
|
If not otherwise stated, recombinant methods were performed
according to Sambrook et al. (14). Inserts coding for full-size
CTAP-III (1–85) and CTAP-III (1–81) to CTAP-III (1–77) were
generated by PCR (2 min, 95°C/2 min, 55°C/3 min, 72°C; 35
cycles). Upon generation of a platelet-derived cDNA library, as
described previously (13), CTAP-III cDNA (corresponding to CTAP-III
mRNA positions 196–450; EMBL accession number M54995) served as
template. While the oligonucleotide used as 5' primer corresponded to
positions 196–219, 3' primers were complementary to positions
434–450, 422–438, 419–435, 416–432, 413–429, and 410–426 to yield
the inserts for CTAP-III (1–85) and CTAP-III (1–81) to CTAP-III
(1–77), respectively. Additionally, 5' and 3' primers at their 5' ends
were provided with a Bgl2 and SalI restriction
site, respectively, which allowed insertion of PCR products into
expression vector pQE-30 (Diagen, Hilden, Germany). These plasmids were
used for transformation of Escherichia coli strain DH5aF'IQ.
Expression plasmids coded for the respective CTAP-III isoforms bearing
an N-terminal histidine affinity tag for purification purposes (His
tag; amino acid sequence, MRGSHHHHHHGS). Plasmids of transformants
were verified by DNA sequencing and subsequently subcloned into
E. coli strain M15[pREP-4]. Resulting transformants were
treated 5 h with isopropyl-ß-D-thiogalactopyranoside
(1 mM) to induce protein overexpression. Subsequently, bacteria were
pelleted and lysed for 1 h in urea-containing buffer (0.1 M
NaH2PO4, 10 mM Tris, 8 M urea, pH 8). The
lysate was loaded onto a Ni2+-nitrilo-triacetic acid column
(Diagen). His-tagged recombinant proteins were subsequently eluted by a
pH shift to 4. The eluate was dialyzed 2 days against PBS to remove
denaturing urea. After precipitated protein was pelleted, the
supernatant was further separated by reversed-phase HPLC, as described
above for the purification of native ß-TG Ag isoforms. rHis-CTAP-III
isoforms eluted as distinct symmetrical protein peaks. Upon N-terminal
sequencing (four steps), all polypeptides exhibited the exclusive
sequence MRGS, identical with the starting sequence of the His tag. As
judged by Western blot staining of overloaded SDS-polyacrylamide gels
using an antiserum to ß-TG Ag (R-ßTG), rHis-CTAP-III isoforms
were devoid of other contaminating ß-TG Ag isoforms. rNAP-2 isoforms
were obtained by digestion of the respective rHis-CTAP-III isoforms
with bovine chymotrypsin, as described previously for the generation of
NAP-2 from native CTAP-III (12). During subsequent HPLC separation
performed as described above, rNAP-2 isoforms eluted as symmetrical
protein peaks. According to N-terminal sequencing (four steps), all
isoforms showed the sequence A-E-L-R, corresponding to the N
terminus of NAP-2. Moreover, masses of the different NAP-2 isoforms, as
determined by matrix-assisted laser desorption/ionization (MALDI) mass
spectroscopy, did not deviate by more than 0.07% from the theoretical
values, confirming the identities of the polypeptides.
Neutrophils: preparation and degranulation assay
PMN were routinely isolated from citrated blood of healthy single donors by gradient centrifugation on Ficoll-Hypaque to a purity greater than 95% in all events, as previously described (11). PMN (1 x 107/ml) suspended in Dulbecco’s PBS (D-PBS)/0.1% BSA (low endotoxin BSA; Serva, Heidelberg, Germany) were preincubated for 10 min with 5 µg/ml cytochalasin B (Sigma, Deisenhofen, Germany), then supplemented with CaCl2 and MgCl2 to yield final concentrations of 0.9 and 0.5 mM, respectively, and 100 µl of cells were then added to 100 µl of test samples. After an incubation period of 30 min at 37°C, cells were pelleted and supernatants were monitored for elastase enzymatic activity, as described elsewhere (4). Release rates were determined as the percentage of total elastase activity in detergent-treated PMN lysates prepared in 0.1% hexadecyl-trimethylammonium bromide. Relative potencies of stimuli to induce elastase release are expressed as the percentage of the potency of full-size NAP-2, according to the following equation: percentage of potency = ([A]full-size NAP-2/[B]NAP-2 variant) x 100. [A] and [B] represent concentrations of stimuli that elicit identical release rates.
