Differential Activation of the Endogenous

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Differential Activation of the Endogenous Leukotriene Biosynthesis
by Two Different Preparations of Granulocyte-Macrophage ColonyStimulating Factor in Healthy Volunteers
By C. Denzlinger, W. Tetzloff, H.H. Gerhartz, R. Pokorny, S. Sagebiel, C. Haberl, and W. Wilmanns
Resultsfrom in vitro investigations and recent data obtained
in patients with drug-induced cytopenia or myelodysplasia
suggest that leukotrienes may be involved in mediating
some of the actions of granulocyte-macrophage colonystimulating factor (GM-CSF). Inthe present study, the possible role of leukotrienes was further characterized in 21
healthy individualsto avoid modification of responseto GMCSF by disease-specific variables. The effects of two different preparations of human recombinant GM-CSF, ie,
glycosylated GM-CSF as expressed in a Chinese hamster
ovary carcinoma (CHO) cell line and nonglycosylated GMCSF obtained from Escherichia coli, were compared. GMCSF was administered subcutaneously at a single dose of
0.7 nmol/kg body weight. Pharmacokinetic parametersand
hematopoietic and adverse effects were monitored by blood
analyses or physical examination, respectively. Leukotriene
generation in vivo was evaluated by determination of leukotriene E4 and N-acetyl-leukotriene E4 in urine. After the
injection of GM-CSF from E coli, serum concentrations increased and decreased more rapidly and reached a 2.3-fold
higher maximum compared with GM-CSF from CHO. GMCSF induced a biphasic change in leukocyte counts that
proceeded considerably faster after the E coli preparation
than after GM-CSF from CHO. The urinary leukotriene concentration increased 1.3- to 14-fold or 2.1 to 44-fold after
the administration of GM-CSF from CHO or Ecoli, respectively. Urinary leukotriene concentrations correlated significantly with the maximum of basophil counts and correlated with the occurrence of some adverse reactions, ie,
flu-like symptoms, bone pain, or dyspnoea. Our data confirm
the conception that leukotrienes may play a significant role
in GM-CSF action in vivo. They especially direct attention
to the possible relevanceof leukotrienesto untoward effects
of GM-CSF treatment.
0 1993by The American Society of Hematology.
G
in GM-CSF action. This may be of interest because native
GM-CSF appears in a variety of forms that differ by the
amount of glyc~sylation~.~
and because either glycosylated
or nonglycosylated GM-CSF have been used in previous
The effects of the two GM-CSF preparations
studies.
on the endogenous LT production in association with pharmacokinetic parameters and biologic effects will be presented.
RANULOCYTE-MACROPHAGE colony-stimulating
factor (GM-CSF) is a hematopoietic cytokine that
stimulates proliferation, differentiation, and functional activity of granulocytic and monocytic cell lines or mature cells,
respectively. These properties suggest possible beneficial
effects of GM-CSF in states of myelosuppression and impaired
host defense, and a number of therapeutic expectations have
our unbeen fulfilled in recent clinical s t ~ d i e s . However,
~.~
derstanding of the mechanism of action of GM-CSF is still
incomplete.
The action of GM-CSF involves generation of lipid mediators. From investigations in vitro it is known that the cytokine stimulates leukotriene (LT) biosynthesis in eosinophils
and neutrophkiX2LTs and their inhibitors have further been
shown to modulate CSF-induced clonal growth of granuloThe impact of LTs in
cyte-macrophage progenitor cells.
GM-CSF action is of great interest as LT biosynthesis and
action are accessible to specific pharmacologic intervention.*-”
We demonstrated recently that GM-CSF enhances LT
biosynthesis in vivo in patients suffering from cytopenia induced by cytostatic drugs or from myelodysplastic syndrome.” However, these disorders affect the bone marrow,
the primary target for GM-CSF action. In the resulting
pathophysiologic situations, distinct changes in the activities
of several cytokines must be expected, making it difficult to
interpret the action of exogenously added GM-CSF. Also,
the appearance of adverse effects of the cytokine may be altered or disguised by an underlying disease.
The present study was conducted in healthy volunteers t o
overcome these problems. Two different preparations of human recombinant GM-CSF were compared glycosylated
GM-CSF derived from Chinese hamster ovary carcinoma
(CHO) cells (molecular mass, 14 to 32 Kd) and GM-CSF
from Escherichia coli that is lacking glycosyl residues ( I 4
Kd). Direct comparison of the two GM-CSF preparations
enables estimation of the impact of the carbohydrate moiety
’,’-’
Blood, Vol81, No 8 (April 15). 1993:pp 2007-201 3
-
334~1245
MATERIALS AND METHODS
Volunteers. The study was conducted with 2 I healthy male individuals without any evidence of disease from anamnestic exploration, physical, or laboratory examination. The age range was 20 to
38 years (median, 28 years). The weight range was 58.0 to 87.1 kg
(median, 76.4 kg). Seven volunteers were randomly selected for
treatment with GM-CSF from CHO, 14 individuals were treated with
GM-CSF from E coli.
