| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, August 10, 1994; and in revised form, January 5, 1995) From the
We have previously described a C-terminally truncated variant of
the chemokine neutrophil-activating peptide 2 (NAP-2) that exhibited
higher neutrophil-stimulating capacity than the full-size polypeptide.
To investigate the impact of the NAP-2 C terminus on biological
activity and receptor binding, we have now purified the novel molecule
to homogeneity. Furthermore, we have cloned, expressed in Escherichia coli, and purified full-size recombinant NAP-2
(rNAP-2-(1-70)) and a series of C-terminally deleted variants
(rNAP-2-(1-69) to rNAP-2-(1-64)). Biochemical and
immunochemical analyses revealed that the natural NAP-2 variant was
structurally identical to the rNAP-2-(1-66) isoform. As compared
with their respective native and recombinant full-size counterparts,
both molecules exhibited
The chemokine neutrophil-activating peptide 2 (NAP-2) ( The
Figure 1:
Recombinant expression of NAP-2 and
variants in E. coli. Shown is a schematic diagram of the
construction of pAXNAP expression vectors. Inserts were generated by
polymerase chain reaction with the 5`-primer FOR and the 3`-primers
BACK70 to BACK64 (Table 1) using platelet-derived NAP-2 cDNA as
template. The 3`-ends of the primers were complementary to the template (openboxes). The primers' 5`-ends were used to
introduce the coding sequence for the factor Xa recognition site (FXa-rs) and the restriction site for BglII (Bgl2) at the 5`-ends as well as a site for SalI (Sal1) and a stop codon (Stop) at the 3`-ends of the
variants' sequences. Inserts were cloned into the multiple
cloning site (MCS) using restriction enzymes BglII
and SalI. The lacZ gene and the
Figure 2:
Comparison of antibody reactivities of
native and recombinant NAP-2 variants. Shown are Western blot analyses
of native NAP-2 (lanesN), NAP-2-(X) (laneX), and recombinant rNAP-2-(1-70) to
rNAP-2-(1-64) (lanes70 to 64,
respectively) separated by SDS-PAGE (20 ng/lane). Blots were
immunochemically stained with the polyclonal antisera R
Figure 3:
Comparison of net charges of native and
recombinant NAP-2 variants. Native NAP-2 (lanes N), NAP-2-(X) (laneX), and recombinant rNAP-2-(1-70) to
rNAP-2-(1-64) (lanes70 to 64,
respectively) were separated by IEF (0.2 µg/lane) and immunoblotted
using polyclonal antiserum R
Figure 4:
Relative potencies of native and
recombinant NAP-2 variants for receptor binding and degranulation. A, relative potencies of native NAP-2 (columnN), NAP-2-(X) (columnX), and
recombinant rNAP-2-(1-70) to rNAP-2-(1-64) (columns70 to 64, respectively) to induce degranulation
in neutrophils (potency of native NAP-2 = 100%); B,
relative potencies of the same variants to compete with 10 nM radiolabeled native NAP-2 for receptor binding (potency of native
unlabeled NAP-2 = 100%). Data are given as means ± S.D.
of three to seven experiments.
Figure 5:
Receptor binding and biological activities
of NAP-2-(X) and rNAP-2-(1-66) in comparison with the full-size
molecules. A, induction of lysosomal elastase release in
cytochalasin B-pretreated PMN by increasing concentrations of NAP-2 and
truncated variants; B, displacement of 10 nM
Corresponding results were obtained in receptor
competition assays. As shown in Fig. 4B, the ability of
recombinant NAP-2 variants to compete for binding with a fixed
concentration of 10 nM
Due to an improved purification
protocol (described under ``Experimental Procedures''), it
was finally possible to isolate sufficient amounts of native NAP-2-(X)
for mass spectroscopic analysis and determination of the C-terminal
sequence. The molecular weight measured by matrix-assisted laser
desorption/ionization mass spectroscopy was 7227 and differed by
<0.1% from the theoretical value(7222) for a NAP-2 molecule
truncated by four residues at the C terminus. In a further approach to
directly identify the C-terminal sequence of NAP-2-(X), digestion of
the protein with endoproteinase Lys-C yielded a peptide fragment with
the sequence KLAGD. This pentapeptide unambiguously corresponds to
positions 62-66 in NAP-2. Thus, both analyses directly identify
NAP-2-(X) as NAP-2-(1-66). As shown in Fig. 5, the
functional properties of NAP-2-(X) and its recombinant homologue
rNAP-2-(1-66) were also in perfect agreement. Identical elastase
release rates were obtained with both polypeptides over a wide range of
concentrations, and this was paralleled by exact alignment of the
ligand binding curves obtained with either molecule in receptor
competition experiments with radiolabeled full-size NAP-2. It may thus
be concluded that enhanced functional activity and receptor binding in
NAP-2-(X) are due to defined proteolytic truncation and not to other
kinds of post-translational modification. This study was brought about by our recent discovery of a