Desensitization of NAP-2-induced degranulation
To determine the desensitizing effect of rCTAP-III isoforms on NAP-2-induced PMN degranulation, the degranulation assay was modified as follows. PMN suspended at 2 x 107 cells/ml in D-PBS/0.1% BSA and preincubated for 10 min with cytochalasin B were added in 50-µl vol to 50 µl of desensitizing agents in the same buffer and pretreated for another 10 min at 37°C under agitation. Thereafter, 100 µl of NAP-2 (final concentration: 40 nM) in D-PBS/0.1% BSA/1.8 mM CaCl2/1 mM MgCl2 was added and the degranulation assay was performed as described above. Desensitization was expressed as the percentage of release rates obtained with control PMN receiving no desensitizing agent. The relative desensitizing potency of isoforms was calculated as the percentage of the desensitizing potency of C-terminally intact rHis-CTAP-III (1–85) according to the equation given above.
Receptor-binding competition assay
The interaction of unlabeled NAP-2 and NAP-2 variants with chemokine receptors on PMN was investigated in a binding competition assay using native radiolabeled NAP-2 as a tracer. Assays were performed as described previously (13), using a constant concentration of 125I-NAP-2 in the absence or presence of increasing concentrations of unlabeled NAP-2 and NAP-2 variants (up to a 100-fold molar excess). Nonspecific binding of 125I-NAP-2 was subtracted. Determination of relative binding potency was performed according to the equation given above for the determination of degranulation potencies. A and B represent concentrations of NAP-2 (variants) that cause 50% competition for binding with labeled NAP-2.
Electrophoreses, immunoblotting, ELISA, Ab reagents
SDS-PAGE was performed under reducing conditions according to
Schägger and von Jagow (15). To assure complete reduction of
proteins, the sample buffer contained 100 mM DTT and 30 µg/ml EDTA.
After incubation for 10 min at 95°C, samples were additionally
treated for 30 min at room temperature with iodoacetamide (2% final
concentration) to prevent reestablishment of disulfide bonds. Samples
were then loaded onto a 13% (w/v) polyacrylamide gel with a 10%
spacer gel and a 4% stacking gel on top. Rainbow Protein Markers
(Amersham Buchler, Braunschweig, Germany) served as molecular mass
markers. Isoelectric focusing on polyacrylamide gels in the presence of
8 M urea, transfer of protein bands onto polyvinylidene difluoride
membranes, and immunochemical detection of ß-TG Ag polypeptides were
conducted as described previously (16). The following rabbit polyclonal
antisera were used: R-ßTG, reacting with all ß-TG Ag known to
date, and R-70, which requires the presence of the C-terminal
amino acid residue aspartic acid for binding to ß-TG Ag isoforms
(16).
Mass spectroscopy
Matrix-assisted laser desorption/ionization (MALDI) mass spectroscopy was applied to determine the m.w. of CTAP-III (1–81) (performed by WITA GmbH, Teltow, Germany) as well as of rNAP-2 isoforms (kindly performed by Dr. B. Lindner, Division of Biophysics, Forschungszentrum Borstel, Borstel, Germany). The m.w. of native NAP-2 (1–63) was determined by ion-spray mass spectroscopy (kindly performed by Dr. A. I. Mallet, St. Thomas Hospital, London, U.K.).
N-terminal amino acid sequencing
N-terminal sequence analyses of native and recombinant ß-TG Ag isoforms suspended in 0.1% TFA were performed by Dr. A. Petersen (Department of Clinical Medicine, Forschungszentrum Borstel) on a gas-phase sequencer (model 473A; Applied Biosystems, Foster City, CA). When sequencing was performed directly from the blot, the membrane area containing the unstained protein of interest was excised, washed with bidistilled water for 1 h, dried, and then sequenced.