GM-CSF. Recombinant human GM-CSF from CHO and recombinant human GM-CSF from E coli was provided by Sandoz
Ltd (Basel, Switzerland). Whereas expression of GM-CSF in CHO
cells results in glycosylation of the molecule and molecular masses
of 14 to 32 Kd, GM-CSF from E coli is devoid of glycosyl residues
and has a molecular mass of 14 Kd. Quantitative specifications of
GM-CSF dose or of GM-CSF serum concentrations will be given on
From the Medizinische Klinik III, Klinikum Grosshadem, LudwigMaximilians Universitat, Munchen; the Iphar CRF Institut fur Klinische Pharmakologie, Hohenkirchen-Siegertsbrunn; and the Institute
of Clinical Hematology, Gesellschaji fur Strahlen-und Umweltjorschung, Munchen, Germany.
Submitted June 1, 1992; accepted November 24, 1992.
Supported by the Deutsche Forschungsgemeinschaft (De 397/1-3).
Address reprint requests to C. Denzlinger, MD, Medizinische Klinik
III, Klinikum Grosshadern, 0-8000 Miinchen, Germany.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1993 by The American Society of Hematology.
0006-4971/93/8108-0028$3.00/0
2007
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
DENZLINGER ET AL
2008
a molar basis throughout this report. GM-CSF from CHO or E coli,
respectively, was injected subcutaneously into the deltoid region of
one upper arm at a dose of 0.7 nmol (correspondingto 10 pg pure
protein) per kilogram of body weight.
Sample collection. Venous blood samples were taken twice before
GM-CSF administration for determination of basal values of GMCSF and blood cell counts. After the injection of GM-CSF, blood
sampling was repeated at intervals of 10 minutes for the first 30
minutes. Thereafter, blood samples were collected every hour for 8
hours, followed by 2- to 4-hour intervals for another 8 hours and 8hour intervals for 32 hours. Additional blood samples were collected
at 96 and at 336 hours after GM-CSF administration. Urine was
obtained from spontaneousmicturition before GM-CSF and sampled
0 to 24 hours and 24 to 48 hours after administration ofthe cytokine.
Urine was stored at -30°C until analysis. Urine samples were screened
(Combur9Test;Boehringer, Mannheim, Germany) for pathologic
constituents (leukocytes, erythrocytes,protein, etc). Creatinine concentrations in urine were determined by the J a E reaction.
GM-CSF serum concentrations. GM-CSF serum concentrations
were determined using a two-antibody sandwich enzyme-linked immunoassay developed at the Sandoz Pharma Laboratories (Basel,
Switzerland). The assay system consisted of a murine monoclonal
anti-GM-CSF antibody raised against GM-CSF from CHO and an
anti-GM-CSF antiserum raised by injection of GM-CSF from E coli
in a mountain sheep. GM-CSF from the serum samples was fixed to
solid phase by the monoclonal antibody and quantitated by binding
of the biotinylated sheep antiserum and subsequent addition of alkaline phosphatase conjugated to streptavidin. The lower detection
limit of the assay was at 10 fmol GM-CSF/mL serum both for GMCSF from CHO and for GM-CSF from E coli. No cross-reactivities
were observed using up to 10 ng of granulocyte-CSF (G-CSF), Interleukin-3 (IL-3), IL-4, or IL-6, respectively.
LT analysis. LT analysis was performed by sequential use of
high-performance liquid chromatography (HPLC) and radioimmunoassay (RIA) as recently described in detail," with a few minor
modifications. Briefly, urine was deproteinized by storage in 80%
methanolic solution at -40°C and subsequent centrifugation at
8,0008. HPLC of deproteinized urine samples was performed on a
C18 Hypersil column (4.6 X 250 mm, 5 pm particles; Shandon,
Runkorn, UK) with a CIS precolumn (Waters, Milford, MA). The
mobile phase consisted of methanol, water, acetic acid (65:35:0.1 by
volume), 1 mmol/L EDTA, pH 5.6, adjusted with ammonium hydroxide. The flow rate was 1 mL/min. The RIA was performed in
the fractions coeluting with standard ['HILTS using a rabbit cysteinyl
LT antiserum, kindly donated by Prof B.A. Peskar (Ruhr-UniversitAt,
Bochum, Germany). Recovery of LTs from urine was tested routinely
by adding defined amounts of standard LTE4 to urine samples followed by processing and analysis by HPLC and RIA as described
above. Reproducibility was assayed by replicate determinations in
urine samples." LTs and ['HILTS were from commercial sources
(Paesel, Frankfurt, Germany: Amersham, Buchnghamshire, UK; or
DuPont, Boston, MA, respectively). N-acetyl LTE, and N-acetyl
['HILTE, were synthesized from LTE, and ['HILTE,, as described.16
Statistics. Student's t-test for paired observations was used to
analyze significance of differences between determinationsbefore and
after GM-CSF treatment. The t-test for unpaired observations was
used to analyze significance of differences between values obtained
with GM-CSF from CHO and from E coli. Correlations were estimated by calculating product-moment correlation coefficients and
rank correlation coefficients (Spearman).