molecular variant of the chemokine NAP-2 (herein termed NAP-2-(X)) that
was truncated at the C terminus and exhibited enhanced biological
activity(23) . These findings indicated that the C terminus
could be important for the function of the chemokine. However, direct
proof was still lacking because it could not be excluded that
post-translational modification other than proteolytic cleavage was
responsible for increased biological activity in NAP-2-(X). A further
incertitude was inferred by the circumstance that the final preparation
of NAP-2-(X) was still contaminated by platelet basic protein and that
the extent of truncation could not be exactly determined. Thus, in the
present study, we first cloned and expressed NAP-2 and a series of
C-terminally truncated variants in E. coli. We then examined
the capacities of the molecule to induce degranulation in PMN and to
bind to specific receptors on these cells. Moreover, the successful
purification to homogeneity of NAP-2-(X) from PBMC-derived culture
supernatants allowed for a direct comparison of this naturally
occurring molecule with the recombinant chemokines, in both structural
and functional respects. Although the biologically active
Functional analyses performed with recombinant C-terminally deleted
NAP-2 variants demonstrated a biphasic impact of truncation on
chemokine activity. Whereas stepwise truncation by up to four residues
successively enhanced degranulation activity and receptor binding,
further shortening reduced these activities even to below the level of
the full-size molecule (Fig. 4). These results demonstrate that
the absence or presence of even a single amino acid residue may
considerably affect the chemokine's function. Dependence of
functional activity on the length of the C terminus has also been
demonstrated for IL-8, where a synthetic analogue truncated by three
residues exhibited slightly enhanced activity, while truncation by six
residues abolished this effect(18) . However, the influence of
single residue deletions was not analyzed. Apart from demonstrating
the impact of C-terminal truncation on the recombinant molecules, we
also found evidence that this principle may be important under
physiological conditions. This was indicated by the occurrence of the
natural isoform NAP-2-(X) purified from PBMC-derived culture
supernatants, which we have shown here to be truncated to the same
extent as the recombinant variant rNAP-2-(1-66). Both molecules
exhibited the same enhanced capacity to stimulate PMN degranulation and
to bind to receptors on these cells, indicating that their improved
function was due to the precise truncation behind Asp There are several possibilities how this truncation could
influence receptor binding and function. Structural analyses of the
related polypeptides IL-8(14) , bovine PF-4(41) , and
MGSA (15) have revealed that the C-terminal stretches of these
molecules are all arranged into amphiphilic The occurrence of precise truncation in NAP-2-(X)
raises questions regarding the mechanism involved in the generation of
such an optimally tailored molecule. The fact that At present, we
can only speculate on the potential physiological role of NAP-2-(X).
Given that this truncated variant is simultaneously formed with NAP-2,
it could participate in the very first line of defense to injury as a
more potent activator of PMN. However, according to our data, it cannot
be excluded that NAP-2-(X) formation is dependent on the presence of
activated leukocytes and would thus be generated at an advanced stage
of inflammation, where other much more potent chemokines such as IL-8
are formed. In such a situation, the prevailing function of NAP-2-(X)
could be to down-regulate the PMN response. As we have previously found
for the full-size chemokine, very low (i.e. nonstimulatory)
concentrations of NAP-2 can desensitize PMN to subsequent challenge
with other chemokines, while IL-8 is inactive in this
respect(10) . Preliminary experiments performed with NAP-2-(X)
indicate that this variant exhibits a desensitizing capacity 4-fold
higher than that of NAP-2.
Volume 270,
Number 11,
Issue of March 17, 1995 pp. 6338-6344
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
COMPARISON OF NATIVE AND RECOMBINANT NAP-2 VARIANTS (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
3-4-fold enhanced potency in the
induction of neutrophil degranulation as well as 3-fold enhanced
binding affinity for specific receptors on these cells. All other
variants were considerably less active. The natural occurrence of a
NAP-2 variant truncated by exactly four residues at the C terminus
suggests that limited and defined proteolysis at this site plays a role
in the regulation of the biological function of the chemokine.
)is a 70-amino acid residue polypeptide (1) that is
formed from platelet-derived precursors by proteolytic processing.
