| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
, Interleukin-6, and Interferon-
) Release from Activated Peripheral Blood Mononuclear Cells or after Minimal to Moderate Surgical Stress1
Division of Clinical Sciences and SCHARR (S.W.), Sheffield University, Sheffield, United Kingdom S5 7AU; and Tromso University (O.K., A.R.), Tromso, Norway
Address all correspondence and requests for reprints to: Dr. Richard J. M. Ross, Clinical Sciences, Northern General Hospital, Sheffield, United Kingdom S5 7AU. E-mail: r.j.ross{at}sheffield.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To confirm the biological activity of GH in our cell culture system we
used a Stat5 functional assay. In this assay GH induced a bell-shaped
curve, with a maximal response at GH levels between 100-1000 ng/mL.
PBMCs from healthy volunteers were incubated with GH in doses from
1–1000 ng/mL for 6–72 h under resting conditions and after activation
with endotoxin and the mixed lymphocyte reaction. Studies were repeated
with PBMCs from six individuals using a GH dose of 100 ng/mL (the level
of GH found after high-dose GH therapy) and an endotoxin dose that gave
a submaximal response (0.01 ng/mL). GH had no effect on cell
proliferation or the production of tumor necrosis factor-
(TNF),
interleukin-6 (IL-6), or interferon-
(IFN). In patients
undergoing laparoscopic cholecystectomy there was a time-related effect
of surgery on cytokine levels. There was a rise in IL-6 and a fall in
TNF at 24 h after surgery; however, high-dose GH therapy had no
effect on the cytokine response. We considered the possibility that
endogenous GH production by PBMCs could influence the cytokine response
in activated PBMCs; however, incubation of PBMCs in the presence of the
GH receptor antagonist, B2036, had no effect on TNF, IL-6, or IFN
production by PBMCs in either the mixed lymphocyte reaction or when
activated by endotoxin.
These results suggest that high-dose GH therapy does not alter the proinflammatory cytokine response to surgery or endotoxin. The results do not exclude an effect of GH on the immune response, but they suggest that the mortality seen in critically ill patients may be due to factors other than immune modulation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The GH receptor (GHR) is a member of the type 1 cytokine receptor family and is expressed on human peripheral blood mononuclear cells (PBMCs) (5). In healthy individuals most circulating B cells and monocytes are GHR positive, and there is low expression in T cells (6). GH-deficient humans show no obvious immune deficit (5); however, GH is locally produced by PBMCs (7), and therefore a pituitary deficiency may not result in immune deficiency (5).
The role of GH in immune regulation of humans is as yet undefined (8),
and studies of GH administration have given varying results.
GH-deficient children had basal tumor necrosis factor- (TNF)
levels similar to controls, but GH administration caused an acute rise
in TNF levels (9). In contrast, in GH-deficient adults, basal TNF
levels were high and fell after prolonged GH administration (10). In
human immunodeficiency virus (HIV) patients treated with GH there was
no significant change in immune function (11), and in surgical patients
treated with GH there was no change in cytokine levels (12); however,
GH treatment of normal human PBMCs reduced the TNF and
interleukin-1ß (IL-1ß) production in response to endotoxin
(13).
The use of GH as an anticatabolic agent and the observation that high-dose GH treatment is associated with an increased mortality due to sepsis in critically ill patients (2) emphasize the importance of investigating the action of GH on the immune system. In this study we asked the specific question of whether high-dose GH treatment promotes proinflammatory cytokine production from resting and activated human PBMCs either in culture or in patients undergoing surgical stress.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Venous blood samples (50 mL) were collected from healthy male adults (aged 20–45 yr). PBMCs were isolated by Lymphoprep, washed twice in normal saline and once in medium, and suspended in medium [RPMI 1640 supplemented with L-glutamine (2 mmol/L), penicillin/streptomycin (10 mg/mL), and 2% heat-inactivated normal human AB serum] to a density of 1 x 106/mL. For the cytokine production assay 2 mL cell suspension were transferred to each well of a 24-well plate, and for the proliferation assay 200 µL were transferred to each well of a 96-well plate. The cells were treated with different concentrations of endotoxin (lipopolysaccharide from Escherichia coli stereotype 0111:B4, Sigma, St. Louis, MO), Phytohemagglutinin (PHA; Sigma), GH (Genotropin, recombinant human GH, Pharmacia Biotech, Uppsala, Sweden), or GH antagonist B2036 (supplied by William Bennett, Sensus Drug Development Corp., Austin, TX). Incubations were performed in a humidified atmosphere containing 5% CO2. After 72-h incubation, media were centrifuged for 10 min. Cells were discarded, and the supernatants were stored at -70 C until assay. Mixed lymphocyte reactions (MLR; allostimulation) were set up using PBMCs from a responder and the same number of allogeneic irradiated (30 Gy) PBMCs from another person as stimulator. For proliferation assays 200 µL of 1 x 106/mL PBMCs were incubated for 5 days in round bottomed microplates, either unstimulated or stimulated with endotoxin, GH, or GH antagonist. For allostimulation, 100 µL PBMC were incubated with 100 µL irradiated PBMC from another person (cell ratio, 1:1). During the last 24 h, 0.1 mCi [3H]thymidine was added, and proliferation assessed from the [3H]thymidine incorporation as counts per min. All materials were screened and were negative for endotoxin.