Ethical approval. Ethical approval for the GM-CSF treatment
studies was obtained from the Ethics Committee, Munchen, incorporated.
RESULTS
GM-CSF pharmacokinetics and hematologic effects. Before GM-CSF administration, GM-CSF serum concentrations
were below the detection limit of the assay system, ie, below
10 fmol/mL serum. Seven volunteers were treated with a
subcutaneous injection of GM-CSF from CHO at a dose of
0.7 nmol/kg body weight; 14 volunteers were treated in the
same way with GM-CSF from E coli. The increase and the
decrease in GM-CSF serum concentration were more rapid
after GM-CSF from E coli compared with GM-CSF from
CHO (Fig 1). After GM-CSF from CHO, maximum GMCSF serum concentrations were obtained at 7.6 k 2.7 hours
and resulted in 221 k 100 pmol GM-CSF/L serum (mean f
SD, n = 7). After GM-CSF from E coli, maximum GM-CSF
serum concentrations were determined already a t 3.9 -t 1.3
hours and resulted in 514 f 136 pmol GM-CSF/L serum (n
= 14). However, due t o a more sustained increase in GMCSF serum concentration, the area under the serum concentration-time curve obtained after GM-CSF from CHO (3.35
k 1.03 nmol/L X hour, mean f SD, n = 7) was not significantly different from the area determined after GM-CSF from
E coli (3.80 k 0.72 nmol/L X hour, n = 14).
The hematologic effects of GM-CSF were characterized by
biphasic changes in leukocyte counts. During the first hour
after injection of GM-CSF, leukocyte counts declined to about
one-third of the counts before treatment (Fig 2). Values returned to the original level within 3 or 4 hours after GMCSF from E coli or CHO, respectively. Maximum leukocyte
concentrations were determined 11 or 27 hours after the respective GM-CSF preparation. They were about 2.4-fold
higher than pretreatment values. The changes in leukocyte
counts occurred significantly faster after GM-CSF from E
coli. Leukocyte counts remained elevated for a significantly
longer period of time after GM-CSF from CHO (Fig 2).
The biphasic shape of leukocyte concentrations was determined mainly by changes in the number of neutrophils,
which almost disappeared in the first phase after GM-CSF
and which strongly increased in the second phase together
with a shift to the left. However, monocyte and eosinophil
counts also declined initially and markedly increased together
-
2 7001
600-
Y
c
0
500-
% 400-
.-
300C
8
200-
€
?
100-
VI
LL
0-
Tine a f t e r GI1-CSF [ h l
Fig 1. Time course of GM-CSF serum concentrations in volunteers treated subcutaneouslywith 0.7 nmol/kg body weight of GMCSF from (0)CHO (n = 7) or from ( 0 )€coli (n = 14). Mean values
f SD are indicated.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
LEUKOTRIENE PRODUCTION IN GM-CSF TREATMENT
2009
20
...
- 20
1
- 18
- 16
18
16
14
- 14
12
- 12
5
10
- 10
“
8
-8
6
-6
u1
L
c
7
4
Fig 2. Time course of leukocyte counts in volunteers treated subcutaneously with 0.7 nmol/kg
body weight of GM-CSF from (0)
CHO (n = 7) or
from (e) E coli (n = 14). Mean values f SD are
indicated.
J
-
4
-2
2
0’
I
I
I
I
I
I
I
I
I
I
I
I
0
4
8
12
16
20
24
28
32
36
40
44
with basophil counts during the second day after GM-CSF.
Monocytes increased to 328% t 259% (mean k SD, n = 7)
or 243% ? 177% (n = 14) of pretreatment values after GMCSF from CHO or E coli, respectively. Maximum eosinophil
counts were 477% f 177%(n = 7) or 320% f 156% (n = 14)
of pretreatment values; maximum basophil counts were 143
f 83 M/L (n = 7) or 91 f 79 M/L (n = 14) after GM-CSF
from CHO or E coli, respectively.
Efect of GM-CSF on urinary cysteinyl LTs. Before
treatment with GM-CSF, healthy volunteers secreted I . 1 to
28 (mean, 9.9) nmol LTE, plus LTE4NAc/mol creatinine
(Fig 3), as determined by combined use of HPLC and RIA.
LTE4NAc constituted a minor, yet significant component
amounting up to 36% (mean, 13%)of LTE4. The concentrations in urine of LTC, and LTD4 were below the detection
limit of our assay system, ie, below 50 pmol/L. There was a
close relation between the urinary LTE, plus LTE4NAcconcentration corrected for the urinary creatinine concentration
and the arl:-ount of LTE, plus LTE4NAc secreted per day.
