Department of Pulmonary Diseases and
* Immunology, University Medical Center and
Utrecht Biotechnology Systems, Utrecht, The Netherlands
Correspondence: Dr. L. Koenderman, Department of Pulmonary Diseases, F.02.333, University Hospital Utrecht, Heidelberglaan 100, NL 3508 GA Utrecht, The Netherlands. E-mail: L.Koenderman{at}hli.azu.nl
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. This induction was time- and
concentration-dependent and optimal at concentrations that are
sufficient for priming functional responses in neutrophils: GM-CSF (10
pM) and TNF- (100 IU/ml). PMNs, isolated from the peripheral blood
of chronic obstructive pulmonary disease (COPD) patients with a
clinical exacerbation, exhibited a partial in vivo primed
phenotype. These antibodies promise to be an ideal tool to monitor
disease activity in whole blood of patients with inflammatory
diseases.
Key Words: priming • detection • antibodies • neutrophils • monocytes
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Neutrophil activation is a multistep process. It is widely accepted now that optimal activation of neutrophils by semiphysiological activators requires priming by chemotaxins or cytokines. This priming of functional responses in vitro has been described in detail during the last decade [6 ]. Priming of granulocytes in vivo has been made plausible for eosinophils isolated from the blood of patients with allergic diseases [7 , 8 ]. Despite the recognition of the importance of priming for neutrophil responses, relatively little is known about the intracellular signals responsible for this process.
Priming is also poorly defined in the context of expression of cell-surface markers. Some studies describe upregulation of Mac-1 and CD66b, and down-regulation of L-selectin in response to cytokines and chemotaxins [9 , 10 ]. The interpretation of these studies is hampered by the fact that priming and activation are poorly defined. Low doses of cytokines (picomolar range) do not cause activation of the respiratory burst and degranulation of azurophil and specific granules. Instead, they preactivate or prime these responses in the context of formyl peptides [11 ] and opsonized particles [12 ]. Another important drawback for the study of neutrophil priming in vitro in the context of cell-surface marker expression is the induction of marker expression caused by isolation artifacts [13 ].
Here we describe the isolation of two phage antibodies that recognize
cytokine-primed neutrophils. These antibodies enable us to monitor
priming in whole blood with very low and priming doses of cytokines
such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and
tumor necrosis factor (TNF)-
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Monoclonal antibodies and cytokines
Polyclonal sheep anti-M13 (Pharmacia) and polyclonal
polycoerythrin (PE)-labeled donkey anti-sheep (Jackson ImmunoResearch,
West Grove, PA) were used to visualize phage antibodies in the
flowcytometric assays. GM-CSF was purchased from Genzyme (Cambridge,
MA), TNF- was bought from Boehringer Mannheim, and interleukin-5
(IL-5) was a kind gift from Dr. M. McKinnon (Cell Biology Unit,
GlaxoWellcome, Stevanage, UK). Monoclonal antibodies (mAbs) directed
against MAC1 (CD11b, clone 44a) and CR1 (CD35, clone 543) were isolated
from the supernatants of hybridomas obtained from American Type Culture
Collection (Rockville, MD).
Chronic obstructive pulmonary disease (COPD) subjects
Nine patients with unstable, moderate-to-severe COPD from the
clinic of the Heart Lung Center (Utrecht, The Netherlands) were used as
subjects. Inclusion criterion for entry was a clinical diagnosis of
COPD according to the European Respiratory Society standard consensus.
Furthermore, the patients had to have a smoking history of at least 10
pack years, an FEV1 < 70% predicted, an
FEV1/FVC < 70%, and a reversibility of < 10%
of the predicted value in stable conditions. All patients were treated
during a hospital admission for a moderate-to-severe clinical
exacerbation with two or all three of the following cardinal symptoms:
1) an increase from baseline of sputum production, 2) sputum purulence
or shortness of breath, or 3) based on the judgment of the clinician.
Patients were allowed to use glucocorticosteroids and bronchodilators
at admission. Patients with uncontrolled, severe disease other than
COPD contributing to the deterioration were excluded. During
hospitalization, subjects were treated optimally according to the
recommendations of the ERS consensus on COPD [14
], and
smoking was forbidden during the time of admission. The study was
approved by an institutional review board.