Assays
Cytokines were measured by enzyme-linked immunosorbent assay
[TNF; R&D Systems, Oxon, UK; interferon- (IFN): Diaclone
Research Kits, Besancon, France; IL-6: R&D Systems]; the intraassay
CVs were 8.7%, 0.59%, and 4.4%, respectively. GH was measured by
immunoradiometric assay (Medix Biochemica, Kauniainen, Finland).
GH treatment in vivo
Nineteen female patients [aged 19–66 yr (4 postmenopause), body mass index, 22–33 kg/m2] underwent elective laparoscopic cholecystectomy. The study design was placebo controlled, randomized, and double blind. Patients received either 13 IU/m2·day rhGH (Genotropin, recombinant human growth hormone, Pharmacia) or placebo (0.9% NaCl) at 0800 h on the day of surgery (before surgery) and then daily for 3 consecutive days. Both groups of patients received standardized total parenteral nutrition containing glutamine. Venous blood samples were collected the day before surgery and on the first, second, and third postoperative days at 1200 h. The protocol was approved by the ethical board of the Clinical Research Center at Tromso University Hospital, and the patients gave informed consent.
Functional assay
Stable clones expressing the full-length human GHR were generated in 293 cells (human kidney embryonal cell line) as previously described (14).Transcription assays were performed in the stable clone 293 cells expressing the GHR and transiently transfected with a reporter construct containing a Stat5-binding element fused to a minimal TK promoter and luciferase. Luciferase activity was measured as previously reported (15).
Statistics
The paired t test was used to compare cytokine levels in media from PBMCs and proliferation of cells. For analysis of serum levels in patients undergoing surgery, repeated measures ANOVA was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)
To confirm the biological activity of our GH preparation in
culture and to determine whether a GH dose response was seen with
physiological levels of GH, we used a previously established functional
assay (15). Essentially 293 cells stably expressing the human GHR and
transiently transfected with a Stat5-luciferase reporter were tested
with GH doses between 1–100,000 ng/mL culture medium (Fig. 1a). There
was a dose response to GH at GH doses between 1–100 ng/mL, with GH
levels of 100-1000 ng/mL giving a maximal response, and there was a
bell-shaped shaped curve, with GH doses greater than 1000 ng/mL
suppressing the Stat5 response.
|
)Before examining the effect of GH on PBMCs activated by endotoxin, we performed a dose response to endotoxin using doses of 0.0001–100 ng/mL. Endotoxin induced production of IL-6 at doses as low as 0.001 ng/mL, and a maximal response was seen with an endotoxin dose of 0.1 ng/mL. Thus, for subsequent experiments a submaximal dose of endotoxin was used (0.01 ng/mL medium).
Effect of GH on resting PBMCs (Figs. 2 and 3)
Resting PBMCs were incubated with and without GH at varying
doses between 1–1000 ng/mL for 3 days, and the medium was then assayed
for cytokines (Fig. 2a). Medium from cells incubated without GH showed
very low levels of TNF, IL-6, and IFN, and there was no induction
of the cytokines with GH doses up to 1000 ng/mL. Endotoxin (0.01 ng/mL)
and PHA (5 µg/mL) induced TNF (20- and 292-fold), IL-6 (36- and
807-fold), and IFN (1.2- and 762-fold). Medium from cells incubated
with GH was also sampled at 3, 6, 12, 24, 48, 72, and 96 h after
the start of the GH incubation, and no change in cytokine level was
found (data not shown). Similarly, there was no change in cell
proliferation in response to increasing doses of GH, although a clear
effect was seen in response to endotoxin and PHA (Fig. 2b). PBMCs from
six separate individuals were then incubated with a GH dose (100 ng/mL)
that we had previously shown to have a maximal effect on GH activation
of Stat5 and was similar to the GH level found in surgical patients
after GH treatment (see below). Again, GH had no effect on production
of TNF, IL-6, or IFN (Fig. 3a).
|
|
)
Using a dose of endotoxin that induced a submaximal IL-6 response
(0.01 ng/mL) and a dose of GH known to induce a maximal response (100
ng/mL), PBMCs from six separate individuals were incubated with and
without GH and endotoxin, alone or in combination. GH had no effect on
the production of TNF, IL-6, and IFN in either the absence or
presence of endotoxin.