On the average, 1 nmol LTE, plus LTE4NAc in urine per
mole creatinine corresponded to 14 (range, 12 to 18) pmol
LTE, plus LTE,NAc secreted per 24 hours. The respective
product moment correlation coefficient ranged from 0.9 1 to
0.99 in the healthy volunteers.
During the first 24 hours after the injection of GM-CSF,
urinary cysteinyl LTs increased 1.3- to 14-fold or 2.1 - to 44fold in volunteers treated with GM-CSF from CHO or E coli,
respectively (Fig 3), resulting in urinary LT concentrations
of 32 f 26 or 161 f 124 nmol LTE4 plus LTE4NAc/mol
creatinine (mean k SD). Enhancement of urinary cysteinyl
LTs was statistically significant (P < .001) with both GMCSF preparations, but GM-CSF from E coli was significantly
(P< .01) more effective in this respect. Twenty-four hours
after administration of GM-CSF from CHO, urinary cysteinyl
LTs returned to pretreatment values (10 -+ 9 nmol LTE, plus
LTE4NAc/mol creatinine, mean f SD), whereas they were
still significantly elevated after GM-CSF from E coli (52 f
35 nmol LTE, plus LTE,NAc/mol creatinine; P < .01). There
were no significantchanges in the proportions of urinary LTE,
and LTE4NAc after GM-CSF administration.
Association of urinary cysteinyl LTs with pharmacokinetic
parameters of GM-CSF. The higher concentrations of urinary cysteinyl LTs obtained after GM-CSF from E coli (Fig
3) were associated with earlier and higher peak serum con-
-0
[//---I-48
96 336
T i m e after GPI-CSF [ h l
centrations of GM-CSF as compared with GM-CSF from
CHO (Fig 1). However, there were no statistically significant
correlations between individual pharmacokinetic parameters
(GM-CSF maximum serum concentration, serum half life,
area under the serum concentration-time curve) and individual changes in urinary cysteinyl LTs in the volunteers
treated with GM-CSF from CHO or E coli, respectively (the
product moment correlation coefficients ranged from -0.2
to 0.3).
Association of urinary cysteinyl LTs with hematologic e$
fects of GM-CSF. Urinary cysteinyl LT concentrations and
their relative changes after GM-CSF administration were
tested for correlation with the cytokine’s effects on total leukocytes, neutrophils, monocytes, eosinophils, basophils,
bands, and youngs. There was a significant positive correlation
between the LT concentrations in the urine sampled during
the first 24 hours after GM-CSF administration and the maximum basophil counts with both the CHO and the E coli
r l
‘T;
10001
f
t
+
w
1 1
0.
J
+
0.5
J
0.1
’
before
0-2th
24-t8h
after Gn-CSF
Fig 3. Urinary cysteinyl LTs in healthy volunteers before, 0 to
24 hours, and 24 to 4%hours after GM-CSF treatment. Analyses
of cysteinyl LTs were performed as described in Materials and
Methods. (0)
Data from individualstreated with GM-CSF from CHO;
(0)data from those treated with GM-CSF from €coli. Solid or dotted
horizontal bars indicate the median values for volunteers treated
with GM-CSF from Ecolior CHO, respectively.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
2010
DENZLINGER ET AL
S
400-
ficients for urinary cysteinyl LTs with maximum leukocyte,
neutrophil, monocyte, or eosinophil counts were 0.15, 0.28,
0.37, and 0.38 after GM-CSF from CHO and 0.42,0.46,0.0,
and 0.4 1 after GM-CSF from E coli, respectively.
Association of urinary cysteinyl LTs with adverse effects
of GM-CSF. Adverse effects of GM-CSF were reported in
17 of the 21 (81%) healthy volunteers. The severity of adverse
effects was WHO grade 1 to 2 with symptoms abating without
pharmacologic intervention. Figure 5 illustrates associations
of adverse effects of GM-CSF with different ranges of urinary
cysteinyl LTs. Low, intermediate, or high LT producers were
classified according to their urinary cysteinyl LTs in the 0to 24-hour fraction after GM-CSF administration. The respective ranges of urinary LT concentrations and of LTs secreted per day are given in Fig 5. The lowest range of urinary
leukotrienes (<30 nmol LTE4 plus LTE4NAc/mol creatinine,
equivalent to <450 pmol LTE4 plus LTE4NAcper day) overlaps with the pretreatment values.
Absence of adverse effects was associated with low urinary
cysteinyl LTs. Flu-like symptoms, including a transient increase in body temperature, nasal congestion and catarrh,
pharyngeal irritation, conjunctival redness and congestion
were observed in 10 of the 21 (48%) volunteers. They were
virtually absent in the low LT producers and increased in
frequency with increasing urinary cysteinyl LTs. Bone pain
was felt in the area of the lumbar and/or thoracic spine and/
or in the sternum. It was characterized as tenderness, as a
feeling of pressure, or as pulsating pain. Eight of the 2 1 (38%)
volunteers complained of bone pain. It was most frequent in
the high LT producers. Dyspnoea was reported as difficulty
of breathing, mainly expiratory. Wheezes and coarse rhonchi
were heard during auscultation. Dyspnoea may thus be classified as an asthmatic response. The two volunteers suffering
from this adverse effect had been treated with GM-CSF from
E coli and had high urinary cysteinyl LT concentrations.