Statistical analyses
Data were statistically analyzed using Microcal Origin 4.10
software (Microcal Software, Northampton, MA). Data obtained for native
C-terminally truncated NAP-2 isoforms, which were tested
simultaneously, were statistically compared using the paired
t test. Data obtained for recombinant NAP-2 and CTAP-III
isoforms, which for practical reasons could not be assayed altogether
at the same time, were statistically compared using the independent
t test. Two populations of data were considered
significantly different at p values 
0.05 (indicated
by *).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Initiated by our recent discovery of a C-terminally truncated
NAP-2 isoform with enhanced biologic activity in supernatants of
mitogen-stimulated, platelet-containing PBMC (13), we systematically
analyzed the same material for the presence of further, so far
unidentified C-terminally truncated ß-TG Ag variants. For this
purpose, total ß-TG Ag was first isolated from supernatant
concentrates by immunoaffinity chromatography using immobilized ß-TG
Ag-specific mAb C24 (data not shown). Further separation of this
material by Mono-S cation-exchange chromatography yielded a wide and
heterogeneous elution pattern consisting of numerous poorly separated
protein peaks that were pooled as indicated in Figure 2
A.
Each of the seven pools formed (A-G) was then subjected to
reversed-phase HPLC on an analytical cyanopropyl column, and eluting
protein peaks (numbered 1-n), as detected by absorbance at = 214 nm
(see Fig. 2B for elution patterns of pools E and G), were
manually collected. To test for the presence of C-terminally truncated
isoforms, two aliquots of each sample were then separated by SDS-PAGE,
blotted, and analyzed in parallel for reactivity with two different
ß-TG Ag-specific antisera. Protein bands staining positive with
antiserum R-ßTG (detecting all ß-TG Ag isoforms known to date)
as well as with antiserum R-70 (requiring the C-terminal Asp in
ß-TG Ag for binding; refer to Fig. 1)
were detected in almost all of the samples, indicating the presence of
C-terminally intact molecules. Only one sample derived from pool E
(peak E1) and two samples derived from pool G (peaks G1 and G5, Fig. 2B) contained protein bands
that failed to react to R-70, indicating the presence of
C-terminally truncated molecules. While the protein contained in peak
G1 was identical to the previously discovered NAP-2 (1–66) (data not
shown), further analyses of the proteins in peaks E1 and G5 led to the
identification of two novel C-terminally truncated ß-TG Ag isoforms.
Characterization of a novel NAP-2 variant
The ß-TG Ag variant contained in cation-exchange pool E and
eluting upon HPLC separation at about 15% n-propanol in
peak E1 (Fig. 2B, left panel) was further
characterized and examined for purity by electrophoreses and
immunoblotting as well as by N-terminal sequence analysis. Upon
SDS-PAGE, a single protein band, migrating slightly faster than
full-size NAP-2 run in parallel, was visualized by staining with
antiserum R-ßTG (Fig. 3A). Its lack of reactivity
with antiserum R-70 as opposed to the blunt staining obtained for
full-size NAP-2 confirmed that it represented a C-terminally truncated
molecule (Fig. 3A). Determination of the N-terminal residues
of the protein in E1 revealed A-E-L-R as the predominant sequence
(
90%), which corresponded to the N terminus of NAP-2 (refer to Fig. 1). These results indicated that the major protein in E1 represented a
C-terminally truncated NAP-2 isoform. A second but minor sequence
(10%) reading L-Y-A-E and corresponding to the N terminus of a
CTAP-III isoform truncated by 13 N-terminal residues (CTAP-III
(14–85)) was also detected. This contamination most likely originated
from incomplete separation of peak E1 from the adjacent peak E2 (refer
to Fig. 2B, left panel), the latter containing a
C-terminally intact ß-TG Ag isoform that we could identify as
CTAP-III (14–85) (data not shown). In fact, analysis by isoelectric
focusing and immunoblotting revealed the presence of two differently
charged proteins in peak E1, the minor one focusing at a pH (8.7)
identical to that of the protein contained in peak E2 (Fig. 3B). Thus, the novel C-terminally truncated NAP-2 isoform in
E1 was obviously represented by the major band that focused at a
considerably more basic pH of 9.5. Due to limited amount of protein in
E1 (50 µg), we did not attempt to further purify the NAP-2
isoform. However, its direct comparison with full-size NAP-2 (pI =
8.9) and the previously discovered NAP-2 (1–66) (pI = 9.4) by
isoelectric focusing demonstrated that it was even more basic than the
latter molecule (Fig. 3B). This result suggested that
truncation in the novel NAP-2 variant was likely to comprise more than
the four C-terminal amino acids missing in NAP-2 (1–66) (two of these
being acidic residues), and included at least the acidic residue
D66 (refer to Fig. 1), which would explain the increase in
positive net charge of the molecule. In fact, the molecular mass of the
main component in E1, as determined by mass spectroscopy, was 6993,
which corresponds by a deviation of less than 0.3% to the theoretical
value (6979) of a NAP-2 variant truncated by exactly seven amino acids.
