|
|
||||||||
CLINICAL STUDY |
Department of Paediatrics, University of Parma, Via Gramsci, 14, 43100 Parma, Italy
(Correspondence should be addressed to M E Street; Email: mariaelisabeth.street{at}unipr.it)
| Abstract |
|---|
|
|
|---|
(TNF
), and insulin.
Methods: Eighteen subjects with CF (mean age 26.6 ± 1.1 years) and 18 controls, comparable for age, sex, and body mass index, were enrolled. Serum IGF-I, IGF-II, IGFBP-2, IGFBP-3, IL-1ß, IL-6, TNF
, insulin and C-peptide were measured. Different molecular forms of IGFBP-2 and IGFBP-3 were investigated by Western immunoblotting. The patients were analysed as a whole and as two subgroups depending on established clinical criteria (SwachmanKulczycki score).
Results: Patients had higher serum concentrations of IL-1ß, IL-6, TNF
and IGFBP-2 than controls. Serum concentrations of IGF-I and IGF-II were significantly lower and insulin and C-peptide levels significantly increased in CF compared with healthy controls whereas IGFBP-3 serum concentrations were similar, with comparable IGF-I/IGFBP-3 and decreased IGF-I/IGFBP-2 and IGF-II/IGFBP-2 molar ratios. From correlation analysis we detected a significant positive correlation between IGFBP-2 and IL-6 and a negative correlation between IGFBP-2 and IGFBP-3.
Conclusions: Our findings suggest that inflammation is an important modulator of the IGF/IGFBP system with an overall reduction in IGF bioactivity in CF.
| Introduction |
|---|
|
|
|---|
IGF-I and IGFBP-3 serum concentrations in CF show a trend to increase and normalize after antibiotic treatment (15), suggesting that infection and inflammation might also play a role in regulating the IGF/IGFBP system. The finding of low IGF-I serum concentrations lead to the hypothesis of peripheral GH resistance. However, in 1999 (16) a study in adult patients with CF reported normal spontaneous GH secretion and release after stimulation with arginine. The report a few years later of low serum IGF-I, increased IGFBP-3 proteolysis and poor growth in transgenic mice overexpressing interleukin (IL)-6 supports the hypothesis that inflammatory mediators per se could modulate the IGF/IGFBP system (17, 18).
In CF the inflammatory response to chronic inflammation is similar to that elicited by lipopolysaccharide and is mediated mainly by cytokines as tumour necrosis factor
(TNF
), IL-1, IL-6 and IL-8 (19). We have previously shown in vitro that IL-1ß and IL-6 modify IGFBP secretion into conditioned media (20), and in vivo, in inflammatory bowel diseases, that IGFBP-2 concentrations are increased together with IL-6, IL-1 and C-reactive protein (21).
The aim of this study was to investigate whether inflammation in CF was related to changes in the IGF/IGFBP system. We investigated whether IL-1ß, IL-6, TNF
or insulin were related to changes in IGF-I, IGF-II, IGFBP-2 and IGFBP-3.
| Materials and methods |
|---|
|
|
|---|
Eighteen consecutive patients in our CF clinic (26.6 ± 1.2 years, 6 males and 12 females) and 18 control subjects (28.0 ± 1.2 years, 4 males and 14 females), were enrolled into the study. Clinical details of patients and controls are shown in Table 1
. The two groups were comparable for chronological age, sex and body mass index (BMI) and all had attained pubertal stage 5 according to the criteria of Marshall and Tanner (22, 23). The diagnosis of CF had been confirmed in all by genetic analysis and a sweat test. All had exogenous pancreatic insufficiency that required enzymatic supplementation and were treated with on-going antibiotic therapy. Two patients had been taking steroids for at least 3 months (0.15 mg/kg per day of prednisone) and none were receiving azitromycin at the time of the study. One patient was excluded from the study for being diabetic and having signs of liver dysfunction. All but one patient tested positive for Pseudomonas aeruginosa infection, and one also had Burkholderia cepacia infection. C-reactive protein serum concentrations were assayed in all patients at the time of the study.
|
Height and weight were measured in all subjects, and BMI was calculated accordingly. BMI, height and target height are expressed as standard deviation scores (SDS) using Italian reference data (24).
Pulmonary function
Standard pulmonary function tests were performed the same day as the blood samples were taken. For statistical analysis the forced expiratory volume in 1 s (FEV1) was expressed as a percentage of the reference value (25).
Subgroups of subjects
The patients were subdivided into two groups according to the ShwachmanKulczycki scoring system (general activity, physical examination, nutrition and X-ray findings) (26). Group 1 were patients (n = 10) with scores of 71100 (mildly affected) and group 2 were patients (n = 8) with scores of 4170 (severely affected) (26). Control subjects were enrolled from young physicians training at our medical school and were healthy at the time of the study.