These precursors are homologous molecules differing in the lengths of
their N termini and are collectively termed 
-thromboglobulin
antigen (TG Ag). At least two of the polypeptides, platelet basic
protein (94 residues) and connective tissue-activating peptide III
(CTAP-III) (85 residues), were directly shown to become converted into
NAP-2 by N-terminal truncation through monocyte and granulocyte
proteases(2, 3) . Mature NAP-2 stimulates various
effector functions of polymorphonuclear neutrophil granulocytes (PMN)
including directed chemotactic migration(4, 5) ,
exocytosis of lysosomal enzymes (5) and secondary granule
contents(6) , and up-regulation of adhesion
receptors(7) . Together with structurally and functionally
related chemokines such as interleukin-8 (IL-8) and melanoma
growth-stimulating activity (MGSA), NAP-2 has been assigned to a
subfamily now termed ``
-chemokines.'' These mediators
are selective activators of granulocyte (but not monocyte) functions
and were found to act on PMN through specific binding to common
receptors, the IL-8 receptors type A and B. Only high affinity binding
was detected for IL-8, while separate high and low affinity binding
sites exist for NAP-2 and
MGSA(4, 8, 9, 10) .-chemokines contain four cysteine residues at highly conserved
positions, which enclose the core region of the molecules. The first
two cysteines are separated by a single amino acid, forming a motif
(CXC) that distinguishes the -chemokine from the
-chemokine subfamily, where these cysteines are in directly
adjacent position (CC) (reviewed in (11, 12, 13) ). Although structural analyses
of chemokines have shown that disulfide bridge formation is essential
for the maintenance of their typical molecular conformation (14, 15) as well as for biological
activity(16, 17) , additional structural features
important for receptor binding and biological function have been found.
Concerning the -chemokines, a common sequence motif of three
successive amino acids (ELR) preceding the CXC grouping was
recognized to form an indispensable prerequisite for the induction of a
transient Ca
influx and biological responses in
PMN(18, 19) . This was directly shown for IL-8 and
MGSA, where substitution of the ELR motif for alanines resulted in a
drastic decrease in PMN-stimulating activity(20, 21) ,
while insertion of the motif conferred enhanced activity to the poorly
active chemokine homologue platelet factor 4 (PF-4)(22) .
Although now it appears clear that defined structural determinants
within the -chemokine N terminus are required for receptor binding
and functional activation of PMN, recent investigations indicate that
further, although not yet precisely defined regions are also important.
Studies on IL-8 suggest that binding to the type A IL-8 receptor also
depends on the presence of certain residues within the core region,
whereas binding to the type B IL-8 receptor requires structures
provided by the C terminus(21) . However, it is not yet clear
whether the same principles apply to other -chemokines. In this
context, our recent finding of a variant NAP-2 molecule (herein termed
NAP-2-(X)) that was truncated at the C terminus and exhibited enhanced
PMN-stimulating potency suggests the involvement of C-terminal
structures in the regulation of NAP-2 function(23) . However,
in the latter study, purification of NAP-2-(X) from culture
supernatants of stimulated peripheral blood mononuclear cells (PBMC)
was difficult, and we were not able to completely separate the molecule
from contaminating platelet basic protein nor could we precisely
determine the extent of C-terminal truncation. In the present work, we
have therefore expressed and characterized recombinant NAP-2 and a
series of C-terminally deleted variants. This approach enabled us to
directly analyze the impact of C-terminal truncation on NAP-2 function
and receptor binding while excluding the potential influence of other
kinds of post-translational modification. We have furthermore now
succeeded in purifying NAP-2-(X) to homogeneity. Comparison of this
molecule with the recombinant polypeptides revealed that strictly
defined C-terminal truncation participates in the regulation of NAP-2
function.
Bacterial Strains and Recombinant DNA
Methods
Competent Escherichia coli cells (MAX
efficiency DH5F`IQ) were obtained from GIBCO BRL (Eggenstein,
Germany). Bacteria were grown in LB medium using ampicillin as the
selective marker. If not otherwise stated, recombinant DNA methods were
performed according to Sambrook et al.(24) . Both
strands of insert-containing plasmid DNA were sequenced with an
automated laser fluorescent DNA sequencer (Pharmacia, Uppsala, Sweden)
using the chain termination method (25) with T7 DNA polymerase
(Pharmacia) and fluorescein isothiocyanate-labeled primers according to
the supplier's manual. Sequencing primers PAX1-FITC
(5`-CCTGGTCTTGCTGGCCAACAT-3`) and PAX2-FITC
(5`-CCCGGCGGCAACCGAGCGTTCT-3`) were derived from the pAX5
(Medac, Hamburg, Germany) plasmid sequence. Sequences were
evaluated on Microgenie software (Release 7.1; Beckmann Instruments,
Munich, Germany). Oligonucleotides were synthesized with an automated
DNA synthesizer (model DNA SM, Beckmann Instruments) using standard
cyanoethylphosphoamidite chemistry.Insert Construction
Poly(A) RNA
was purified from enriched platelet preparations (obtained from Dr. B.
Katzmann (Institute of Immunology and Transfusion Medicine, Medical
University of Lübeck, Lübeck,
Germany)) by oligo(dT)
Dynabeads (Dynal, Oslo, Norway)
according to the manufacturer's manual. The cDNA was prepared by
reverse transcription with oligo(dT) using the Superscript kit (GIBCO
BRL). Inserts were generated by polymerase chain reaction employing
NAP-2 cDNA as template (corresponding to CTAP-III mRNA (positions
241-450), EMBL accession number M54995(26) ). Polymerase
chain reaction conditions were chosen as described
previously(27) . The following oligonucleotide termed FOR
served as the 5`-primer:
5`-ATATAGATCTTGATCGAGGGTAGGGCTGAACTCCGCTGCATGTGTATAAAG-3`.