Cell isolations
Blood was obtained from healthy donors from the Red Cross
Bloodbank (Utrecht, The Netherlands) or from healthy, nonallergic
donors from the laboratory staff. Unprimed leukocytes were isolated
from heparinized blood as follows: After venapuncture, the blood was
directly cooled to 0°C, and cells were kept on ice. Erythrocytes were
lysed via ice-cold lysis with NH4Cl, as previously
described [10
]. Remaining leukocytes were washed with
ice-cold phosphate-buffered saline (PBS), supplemented with HSA (0.5%,
wt/vol.). Total leukocyte preparations were used for the isolation of
phage antibodies and in flowcytometric analysis. Lymphocytes and
neutrophils were identified according their specific side-scatter and
forward-scatter signals [10
].
Phage antibody library
The semisynthetic phage antibody display library of human scFv
antibody fragments has been described in detail elsewhere
[15
]. Briefly, 49 germline VH genes were fused to
semirandomized, synthetic, heavy-chain CDR3 regions, varying in length
between 6 and 15 amino acid residues. The resulting products were
inserted into phagemid vectors, containing seven different light chains
of 
and 
subclasses, resulting in a library of 3.6 x
108 MoPhabs.
Strategy for the isolation of phage antibodies directed against
primed neutrophils
Isolation of phages directed against primed granulocytes was
performed as follows. In short, the phage library (
1011
phage particles) was precleared with resting/unprimed leukocytes from a
nonallergic healthy donor (70x106 cells in 10 ml PBS in
the presence of 1% milk) during 90 min at 4°C on a rotating wheel to
deplete the library from all phages recognizing epitopes present on
unprimed cells. Subsequently, the precleared library was mixed with
GM-CSF-primed eosinophils (20x106 cells in 10 ml PBS/1%
milk) for 90 min at 4°C on a rotating wheel. We used eosinophil
rather than neutrophils, because these latter cells are very sensitive
for a specific priming caused by isolation artifacts. By using
eosinophils, a better chance was foreseen for obtaining antibodies
directed against cytokine-induced priming epitopes. The cell-associated
phages were isolated from nonbinding phages via two wash steps and a
subsequent centrifugation over isotonic Percoll (d 1.030 g/ml, during
20 min, 1000 g at 4°C). Phages were eluted from the cells
by incubation in 76 mM citric acid, pH 2.5, during 5 min at room
temperature. This whole procedure was performed three times.
Subsequently, the positive phages were expanded, essentially as
described before, and screened for epitopes on primed cells, which are
absent on resting/unprimed cells.
Procedure for staining neutrophils with phages A17 and A27
Blood was collected immediately after venapuncture, kept on
37°C, and immediately treated with buffer (control) or with different
amounts of cytokines during different periods of time, as indicated in
the figures. Hereafter, the blood was chilled to 4°C, and red cells
were lysed in ice-cold isotonic NH4Cl [10
].
Then, the cells were washed twice and resuspended in incubation medium.
For staining of leukocytes, 25 µl of MoPhab was blocked by adding 75
µl of PBS/4% milk powder for 15 min on ice. Leukocytes
(5x105) in 100 µl of PBS/HSA (1%, wt/vol) were added
and incubated on ice for 90 min. The cells were washed twice in
ice-cold PBS/1% HSA. To detect cell-bound phages, the cells were
incubated in 50 µl of a 1:200 dilution of sheep anti-M13 polyclonal
antibody (Pharmacia) for 45 min on ice, washed twice, and incubated in
50 µl of a PE-labeled donkey anti-sheep polyclonal antibody solution
(20 µg/ml) (Jackson ImmunoResearch) for 20 min on ice. After a final
wash step, the cells were analyzed in a FACSvantage Flowcytometer
(Becton & Dickinson, Mountain View, CA). Neutrophils were identified
according to their specific side-scatter and forward-scatter signals
[10
].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (48K): [in a new window] |
Figure 1. Histograms describing the expression of the epitopes on neutrophils
(N), monocytes (M), and lymphocytes (L) recognized by the MoPhab A17.