Gastrointestinal symptoms included nausea, vomiting, ab-
0
\
2
300-
C
Y
m
200-
YU
$
100-
W
I-
J
+
0I
~
0
I
100
'
I
'
200
I
300
'
I
400
Ilaximum of basophil counts [Il/ll
Fig 4. Correlationsbetween the LT concentrations in urine sampled 0 to 24 hours after GM-CSF administration and the maximum
of basophil counts. ( 0 )Data from individuals treated with GM-CSF
from E coli; (0)data from those treated with GM-CSF from CHO.
The continuous and the broken lines represent the lineary regression
curves of the data obtained with GM-CSF from E coli or CHO, respectively. Product moment correlation coefficients were 0.70 or
0.85 and rank correlation coefficients were 0.70 or 0.50 for the
data obtained with GM-CSF from Ecolior from CHO, respectively.
GM-CSF preparations. The corresponding product moment
correlation coefficients were 0.85 and 0.70, respectively (Fig
4). Correlations of urinary cysteinyl LTs and maximum basophil counts differed with respect to their slopes in the groups
of volunteers treated with GM-CSF from CHO or E coli. A
similar increase in basophil counts in volunteers treated with
either GM-CSF preparation was paralleled by significantly
higher urinary LTs in the group of individuals treated with
GM-CSF from E coli (Fig 4).
No consistent correlations between urinary cysteinyl LTs
and hematologic parameters other than basophils were found
in the healthy volunteers. Product moment correlation coefVI
z
loo
NO ADVERSE
MI[ REACTIONS
0
IOt
40
20
20
0
$
4
loo
DYSPNOEA
loo
SKIN
INTESTINAL
SYMPTOMS
x
60
MI
2oM
20
0
LL
I:
II: 30
Ill:
REACTIONS
0
LEUKOTRIENES IN URINE
< 30 nmol
-
150
>150
LTE4+LTE4NAc/mol
I'
II
I1
II
< 450 pmol/day (n =
1.e. 450
1900 pmol/day
Le. > 1900 pmol/day (n = 7)
Crea !.e.
'I
'I
'I
-
Fig 5. Frequency of adverse
reactions of GM-CSF correlated
with the endogenous LT production as assessed by determination of LTs in urine. (El]Data
obtained with GM-CSF from
CHO; (w) data obtained with
GM-CSF from €coli. Volunteers
were ranked as low (I),intermediate (11). or high (111) LT producers according to the LT concentration in their 0- to 24-hour
urine fraction after GM-CSF administration. The ranges for the
respective urinary LT concentrations as well as the corresponding absolute amounts of
cysteinyl LTs secreted per day
are given (n = number of individuals).
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
LEUKOTRIENE PRODUCTION IN GM-CSF TREATMENT
domina1 pain, and diarrhea. Thirty-eight percent of the volunteers complained of these symptoms. There was no obvious
relation of gastrointestinal symptoms with urinary cysteinyl
LTs. Skin reactions included erythema, scattered pustulation,
or pustular eruption at the injection site. These symptoms
were observed in 48% of the volunteers. They also do not
seem to be related to increased urinary LTs. Arthralgia in
the olecranon and knee joints combined with pain in neighboring muscles occurred in one volunteer treated with GMCSF from E coli. The LT generation was ranked intermediate
in this individual.
Some volunteers suffered from more than one adverse effect. Combinations of up to four untoward effects were observed without a detectable interrelation. Some participants
had additional complaints of mild to moderate headache,
tiredness, heat sensations, or feeling of cold extremities.
However, the relation of these latter symptoms to GM-CSF
treatment was uncertain.
DISCUSSION
In the present study, we demonstrated that GM-CSF activates the endogenous LT production in healthy volunteers.
Urinary LTE, plus LTE,NAc corrected for urinary creatinine
was used as a measure for the endogenous LT biosynthesis.
Because in some individuals a significant portion of LTE,
may be acetylated to LTE,NAC,~','~
determination of both
LTE, plus LTE,NAc appears to be superior to determination
of LTE, alone for a survey of the generation of these lipid
mediators. LTE4 in urine is by now widely accepted as a
parameter for the endogenous LT prod~ction.",'~-~~
The close
correlation between urinary LT concentrations corrected for
urinary creatinine and the renal LT elimination rate confirms
earlier observation^,'^ suggesting that the renal elimination
of LTs proceeds via glomerular filtration without appreciable
secretion or reabsorption.