The sequence written in boldface is consistent with the first 27 bases
coding for NAP-2. To permit generation of a definite N terminus by
site-specific proteolysis, a sequence (written in italics) coding for
the endoproteinase factor Xa recognition site Ile-Glu-Gly-Arg was
introduced at the 5`-end of the NAP-2 cDNA sequence. A BglII
restriction site (underlined) was added to allow cloning into the
pAX5 vector. To create different C-terminally deleted
variants of NAP-2, disparate 3`-primers (BACK70 to BACK64) (Table 1) were used. The 17 bases at the primers' 3`-ends
matched the NAP-2 cDNA sequence, followed by a nonsense codon and a SalI restriction site for cloning into pAX5.
The 3`-end of BACK70 was complementary to the last 17 bases at the
3`-end of the NAP-2 sequence, whereas the complementary regions of
3`-primers BACK69 to BACK64 were successively shifted upstream by one
triplet to cause stepwise deletion of C-terminal amino acids at the
protein level.
Plasmid Construction
Inserts were ligated into
vector pAX5, leading to expression plasmids pAXNAP70
to pAXNAP64 (see Fig. 1), coding for a tripartite fusion protein
consisting of -galactosidase, a collagen fragment, and the
respective NAP-2 variant with an N-terminal factor Xa recognition site.
Expression plasmids were used for transformation of competent E.
coli cells. Transformed clones were screened for inserts by
polymerase chain reaction (28) using primer PAX1-FITC in
combination with BACK70 to BACK64, respectively. Plasmids of
transformants used for protein expression were verified by DNA
sequencing.
t
terminator are shown as blackboxes. The
Amp
antibiotic resistance marker and the collagen fragment (CS) gene are shown as grayboxes. The lacZ gene is preceded by its specific promoter
(P
). Inset, Western blot analyses using mAb C-24
of fusion protein from enriched inclusion bodies (lane1) and of the peptide released upon factor Xa proteolysis (lane2). Lane3 shows silver
staining of the immunopurified peptide upon
SDS-PAGE.
Expression and Purification of Recombinant NAP-2
Variants
After fermentation to a culture density of A
= 1.0, transformants were induced to
express fusion protein with
isopropyl--D-thiogalactopyranoside at a final
concentration of 1 mM. Upon 5 h of incubation at 37 °C,
cells were harvested, and the pellet was resuspended in lysis buffer
(50 mM Tris-HCl, pH 7.4, 10 mM -mercaptoethanol,
10 mM EDTA, 10 mM MgCl
, 10 µg/ml
DNase I (Boehringer, Mannheim, Germany)) and sonicated. Expressed
fusion protein stored in inclusion bodies was purified according to
Schoner et al.(29) and subsequently refolded
following a protocol optimized for the refolding of
-galactosidase(30) . Upon digestion with factor
Xa(31) , the solution was acidified with trifluoroacetic acid
to pH 2.1 and afterwards was reneutralized to pH 7.2 by sodium
hydroxide. Precipitated protein was pelleted, and the supernatant was
applied to an immunoaffinity column coated with TG Ag-specific
monoclonal antibody C-24 (see below)(3) . Immunopurified
peptides were stored in 0.1% trifluoroacetic acid at -20 °C.Purification of Native Cytokines
Native NAP-2 was
purified to homogeneity from culture supernatants of stimulated PBMC
using sequential immunoaffinity chromatography, cation-exchange
chromatography, and reversed-phase HPLC as described
previously(3, 32) . NAP-2-(X) was purified to
homogeneity from the same source. After its separation from NAP-2 by
cation-exchange chromatography as described previously(23) ,
further separation from platelet basic protein and other contaminants
was achieved by reversed-phase HPLC on an analytic cyanopropyl column
(4.6 
250 mm, 5 µm, wide pore; J. T. Baker Inc.). Samples
acidified with trifluoroacetic acid were directly loaded and eluted at
0.5 ml/min with a gradient of 0-35% 1-propanol in 0.1%
trifluoroacetic acid. NAP-2-(X) eluted at 12% 1-propanol as a distinct
symmetrical protein peak. A single sequence (AELRXMXI
. . . , where X stands for an unidentified residue, probably
C, since samples were not reduced) was obtained upon N-terminal amino
acid sequencing, and a single band was detectable by silver stain
analysis of SDS-polyacrylamide gels as well as on immunoblots of
SDS-polyacrylamide and IEF gels stained with TG Ag-specific
antibodies (see below). In comparison with a control of NAP-2 run in
parallel, NAP-2-(X) migrated slightly faster on SDS-PAGE and focused at