From these data, we conclude that besides NAP-2 (1–66), there exists a
further natural NAP-2 variant being truncated at the C terminus, namely
NAP-2 (1–63).
|
C and
4A) and about 3.5-fold
enhanced potency to compete with 10 nM radiolabeled full-size NAP-2 for
the binding to specific receptors on PMN (Fig. 4C). Its
enhanced potency was clearly not due to the contaminating CTAP-III
(14–85), which as a homogeneous protein was considerably less potent
than full-size NAP-2 (20% of that of NAP-2, as determined in the
degranulation assay; data not shown). Thus, it rather had to be taken
into account that contamination with this less active protein would
instead lead to underestimation of the potency of the NAP-2 (1–63)
preparation. Nevertheless, this preparation was not only more potent
than NAP-2, but moreover reproducibly proved to be even slightly more
potent than homogeneous NAP-2 (1–66) assayed in parallel (Figs. 3C and 4, A and C).
|
To obtain homogeneous NAP-2 (1–63) and to investigate the
functional impact of single amino acid residue deletions at the NAP-2 C
terminus, we recombinantly expressed in E. coli and purified
full-size NAP-2 (rNAP-2 (1–70)) as well as NAP-2 isoforms that were
successively truncated by four to eight residues (rNAP-2 (1–66) to
rNAP-2 (1–62)). We did not generate the longer isoforms truncated by
only one to three residues (rNAP-2 (1–69) to rNAP-2 (1–67)), because
these were already expressed and analyzed in our previous study dealing
with the structure-activity relationship of NAP-2 (1–66) (13).
Different from that former study, we herein used a modified expression
system (refer to Materials and Methods), allowing for easier
handling and higher gains of products. After their purification to
homogeneity, N-terminal sequence analyses of every single recombinant
polypeptide yielded the sequence A-E-L-R, thereby confirming their
identity as NAP-2 isoforms. First evidence for correct C-terminal
truncation of the isoforms was obtained by isoelectric focusing, in
which a successive shift to a more basic pI was observed with every
isoform that was expected to have lost an additional acidic residue
(refer to Fig. 1). Thus, in contrast to full-size NAP-2 (pI
8.9), NAP-2 (1–66), lacking two acidic residues focused at pH
9.4, while the isoforms devoid of a further acidic residue (NAP-2
(1–65) to NAP-2 (1–62)) focused at pH 9.5 (data not shown).
In fact, mass-spectroscopic analyses finally confirmed the identities
of the polypeptides, yielding masses that deviated by less than 0.07%
from the expected theoretical values. (Refer to Materials and
Methods.)
Functional analyses in the degranulation assay and in the receptor
competition assay showed that, in accordance with the results obtained
with recombinant isoforms produced in the former expression system
(13), rNAP-2 (1–70) as well as rNAP-2 (1–66) were as active as their
native counterparts, with rNAP-2 (1–66) being threefold more potent
than full-size rNAP-2 (Fig. 4). Further deletion of NAP-2 by the fifth
residue (D66) in rNAP-2 (1–65) led to another increase in
potency to induce degranulation to about fivefold as compared with
full-size NAP-2 (Fig. 4B). As observed with rNAP-2 (1–64)
and rNAP-2 (1–63), this enhanced level of activity was maintained upon
deletion of two further residues. Thus, rNAP-2 (1–63) proved to be
slightly more potent than the native NAP-2 (1–63) preparation,
pointing to the fact that this isoform indeed represents the most
potent naturally ocurring NAP-2 isoform known to date (Fig. 4, A and B). A subsequent decrease in potency to
twofold of that of full-size NAP-2 was observed with NAP-2 (1–62)
(truncated by eight C-terminal residues), which isoform additionally
missed the residue L63. Interestingly, a similar change in
potencies was observed when analyzing the NAP-2 variants for their
ability to compete the binding of 125I-NAP-2 to PMN in the
receptor competition assay (Fig. 4D). While rNAP-2 (1–66)
competed threefold more potently than full-size rNAP-2 (1–70),
further truncation by five to seven residues led to another increase in
potency up to 4.5-fold. As observed with rNAP-2 (1–62), deletion of
L63 again resulted in a decrease of potency down to
twofold as compared with that of full-size NAP-2. These results
suggest that the differences in functional potency of the different
NAP-2 isoforms are due to changes in the affinity to specific NAP-2
binding sites on PMN.