Data obtained in vitro suggest that GM-CSF is not a direct
stimulator of LT production, but rather primes leukocytes
for enhanced LT production elicited by succeeding
If this is the case, our results suggest that such stimuli are
constitutively present or concomitantly induced by GM-CSF
administration, not only in cytopenic or myelodysplastic patients," but also in healthy volunteers (Fig 3).
Our results encourage speculations on causal relations between hematologic effects, adverse reactions, and the elevation
of endogenous LTs after GM-CSF administration. Indeed,
LTs may play a role in both aspects of GM-CSF action.
LTs may be involved in the proliferative response to GMCSF, as LTs and LT inhibitors have been shown to modulate
the proliferation of myeloic progenitor cells and malignant
blasts in vitro.1,5-7,26,27
In cytopenic patients, a lineary correlation was found between the increase in total leukocyte
counts and the increase in urinary LTs after GM-CSF administration." The corresponding correlation was not statistically significant in the present study in healthy volunteers
who had normal blood cell counts before treatment with the
cytokine. However, a more rapid increase in leukocyte counts
(Fig 2) was associated with a higher LT production (Fig 3)
in the volunteers treated with GM-CSF from E coli as compared with those treated with GM-CSF from CHO. This may
201 1
be interpreted as an argument for a role of LTs in the proliferative response after GM-CSF administration in vivo.
LTs may also be involved in the GM-CSF-induced stimulation of leukocytes. The initial decrease in leukocyte counts
to about one-third of pretreatment values (Fig 2) is most
likely caused by activation of leukocytes. This effect was found
to be associated with an upregulation of adhesion molecules,
promoting adherence of leukocytes to the blood vessel
LTs induce leukocyte adhesion to endothelial cells via the
same type of adhesion molecules (CD1 1/CD18)28329
and leukocyte adhesion can be antagonized by LT biosynthesis inhibitors as demonstrated in a model of reperfusion injury.3o
It seems possible, therefore, that the GM-CSF-induced initial
decrease in leukocyte counts is mediated by LTs.
Functional activation of leukocytes, including increased
generation of LTs, may also explain some of the adverse
effects caused by GM-CSF. After GM-CSF treatment, a significant positive correlation was found between the urinary
cysteinyl LTs and the basophils (Fig 4), which are known to
be central components in hypersensitivity reactions. A positive relation was also found between enhanced LT production and flu-like symptoms, bone pain, or asthmatic reactions
(Fig 5). Flu-like symptoms associated with enhanced vascular
permeability and inflammatory reactions in mucous membranes might well be explained by known actions of LTs.~',~*
Involvement of LTs in asthmatic responses is widely accepted.8-'o.'8.19~31~32
No obvious correlation was observed between GM-CSF-induced LT production and gastrointestinal
symptoms or skin reactions (Fig 5).
Comparison of the effects of the two different GM-CSF
preparations showed several differences. Structurally, GMCSF from E coli differs from GM-CSF from CHO in the
absence of glycosyl residue^.^,^ Pharmacokinetic consequences
of this difference were a more rapid increase and decrease in
serum concentration and a higher maximum serum concentration of GM-CSF from E coli (Fig 1). The data demonstrated are in line with results derived from a larger number
of healthy volunteers (R. Pokorny et al, unpublished observations). Differences in the pharmacokinetics between glycosylated and nonglycosylated GM-CSF also appear likely
when results from recent studies using either preparation are
~ o m p a r e d . ' ~ Pharmacodynamic
~'~,'~
consequences of the differences between the two GM-CSF preparations were a faster
increase and a faster decrease in leukocyte counts (Fig 2),
higher urinary cysteinyl LTs (Fig 3), and a different pattern
of side effects (Fig 5 ) after administration of GM-CSF from
E coli compared with GM-CSF from CHO. GM-CSF from
CHO caused a number of pathologic skin reactions, whereas
the other side effects (flu-like symptoms, bone pain, dyspnoea,
gastrointestinal symptoms) clearly predominated in the volunteers treated with GM-CSF from E coli (Fig 5).
Our results suggest that differences in pharmacokinetics
between the two GM-CSF preparations may explain some
of the differences in their pharmacodynamics: a more sustained increase in serum GM-CSF after GM-CSF from CHO
(Fig 1) corresponded to a more sustained increase in leukocyte
counts (Fig 2); a higher maximum GM-CSF serum concentration after GM-CSF from E coli (Fig 1) corresponded to a
higher LT production (Fig 3) and to more severe side effects
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
2012
DENZLINGER ET AL
(Fig 5). The GM-CSF dose, route, and velocity of administration are empirical determinants of its t o x i ~ i t y , ~and
, ~ , 'the
~
better tolerated GM-CSF from CHO behaves like a prolonged-release form compared with GM-CSF from E coli
(Fig 1).
GM-CSF-induced side effects are usually mild to moderate.
