| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
Department of Pediatrics, Division of Endocrinology, University of North Carolina, Chapel Hill, North Carolina 27599
Address all correspondence and requests for reprints to: Dr. L. Sävendahl, Department of Pediatrics, Umea University, S-901 85 Umea, Sweden. E-mail: lars.savendahl{at}histocel.umu.se
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Insulin-like growth factor I (IGF-I), which is known to exert a wide
array of biological effects on a variety of cell types, is suppressed
by acute or chronic protein-calorie deprivation (4). IGF type 1
receptors are present on monocytes, natural killer (NK) cells, and T
helper cells (5), and IGF is believed to be essential for T cell
proliferation (6, 7). Therefore, IGF-I might exert effects on the
immune system. Several cytokines, including interleukin-1ß (IL-1ß)
and IL-2, are important for the immune system to work properly (8).
Tumor necrosis factor-
(TNF
) is increased in malnourished
patients (2, 9), but the extent that other cytokines are influenced by
nutritional status is not clear. Although quantification of serum
concentrations of cytokines is difficult, the measurement of cytokines
produced by peripheral blood mononuclear cells (PBMCs) in
vitro can be used to overcome these problems and has been proposed
to reflect cytokine production in vivo (2).
Our working hypothesis is that nutritional deprivation reduces not only
serum concentrations of IGF-I, but also the production of cytokines
important for immune function. To test our hypothesis, normal
volunteers were subjected to 7 days of fasting, and PBMC production of
IL-1ß, IL-2, TNF
, PGE2, IGF-I, and IGF-binding
protein-1 (IGFBP-1), -2, and -3 was studied. Serum concentrations of
IL-1ß, TNF
, IGF-I, and IGFBP-1, -2, and -3 were also followed.
Although this study model may not mimic precisely the effects of
prolonged protein-calorie malnutrition, it allows study of in
vivo/in vitro events related to undernutrition without
interference from various factors that complicate studies of patients
with chronic malnutrition secondary to intercurrent illness.
| Subjects and Methods |
|---|
|
|
|---|
Ten healthy nonsmoking volunteers (four women and six men) who
were taking no medication completed the study (Table 1
).
After obtaining informed consent, subjects were housed in private rooms
at the General Clinical Research Center at the University of North
Carolina, and their activities were strictly supervised. Oral intake,
from 2300 h on day 0 (admission) through day 8 (discharge), was
limited to mineral water (minimum, 2000 mL daily) and one daily tablet
of multivitamins with minerals (Theragran M, Apothecon, Princeton, NJ).
Blood samples were collected between 08000845 h on days 1 (overnight
fast = prefast), 2, 4, 6, and 8 (7 days of fast = postfast)
for hematological, immunological, and biochemical testing. Within
48 h, each subject developed ketonuria and began to lose weight.
The average weight loss between days 18 was 5.5 ± 0.2 kg. All
subjects experienced hunger, which lessened by day 3; two experienced
nausea (days 23); one developed orthostatic hypotension, which
resolved with increased water intake (day 2); one experienced nightly
muscle cramps relieved by walking (day 4; normal electrolytes). Social
and psychological problems and/or hunger caused four female subjects to
withdraw from the study (days 14; data not included). Three
age-matched healthy volunteers (women, 24 and 30 yr, and a men, 27 yr),
weighing 95115% of ideal body weight, were recruited as nonfasting
control subjects.
|
PBMCs were separated over Ficoll-Paque (Pharmacia, NJ) by a
modification of a technique described by van der Meer et al.
(10). The interphase was washed twice (225 x g, 8 min)
in RPMI 1640, and differential counts revealed more than 90%
mononuclear blood cells. PBMCs were diluted with serum-free RPMI 1640
culture medium supplemented with 2 mmol/L L-glutamine, 50
U/mL penicillin, 50 mg/mL streptomycin, and 1% Nutridoma HU
(Boehringer Mannheim Biochemicals, Indianapolis, IN), and 200-µL
samples were aliquoted to each well of a flat-bottom, tissue
culture-treated 96-well cell culture cluster (Costar, Cambridge, MA).
PBMCs were challenged with phytohemagglutinin P (PHA-P; 0.1100
µg/mL; Difco, Detroit, MI) or with endotoxin [lipopolysaccharide
(LPS) extracted from Escherichia coli 055:B5; 33000 pg/mL;
Sigma Chemical Co., St. Louis, MO] at 37 C in 5% CO2.