Effect of GH on PBMCs activated by MLR (Fig. 3b)
PBMCs were activated by coculture with MHC-incompatible
(allogeneic) stimulator cells (MLR) and were incubated with and without
GH. The effects of GH were studied in experiments lasting for 1–5 days
in six separate individuals with doses of GH ranging from 1–1000
ng/mL. GH had no effect on the MLR induction of TNF, IL-6, and
IFN. MLR augmented PBMC proliferation, but this was not altered by
GH (PBMCs only, 923 ± 178; MLR only, 3867 ± 372; MLR plus
GH, 3594 ± 532 cpm).
The GH antagonist, B2036 (Fig. 4)
In case endogenous production of GH was masking any GH action, experiments were repeated with a dose of GH antagonist (B2036, 1000 ng/mL) known to completely block receptor signaling (14). The GH antagonist had no effect on cytokine production in either resting PBMCs or cells activated by incubation with endotoxin.
|
)
Data was collected over 4 days for the 19 patients (10 GH and 9
placebo) who underwent laparascopic cholecystectomy. There was no
effect of GH on any of the cytokine levels despite GH levels of
116 ± 22 ng/mL in the GH group the day after surgery and 1.7
± 0.4 ng/mL in the placebo group. However, there was statistical
evidence of a time effect after surgery for all cytokines (IL-6,
P < 0.001; TNF, P < 0.001). The
IL-6 and TNF responses followed a quadratic shape.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, IL-6, and IFN. Similarly, high-dose GH
treatment of patients undergoing surgical stress had no effect on the
surgically induced changes in TNF and IL-6. In these studies we were
very careful in our choice of GH dose and controls. In simultaneous
experiments we showed that GH, at the doses used, was biologically
active in a functional assay of Stat5 induction, and that a GH level of
100 ng/mL had a maximal effect on Stat5 induction (the dose response
was a bell-shaped curve, as it is now well recognized that high-doses
of GH block receptor signaling) (16). All of the PBMC incubations had
controls with endotoxin or PHA. PBMCs were very sensitive to incubation
with endotoxin, and a level of 0.01 ng/mL induced a clear response of
cell proliferation and cytokine production. The doses of GH used in the
PBMC incubations were chosen to reflect those found in patients treated
with high-dose GH. In the surgical patients a GH level of 116 ±
22 ng/mL (mean ± SEM) was found after GH
treatment with 13 IU/m2·day GH. This dose of GH
is 10–20 times the normal dose for GH replacement in adults and is
similar to the dose of GH used in many studies of its anticatabolic
therapy (17), including the recent study in critically ill patients
(2). Our results, therefore, suggest that high-dose GH therapy is
unlikely to induce PBMC proliferation or alter the production of
TNF, IL-6, or IFN. We chose to study these cytokines because they
are proinflammatory, they modulate the activation of macrophages, and
they are implicated in inducing septic shock and multiple organ failure
(18).
The changes in cytokine levels following surgery are determined by a
number of factors, including the type of surgery and anesthetic. For
example, different changes are seen in patients undergoing open
vs. laparoscopic cholecystectomy (19). We chose to study
patients undergoing laparoscopic cholecystectomy because these patients
represent a relatively homogenous group with, therefore, a more
reproducible physiological response in whom we would be likely to
detect any influence of GH treatment. The surgery had a relatively
minor effect on cytokine production, inducing a rise in IL-6 levels at
24 h and a fall in TNF at 24 h. These results are broadly
similar to previous observations in this patient group (19, 20),
although in one study TNF levels were higher at 24 h (21). The
finding that GH had no effect on the cytokine response to surgery is
consistent with results in another patient group. Patients undergoing
abdominal aortic aneurysm repair were treated with high-dose GH or
placebo for 6 days before surgery, and there was no difference in their
IL-6 rise after surgery (12). Body mass index and estrogen status may
affect cytokine levels; however, as cytokine measurements were made
within the same patient before and after surgery, these variables would
be unlikely to have influenced our results.