Detection of a C-terminally truncated NAP-2 precursor
In addition to NAP-2 (1–66) and NAP-2 (1–63), we discovered in
the same PBMC supernatants another C-terminally truncated ß-TG Ag
isoform. This isoform eluted during HPLC separation of cation-exchange
pool G in peak 5 (G5) at about 22% n-propanol (Fig. 2B, right panel). Upon separation by SDS-PAGE and
subsequent Western blotting using antiserum R-ßTG for detection,
G5 turned out to contain at least two different ß-Tg Ag isoforms,
with apparent m.w. of about 9 and 10 kDa, respectively (Fig. 5). Only the 9-kDa protein was
C-terminally truncated according to its lack of reactivity with
antiserum R-70 (Fig. 5, arrowed band). Its apparent m.w. being
higher than that of NAP-2 run in parallel indicated that this isoform
represented a C-terminally truncated NAP-2 precursor molecule. In fact,
N-terminal sequencing of this protein band directly from the blot
revealed the sequence N-L-A-K, unambiguously identifying this isoform
as a C-terminally truncated CTAP-III variant (refer to Fig. 1). This
finding was consistent with the observation that in SDS-PAGE the 9-kDa
protein migrated slightly faster than full-size CTAP-III run in
parallel (Fig. 5). Interestingly, mass-spectroscopic analyses of
fraction G5 revealed the presence of four components, only two of
which, having masses of 8433 and 8900, were smaller than full-size
CTAP-III (theoretical mass = 9289), and thus could theoretically
represent C-terminally truncated CTAP-III isoforms. The fact that the
molecule with a mass of 8900 gave the by far most intense signal
indicated that it represented the predominant CTAP-III isoform in
fraction G5. Since its mass deviated by less than 0.2% from the
theoretical value (8887) for a CTAP-III isoform truncated by exactly
four C-terminal residues, we conclude that the predominant C-terminally
truncated CTAP-III isoform in G5 is identical to CTAP-III (1–81).
Interestingly, this isoform would represent the potential precursor
molecule for NAP-2 (1–66).
|
We have shown previously that a major impact of CTAP-III on PMN
function is the desensitization of chemokine-induced degranulation (10)
(see above). Since this antiinflammatory effect is not induced
directly, but finally relies on the binding to and desensitization of
predominantly CXCR-2 by NAP-2 generated during coincubation of PMN and
CTAP-III, it could be envisaged that C-terminally truncated precursors
such as CTAP-III (1–81) would show enhanced desensitizing potency due
to the formation of the more potent NAP-2 (1–66). To test this
hypothesis, we generated rCTAP-III isoforms that were C-terminally
intact (His-CTAP-III (1–85)) or truncated by four to eight amino acids
(His-CTAP-III (1–81) to His-CTAP-III (1–77)). All rCTAP-III isoforms
contained an additional histidine-rich sequence tag (His) at the N
terminus for purification purposes (refer to Materials and
Methods). As tested with His-CTAP-III (1–85), the presence of the
His tag did not interfere with the proteolytic conversion of the
precursor into intact NAP-2 by PMN (data not shown). Thus, C-terminally
truncated CTAP-III isoforms were then analyzed for their capacities to
desensitize the NAP-2-induced degranulation response of PMN in
comparison with the full-size precursor molecule His-CTAP-III (1–85).