Both secretagogues gave similar results (LPS results not shown). After
culture, supernatants from 313 replicate wells were pooled and frozen
(-80 C) for determinations of IL-1ß, IL-2, TNF
, PGE2,
IGF-I, and IGFBP-1, -2, and -3. Cell viability was consistently greater
than 99% throughout the culture period when tested by exclusion of
trypan blue (11).
From three nonfasting control subjects, different concentrations of
PBMCs (0.510 x 106 cells) were cultured for 24
h in the presence of PHA-P (10 µg/mL). The stimulation of IL-1ß,
IL-2, TNF
, and PGE2 production was optimal when the
concentration of PBMCs was 2.5 x 106 cells/mL (Fig. 1
). To determine the optimal time of culture, PBMCs
(2.5 x 106 cells/mL) were harvested after 4, 8, 16,
and 24 h and daily on days 27. Peak concentrations of IL-2,
TNF
, and PGE2 were reached after about 24 h of
culture (Fig. 2
). IL-1ß was maximal after 3 days in
culture (Fig. 2
). In all subsequent experiments, we chose to culture
2.5 x 106 PBMCs/mL for 24 h. For each of the
three nonfasting control subjects, PBMC production of IL-1ß, L-2,
TNF
, and PGE2 was very stable when exposed to various
concentrations of LPS or PHA-P 1 week apart [P = 0.99
when comparing dose (1100 µg/mL PHA-P)-response (IL-2)
curves].
|
|
IL-1ß, IL-2, and TNF
were determined in the culture medium
using the Quantikine Immuno Assays System (R&D Systems, Minneapolis,
MN), which uses monoclonal antibodies that are specific and sensitive
to levels of 0.3 pg/mL (IL-1ß), 6.0 pg/mL (IL-2), and 4.4 pg/mL
(TNF
). Serum concentrations of IL-1ß and TNF
(measures free and
bound TNF) were determined by specific high sensitivity immunoassays
(R&D Systems) using monoclonal antibodies sensitive to levels less than
0.05 pg/mL (IL-1ß) or less than 0.18 pg/mL (TNF
). Immunoreactive
IGF-I was measured by a highly specific nonequilibrium RIA (12) after
removal of IGFBPs by C18 cartridge chromatography (Sep-Pak,
Waters Associates, Milford, MA) (13). In this separating procedure,
bound IGF-I is separated from the IGFBPs. Culture medium was
concentrated 10 times before assay of IGF-I. IGFBP-1, -2, and -3 were
measured using sensitive double antibody RIAs (14, 15, 16) with lower
limits of detection of 0.2, 0.08, and 0.5 ng/mL, respectively.
Statistical analysis
Statistical analysis was performed using repeat measure one-way ANOVAs (time-course data), repeat measure two-way ANOVAs (dose-response data; SuperANOVA software from Abacus Concepts, Berkeley, CA), or paired Students t test (comparisons between prefast and postfast serum levels). Five levels of significance were tested: P < 0.05, P < 0.01, P < 0.005, P < 0.001, and P < 0.0001.
| Results |
|---|
|
|
|---|
were similar prefast (0.67 ±
0.20 and 1.92 ± 0.24 pg/mL, respectively) and postfast (0.76
± 0.22 and 2.07 ± 0.29 pg/mL, respectively). Serum PRL decreased
from 20.6 ± 3.6 ng/mL (prefast) to 14.5 ± 3.8 ng/mL
(postfast; P = 0.05). Serum IGF-I declined from
247 ± 29 (prefast) to 87 ± 10 ng/mL (postfast;
P < 0.0001). Serum IGFBP-1 increased from 45 ± 6
ng/mL (prefast) to 248 ± 26 ng/mL (postfast; P <
0.0001), and IGFBP-2 increased from 184 ± 21 ng/mL (prefast) to
303 ± 27 ng/mL (postfast; P < 0.0001). Serum
IGFBP-3 decreased from 3.1 ± 0.1 ng/mL (prefast) to 2.4 ±
0.1 ng/mL (postfast; P < 0.001).
|
when PBMCs were challenged with various concentrations of PHA-P or LPS
(data not shown). Fasting tended to increase PBMC release of
PGE2 when stimulated with low concentrations of PHA-P (0.1,
0.3, and 1 µg/mL) or LPS (3, 10, 30, and 100 pg/mL), but this did not
reach statistical significance (data not shown). When stimulated with
higher concentrations of PHA-P (3, 10, and 100 µg/mL) or LPS (300 and
1000 pg/mL), PGE2 release was not affected by fasting. The
concentrations of IGF-I and IGFBP-1, -2, and -3 were not measurable in
culture medium by RIA.
|
| Discussion |
|---|
|
|
|---|
Fasting increased the release of IL-1ß from stimulated PBMCs.