Our results do not preclude the possibility that high-dose GH has a detrimental effect on another aspect of the immune system that we have not studied and that could be implicated in the increase in mortality seen in critically ill patients (2). In addition, critically ill patients may well have other confounding factors, such as sepsis or drugs, which may interact with GH and the immune system. Previous reports, however, have generally been consistent with GH having a beneficial effect on the immune response. In patients with moderately advanced HIV infection, high-dose GH therapy was reasonably well tolerated, and there was a modest improvement in HIV-specific immune function (11). In patients undergoing cholecystectomy, high-dose GH treatment was associated with improved cell-mediated immunity and a reduced incidence of postoperative wound infection (22), although in this study the incidence of infection was surprisingly high in the placebo-treated group. Consistent with GH not influencing the immune response was the lack of effect GH had on septic episodes in children with burns (23) or on sepsis score in patients with sepsis (24). Overall, our results and the above studies suggest that the increase in mortality seen in septic animals (3, 4) and critically ill humans (2) treated with GH is not related to an effect of GH on the inflammatory response. It seems likely that the detrimental effect of GH in the presence of sepsis may well be associated with changes in the metabolic response (4).
The effects of GH on immune function in animal models of inflammation
and in vitro cultures of mononuclear cells have been varied.
This probably relates to experimental conditions, as in a number of
studies doses of GH and endotoxin were used that greatly exceeded those
found in either normal subjects or septic patients, in whom high-dose
GH treatment has been used. GH has been shown to prime macrophages (25)
and enhance the production of TNF in response to endotoxin (26), but
in calves GH reduced the TNF response to endotoxin (27). GH enhanced
cytokine responses and improved the survival of septic mice (28), in
contrast to the previously discussed increase in mortality seen in
other animal models (4). In studies of splenic lymphocytes GH at an
extremely high-dose (10,000 ng/mL) increased cell proliferation and
production of IFN in mice with burn injury (29), and GH at a very
low dose (0.1 ng/mL) reduced the IFN response to treatment with
staphylococcal A toxin (30), but increased the response to a very
high-dose of lipopolysaccharide (500 ng/mL) (31). In human PBMCs a very
high-dose of GH (500 ng/mL) inhibited the production of TNF in
response to a very high-dose of endotoxin (10 ng/mL). In our studies we
used GH concentrations (100 ng/mL) that are encountered in patients
receiving high-dose GH treatment and levels of endotoxin that induced a
submaximal response. Under these conditions GH had no effect on
cytokine production or cell proliferation. The studies were also
repeated using another condition of priming and activating PBMCs, the
MLR; again, under these conditions there was no effect of GH.
In view of the observation that normal replacement doses of GH in
GH-deficient children (9) increased TNF and decreased TNF levels
in adults (10), we considered the possibility that low levels of
endogenous GH produced by PBMCs (7) could alter cytokine responses.
To test this hypothesis we used the recently synthesized GHR
antagonist, B2036. This GH mutant blocks GH-stimulated cell
proliferation (16, 32, 33, 34, 35) and GH signaling (14). At doses that
completely block signaling (14), the antagonist had no effect on either
basal or activated PBMC cytokine production or proliferation.
In conclusion, we examined whether high-dose GH therapy induces proinflammatory cytokines and PBMC proliferation. In cell culture, under conditions that simulate high-dose GH therapy, GH had no effect on cytokine production of resting or activated PBMC. Similarly, GH had no effect on the cytokine response to surgery. These studies do not exclude the possibility that the adverse effect of GH in critically ill patients is due to a change in the immune response, but suggest that this detrimental action of GH is not due to activation of proinflammatory cytokines.
|
Footnotes |
|---|
Received January 8, 2000.
Revised May 12, 2000.
Accepted June 7, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and interleukin-1ß. Eur J Endocrinol. 138:640–643.[Abstract]
, cortisol, and thromboxane-B2 responses to endotoxin
in vivo. Endocrinology. 134:1082–1088. and
IFN- release by murine splenocytes activated by LPS or porins of
Salmonella typhimurium. J Med Microbiol. 45:40–47.This article has been cited by other articles:
 |
|
 |
|
  T. K. Hansen, S. Fisker, R. Dall, T. Ledet, J. O. L. Jorgensen, and L. M. Rasmussen Growth Hormone Increases Vascular Cell Adhesion Molecule 1 Expression: in Vivo and in Vitro Evidence J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 909 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Mesotten, G. Van den Berghe, C. Liddle, S. Coulter, F. McDougall, R. C. Baxter, and P. J. D. Delhanty Growth Hormone Modulation of the Rat Hepatic Bile Transporter System in Endotoxin-Induced Cholestasis Endocrinology, September 1, 2003; 144(9): 4008 - 4017. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Zarkesh-Esfahani, G. Pockley, R. A. Metcalfe, M. Bidlingmaier, Z. Wu, A. Ajami, A. P. Weetman, C. J. Strasburger, and R. J. M. Ross High-Dose Leptin Activates Human Leukocytes Via Receptor Expression on Monocytes J. Immunol., October 15, 2001; 167(8): 4593 - 4599. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
