OUP user menu

Interleukin-10 and sudden infant death syndrome

Sophia M. Moscovis, Ann E. Gordon, Osama M. AL Madani, Maree Gleeson, Rodney J. Scott, June Roberts-Thomson, Sharron T. Hall, Donald M. Weir, Anthony Busuttil, C. Caroline Blackwell
DOI: http://dx.doi.org/10.1016/j.femsim.2004.06.020 130-138 First published online: 1 September 2004

Abstract

Uncontrolled pro-inflammatory responses to infections or bacterial toxins have been suggested to play a role in triggering the physiological events leading to sudden infant death syndrome (SIDS). We tested the hypothesis that these uncontrolled responses might be due to interactions between the gene polymorphisms inducing low levels of IL-10 and exposure to cigarette smoke. In vitro, the IL-10 (G-1082A) polymorphism was associated with low IL-10 levels and the –1082G allele was associated with high levels. The first objective was to assess the distribution of this polymorphism among SIDS infants, parents of SIDS infants and controls, and two ethnic groups: Aboriginal Australians who have a high incidence of SIDS; and Bangladeshis who in Britain have a low incidence of SIDS compared with Europeans. The second objective was to assess effects of human recombinant IL-10 on interleukin-6 (IL-6) and tumour necrosis factor-α (TNF-α) responses of human leukocytes to staphylococcal toxins implicated in SIDS. The third objective was to assess IL-10 responses to endotoxin and toxic shock syndrome toxin (TSST) from leukocytes of smokers and non-smokers in relation to the IL-10 (G-1082A) polymorphism. There were major differences in the distributions of these polymorphisms between Europeans and Bangladeshis (p=0.00) and between Europeans and Aboriginal Australians (p=0.00); however, they were similar for the Bangladeshi and Aboriginal Australian subjects. There were no significant differences in the distribution of these polymorphisms among SIDS infants or parents of SIDS infants compared to control groups. IL-10 significantly reduced IL-6 and TNF-α responses to TSST and staphylococcal enterotoxins A and C. At 50 ngml−1, IL-10 significantly increased TNF-α but not IL-6 responses to TSST and enterotoxin A. Although IL-10 responses to endotoxin were lower from leukocytes of smokers who were homozygous for the G allele, the differences were not significant; however, significantly lower IL-10 responses were found for smokers who were homozygous for the A allele (p=0.01) and heterozygotes (p=0.04). The pooled data found smokers had significantly lower levels of IL-10 responses to TSST, but there were no significant differences for smokers compared with non-smokers for the three genotypes. The high incidence of SIDS and serious respiratory infections among Aboriginal Australian infants and the low incidence of these conditions among Bangladeshi infants might be explained in part by our findings of differences in IL-10 responses between smokers and non-smokers. The lowest levels of IL-10 responses were observed among smokers who were homozygous for the A allele which is most prevalent among the Aboriginal Australians (83%) and Bangladeshis (84%). The major difference between the risk factors for SIDS in these two groups is the level of exposure of infants to cigarette smoke associated with maternal smoking.

Keywords
  • Interleukin-10
  • Gene polymorphisms
  • SIDS
  • Cigarette smoke
  • Ethnic groups

1 Introduction

It has been suggested that uncontrolled inflammatory responses to bacterial toxins play a role in triggering the events leading to sudden infant death syndrome (SIDS) [1,2]. Several pro-inflammatory cytokines such as interleukin 1β (IL-1β), tumour necrosis factor α (TNF-α) and interleukin-6 (IL-6) are elicited by infectious agents and some of these inflammatory mediators have been identified in increased levels in body fluids of SIDS infants [3,4]. These cytokines have powerful effects on physiological functions proposed to lead to death among SIDS infants: respiratory control; cardiac arrhythmia; hypoglycaemia; hyperthermia; anaphylaxis; vascular shock [2,5].