IL-1ß, mainly produced by macrophages and monocytes, is one of the
key mediators of the immune response to microbial invasion,
immunological reactions, inflammatory responses, and tissue injury (8).
A functional defect in the capacity of T lymphocytes to respond to
exogenous IL-1 has been reported in malnourished rabbits (20).
Therefore, it is possible that the increased production of IL-1ß in
fasting subjects is an adaptive mechanism to counteract a decreased T
cell responsiveness. Malnutrition has been reported to enhance the
production of TNF
in humans (2, 9). This could not be confirmed in
our study, but the lack of such an effect might be explained by the
relatively short period of starvation.
Fasting decreased serum concentrations of IGF-I and PRL, hormones known to stimulate the proliferation of T cells (6, 7, 21). The T cell suppression observed in fasted subjects could, therefore, be secondary to reduced concentrations of IGF-I and PRL in serum. PRL also increases the expression of IL-2 receptors on T helper cells (22), further linking a classical hormone to the cytokine/immune system. The marked increase in serum concentrations of IGFBP-1 and IGFBP-2 during fasting could serve as an adaptive mechanism to increase the bioavailability of IGF-I, thereby preserving T cell proliferation during conditions of nutritional deprivation.
In malnourished patients, the in vitro basal cytotoxicity of NK cells is decreased, but can be stimulated by IL-2 or IGF-I (23, 24, 25). Our results showing that IL-2 and IGF-I are decreased in fasting subjects further supports an important role for IL-2 and IGF-I in maintaining a normal NK cell cytotoxicity. A relatively high density of type I IGF receptors has been detected on NK cells (5, 24), suggesting that the type I receptor on this cell type is functional, and IGF-I could act as a mediator of hormonal action on the immune system.
It seems unlikely that the approximately 3-fold suppression of IL-2 and the decreased IGF-I we observed could cause immunosuppression, as our model undoubtedly does not faithfully mimic chronic starvation. The prolongation of nutrient denial, however, could continue to reduce IL-2 and IGF-I so as to eventually promote immunosuppression. IGF-I and IL-2 are available for human use, thereby raising the possibility of developing a therapeutic regimen for treatment of the immunodeficiency that occurs in malnourished patients.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Recipient of a research fellowship from the European Society for
Pediatric Endocrinology. ![]()
Received June 18, 1996.
Revised October 18, 1996.
Accepted December 26, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. Papathanassoglou, K. El-Haschimi, X. C. Li, G. Matarese, T. Strom, and C. Mantzoros Leptin receptor expression and signaling in lymphocytes: kinetics during lymphocyte activation, role in lymphocyte survival, and response to high fat diet in mice. J. Immunol., June 15, 2006; 176(12): 7745 - 7752. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Fantuzzi and R. Faggioni Leptin in the regulation of immunity, inflammation, and hematopoiesis J. Leukoc. Biol., October 1, 2000; 68(4): 437 - 446. [Abstract] [Full Text] |
||||
![]() |
L. Sävendahl and L. E. Underwood Fasting Increases Serum Total Cholesterol, LDL Cholesterol and Apolipoprotein B in Healthy, Nonobese Humans J. Nutr., November 1, 1999; 129(11): 2005 - 2008. [Abstract] [Full Text] |
||||
![]() |
M. Haugen, D. Fraser, and O. Forre Diet therapy for the patient with rheumatoid arthritis? Rheumatology, November 1, 1999; 38(11): 1039 - 1044. [Full Text] [PDF] |
||||
![]() |
T. Nagata, W. Tobitani, N. Kiriike, T. Iketani, and S. Yamagami Capacity to Produce Cytokines During Weight Restoration in Patients With Anorexia Nervosa Psychosom Med, May 1, 1999; 61(3): 371 - 377. [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 |