a clearly higher pI (8.9 versus 9.4) on isoelectric focusing
(IEF).Mass Spectroscopy
Determination of the molecular
weight of NAP-2-(X) was performed by matrix-assisted laser
desorption/ionization mass spectroscopy (33) by Eurogentec
(Seraing, Belgium).N-terminal Amino Acid Sequencing
N-terminal
sequence analyses of native and recombinant NAP-2 and its variants were
performed by Dr. A. Petersen (Department of Clinical Medicine,
Forschungsinstitut Borstel, Borstel, Germany) on a gas-phase sequencer
(Model 473A, Applied Biosystems Inc., Foster City, CA).Identification of the NAP-2-(X) C-terminal
Sequence
Purified NAP-2-(X) was subjected to digestion with
endoproteinase Lys-C. Upon separation of the resulting peptides by
reversed-phase HPLC on a MicroRPC C-C
column
(Pharmacia), the fragment representing the C terminus of NAP-2-(X) was
identified by complete sequence analysis by Edman
degradation(34) . All procedures were performed as a custom
service by Eurogentec.Electrophoresis, Immunoblotting, and Antibody
Reagents
SDS-PAGE under reducing conditions, IEF on
polyacrylamide gels in the presence of 8 M urea, transfer of
protein bands onto polyvinylidene difluoride membranes, and
immunochemical detection of TG Ag polypeptides were carried out as
described previously(23) . A monoclonal antibody (mAb C-24)
reacting with all variably truncated isoforms of TG Ag that are
presently known was induced in mice, cloned, and purified as described
previously (3) . Furthermore, the following rabbit polyclonal
antisera were used: R-TG, raised against a purified
preparation of native TG Ag, and R-NAPII/55-70, raised
against a synthetic peptide consistent with the 16 C-terminal amino
acids of NAP-2 (Ile
-Asp
)(23) .
This antiserum reacted against different epitopes within the peptide
structure, one located at the ultimate C terminus
(Glu
-Asp) and other(s) located more
upstream. Separation of R-NAPII/55-70 by sequential affinity
chromatography with immobilized peptides (NAP-2
Ile-Ala
and
Ile-Asp, respectively) yielded an
antibody fraction (R-70) that bound to the
Glu-Asp motif and indispensably
required the presence of the ultimate C-terminal amino acid in NAP-2
(Asp) for binding, as described previously(23) .Neutrophils: Preparation and Degranulation
Assay
Human PMN were isolated from citrated blood of single
healthy donors by gradient centrifugation on Ficoll-Hypaque as
described previously (3) to a purity >95% in all events.
Activities of stimuli (tested in 2-fold serial dilutions) were assessed
by their ability to induce the release of lysosomal marker elastase
from cytochalasin B-treated cells. Elastase activity released into the
supernatant was measured as described previously(3) . Release
rates for elastase are expressed as the percentage of total content in
detergent-treated PMN lysates prepared in 0.1%
hexadecyltrimethylammonium bromide. Relative potencies of NAP-2
variants are expressed as percentages of the potency of native
full-size NAP-2 according to the following equation: % potency =
([A]
/[B]
) 100. [A] and [B] represent
concentrations of stimuli that elicit identical release rates.Receptor Binding Competition Assay
The interaction
of unlabeled NAP-2 and NAP-2 variants with chemokine receptors on PMN
was investigated in binding competition assays using native
radiolabeled NAP-2 as a tracer. NAP-2 was chemically modified by the
introduction of additional tyrosine residues prior to labeling with 
I, exactly as recently published(10) . The
specific radioactivity of I-NAP-2 was 565 Ci/mmol.
Receptor binding competition assays were performed as
described(10) , using a constant concentration of I-NAP-2 in the absence and presence of increasing
concentrations of unlabeled NAP-2 and NAP-2 variants (up to a 100-fold
molar excess). Nonspecific binding of I-NAP-2 was
subtracted. Competition by full-size NAP-2 up to a 100-fold molar
excess was measured in every assay, serving as a reference.
Determination of relative binding potency was performed according to
the equation described above. In the receptor binding competition
assay, [A] and [B] represent concentrations that
cause 50% competition with labeled NAP-2.
Cloning, Expression, and Purification of Recombinant
NAP-2 and Its Variants
We have established a prokaryotic
expression system in E. coli for recombinant full-size NAP-2
(rNAP-2-(1-70)) and NAP-2 variants deleted by up to six
C-terminal amino acids (rNAP-2-(1-69) to rNAP-2-(1-64)).