Whole blood was incubated with buffer (control) or TNF- (100 IU/ml)
during 30 min at 37°C. Hereafter, the red cells were lysed in
ice-cold isotonic NH4Cl. Subsequently, the white blood
cells were washed, stained with the phage antibody A17, and analyzed by
flow cytometry. Neutrophils (N), monocytes (M), and lymphocytes (L)
were identified according their forward-scatter and side-scatter
characteristics. As control, the TNF- signal of an irrelevant MoPhab
is shown in the neutrophil panel N (striped line). The histograms shown
are representative of 24 experiments.
|
 View larger version (11K): [in a new window] |
Figure 2. The effect of priming by cytokines on the expression of epitopes
on neutrophils recognized by the phage antibodies A27 and A17. Whole
blood was treated with different cytokines (100 pM IL-5, 10 pM GM-CSF,
and 100 IU/ml TNF-), and leukocytes were stained with A27 (A) and
A17 (B), as described in the legend of Figure 1
. The median values of
24 different experiments were expressed as means ±
SE. *Values differ significantly from the control (37°C)
value (P<0.001).
|
 View larger version (14K): [in a new window] |
Figure 3. Dose-response curves of GM-CSF-induced expression of epitopes expressed
on neutrophils recognized by MoPhabs A27 and A17. Whole blood was
treated with different amounts of GM-CSF for 30 min at 37°C.
Hereafter, red cells were lysed, and the leukocytes were stained with
MoPhabs A27 (A) and A17 (B). Data are expressed as means ±
SE of eight different experiments.
|
 View larger version (13K): [in a new window] |
Figure 4. Dose-response curves of TNF--induced expression of epitopes
expressed on neutrophils recognized by MoPhabs A27 and A17. Whole blood
was treated with different amounts of TNF- for 30 min at 37°C.
Hereafter, red cells were lysed, and the leukocytes were stained with
MoPhabs A27 (A) or phage A17 (B). Data are expressed as means ±
SE of eight different experiments.
|
 View larger version (20K): [in a new window] |
Figure 5. Time courses of TNF-- (100 IU/ml) and GM-CSF- (10 pM) induced
expression of epitopes recognized by MoPhabs A27 and A17. Whole blood
was treated with buffer (A, B), GM-CSF (C, D), or TNF- (E, F) for
different time periods at 37°C. Hereafter, red cells were lysed, and
the leukocytes were stained with MoPhabs A17 (A, C, E) and A27 (B, D,
F). Data are expressed as means ± SE of eight
different experiments.
|
 View larger version (11K): [in a new window] |
Figure 6. Comparison of kinetics of cytokine/chemotaxin-induced expression of
priming epitopes recognized by the phage antibodies A17 and A27, and
complement receptor 1 (CR1, CD35, Mab 543) and Mac-1 (CD11b, Mab 44A).
Whole blood was treated with TNF- (100 IU/ml), GM-CSF (10 pM), and
fMLP (1 µM) at 37°C for the indicated timepoints. Hereafter, the
samples were immediately chilled on ice, red cells were lysed, and the
leukocytes were stained with MoPhab A17 (solid diamonds), MoPhab A27
(open triangles), CD11b (open circles), and CD35 (solid triangles).
Data are expressed as percentage of maximum expression to allow proper
comparison of the kinetics. Maximally induced values were 1823 ±
254, 548 ± 16, 53 ± 4, and 270 ± 45 for A17, A27,
CD11b, and CD35, respectively. Data are expressed as means ±
SE of three independent experiments.
|
 View larger version (18K): [in a new window] |
Figure 7. In vivo priming of neutrophils in the peripheral blood of
patients with a clinical exacerbation of COPD. Whole blood was drawn
from normal volunteers or patients with COPD, and part of the blood was
immediately chilled on melting ice. Hereafter, the cells were analyzed
with MoPhabs A17 (A) and A27 (B), as described in the legend of Figure 1
. Data are expressed as means ± SEM of 20 normal controls
compared with 11 patients with a clinical exacerbation of COPD.
*<0.001 compared with control value (Student’s t-test for
independent samples)
|
(100 IU/ml; n=20). The control incubation of the
blood at 37°C (30 min) did not induce a significant change in
expression of both epitopes when compared with the control immediately
chilled upon venapuncture. Cytokine-induced expression of the epitope
recognized by MoPhab A27 was always higher on resting neutrophils
compared with the epitope recognized by MoPhab A17.