As shown in Figure 6, C-terminal
truncation changed the desensitizing potency of CTAP-III, in that
deletion of four residues enhanced the potency to threefold, and to
sixfold, with deletion of five, six, or seven residues. A subsequent
decrease in potency to twofold of that of full-size NAP-2 occurred
with the deletion of eight residues. These results show that C-terminal
truncation alters the desensitizing function of CTAP-III in a similar
fashion as previously observed for the activating functions of NAP-2,
which were likewise enhanced upon deletion of four and five residues,
and decreased again upon deletion of eight residues (Fig. 4, C and D). Thus, our data suggest that natural
CTAP-III (1–81) could indeed act as an agent with enhanced
desensitizing capacity as compared with full-size CTAP-III.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to the previously described NAP-2 (1–66), we in fact
discovered and identified two novel variants, namely a NAP-2 molecule
truncated by seven C-terminal residues, the NAP-2 (1–63), which,
according to the yields upon purification, represents about 10% of all
NAP-2 isoforms isolated from the culture supernatants (data not shown),
and a precursor molecule truncated by four residues, the CTAP-III
(1–81). Interestingly, there was no evidence for the occurrence of
other C-terminally truncated variants, suggesting a strong preference
for the generation of isoforms truncated by either four or seven
C-terminal residues. This phenomenon is most likely due to the
three-dimensional arrangement of the ultimate C-terminal residues.
While the major part of the C terminus of ß-TG Ag is arranged into an
amphiphilic -helix mounted on top of a three-stranded ß-sheet (18, 19), which structure may largely protect the molecule from
proteolytic attack, it was reported that the four ultimate C-terminal
residues in dimeric NAP-2 do not participate in the -helix, but form
fraying ends with a high degree of mobility (19, 20). These residues
could represent readily accessible targets for protease(s), which
circumstance would explain the preferential occurrence of ß-TG Ag
isoforms NAP (1–66) and CTAP-III (1–81), both of which are truncated
by four residues. A similar situation can be imagined for the formation
of NAP-2 (1–63). The ultimate residue in this isoform, the
L63, is highly conserved in CXC chemokines, and structural
analyses on monomeric IL-8 have shown that it is positioned at the end
of the C-terminal -helix (21). Since NAP-2 also predominantly occurs
as a monomer at physiologic concentrations (17, 22), it can be
envisaged that in this molecule the conserved L63 likewise
determines the termination of the helix, rendering the following seven
residues more susceptible to proteolysis. The resulting NAP-2 (1–63),
bearing extensive truncations at the C terminus as well as at the N
terminus, may well represent the stable core structure of ß-TG Ag,
being largely resistant to further proteolytic attack due to
conformational restriction.
Interestingly, NAP-2 (1–63) is not only the most extensively truncated
natural ß-TG Ag molecule discovered to date, but in terms of biologic
activity, also proved to be the most potent neutrophil-activating
isoform. While in agreement with our previous studies (13) that NAP-2
(1–66) was about three times more potent than full-size NAP-2 to
induce neutrophil degranulation and to compete for receptor binding on
neutrophils, NAP-2 (1–63) turned out to be even five times more potent
in these respects. Further analyses using a series of rNAP-2 variants
bearing C-terminal deletions of up to eight amino acid residues
demonstrated that the increase in biologic activity was not strictly
correlated to the number of residues missing, but that the
physicochemical properties of these residues were important. Thus,
deletion of only one additional residue from rNAP-2 (1–66), the
D66, generated a molecule (rNAP-2 (1–65)) considerably
higher in potency than rNAP-2 (1–66) (fivefold versus threefold that
of full-size NAP-2), while subsequent deletions by up to seven residues
(G65 and A64) did not lead to further changes.
Thus, NAP-2 (1–63), the most active natural isoform, as well as its
recombinant counterpart, were practically equal in potency to isoforms
being one or two residues longer at the C terminus. The observation
that deletion of an acidic amino acid, the D66, was
critical for biologic activity, while the removal of uncharged residues
had no effect, was in good accordance with previous results obtained
during the analysis of the structure-activity relationship in NAP-2
(1–66) (13). Together with those former results, indicating that
particularly the loss of the acidic residues D70 and
E67 caused subsequent increases in biologic activity and
receptor-binding affinity of NAP-2, our present observations now permit
the conclusion that all three negatively charged residues at the C
terminus play critical roles for the neutrophil-stimulating capacity of
NAP-2 and its isoforms. The fact that in our previous study we did not
find enhanced activity with rNAP-2 (1–65) and rNAP-2 (1–64) that were
produced in a different expression system, is most likely due to
difficulties to achieve correct refolding of these molecules in that
system. Most interestingly, others have reported a similar
structure-activity relationship for IL-8. Using synthetic C-terminal
IL-8 variants successively truncated by tripeptides, they observed an
increase in activity when acidic residues were removed and a decrease
upon the loss of basic residues (23). Indeed, it is conceivable that
the charge distribution at the C terminus of CXC chemokines could be
important for receptor binding. Taking into account that the
extracellular N termini of the two ELR-CXC chemokine receptors CXCR-1
and CXCR-2 bear an excess of acidic residues, whereas all of their
ligands contain clusters of basic residues within their C-terminal
-helices, it is likely that these complementary charges contribute
to the ligand-receptor interaction (1, 24). This interaction is
probable to be disturbed by acidic residues within the ligand,
especially by those located in proximity to the basic residues at the C
terminus of the ligand.