For this purpose, we used expression vectors pAXNAP70 to pAXNAP64,
which were constructed as depicted in Fig. 1and as described
under ``Experimental Procedures.'' The insert sequences in
plasmids of transformed E. coli subsequently used for protein
expression were verified by DNA sequencing. Induction of
pAXNAP70-containing bacteria led to the high level expression of a
protein that proved to be reactive with TG Ag-specific mAb C-24
upon SDS-PAGE and Western blotting (Fig. 1, inset, lane1). The apparent M of
132,000 was consistent with that expected for the tripartite
fusion protein. The major part of this immunoreactive protein
(70%) was insoluble in lysis buffer and, according to
phase-contrast microscopy, was stored in inclusion bodies. Enrichment
of inclusion bodies by differential centrifugation and subsequent
solubilization by 8 M urea, 1% -mercaptoethanol led to a
fusion protein solution of 80% purity as estimated by
semiquantitative analysis of Coomassie Blue-stained SDS-polyacrylamide
gels. Upon refolding by dialysis, digestion of enriched fusion protein
with endoproteinase factor Xa released several mAb C-24-reactive
protein fragments visible on Western blots (data not shown). The
smallest evolving fragment had a size of 8 kDa, consistent with
the molecular mass of native NAP-2. Subsequent acidification and
reneutralization of the solution led to precipitation of all
immunoreactive protein, except for the 8-kDa fragment (Fig. 1, inset, lane2). After immunopurification on
mAb C-24-Sepharose, the 8-kDa polypeptide was homogeneous as confirmed
by silver stain analysis of SDS-polyacrylamide gels (Fig. 1, inset, lane3) and by N-terminal sequence
analysis, where only the expected sequence AELRXMXI .
. . , consistent with the N terminus of native NAP-2, was obtained.
Expression and purification to homogeneity of the C-terminally deleted
variants rNAP-2-(1-69) to rNAP-2-(1-64) were achieved by
the same methods. The final products all exhibited a correct N
terminus.Comparison of Native and Recombinant Full-size
NAP-2
To investigate whether native NAP-2 and
rNAP-2-(1-70) were structurally and functionally equivalent, the
polypeptides were directly compared in a set of different assays.
According to SDS-PAGE and Western blot analyses, the molecules were of
the same size, migrating at identical positions corresponding to 8
kDa (Fig. 2, lanesN and 70).
Furthermore, comparable reactivities with the three different antisera
used for detection indicated the presence of identical epitopes in both
molecules. Especially, positive reactivity with R-70, an antiserum
dependent on the presence of the ultimate residue (Asp) in
native NAP-2, confirmed that the recombinant polypeptide had an intact
C terminus. Finally, as seen by IEF and subsequent immunoblotting,
native and recombinant NAP-2 exhibited an identical pI of 
8.9 (Fig. 3, lanesN and 70), providing
further evidence that these molecules are structurally equivalent.
Results from functional analyses of the polypeptides paralleled these
findings. Thus, rNAP-2-(1-70) and its native counterpart
exhibited comparable biological activities as seen by their identical
potencies to stimulate the release of lysosomal elastase from
cytochalasin B-treated PMN (Fig. 4A) and by practically
identical dose-response curves for a wide range of concentrations (Fig. 5A). On the other hand, degranulation in response
to a high concentration (80 nM) of either polypeptide was
dose-dependently inhibited in the presence of TG Ag-specific mAb
C-24. Total inhibition occurred at a 2-fold molar excess of the
antibody over native NAP-2 as well as rNAP-2-(1-70) (data not
shown). These results confirmed that the PMN-stimulating activity
observed with rNAP-2-(1-70) was associated with the polypeptide
and was not due to potential contamination with bacterial formylated
peptides. Corresponding results were obtained in receptor binding
assays where rNAP-2-(1-70) exhibited the same potency as native
NAP-2 in competing with radiolabeled native NAP-2 for binding to
specific receptors on neutrophils. As shown in Fig. 5B,
a 10 nM concentration of either unlabeled polypeptide reduced
the specific binding of 10 nMI-NAP-2 by 50%.
-TG (upperpanel), R-70 (centerpanel), and R-NAPII/55-70 (lowerpanel).
-TG as the detecting reagent. pH
values in the gel were determined by means of a
microelectrode.
I-NAP-2 binding to PMN by the same polypeptides (dashedline indicates 50% competition). In A and B, one representative experiment (out of five) is
shown.
Impact of C-terminal Truncation on Biological Function of
NAP-2
The successful preparation of a recombinant NAP-2 molecule
that was functionally equivalent to native NAP-2 provided an adequate
basis for further studies aimed at elucidating the potential role of
the chemokine's C terminus. Comparison of the biological
activities of the full-size molecules with those of C-terminally
truncated variants yielded the results shown in Fig. 4A: deletion of up to three amino acids
(rNAP-2-(1-69) to rNAP-2-(1-67)) led to a slight but
reproducible increase in the chemokine's potency to stimulate
neutrophil degranulation (to 190% of control). A more prominent
increase in potency to 400% was observed with a polypeptide
truncated by four residues (rNAP-2-(1-66)), while further
truncated variants (rNAP-2-(1-65) and rNAP-2-(1-64))
exhibited activities even lower than those of the full-size chemokines.