Characterization of neutrophil priming by GM-CSF and TNF- in
the context of expression of the priming epitopes recognized by the
MoPhabs A17 and A27
Figures 3
and 4
show dose response curves of
the expression of priming epitopes induced by GM-CSF (Fig. 3)
and TNF- (Fig. 4) . Whole blood was treated with different amounts of the cytokines at
37°C. After 30 min, the blood was chilled, erythrocytes were lysed,
and the total leukocyte preparation was treated with the MoPhabs A17
and A27. TNF- priming of neutrophils was optimal at 10 IU/ml and 100
IU/ml for MoPhabs A27 and A17, respectively. GM-CSF priming of
neutrophils was optimal at 100 pM and 1 nM for phages A17 and A27,
respectively.
Figure 5
shows a time course of GM-CSF (100 pM, C, D) and TNF- (100 IU/ml, E,
F) induced induction of the priming epitopes on neutrophils recognized
by A17 and A27. TNF--induced expression of the priming epitopes
expressed on neutrophils was optimal after 30 min and remained high up
to at least 60 min. Similar kinetics was found for GM-CSF-treated
leukocytes.
The next series of experiments was designed to compare the
kinetics with other granule-associated markers. As can be seen in
Figure 6
, cytokine-induced (GM-CSF and TNF-) and chemotaxin-induced (fMLP)
expression of the epitopes recognized by A17 and A27 coincide with the
expression of MAC-1 (CD11b/CD18) and CR1 (CD35), albeit with a superior
dynamic range.
MoPhabs A17 and A27 recognize primed neutrophils in the blood of
COPD patients with a clinical exacerbation
The next series of experiments was set out to evaluate whether
MoPhabs A17 and A27 were able to detect neutrophil priming in
vivo in patients with a neutrophil-driven inflammatory process. As
is shown in Figure 7
, neutrophils isolated from patients suffering from COPD with an acute
clinical exacerbation exhibited a primed phenotype. A significant
difference in binding MoPhabs A17 and A27 to these cells was observed
compared with cells from healthy controls. These data demonstrate that
the neutrophil compartment of these patients exhibits a phenotype
comparable with cells that have interacted with cytokines.
After activation of whole blood with fMLP, a clear, increased binding of A17 and A27 was observed in cells of healthy control subjects and COPD patients (data not shown). These data are consistent with a partial-primed phenotype of the cells obtained from COPD patients. Moreover, these data clearly demonstrate that the MoPhabs A17 and A27 can be used to monitor priming in vivo in patients with a neutrophil-driven inflammatory disease such as COPD.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
m chain of MAC-1 and the
glycoproteins recognized by mAbs clustered in CD63 and CD66b. Moreover,
they studied the shedding of L-selectin, which is a sensitive marker
for neutrophil activation. Unfortunately, these markers turned out to
be poor monitors for priming processes in whole blood induced by low
amounts of cytokine [10, 17, and unpublished results]. Likely, this
is caused by the fact that these processes are, in part, linked to the
degranulation of specific granules in neutrophils. CD66b is present in
specific granules, and fusion of these granules with the plasma
membrane caused upregulation of the protein. So far, no evidence has
emerged to support the fact that degranulation of azurophilic or
specific granules occurs in peripheral blood of patients with chronic
inflammatory pulmonary diseases.
We set out experiments to develop mAbs for priming epitopes from a
synthetic phage-antibody library for various reasons. First, until now,
no mAbs have been described that recognize primed neutrophils with an
appreciable dynamic range in vitro (i.e., large difference
in expression between nonprimed and primed cells). Second, the putative
priming epitope might not be very immunogenic, which precludes
development of useful m
Abs. The synthetic phage library circumvents
this latter problem because of the partial randomization of the
complementarity-determining region (CDR) III. Finally, the speed of the
synthetic-phage technology accelerates the development of new
antibodies dramatically. Indeed, we were able to isolate two different
phage antibodies that specifically recognize primed cells. The
characteristics of these antibodies are superior to the pattern of
recognition by established activation markers for neutrophils such as
MAC1, CD63, CD66b, and ICAM-1 (unpublished results).