However, a case of a capillary leak syndrome and a case of
an adult respiratory distress syndrome (ARDS), probably
triggered by GM-CSF, have been presented recently.33s34
Involvement of LTs in these pathophysiologic situations is
likely. Enhanced vascular permeability resulting in macromolecular leakage is a well-characterized feature of LT act i ~ n . ' ' . ~Recent
~
studies on endogenous LT production in
ARDS provide additional evidence for a role of leukotrienes
in this
Increased leukocyte activity that is associated with enhanced LT biosynthesis and adverse effects is clearly undesirable in healthy individuals and may be dangerous in certain
clinical situations. However, a limited increase in LT biosynthesis may be beneficial in immunocompromised patients.
Besides a possible involvement in the proliferative response
to GM-CSF, enhanced LT biosynthesis may be involved in
the upregulation of host defense, including upregulation of
immune functions, enhanced antimicrobial activity, and enhanced antitumor cyto~tasis.~~-~'
The more pronounced increase in the endogenous LT production after GM-CSF from
E coli is, therefore, not necessarily disadvantageous in a therapeutic situation.
Future trials will show whether drugs affecting LT biosynthesis or action can be used to optimize GM-CSF treatment.
ACKNOWLEDGMENT
We are indebted to Prof Dr B.A. Peskar (Bochum, Germany) for
providing the leukotriene antibody. We thank the staffof the Institute
for Clinical Chemistry at the Klinikum Grosshadem (Prof Dr D.
Seidel, director) for performing measurements of urinary creatinine
concentrations in the volunteers included in this study.
REFERENCES
I . DiPersio JF: Colony-stimulating factors: Enhancement of effector cell function. Cancer Surv 9231, 1990
2. Gasson J C Molecular physiology of granulocyte-macrophage
colony-stimulating factor. Blood 77: 1 131, 1991
3. Moore MAS: The clinical use of colony stimulating factors.
Ann Rev lmmunol9:159, 1991
4. Grant SM, Heel R C Recombinant granulocyte-macrophage
colony-stimulating factor (GM-CSF). A review of its pharmacological
properties and prospective role in the management of myelosuppression. Drugs 43516, 1992
5. Ziboh VA, Wong T, Wu M-C. Yunis AA Modulation ofcolony
stimulating factor-induced murine myeloid colony formation by Speptido-lipoxygenase products. Cancer Res 46:600, I986
6 . Miller AM, Weiner RS, Ziboh VA: Evidence for the role of
leukotrienes C4 and D4 as essential intermediates in CSF-stimulated
human myeloid colony formation. Exp Hematol 14:760, 1986
7. Estrov 2,Halperin DS, Coceani F, Freedman MH: Modulation
of human marrow haematopoiesis by leucotrienes in vitro. Br J Haematol69:321, 1988
8. Gardiner PJ: Inhibitors of the biosynthesis or actions of the
leukotrienes. Asthma Rev 2:75, 1989
9. Snyder DW, Fleisch JH: Leukotriene receptor antagonists as
potential therapeutic agents. Ann Rev Pharmacol Toxicol 29: 123,
1989
10. Salmon JA, Garland LG: Leukotriene antagonists and inhibitors of leukotriene biosynthesis as potential therapeutic agents. Prog
Drug Res 37:9, 1991
1 1. Denzlinger C, Kapp A, Grimberg M, Gerhartz HH, Wilmanns
W Enhanced endogenous leukotriene biosynthesis in patients treated
with granulocyte-macrophage colony-stimulating factor. Blood 76:
1765, 1990
12. Cebon JS, Bury RW, Lieschke GJ, Mostyn G: The effects of
dose and route of administration on the pharmacokinetics of granulocyte-macrophagecolony-stimulatingfactor. Eur J Cancer 26: 1064,
1990
13. Schwartz GK, Collins JJ, Galazka A, Nessi PA, Lehrer D,
Baldwin Y, Mandeli J, Holland J F Phase I study of subcutaneouslyadministered bacterially-synthesised recombinant human granulocyte-macrophage colony-stimulating factor. Eur J Cancer 28:470,
1992
14. Cebon JS, Lieschke GJ, Bury RW, Morstyn G The dissociation
of GM-CSF efficacy from toxicity according to route of administration: A pharmacodynamic study. Br J Haematol 80:144, 1992
15. Hovgaard D, Mortensen BT, Shifter S , Nissen NI: Clinical
pharmacokinetic studies of a human haemopoietic growth factor,
GM-CSF. Eur J Clin Invest 22:45, 1992
16. Hagmann W, Denzlinger C, Rapp S , Weckbecker G, Keppler
D Identification of the major endogenous leukotriene metabolite in
the bile of rats as N-acetyl leukotnene E4. Prostaglandins 31:239,
I986
17. Huber M, K;istner S, SchBlmerich J, Gerok W, Keppler D:
Analysis of cysteinyl leukotrienesin human urine: Enhanced excretion