Interestingly, the still undefined native variant NAP-2-(X) (purified
to homogeneity from PBMC culture supernatants; see ``Experimental
Procedures'') was as potent as the most active variant,
rNAP-2-(1-66).I-NAP-2 increased with
C-terminal deletions of up to four amino acids and sharply declined
upon further truncation. The very prominent increase in binding potency
obtained with rNAP-2-(1-66) was practically identical to that of
NAP-2-(X), with both polypeptides exhibiting 280% of the potency
observed with the full-size chemokine. Improved binding of NAP-2-(X) to
PMN could not be ascribed to a selective increase in affinity for one
of the two binding sites known for NAP-2 (determined to K
0.4 and 20 nM, as described
previously(10) ). In competition assays performed with 1 nMI-NAP-2, a concentration selectively addressing the
high affinity binding site, the binding potency of rNAP-2-(1-66)
over the full-size molecules was 180%. When using a concentration
of 20 nMI-NAP-2 involving both binding sites,
the potency of rNAP-2-(1-66) increased to 300% (data not
shown), indicating that enhanced binding seen with C-terminal
truncation is a phenomenon mediated by both sites.Structure of NAP-2-(X): Comparison with Recombinant
C-terminally Deleted NAP-2 Variants and Direct Analyses
The
similarities of natural NAP-2-(X) and recombinant rNAP-2-(1-66)
in biological activity and receptor binding led us to examine whether
these molecules were structurally identical. This was first done by
comparing the biochemical and immunochemical characteristics of
NAP-2-(X) with those of variants rNAP-2-(1-69) to
rNAP-2-(1-64) and the full-size chemokines. In immunoblots of
SDS-polyacrylamide gels developed with R-TG antiserum,
NAP-2-(X) and the recombinant variants all migrated slightly faster
than the full-size polypeptides (Fig. 2). The failure of
antiserum R-70 (specific for the ultimate TG C terminus) to
detect the recombinant variants as well as NAP-2-(X) reconfirmed that
all these molecules were truncated at the C terminus. More relevant
information could be deduced from the reactivity pattern of
polypeptides with an antiserum (R-NAPII/55-70) that reacted
to an additional epitope located more upstream within the NAP-2
C-terminal sequence. As seen in Fig. 2, this antiserum detected
not only full-size NAP-2, but also part of the recombinant variants,
namely those deleted by up to four residues (rNAP-2-(1-69) to
rNAP-2-(1-66)). By contrast, immunoreactivity with further
truncated variants was almost completely abolished, indicating that the
epitope for antibody recognition had become destroyed in these
molecules. Interestingly, reactivity with NAP-2-(X) was fully
preserved, suggesting that this polypeptide is truncated by at least
one but maximally four residues. A more precise estimate on the number
of residues missing in NAP-2-(X) was possible after further comparative
analyses by IEF and immunoblotting. As depicted in Fig. 3, all
truncated variants focused at more basic pI values than the full-size
molecules. A first shift in net charge was observed with variants
rNAP-2-(1-69) to rNAP-2-(1-67), probably due to the
elimination of one negatively charged amino acid, the ultimate
Asp (C-terminal sequence in NAP-2, . . .
Lys
-Lys
-Leu
-Ala
-Gly
-Asp
-Glu
-Ser
-Ala
-Asp
(negatively charged residues are depicted in boldface)). A further
increase in positive net charge resulted upon deletion of Glu with rNAP-2-(1-66), while no additional changes were
detectable with variants rNAP-2-(1-65) and rNAP-2-(1-64).
The band representing NAP-2-(X) focused at a pI value identical to that
of variants rNAP-2-(1-66) to rNAP-2-(1-64). According to
these results, NAP-2-(X) should represent a molecule lacking at
least four residues. Since truncation of the molecule by maximally four residues was obvious from its reactivity
patterns with the different TG Ag antisera (as shown above in Fig. 2), NAP-2-(X) should represent a NAP-2 isoform truncated by
exactly four C-terminal amino acids.
-chemokines IL-8(35) , MGSA(36) , and PF-4 (37, 38) have been successfully expressed in
prokaryotes, no such reports exist for NAP-2. A possible reason for
this could be problems in establishing stable rNAP-2-producing E.
coli clones, as was reported by others(39) . We
encountered similar problems with E. coli strain JM109, ()but successful expression of NAP-2 as part of a fusion
protein was achieved in strain DH5F`IQ. The chemokine could then
easily be released by digestion with endoproteinase factor Xa.
Comparison of purified recombinant NAP-2 (rNAP-2-(1-70)) with
native NAP-2 revealed identity of the molecules in all biochemical and
biological parameters analyzed (Fig. 2, 3, and 5). Apparently,
the chemokine's structure responsible for degranulation activity
as well as receptor binding on PMN is exclusively determined by its
primary amino acid sequence and not by potential post-translational
modifications. These results are in accordance with those obtained for
synthetic NAP-2 by Clark-Lewis et al.(39) . In fact,
the only modification observed in TG Ag consisted of the
nonenzymatic addition of glucose to variable lysine side
chains(40) . Although the NAP-2 sequence contains a potential
phosphorylation site, such modification has never been detected. 