The identity of the epitopes remains to be elucidated. Unfortunately, single-chain Fv fragments of phages A17 and A27 turned out be very poor blotting antibodies and do not immunoprecipitate a protein from membrane preparations. This precludes straightforward-expression cloning of this epitope. Experiments to monitor priming of granulocytes in whole blood performed with these scFv preparations lead to similar conclusions as the experiments performed with MoPhabs (D. Kanters, unpublished results).
The observation that phagoyctes (neutrophils and monocytes), in marked
contrast to lymphocytes, express the priming epitopes makes it tempting
to speculate on a granular localization of the epitope. As is mentioned
above, it is not very likely that the epitope is only present in
azurophilic and/or specific granules. A third compartment consists of
the small, secretory vesicles localized in the vicinity of the plasma
membrane, as described by Borregaard and coworkers [18
].
This compartment also contains elements of the NADPH oxidase and part
of the MAC-1 and CR1 molecules [18
]. These vesicles are
thought to associate with the plasma membrane in response to cytokines
and heat shock [for a review, see ref. 19
]. To explore this
hypothesis further, we studied the kinetics of
cytokine/chemotaxin-induced expression of the epitopes recognized by
A17 and A27 in comparison with the expression profile of CR1 and
MAC1/CR3. As can be seen from Figure 6
, the kinetics of expression of
all these markers is remarkably similar. Also, heat shock induces the
expression of these markers (D. Kanters, unpublished results). These
data further support the hypothesis that these markers are present
within a similar localization in the neutrophil. The expression range
of the phage antibodies is 100 times better compared with the CD35
and CD11b antibodies. Detailed immunohistochemistry with scFv fragments
will elucidate the precise localization of the epitopes within the
cell.
Antibodies that recognize neutrophil-priming epitopes may be clinically relevant, because they allow the monitoring of cytokine/chemokine action in the peripheral blood. Priming occurs in vivo in the peripheral blood by proinflammatory cytokines as a first step to recruit granulocytes to the side of inflammation.
Several lines of evidence suggest that priming precedes extravasation and activation of granulocytes in the tissues. Integrins need priming signals for upregulation of their functions [for recent reviews, see refs. 20 and 21]. For eosinophils, indeed it has been made plausible that these cells are primed in the peripheral blood of allergic patients in the context of migration and activation of the respiratory burst before actual activation [7 , 8 ]. The occurrence of primed granulocytes in peripheral blood in patients with different diseases associated with inflammation might predict a possible worsening of the clinical condition. We set out a proof-of-concept study to evaluate the possibility that in vivo priming of neutrophils occurs in vivo in COPD, because of the indications that a neutrophil-mediated inflammatory process contributes to the pathogenesis of this disease [22, and refs. therein]. Our experiments with the cells from COPD patients show a clear, in vivo, partially primed phenotype during a severe clinical exacerbation. This supports the concept that circulating cytokines prime inflammatory cells in the peripheral blood in inflammatory lung diseases. It is tempting to speculate now that priming in vivo is a multistep process and that only the first steps of priming occur in the peripheral blood. Fully primed cells will marginate and extravasate into the tissues and thereby leave the peripheral blood. Flow cytometry with labeled antibodies directed against priming epitopes will contribute to the monitoring of the inflammatory processes in these patients in a rapid and reliable fashion.
Received August 6, 1999; revised February 22, 2000; accepted February 24, 2000.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E-J D Oudijk, E H J Nijhuis, M D Zwank, E A van de Graaf, H J Mager, P J Coffer, J-W J Lammers, and L Koenderman Systemic inflammation in COPD visualised by gene profiling in peripheral blood neutrophils Thorax, July 1, 2005; 60(7): 538 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Galibert, G. S. Diemer, Z. Liu, R. S. Johnson, J. L. Smith, T. Walzer, M. R. Comeau, C. T. Rauch, M. F. Wolfson, R. A. Sorensen, et al. Nectin-like Protein 2 Defines a Subset of T-cell Zone Dendritic Cells and Is a Ligand for Class-I-restricted T-cell-associated Molecule J. Biol. Chem., June 10, 2005; 280(23): 21955 - 21964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E{-}J.D. Oudijk, J{-}W.J. Lammers, and L. Koenderman Systemic inflammation in chronic obstructive pulmonary disease Eur. Respir. J., November 2, 2003; 22(46_suppl): 5s - 13s. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