in patients with liver cirrhosis and hepatorenal syndrome. Eur J Clin
Invest 1953, 1989
18. Taylor GW, Taylor I, Black P, Maltby NH, Tumer N, Fuller
RW, Dollery CT: Urinary leukotriene E4 after antigen challenge and
in acute asthma and allergic rhinitis. Lancet 1584, 1989
19. Manning PJ, Rokach J, Malo J-L, Ethier D, Cartier A, Girard
Y, Charleson S, OByme PM: Urinary leukotriene E4 levels during
early and late asthmatic responses. J Allergy Clin Immunol 86:211,
I990
20. Bernard GR, Korley V, Chee P, Swindell B, Ford-Hutchinson
AW, Tagari P Persistent generation of peptido leukotrienes in patients
with the adult respiratory distress syndrome. Am Rev Respir Dis
144:263, 1991
21. Fauler J, Tsikas D, Holch M, Seekamp A, Nerlich ML, Sturm
J, Frolich J C Enhanced urinary excretion of leukotriene E4by patients
with multiple trauma with or without adult respiratory distress syndrome. Clin Sci 80:497, I99 1
22. Westcott JY, Thomas RB, Voelkel NF: Elevated urinary leukotriene E4 excretion in patients with ARDS and severe burns. Prostaglandins Leukotrienes Essent Fatty Acids 43:15 I , 1991
23. Moore KP, Sheron N, Ward P, Taylor GW, Alexander GJM,
Williams R: Leukotriene and prostaglandin production after infusion
of tumour necrosis factor in man. Eicosanoids 4: I 15, 1991
24. Hackshaw KV, Voelkel NF, Thomas RB, Westcott JY: Urine
leukotriene E4 levels are elevated in patients with active lupus erythematosus. J Rheumatol 19:252, 1992
25. Carry M, Korley V, Willerson JT, Weigelt L, Ford-Hutchinson
AW, Tagari P: Increased urinary leukotriene excretion in patients
with cardiac ischemia. In vivo evidence for 5-lipoxygenaseactivation.
Circulation 83230, 1992
26. Tsukada T, Nakashima K, Shirakawa S: Arachidonate 5 4 poxygenase inhibitors show potent antiproliferative effects on human
leukemia cell lines. Biochem Biophys Res Commun 140332, 1986
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
LEUKOTRIENE PRODUCTION IN GM-CSF TREATMENT
27. Snyder DS, Castro R, Desforges JF: Antiproliferative effects
of lipoxygenax inhibitors on malignant human hematopoietic cell
lines. Exp Hematol 17:6, 1989
28. Lindstrom P, Lemer R, Palmblad J, Patarroyo M: Rapid adhesive responses of endothelial cells and of neutrophils induced by
leukotriene B4 are mediated by leucocytic adhesion protein CD IS.
S a n d J Immunol31:737, 1990
29. Goldman G, Welbourn R, Valeri CR, Shepro D, Hechtman
H B Thromboxane A2 induces leukotriene B4 synthesis that in turn
mediates neutrophil diapedesis via CD 18 activation. Microvasc Res
41:367, 1991
30. Lehr HA, Guhlmann A, Nolte D, Keppler D, Messmer K:
Leukotrienes as mediators in ischemia-reperfusion injury in a miIClin Invest 87:2036, 1991
crocirculation model in the hamster. .
31. Brain SD, Williams TJ: Leukotrienes and inflammation.
Pharmacol Ther 4657, 1990
32. Lewis RA, Austen KF, Soberman RJ: Leukotrienes and
other products of the 5-lipoxygenase pathway. Biochemistry and
2013
relation to pathobiology in human diseases. N Engl J Med 323:
645, 1990
33. Emminger W, Emminger-Schmidmeier W, Peters C, Susani
M, Hawliczek R, Hijcker P, Gadner H: Capillary leak syndrome
during low dose granulocyte-macrophage colony-stimulating factor
(rh GM-CSF) treatment of a patient in a continuous febrile state.
Blut 61:219, 1990
34. Verhoef G, Boogaerts M: Treatment with granulocyte-macrophage colony stimulating factor and the adult respiratory distress
syndrome. Am J Hematol 36:285, 1991
35. Demitsu T, Katayama H, Saito-Tab T, Yaoita H, Nakano
M: Phagocytosis and bactericidal action of mouse peritoneal macrophages treated with leukotriene B4. Int J Immunopharmacol 11:
801, 1989
36. Rola-Pleszczynski M: Leukotrienes and the immune system.
J Lipid Mediators 1: 149, 1989
37. Bonta IL, Ben-Efraim S: Interactions between inflammatory
mediators in expression of antitumor cytostatic activity of macrophages. Immunol Lett 25295, 1990
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1993 81: 2007-2013
Differential activation of the endogenous leukotriene biosynthesis by
two different preparations of granulocyte-macrophage
colony-stimulating factor in healthy volunteers
C Denzlinger, W Tetzloff, HH Gerhartz, R Pokorny, S Sagebiel, C Haberl and W Wilmanns
Updated information and services can be found at:
http://www.bloodjournal.org/content/81/8/2007.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.