Assays
Blood samples were drawn after an overnight fast between 0800 and 0900 h. The blood was kept on ice until centrifugation at 2000 g/min for 10 min at 4 °C. The serum was then aliquoted and stored at 80 °C until assayed. Total serum IGF-I was measured using a RIA method (Nichols Institute Diagnostics, CA, USA). The coefficients of variation (CVs) were 3% (intra-assay) and 8.4% (inter-assay), respectively. Prior to assaying, serum samples were acid-ethanol-precipitated to avoid interference of binding proteins. Serum IGF-II and IGFBP-3 were measured using an immunoradiometric assay method (Diagnostic System Laboratories, Webster, TX, USA). The CVs were 5.3 and 1.9% (intra-assay) and 8.7% and 3.9% (inter-assay), respectively. Serum IGFBP-2 was assayed using an RIA method (Diagnostic System Laboratories, Inc. Webster, Texas, USA). The CVs were 4.7% (intra-assay) and 7.2% (inter-assay).
To transform IGF-I, IGF-II, IGFBP-2 and IGFBP-3 from ng/ml to nM we multiplied the single values by 0.131, 0.134, 0.035 and 0.035, respectively. Serum IL-6 and IL-1ß were measured using an ultrasensitive ELISA method (Quantikine HS; R&D Systems, Minnea-polis, MN, USA). The CVs were 3.8 and 6.9% (intra-assay) and 9.9% and 10.3% (inter-assay).
Serum TNF
was measured using an ultrasensitive ELISA method (Biosource International, Camarillo, CA, USA). The CVs were 11.9% (intra-assay) and 12.1% (inter-assay). Serum insulin and C-peptide were measured using a chemiluminescence method by Diagnostic Products Corporation (Los Angeles, CA, USA) for reading by Immulite2000. The intra-assay CVs were 6.5 and 7.8%, respectively. The inter-assay CV for insulin was 7.1%. The inter-assay CV for C-peptide was not calculated.
Different molecular forms of IGFBP-2 and -3 in the sera from all the patients and controls were assessed by Western immunoblotting. Equal aliquots of serum were subjected to electrophoresis on 12.5% acrylamide gels (27). The proteins were then transferred to nitro-cellulose membranes and probed with specific antibodies against IGFBP-2 and -3 (Upstate Biotechnology , NY, USA; 1:2000) at 4 °C overnight as previously described (20, 21). The bands were visualized by enhanced chemiluminescence as recommended by the manufacturer (Amersham Pharmacia Biotech, Little Chalfont, Bucks, UK).
Statistical analysis
Statistical analysis was performed using a one-way analysis of variance (ANOVA) test followed by Scheffés test for the growth factors. The IL data were not normally distributed and were analysed using KruskalWallis test. The correlation analysis was performed using Pearsons or Spearmans linear regression analyses. P < 0.05 was considered significant. Data are expressed as means ± S.E.M. unless otherwise stated.
Ethical approval
Informed consent was obtained from the subjects and/or the parents as appropriate. The study was approved by the local Ethics Committee (University of Parma Medical School).
| Results |
|---|
|
|
|---|
, insulin, C-peptide, IGF-I, IGF-II, IGFBP-2 and IGFBP-3 are shown in Table 2
levels were significantly higher in CF patients compared with control subjects.
|
Insulin was significantly higher in the patients with respect to controls, and in the patient groups levels were higher in group 2 patients (severely affected, with lower ShwachmanKulczycki scores) compared with group 1 (mildly affected): 18.8 ± 4.8 versus 7.1 ± 1.5 IU/l, respectively (P < 0.05). All controls had normal values (normal range, 425 IU/l; Fig. 1
). Serum C-peptide, similarly to insulin, was higher in the patients than in the controls, with higher values in the group 2 compared with the group 1 patients: 4.3 ± 1.2 versus 2.0 ± 0.3 mg/l, respectively (P < 0.05).
|
|
The IGF-I/IGFBP-2 and IGF-II/IGFBP-2 molar ratios were significantly lower in the patient group: 11.6 ± 2.1 and 3.2 ± 0.6 respectively versus 23.1 ± 4.2 and 5.4 ± 0.7 in controls (P < 0.05).
Correlation analysis
Results from the whole group are reported in Table 3
. In the control subjects, a positive correlation between IGF-I and IGFBP-3 was shown (P = 0.022; r = 0.57).
|
|
| Discussion |
|---|
|
|
|---|
and IGFBP-2 compared with normal controls. Serum concentrations of IGF-I and IGF-II were significantly lower than in healthy controls, with insulin and C-peptide levels being increased. We also detected a significant positive correlation between serum IL-6 and IGFBP-2 concentrations. For the first time we describe relationships between cytokines and the IGF/IGFBP system in CF patients, even if these must be regarded as preliminary data owing to the small number of patients enrolled in the study. Moreover, this is a cross-sectional study so we cannot comment on the natural progression of these relationships over time, and can put forward only hypotheses.