One of the functions of the anti-inflammatory cytokine interleukin 10 (IL-10) is to suppress the activity of pro-inflammatory cytokines. IL-10 is important in reducing the lethality of staphylococcal toxins in animal models [6], and one or more staphylococcal toxins have been demonstrated in tissues from over half of 105 SIDS infants tested [2,7]. Evidence obtained from a small series of SIDS infants suggested an IL-10 polymorphisms associated with lower levels of IL-10 responses was over-represented among SIDS infants [8,9]; however, another study in Scandinavia on a larger number of SIDS infants and infants who died of infection or other causes found no association with any IL-10 polymorphisms among the SIDS infants tested [10,11]. All the SIDS populations tested to date were from northern Europe. These studies need to be extended to other populations, as there are significant differences in the incidence of SIDS among ethnic groups. The highest incidences are among indigenous groups (e.g., 6.1/1000 live births for Aboriginal Australians) [12], but one of the lowest reported was among Bangladeshi families living in Britain (0.3/1000 live births) [13].

In an in vitro model using human leukocytes, we previously demonstrated that the unstimulated baseline IL-10 levels of cells from SIDS parents were higher than those of controls, but there were no significant differences between the SIDS or control parents in their IL-10 responses to either endotoxin or toxic shock syndrome toxin (TSST). More importantly, we found that levels of IL-10 were significantly reduced in smokers for both the baseline levels before stimulation with bacterial toxins and the responses to both endotoxin and TSST [2].

In this study, we chose to examine the IL-10 (G-1082A) polymorphism because it has been associated with severity of disease among patients with Epstein–Barr Virus (EBV) infection [14]. In a paper by Hajeer et al. [15], the –1082A allele was associated with low IL-10 levels and the –1082G allele was associated with high in vitro production of IL-10. Peripheral blood mononuclear cells from individuals with the –1082A allele showed significantly lower inducible IL-10 levels compared with individuals with the –1082G allele, and this was independent of the other polymorphisms at positions −819 and −592 [16].

Our hypothesis was that the severity of the responses to infections or toxins in SIDS might be due to interactions between the gene polymorphisms inducing low levels of IL-10 and exposure to cigarette smoke. The first objective of this study was to assess the IL-10 (G-1082A) polymorphism among the following groups: SIDS infants from different countries; SIDS parents and control parents from different countries; ethnic groups with high (Aboriginal Australians), medium (European) and low (Bangladeshi) incidences of SIDS. The second objective was to assess the anti-inflammatory effects of different levels of IL-10 on induction of two pro-inflammatory cytokines implicated in SIDS (TNF-α and IL-6) by staphylococcal toxins identified in tissues of SIDS infants. The third objective was to determine if cigarette smoke affected IL-10 responses to bacterial toxins by cells from donors with different genotypes of the IL-10 (G-1082A) polymorphism.

2 Materials and methods

2.1 Assessment of cytokine gene polymorphisms

Approval for the study was obtained from the Lothian Health Ethics Committee (UK), Hunter Area Research Ethics Committee and the University of Newcastle Human Research Ethics Committee (Australia). Buccal epithelial cells were collected from parents of SIDS infants from Britain (n=37) and Australia (n=83) and their matched controls (Britain=58, Australia=60). Fixed samples of tissue from SIDS infants were obtained from Australia (n=17), Hungary (n=21), and Germany (n=47). Stored frozen blood samples from Aboriginal Australians (n=123) and buccal epithelial cells from Bangladeshis (n=32) were used as DNA sources for comparisons between ethnic groups.

2.2 Extraction of DNA from epithelial cells and tissues

DNA from buccal swabs was extracted using the QIAamp® DNA Mini Kit (QIAGEN GmbH, Germany) according to the manufacturer's instructions for isolation of genomic DNA from buccal swabs.

DNA was extracted from previously homogenised tissue sample suspensions which had been screened for the presence of staphylococcal toxins by the salt precipitation method [17] with minor modifications. The homogenised frozen tissue sample in phosphate buffered saline (PBS) (100 µl) was digested in 1 ml nuclei lysis buffer. The buffer contained 100 mM Tris–HCl (pH 8.0), 2 mM ethylenediamine tetra acetic acid (EDTA) and 100 mM NaCl to which 50 µl of 20 mgml−1 proteinase K and 50 µl 20% (v/v) sodium dodecyl sulphate (SDS) were added before incubation. The solution was mixed by inverting the tube and incubated over-night at 56 °C in a shaking water bath. An additional 10 µl of proteinase K was added to undigested tissue and incubated until all tissue was digested.

To extract DNA from tissues from Hungarian infants, 333 µl of 6 M NaCl was added to the lysed tissue solution and vortexed for 30 s. To pellet the precipitate, the solution was centrifuged at 12,500g for 30 min at 4 °C. The supernatant was decanted into a 15 ml capped tube to which 5 ml cold absolute ethanol was added. The DNA was precipitated in solution by gentle inversions. The sample was centrifuged at 2,500g for 10 min at room temperature and the supernatant discarded. The pellet was vortexed in the remaining supernatant. DNA was washed in 1 ml of 70% (v/v) ethanol by inversion. To pellet the DNA, the solution was centrifuged at 12,500g for 2 min and the remaining supernatant removed using a pipette tip. The pellet was centrifuged again at 12,500g for 1 min and the remaining ethanol was removed with a pipette. The samples were left to air dry for 1 h to evaporate any remaining ethanol. The DNA pellet was reconstituted in 100 µl sterile 1× Tris–EDTA buffer which contained 1 mM Tris–HCl (pH 8.0), 0.1 mM EDTA (pH 8.0) and incubated in a 56 °C shaking water bath over night. Eluted DNA was stored at 4 °C until genotyping was performed.

DNA was extracted from tissues from German infants according to the manufacturer's instructions with the QIAamp® DNA Mini Kit (QIAGEN GmbH) following the protocol for the salt precipitation method. Eluted DNA was stored at 4 °C until genotyping was performed.

DNA was extracted from the Australian tissue samples on a BioRobot® M48 machine (QIAGEN GmbH); 200 µl of lysed tissue solution was placed in microfuge tubes and DNA extracted according to the manufacturer's instructions. The smallest elution volume (50 µl) was used due to the small amounts of DNA in Australian tissue samples. Eluted DNA samples were collected from the cooling block and stored at 4 °C until genotyping was performed.

2.3 Analysis of the IL-10 (G-1082A) polymorphism by real time polymerase chain reaction

A real time polymerase chain reaction (PCR) allelic discrimination assay was developed to test for the presence of the IL-10 (G-1082A) polymorphism. The primers (Invitrogen, Frederick, USA) used to amplify the sequence surrounding the SNP were: sense, 5′ CACAAATCCAAG ACAACACTACTAAGG 3′; antisense, 5′ TCCATGGAGGCTGGATAGGA 3′. Minor groove binding (MGB) probes (PE Applied Biosystems, Foster City, USA) used for detection of the single nucleotide polymorphism (SNP) were designed in Primer Express® Version 1.5 (PE Applied Biosystems) with genomic sequences obtained from the GenBank database using a BLAST search (National Center of Biotechnology Information, National Institutes of Health, Bethesda, MD, USA) (Accession Number AF 295024).

The assay used two specific, fluorescent-labelled probes to identify each allele. The probe for the wildtype (G) allele was labelled with 6-carboxyfluorescein (6-FAM) (5′ 6-FAM-CTTCCCCCTCCCAAA 3′). The variant (A) allele was labelled with VIC™ (5′ VIC-CCTCCCCTTCCCAAA 3′).

DNA samples were analysed in 96-well optical reaction plates (PE Applied Biosystems). Each PCR mixture contained the following components: 50 ng of sample DNA; 200 nM of each MGB probe; 300 nM of each primer; and 1× TaqMan® Universal PCR Master Mix (PE Applied Biosystems) made up to a final volume of 25 µl with sterilized MilliQ water (Millipore, Sydney, Australia). The reaction was conducted with the ABI PRISM® 7900HT sequence detection system (PE Applied Biosystems). The following thermal cycling conditions were used for 40 cycles: 50 °C for 2 min; 95 °C for 10 min; 92 °C for 15 s; and 55 °C for 1 min. Fluorescence emitted from the sequence specific probes is proportional to the level of sequence specific PCR product and was detected spectrophotometrically at intervals during the cycle.

2.4 Effect of IL-10 levels on pro-inflammatory cytokine responses to staphylococcal toxins

The assay using human leukocytes for assessment of cytokine responses to TSST, staphylococcal enterotoxin A (SEA) and staphylococcal enterotoxin C (SEC) has been described previously [18]. Because different staphylococcal toxins were identified in different populations of SIDS infants tested (e.g., TSST in Australia, SEC and staphylococcal enterotoxin B (SEB) in Britain, SEA in Hungary) [2], the effects of IL-10 on induction of TNF-α and IL-6 by SEA, SEC and TSST were examined. The method is described briefly.

One day old buffy coats (50 ml) from donors of blood group O were obtained from the Scottish National Blood Transfusion Service, Royal Infirmary of Edinburgh. The samples were collected between 9:00 and 16:30 h to minimise effects of circadian rhythm on inflammatory responses [19]. The cells were diluted 1 in 4 in sterile PBS under aseptic conditions. The diluted blood (40 ml) was layered on Histopaque (12 ml) (Sigma, Poole, Dorset, UK) in sterile plastic 50 ml conical tubes and centrifuged for 30 min at 300g. Leukocytes were collected from the interface and washed twice in Dulbecco's modified essential medium (DMEM) (Sigma) at 250g for 10 min. Viable cells were enumerated microscopically by the trypan blue exclusion method [20] with a Neubauer haemocytometer and diluted to 2×106 cells ml−1 in DMEM with 10% (v/v) serum obtained from the buffy coat of the individual donor.

Individual donor cells (2×106 ml −1, 500 µl/well) were placed in 24 well tissue culture plates and 250 µl of TSST, SEA or SEC (Toxin Technology, Sarasota, Fla., USA) were added to obtain final concentrations of 0.05, 0.1, 0.5, or 1 µgml−1. All conditions were tested using triplicate samples. The cells were incubated at 37 °C in a humidified incubator with 5% CO2. The cell supernatants were collected at different times (0, 4, 8, 16, 24, 36, and 72 h) in sterile tubes and centrifuged at 250g for 10 min. The supernatants were stored at −20 °C until assayed for cytokines.

To determine the effect of IL-10 on inflammatory responses, human leukocytes (2×106, 500 µl/well) were placed in 24 well tissue culture plates and 250 µl of TSST, SEA or SEC were added to the cells to obtain a final concentration of 0.1 µgml−1. These were incubated at 37 °C for 30 min and 250 µl of different concentrations of human recombinant IL-10 (R&D Systems, Minneapolis, MN, USA) diluted in DMEM were added to give a final concentration of 800, 400, 200, 100, 50, 25, 12.5, 6.25, 3 or 1.5 ng/ml−1. The positive control contained toxin but DMEM replaced IL-10 and for the negative control toxin was replaced by DMEM. The cells were incubated at 37 °C for 16 h in a humidified incubator with 5% CO2. The cell supernatants were collected in sterile tubes and centrifuged at 250g for 10 min. The supernatants were stored at −20 °C until assayed for cytokine production.

2.5 Assessment of inflammatory cytokine responses of smokers and non-smokers

Blood samples (10–20 ml) were collected from British parents of SIDS infants and control individuals who had no family history of SIDS. The leukocytes were assessed for inflammatory responses to TSST and endotoxin. The blood samples were collected in the morning to limit the effect of circadian variation on cytokine production. The blood was transported to the laboratory at room temperature. All samples were coded and tested without knowing the smoking status of the donors.

The numbers of white blood cells in each sample were determined microscopically and the blood diluted in RPMI 1640 medium (Sigma) to a final concentration of 2×106 white blood cells ml−1. In sterile 24 well tissue culture plates, 1 ml of diluted blood was incubated with 1 ml of RPMI medium (unstimulated control), 1 ml of endotoxin from Escherichia coli (Sigma) diluted in RPMI medium (final concentration of 0.01 µgml−1) or 1 ml of TSST diluted in RPMI medium (final concentration 0.5 µgml−1). Triplicate samples for each condition tested were incubated for 24 h at 37 °C in 5% CO2. The samples were transferred to a sterile tube, centrifuged at 400g for 20 min at room temperature and the supernatants stored at −20 °C until tested for cytokine production [20].

2.6 Quantitative assessment of cytokines

The L929 mouse fibroblast cell line was used for the bioassay for TNF-α as described previously and the results for the samples for each experiment expressed as IU ml−1 derived from the curve obtained with the human TNF-α standard obtained from the National Institute for Biological Standards and Controls (NIBSC) (Herefordshire, UK). IL-6 and IL-10 were assessed by enzyme linked immunosorbent assays (ELISA) as described previously and the results for the test samples for each experiment expressed as ngml−1 derived from the standard curves obtained with the recombinant human IL-6 or IL-10 standards [18,20].

2.7 Statistical analyses

The cytokine gene polymorphisms were analysed using the statistical software package Statistics/Data Analysis™ (STATA) Version 7.0 (Stata Corporation, Texas, USA). The Fisher's exact test was used to assess the distribution of cytokine gene polymorphisms in SIDS infants, parents of SIDS infants compared to control individuals, and between ethnic groups.

The Student's t-test for paired samples was used to assess differences between TNF-α and IL-6 cytokine responses to the staphylococcal toxins in the presence of IL-10. The cytokine levels were log transformed to assess the effect of recombinant IL-10 on the pro-inflammatory cytokine responses. The Student's t-test was also used to assess differences in the cytokine responses of smokers and non-smokers in relation to genotype.

The significance level for all tests was set at p<0.05.

3 Results

The distribution of the allele frequency of the IL-10 (G-1082A) polymorphism was examined in SIDS infants from different countries, parents of SIDS infants, unrelated control parents who had not had a SIDS infant, and randomly selected individuals from different ethnic groups. The numbers of individuals in each category and the allele frequencies are summarised in Table 1.

View this table:
Table 1

IL-10 (G-1082A) SNP allele frequency distribution in the study populations

EthnicityGroupAllele frequency (%)Sample size (n)
G/GG/AA/A
BritishSIDSParents29.754.116.237
BritishControlParents25.941.432.858
AustralianSIDSParents28.936.134.983
AustralianControlParents33.336.730.060
AustralianSIDSInfants47.111.841.217
HungarianSIDSInfants19.028.652.421
GermanSIDSInfants27.746.825.547
CombinedSIDSInfants29.435.335.385
BangladeshiAdults015.684.432
AboriginalAustralianAdults017.182.9123
EuropeanAdults29.739.031.4118

3.1 Allele frequency distribution of the IL-10 (G-1082A) polymorphism among SIDS infants from different countries

The combined results for the 85 SIDS infants showed approximately equal proportion of the three genotypes for the IL-10 (G-1082A) SNP, similar to distribution for the European controls (p=0.81). The distribution of allele frequencies among Australian SIDS infants varied from the distribution found for the Australian control parents; however the differences were not significant (p=0.15).

Although differences in the distribution of the allele frequency were observed among SIDS infants from different countries, due to small sample sizes, only the results for the Australian and German SIDS infants were significantly different (p=0.03) (Table 1).

3.2 Allele frequency distribution of the IL-10 (G-1082A) SNP among parents of SIDS infants and controls

The distribution of the allele frequency between British parents of SIDS infants and their control subjects were not significantly different (p=0.21). Parents of SIDS infants had a higher frequency of the heterozygous variant (GA) genotype, and a lower frequency of individuals with the homozygous variant (AA) genotype than their control group.

Allele frequency distributions within Australian parents of SIDS infants and their control subjects were not significantly different (p=0.80). The distribution of the alleles in both groups was in equal proportions, with approximately 30% of individuals with each genotype (Table 1).

3.3 Allele frequency distribution of the IL-10 (G-1082A) SNP in different ethnic groups

The distribution of the allele frequencies (Table 1) were significantly different among the ethnic groups tested (p=0.00). Individuals from Bangladeshi and Aboriginal Australian ethnic backgrounds were significantly different from those of Caucasian/European descent living in Australia and UK (p=0.00). There were no statistically significant differences in the allele frequency distributions between the Aboriginal Australian and Bangladeshi individuals (p=1.00). In these groups approximately 80% of individuals possessed the AA genotype, and approximately 15% possessed the GA genotype. In both the Australian Aboriginal and Bangladeshi populations there were no individuals possessing the homozygous wildtype (GG) genotype.

Distribution of these genotypes between the populations of European descent, (non-Aboriginal Australians and British subjects), were not significantly different (p=0.70).

3.4 Effect of IL-10 levels on induction of TNF-α from human leukocytes by TSST-1, SEC or SEA

In preliminary experiments, the maximum maximum production of both TNF-α and IL-6 was observed at 16 h following addition of 0.1 µgml−1 of each of the staphylococcal toxins. The peak production of IL-10 was obtained with 0.1 µgml−1 of the toxins at 24 h. This concentration of each toxin and an incubation time of 16 h were used in subsequent experiments.

IL-10 levels ranging from 800–1.5 ngml−1 were tested for their effect on induction of TNF-α by the three toxins from leukocytes of randomly selected blood donors of blood group O. The TNF-α responses to toxins in the presence of different concentrations of IL-10 are summarised in Table 2. Compared with the positive control containing toxin alone, there was a significant reduction in TNF-α induced by TSST-1 with concentrations of IL-10 between 800 and 100 ngml−1. In response to SEC, IL-10 significantly reduced induction of TNF-α at concentrations between 800 and 100 ngml−1 (Table 2). At 50 ngml−1 of IL-10, there was a significant enhancement of production TNF-α induced by both TSST-1 (p=0.005) and SEC (p=0.03). There was a significant reduction in TNF-α induced by SEA at concentrations of IL-10 between 800 and 50 ngml−1. In contrast to the results with TSST-1 and SEC, there was no significant enhancement with 50 ngml−1 for cells stimulated with SEA.

View this table:
Table 2

Effect of increasing concentrations of recombinant IL-10 on TNF-α responses (IU ml−1) induced from human leukocytes by TSST-1, SEC and SEA (n=5)

ToxinTNF-α responses (IU/ml−1)
TSST-1SECSEA
Mean95% CIpMean95% CIpMean95% CIp
Cells×1061.51.751.25
Toxin21.256.6, 32.80.023719, 51.40.00628.2541.2, 12.70.009
IL-10 (ngml−1)
8008−23.4, −3.070.0319−30.9, −50.0317.75−17.9, −30.02
4009−21.8, −2.660.0321.5−30.9, −0.040.0518.5−15.8, −3.60.02
20012.7−13.9, −3.060.0225−22.7, −1.20.0419−15.2, −3.20.02
10015.5−11.3, −0.180.0529−12.3, −2.10.0220.5−13.8, −1.60.03
50273.3, 8.10.005410.7, 8.70.0422.75−10.77, −0.220.05
2518.5−5.13, −0.30.0432.75−5.7, −2.70.00322.25−0.75, 12.70.07
12.519.75−.091, 3.10.0633−2.3, 10.30.1422.75−1.1, 12.180.08
6.2520−0.75, 3.20.1435.75−0.2, 2.70.0824.25−1.5, 9.50.10
320.75−1.09, 2.090.3935.75−0.27, 2.70.0824.5−0.4, 7.90.07
1.520.75−1.09, 2.090.3935.75−0.27, 2.70.0825.2−0.6, 6.60.081
  • Concentration of TNF-α (IU ml−1) compared with the negative control with no toxin.

  • Concentration of TNF-α (IU ml−1) compared with the positive control with toxin but no added IL-10.

3.5 Effect of IL-10 levels on induction of IL-6 from human leukocytes by TSST-1, SEC or SEA

Compared with the positive control containing toxin alone, there was a significant reduction in IL-6 induced by TSST-1, SEC or SEA at concentrations of IL-10 between 800 and 25 ngml−1 (Table 3). The significant increase in TNF-α observed for cells stimulated with TSST-1 or SEC in the presence of 50 ngml−1 of IL-10 was not observed for IL-6.

View this table:
Table 3

Effect of increasing concentrations of human recombinant IL-10 on IL-6 responses (ngml−1) induced from human leukocytes by TSST-1, SEC and SEA (n=5)

ToxinIL-6 responses (ngm−1)
TSST-1SECSEA
Mean95% CIpMean95% CIpMean95% CIp
Cells×1061.51.381.6
Toxin3210.8, 50.10.012159.9, 17.20.0013120.2, 38.50.002
IL-10 (ngml−1)
8006.5−42.6, −8.30.026.25−13.3, −4.170.0113−28.3, −7.60.02
4008−40, −7.90.027.75−10.7, −3.720.0113−28.3, −7.60.02
20011−35.29, −6.70.028−11.6, −2.30.0215−28.5, −3.470.03
10011.5−33.5, −7.40.0210−7.5, −2.40.0117−25.1, −2.820.03
5015.25−29.3, −4.190.037.5−12.9, −2.060.0218.75−22.2, −2.230.03
2521−19, 2.990.0212.25−5.1, −0.360.0424.5−11.09, −1.90.02
12.526.7−10.5, −0.0070.0513.75−3.2, 0.70.1427.5−6.54, −0.450.04
6.2528.75−0.9, 7.40.0911−3.6, 11.60.1932.25−2.7, 0.270.08
331.25−2.7, 4.20.514−0.29, 2.290.0930−0.29, 2.290.09
1.531−1.9, 3.90.313.5−0.55, 3.50.129.5−0.5, 3.50.1
  • IL-6 concentration (ngml−1) compared with the negative control with no toxin.

  • IL-6 concentration (ngml−1) compared with the positive control with toxin but no added IL-10.

3.6 Effect of smoking on IL-10 responses by genotypes for the IL-10 (G-1082A) SNP

Endotoxin induced significantly lower median levels of IL-10 from leukocytes from smokers (25.6 ngml−1, range 1–171.4 ngml−1) than from non-smokers (57.7 ngml−1, range 1–1608.0 ngml−1) (p=0.00). There were no significant differences for IL-10 responses from leukocytes from non-smokers to endotoxin for the three genotypes of the IL-10 (G-1082A) SNP. There were significant differences between smokers and non-smokers for individuals with the GA genotype (p=0.04) and the AA genotype (p=0.01). The difference between smokers and non-smokers for the GG genotype was not significant (p=0.09) (Fig. 1).

Figure 1

Median IL-10 responses of smokers and non-smokers to 0.01 mgml−1 LPS assessed for genotypes of the IL-10 (G-1082A) SNP. The number of subjects with each allele is indicated below under the smoking category.

Leukocytes from smokers had significantly lower levels of IL-10 responses to TSST (80.9 ngml−1, range 8.8–351.1 ngml−1) than leukocytes from non-smokers (92.8 ngml−1, range 8–1700 ngml−1) (p=0.04). Although the responses for smokers were lower for each genotype, the differences were not significant.

4 Discussion

This study indicates that while there are significant differences in the distribution of the IL-10 (G-1082A) polymorphism among different ethnic groups, there were no significant differences in the distribution of the SNP between SIDS infants or between parents of SIDS infants and their control groups. As with other studies [9,10], the findings for the 85 SIDS infants did not reveal a significant difference in the distribution of the SNP in populations of European origin. There is a need to examine DNA samples from additional SIDS infants from different geographic areas and ethnic groups; however, these studies are difficult to carry out because of the cultural and legal restrictions on use of autopsy material for research purposes. There were differences in the distribution of the SNP in the three populations of SIDS infants examined, but only the Australian and German infants were significantly different from each other. The ethnic origin of the Australian SIDS infants was not known, but it was unlikely that there was a significant proportion of indigenous infants (personal communication).

There were no significant differences in the distribution of the IL-10 (G-1082A) SNP between SIDS parents and their respective controls for groups in either Australia or Britain.

Major differences were observed in the frequency of the IL-10 (G-1082A) SNP among the different ethnic groups. Distribution of the allele frequency among the non-Aboriginal Australian and British populations was not significantly different, consistent with the origins of the non-indigenous population in Australia. The distribution of the allele frequency did not differ significantly between the Aboriginal Australian and Bangladeshi groups, but both the Bangladeshi and Aboriginal Australians differed significantly from those for the subjects of European heritage. The similarity of the allele distributions between the Bangladeshi and Aboriginal Australians indicated a shared ethnic origin. The lack of the GG genotype indicates that in the evolutionary process these ethnic groups have not been exposed to the pathogens that have led to this genotype becoming dominant in the groups of European ethnicity.

If differences in allele distribution are important in the aetiology of SIDS, it would be predicted that the incidence of SIDS would be similar for the Bangladeshi and Aboriginal Australians; however, the low incidence for Bangladeshi infants living in Britain (0.3/1000 live births) [13] contrasts with the high incidence (6.1/1000 live births) among Aboriginal Australians [12]. Both groups share many of the risk factors for SIDS; however, a major difference is the proportion of mothers who smoke is low among Bangladeshi women in Britain (3%) [21] compared with Aboriginal Australian mothers (75%) [22]. The interactions between the IL-10 (G-1082A) SNP and cigarette smoking on IL-10 responses to bacterial toxins was, therefore, assessed.

Since there are differences in the types of staphylococcal toxins identified in SIDS infants from different countries, we assessed responses to TSST, SEC and SEA and the effects of different concentrations of IL-10 on the IL-6 and TNF-α responses to the toxins. Other studies had examined the effects of IL-10, interleukin-4 (IL-4) and dexamethasone on cytokine production from human peripheral blood mononuclear cells induced by TSST, SEA, or SEB [23,24]. IL-10 was found to have a more potent effect on suppression of TNF-α, IL-6, interleukin-2 (IL-2) and interferon-γ than IL-4; however, the suppressive effect of IL-10 was not as strong as that observed with dexamethasone [23,24]. In the human leukocyte model system used for the studies reported here, IL-10 reduced induction of both TNF-α and IL-6 to all three staphylococcal toxins tested in a dose dependent pattern. Greater suppression of both pro-inflammatory mediators was observed with higher concentrations of IL-10. The only exception was a significant increase in the TNF-α responses to TSST and SEC for a low concentration of IL-10 (50 ngml−1). This increase was not noted for TNF-α responses to SEA or for IL-6 responses to any of the three toxins tested.

The results of the in vitro studies indicate that in humans, as in animal models [6], IL-10 plays a role in controlling pro-inflammatory responses induced by the staphylococcal toxins identified in SIDS infants. It must be considered, however, that some low levels of IL-10 might enhance TNF-α responses to some toxins. The majority of individuals will respond to endotoxin, but responses to TSST will be greater for individuals expressing the Vβ2 T cell receptor to which the toxin binds. This receptor will be present in only a proportion of the population examined and we did not assess this.

Blood endotoxin levels in SIDS infants were higher in those in whom there was histological evidence of mild to moderate inflammation [25]. There were no significant differences between IL-10 responses to endotoxin observed with cells from non-smokers for the three genotypes of the IL-10 (G-1082A) SNP. Smoking had a significant effect on IL-10 responses of two of the three genotypes to endotoxin. This is consistent with our previous study indicting that smokers had lower levels of IL-10 responses to TSST and endotoxin [2]. Although IL-10 responses to endotoxin were lower from leukocytes of donors with the GG genotype, the differences were not significant; however, the cells from smokers who were heterozygotes had significantly lower IL-10 responses and leukocytes from donors of the AA genotype had the lowest IL-10 responses. While the pooled data found smokers had significantly lower levels of IL-10 responses to TSST, the patterns observed for IL-10 responses of smokers and non-smokers to endotoxin for the three genotypes were not observed for TSST.

There is evidence that some infants have levels of the nicotine metabolite cotinine equivalent to those found in active smokers [26]. The risk of SIDS increases with increased exposure to cigarette smoke [27]. The ability of infants to suppress inflammatory responses induced by viruses or bacterial toxins might be significantly impaired if their IL-10 levels were constitutively low and further reduced by interactions with components of cigarette smoke. The high incidence of SIDS and serious respiratory infections among Aboriginal Australian infants [28] and the low incidence of these conditions among Bangladeshi infants [13] might be explained in part by our findings of differences in IL-10 responses between smokers and non-smokers. The lowest levels of IL-10 responses were observed among smokers with the AA genotype which is most prevalent among the Aboriginal Australians and Bangladeshi. The major difference between the risk factors for SIDS in these two groups is the level of exposure of infants to cigarette smoke associated with maternal smoking [21,22].

Levels of cigarette smoke contributing to a suppressive effect on IL-10 in relation to ranges of cotinine observed in children exposed to cigarette smoke need to be examined. The findings of this study have implications not only for SIDS but also for associations observed between cigarette smoke and susceptibility to or severity of a number of infectious diseases.

The distributions of the IL-10 (G-1082A) SNP need to be examined in larger numbers of SIDS infants from different geographic regions and ethnic groups, particularly among high risk groups such as indigenous populations and low risk groups such as Asians in Britain or Japan. The relevance of the ethnic differences on ability to respond to infections needs to be investigated. The relationships between the IL-10 polymorphism, the (C-511T) polymorphism in the IL-1β gene, and the (T+2018C) polymorphism in the IL-1RN gene are assessed in the accompanying paper [29].

Acknowledgements

We are grateful to Sally Gulliver and Doris MacKenzie for their assistance with recruitment of Australian and British families into the study. Samples of tissues were kindly supplied by Dr. R. Amberg, Dr. J. Hilton and Dr. K. Törö. We thank Dr. C. Meldrum for his assistance with method development in the early stages of the genetic analyses. The studies were supported by: Babes in Arms, UK; The Meningitis Association of Scotland; The Scottish Cot Death Trust; SIDS and Kids Hunter Region; The Hunter Area Pathology Charitable Trust; and Hunter Medical Research Institute, Newcastle, Australia.

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
View Abstract