Interestingly, we observed that deletion of the leucine on position 63
in NAP-2 was associated with a threefold reduction of the capacity
of rNAP-2 (1–63) to bind to and activate PMN. Thus, L63
appears to represent a functionally important residue within the
structure of NAP-2. As mentioned above, this leucine is highly
conserved within CXC chemokines, and thus might also be important for
the functions of other members of this cytokine family. In fact,
functional studies on IL-8 tripeptide deletion mutants have shown (23)
that binding to and activation of PMN are reduced by 10-fold upon
truncation of IL-8 to beyond the conserved leucine. Interestingly, this
conserved residue represents the ultimate amino acid being arranged
into the -helix of monomeric IL-8 (21), which possibly also holds
true for monomeric NAP-2. It can be imagined that loss of this residue
is associated with a disturbance of the -helical structure. This
might lead to deterioration of the core structure of CXC chemokines,
and thus affect their binding and functional activity toward PMN. Taken
together, our data indicate that NAP-2 (1–63) represents a ß-TG Ag
isoform that by N- as well as C-terminal processing appears to be
optimally tailored to act as a neutrophil agonist.
Nevertheless, there remains the question as to whether the only physiologic role of C-terminal truncation in ß-TG Ag really is to enhance the proinflammatory impact of this chemokine. As we have shown recently, preincubation of PMN with CTAP-III (a situation mimicking the conditions encountered upon platelet activation) leads to down-regulation of responsiveness of the cells toward NAP-2 and other CXC chemokines (10). This desensitizing effect of CTAP-III relies on the circumstance that PMN themselves are capable of converting the inactive precursor into initially low, i.e., substimulatory amounts of NAP-2, which nevertheless bind to and efficiently down-regulate CXCR-2 from the cell surface (4, 10, 12). As we could show in the present study, C-terminal truncation modulates the desensitizing potency of CTAP-III in just the same way as it alters the stimulatory capacity of NAP-2, i.e., in both systems, deletion of acidic residues tends to increase the respective potencies, while deletion of the conserved leucine is correlated to a drastic decrease. In this respect, the natural occurrence of CTAP-III (1–81), a molecule that could represent the precursor for NAP-2 (1–66), i.e., a molecule having enhanced capacity to bind to receptors on PMN, was an interesting finding. It thus appears indeed possible that C-terminal truncation in ß-TG Ag might as well play a role in the down-modulation of PMN reactivity, which would be important, e.g., in the termination or dampening of an inflammatory reaction. The circumstance that we detect only small amounts of CTAP-III (1–81) could possibly be due to its conversion to NAP-2 (1–66) by PBMC-mediated N-terminal processing. Moreover, it has to be taken into account that C-terminally truncated ß-TG Ag isoforms as found in the PBMC supernatants represent just a minor portion of total ß-TG Ag, indicating that the given conditions are far from optimal to cause C-terminal processing. Therefore, one of our future aims will be to clarify the underlying enzymatic mechanism and to look for physiologic conditions leading to favored C-terminal processing of ß-TG Ag.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. J. E. Ehlert, Division of Biological Chemistry, Department of Immunology and Cell Biology, Forschungszentrum Borstel, Parkallee 22, D-23845 Borstel, Germany. E-mail address:
3 Abbreviations used in this paper: NAP-2, neutrophil-activating peptide-2; ß-TG, ß-thromboglobulin; CTAP-III, connective tissue-activating peptide-III; D-PBS, Dulbecco’s phosphate-buffered saline; 125I-NAP-2, 125I-labeled neutrophil-activating peptide-2; pI, isoelectric point; PMN, polymorphonuclear neutrophil granulocyte; TFA, trifluoroacetic acid.