The serum concentrations of pro-inflammatory cytokines in our patients were similar to previously published data (19). Serum IGF-I levels were significantly lower in CF patients than in controls. At variance with previous studies this did not seem to be dependent on malnutrition or anorexia (9, 14, 28, 29) as the BMI values of these patients were normal and similar to those of controls, and total serum protein concentration was normal. Other authors have hypothesized that low serum IGF-I concentrations could be dependent on insulin hyposecretion in CF. Taylor et al. (28) described a progressive reduction of IGF-I and IGFBP-3 related to declining BMI, and deteriorating pulmonary function, and suggested that progressive insulin deficiency might be contributing to this. Ripa et al. (29) further supported this hypothesis by showing a high frequency of insulin hyposecretion with increased serum IGFBP-1 levels, normal IGF-I and reduced growth velocity in CF. However, this group of patients had basal insulin levels similar or greater than controls. Furthermore, the low IGF-I serum levels cannot be ascribed to impaired liver function (14, 16) as liver function tests were normal in all the subjects. In the case of impaired liver function insulin clearance would be reduced; however, C-peptide, which is metabolized by the kidneys, showed a similar pattern to insulin, suggesting that this is very unlikely. Thus, a state of latent insulin resistance that worsens with deteriorating clinical conditions is possible (30). Insulin concentrations in our patients could be increased because of increased serum levels of TNF
reducing the function of the glucose transporter GLUT4 (31), which is associated with reduced glucose disposal rate and insulin resistance. However, we did not have specific data on glucose tolerance, as this was not the aim of this study. Alternatively, other mechanisms leading to altered insulin receptor signalling that have been reported could be involved: IL-6 has also been suggested to interfere with insulin action (32). Furthermore, based on very recent literature, IGF-I concentrations seem to affect insulin sensitivity, and GH and IGF-I replacement seem to be capable of improving insulin sensitivity in diabetes and CF (3335).
Recently, it has been observed that IGF-I concentrations progressively decrease with age whereas IL-6 tends to increase, with a possible link to morbidity in the elderly (36). In light of this, data from our study suggest that this combination could be harmful also in subjects with CF, who present similar concentrations of IL-6 and IGF-I at a much younger age. Serum IGF-II levels have not been investigated previously and showed a similar pattern as IGF-I. IGFBP-2 concentrations were definitely increased in the patients and correlated positively with IL-6. Furthermore, the pattern was similar to that of TNF
. We previously described this increase in patients with inflammatory bowel diseases (21). These data suggest that IGFBP-2 on the one hand behaves as an acute phase protein, and on the other that the IGF/IGFBP system is modulated by some proinflammatory cytokines. It could be suggested that the serum levels could be altered in the two patients treated with oral steroids; however, it has been shown that 3 months of treatment with steroids reduce, not increase, IGFBP-2 serum levels (37).
IGFBP-3 serum concentrations were similar in the patients and controls although they tended to be lower in patients with the most severe clinical state, and we did not find any relationship with insulin at variance with previously published data (28, 29). This helps to explain why the IGF-I/IGFBP-3 molar ratios were similar in CF and healthy subjects. The IGF-I/IGFBP-2 and IGF-II/IGFBP-2 molar ratios were strongly reduced in CF, which would confirm a possible impaired compensatory mechanism for free IGF-I and IGF-II in CF.
Moreover, we did not find different molecular forms of IGFBP-2 and IGFBP-3 in patients and controls that could suggest increased proteolysis. The negative relationship between IGFBP-3 and IGFBP-2, both in the whole group of subjects enrolled in the study and in the patients, further strengthens the hypothesis that overall IGF bioactivity is blunted as the two peptides seem to go in opposite directions; if IGFBP-3 decreases, possibly increasing free IGF-I, IGFBP-2 increases, binding and reducing IGF-I bioavailability. The data of this study therefore suggest that the IGF/IGFBP system is modulated to reduce IGF-I and IGF-II bioactivity and thus their anabolic and pro-insulin activity.
In conclusion, it can be speculated that chronic inflammation is an important modulator of the IGF/IGFBP system in CF, and represents a possible mechanism for peripheral resistance to GH. This is in consideration of the fact that the ShwachmanKulczycki score uses general activity, a physical examination, nutrition and X-ray findings. Nutritional status in this patient group was very good and cannot have played an important role, whereas inflammatory cytokines were significantly increased and correlated with growth factors. The overall data suggest a dysregulation which influences bioactivity of IGFs. Furthermore, these data prompt considerations of alternative or additional treatments in CF. They support the use of anti-inflammatory drugs such as ibuprofen (38), possibly new anti-cytokine drugs (39), and even GH and IGF-I.
| Acknowledgements |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. E Street, P. Seghini, S. Fieni, M. A. Ziveri, C. Volta, D. Martorana, I. Viani, D. Gramellini, and S. Bernasconi Changes in interleukin-6 and IGF system and their relationships in placenta and cord blood in newborns with fetal growth restriction compared with controls. Eur. J. Endocrinol., October 1, 2006; 155(4): 567 - 574. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |