Presented at the Neonatal Society 2003 Spring Meeting.
Hamilton K1,2, Turnbull W2, Modi N3, Carr R1
1 Department of Haematology, King’s College, Guy’s Hospital, London, UK
2 Department of Immunobiology, King’s College, Guy’s Hospital, London, UK
3 Section of Paediatrics & Neonatal Medicine, Division of Paediatrics, Obstetrics & Gynaecology, Faculty of Medicine, Imperial College, London, UK
Introduction: Sepsis is a major cause of death and impairment in preterm neonates. Recent attention has focused on the role of individual cytokines and the balance between TH1 (“pro-inflammatory”) and TH2 (“anti-inflammatory”) cytokines in the protection against bacterial invasion and associated tissue damage. Interleukin-2 (IL-2) drives lymphocytes towards TH1 polarisation; interferon-g (IFNg) supports phagocyte antibacterial activity and is used therapeutically to correct the neutrophil dysfunction of chronic granulomatous disease; tumour necrosis factor-a (TNFa) is believed to be a mediator of tissue damage associated with the systemic inflammatory response syndrome. Aims: i) To develop a whole blood intracellular cytokine assay suitable for very small sample volumes; ii) to examine cytokine production by lymphocytes from growth restricted preterm neonates; iii) to study cytokine responses in infants recruited into PROGRAMS, a randomised controlled trial of prophylactic granulocyte-macrophage colony stimulating factor (GM-CSF) to reduce sepsis in growth restricted preterm neonates born = 31 weeks gestation.
Methods: The proportion of each lymphocyte subtype producing each cytokine is assayed prior to trial entry (age <72h), after 5 days prophylactic GM-CSF treatment or equivalent time in controls, and 21 to 28 days after trial entry. This study has been approved by the UK multicentre research ethics committee.
Results: We have developed a whole blood intracellular cytokine assay capable of simultaneously quantifying up to 6 different cytokines in 4 different T lymphocyte sub-populations in a 600µl blood sample. Samples are incubated for 16 hours with PMA and ionomycin to stimulate cytokine production, together with the Golgi channel blockers monensin and brefeldin A to inhibit cytokine secretion. Lymphocyte populations are identified by MoAbs to CD3, CD8 and CD45RA then fixed and permeabilised, stained for intracellular cytokines and read on a 6 colour flow cytometer (Mo-Flo). To date, a total of 20 samples from nine trial infants have been studied. The proportion of “memory” T helper cells (CD4+CD45RO) was 41.4 ± 36.6% (mean ± SD) at birth. This increased in all infants over time. IL-2 and IFNg were produced by significantly fewer memory cells from neonates <72h old compared with adults ( IL-2: 12.1 ± 11.3% vs 57.9 ± 11.4%, p<0.001; IFNg: 3.4 ± 4.7% vs 34.5 ± 6.8%, p<0.001, Students t-test). IL-2 production increased with postnatal maturation, but virtually no IFNg producing T cells were detectable even a month after birth. TNFa was also produced by fewer newborn infant CD4 memory cells than adults (16.7 ± 11.4% vs adults 59.3 ± 14.5%, p<0.001). By the fourth postnatal week the number of CD4 memory T helper cells producing TNFa overlapped the lower adult range (mean for all trial babies 28.1 ± 18%, p<0.01 vs adults). However at this time point the number of naïve CD4 T cells (CD45RA) producing TNFa was equivalent to the adult population (mean for all trial babies 35.2 ± 23.4% vs adults 31.9 ± 19.1%, ns). In blood samples taken at the end of five days of GM-CSF treatment there was no evidence of an increase in the number of IL-2, IFNg or TNFa producing cells in treated compared with control infants.
Conclusions: i) This novel micro-assay allows sequential cytokine profiling in different lymphocyte populations during postnatal development. ii) These data suggest that growth restricted preterm neonates have minimal T helper cell production at birth of the TH1 cytokines IL-2 and IFNg. This is further evidence of poor antibacterial defences in preterm neonates. iii) By contrast, production of TNFa appears well developed, which may be of relevance to perinatal sepsis-related tissue injury. iv) This preliminary data does not support the hypothesis that GM-CSF treatment in neonates upregulates production of IL-2, IFNg or TNFa in vivo.