Received for publication March 20, 1998. Accepted for publication June 25, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and neutrophil-activating peptide 2 to interleukin 8 receptors on human neutrophils. Proc. Natl. Acad. Sci. USA 89:10542. and ß IL-8 receptors. J. Immunol. 151:6418.[Abstract]
and NAP-2. FEBS Lett. 341:187.[Medline]
This article has been cited by other articles:
|
|
 |
|
  P. Proost, T. Loos, A. Mortier, E. Schutyser, M. Gouwy, S. Noppen, C. Dillen, I. Ronsse, R. Conings, S. Struyf, et al. Citrullination of CXCL8 by peptidylarginine deiminase alters receptor usage, prevents proteolysis, and dampens tissue inflammation J. Exp. Med., September 1, 2008; 205(9): 2085 - 2097. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hasan, L. Mazzucchelli, M. Liebi, M. Lis, R. E. Hunger, A. Tester, C. M. Overall, and M. Wolf Function of Liver Activation-Regulated Chemokine/CC Chemokine Ligand 20 Is Differently Affected by Cathepsin B and Cathepsin D Processing. J. Immunol., June 1, 2006; 176(11): 6512 - 6522. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J.J.M. Rennen, C. Frielink, E. Brandt, S. A.J. Zaat, O. C. Boerman, W. J.G. Oyen, and F. H.M. Corstens Relationship Between Neutrophil-Binding Affinity and Suitability for Infection Imaging: Comparison of 99mTc-Labeled NAP-2 (CXCL-7) and 3 C-Terminally Truncated Isoforms J. Nucl. Med., July 1, 2004; 45(7): 1217 - 1223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Hensbergen, D. Verzijl, C. I. A. Balog, R. Dijkman, R. C. van der Schors, E. M. H. van der Raaij-Helmer, M. J. A. van der Plas, R. Leurs, A. M. Deelder, M. J. Smit, et al. Furin Is a Chemokine-modifying Enzyme: IN VITRO AND IN VIVO PROCESSING OF CXCL10 GENERATES A C-TERMINALLY TRUNCATED CHEMOKINE RETAINING FULL ACTIVITY J. Biol. Chem., April 2, 2004; 279(14): 13402 - 13411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. De La Luz Sierra, F. Yang, M. Narazaki, O. Salvucci, D. Davis, R. Yarchoan, H. H. Zhang, H. Fales, and G. Tosato Differential processing of stromal-derived factor-1{alpha} and stromal-derived factor-1{beta} explains functional diversity Blood, April 1, 2004; 103(7): 2452 - 2459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. I. Schenk, F. Petersen, H.-D. Flad, and E. Brandt Platelet-Derived Chemokines CXC Chemokine Ligand (CXCL)7, Connective Tissue-Activating Peptide III, and CXCL4 Differentially Affect and Cross-Regulate Neutrophil Adhesion and Transendothelial Migration J. Immunol., September 1, 2002; 169(5): 2602 - 2610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Ehlert, A. Ludwig, T. A. Grimm, B. Lindner, H.-D. Flad, and E. Brandt Down-regulation of neutrophil functions by the ELR+ CXC chemokine platelet basic protein Blood, November 1, 2000; 96(9): 2965 - 2972. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Wuyts, A. D'Haese, V. Cremers, P. Menten, J.-P. Lenaerts, A. De Loof, H. Heremans, P. Proost, and J. Van Damme NH2- and COOH-Terminal Truncations of Murine Granulocyte Chemotactic Protein-2 Augment the In Vitro and In Vivo Neutrophil Chemotactic Potency J. Immunol., December 1, 1999; 163(11): 6155 - 6163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Stoeckelhuber, P. Dobner, P. Baumgartner, J. Ehlert, E. Brandt, R. Mentele, D. Adam, and B. Engelmann Stimulation of Cellular Sphingomyelin Import by the Chemokine Connective Tissue-activating Peptide III J. Biol. Chem., November 22, 2000; 275(48): 37365 - 37372. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