and
not to other post-translational modifications. Our present finding that
NAP-2-(X) was truncated by four residues is not consistent with results
from our previous work(23) , where we assumed that maximally
three residues were missing. This difference is probably due to the
circumstance that the epitope specificity of R-NAPII/55-70
in that study was determined by means of short synthetic peptides,
which were not long enough to include all the epitopes required for
binding.-helices that interact
with stretches of underlying -sheets. Homologies in primary and
secondary structure allow these conformational features to be
superimposed on NAP-2. Previous studies on IL-8 showing that a molecule
lacking the complete C-terminal -helix was still active while a
corresponding C-terminal synthetic peptide was not suggested that this
region stabilizes the chemokine's tertiary structure and does not
directly interact with the receptor (18) . Although our own
studies (
)showing that a synthetic peptide homologous to the
C terminus of NAP-2 (residues 48-70) was likewise inactive are
consistent with the above model, our results obtained with the stepwise
truncated polypeptides favor an alternative hypothesis. The biphasic
course of binding affinities peaking with the removal of exactly four
residues indicates that there may exist an optimal size of the
molecule's C terminus with best fit into the receptors. As
indicated by a concomitant increase in biological activity, this
mechanism may lead to improved interaction of functionally important
sites (e.g. the ELR motif(18) ) with their
counterstructures within the receptors. Our observation that binding of
NAP-2-(X) and rNAP-2-(1-66) was enhanced to both the high and low
affinity receptors to a similar extent is in agreement with our
previous findings that simultaneous interaction of NAP-2 with both
receptors is required to induce an optimal degranulation response in
PMN(10) .-helical
structures are generally highly resistant to proteolytic attack would
preclude cleavage at the NAP-2 C terminus. In this respect, it is
interesting to note that the ultimate amino acids in IL-8 (one
residue)(42) , bovine PF-4 (three residues)(41) , and
MGSA (four residues) (15) have been found to form fraying ends
with high mobility instead of participating in the rigid -helical
structure. Although no such structural analyses have been performed on
NAP-2, its sequence homology with bovine PF-4 at the C terminus (NAP-2,
. . . GDESAD; and PF-4, . . . GDES) implies that a corresponding
fraying end comprising five residues may follow the -helix in the
former chemokine. Such a disordered structure would facilitate the
access of proteolytic enzymes, as has previously been reported for
MGSA(43) . Nevertheless, the conditions under which NAP-2-(X)
is formed are not yet clear. Since we discovered NAP-2-(X) in
stimulated PBMC culture supernatants(23) , but not upon
coincubation with purified PMN(3) , its formation probably
depends on proteolytic enzymes associated with leukocytes other than
neutrophils. Apart from various carboxypeptidases, a likely candidate
would be an endoproteinase termed granzyme B, which is released by
activated cytotoxic T-cells. The unique specificity of this enzyme,
which is one of the few proteases known to preferentially attack
peptide bonds C-terminal to aspartic acid residues(44) , would
most easily explain the generation of NAP-2-(X) in stimulated PBMC
cultures. Experiments to verify this are underway. Thus, the truncated molecule
could participate in the termination of an inflammatory response.
)TG
Ag, -thromboglobulin antigen; CTAP-III, connective
tissue-activating peptide III; PMN, polymorphonuclear neutrophil
granulocyte(s); IL-8, interleukin-8; MGSA, melanoma growth-stimulating
activity; PF-4, platelet factor 4; PBMC, peripheral blood mononuclear
cell(s); HPLC, high pressure liquid chromatography; IEF, isoelectric
focusing; PAGE, polyacrylamide gel electrophoresis; mAb, monoclonal
antibody.))
We thank Dr. A. Petersen for performing sequence
analyses of native and recombinant NAP-2 and variants, C. Wohlenberg
for performing sequence analyses of plasmid DNA, and Dr. B. Katzmann
for providing platelet preparations. We especially thank C. Pongratz
and G. Hucß for perfect technical assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
|
|
|
|
J. E. Ehlert, C. A. Addison, M. D. Burdick, S. L. Kunkel, and R. M. Strieter Identification and Partial Characterization of a Variant of Human CXCR3 Generated by Posttranscriptional Exon Skipping J. Immunol., November 15, 2004; 173(10): 6234 - 6240. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
J. E. Ehlert, J. Gerdes, H.-D. Flad, and E. Brandt Novel C-Terminally Truncated Isoforms of the CXC Chemokine {beta}-Thromboglobulin and Their Impact on Neutrophil Functions J. Immunol., November 1, 1998; 161(9): 4975 - 4982. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ludwig, F. Petersen, S. Zahn, O. Gotze, J.-M. Schroder, H.-D. Flad, and E. Brandt The CXC-Chemokine Neutrophil-Activating Peptide-2 Induces Two Distinct Optima of Neutrophil Chemotaxis by Differential Interaction With Interleukin-8 Receptors CXCR-1 and CXCR-2 Blood, December 1, 1997; 90(11): 4588 - 4597. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||
