NF-κB activation in peripheral blood mononuclear cells in neonatal asphyxia

Abstract

Neonatal asphyxia results in hypoxic–ischaemic encephalopathy. Previous studies have demonstrated that brain hypoxia and ischaemia lead to the production of proinflammatory cytokines, including tumour necrosis factor-α (TNF-α), interleukin-1 (IL-1) and IL-6. Transcription factor NF-κB is essential for the expression of these cytokines. We examined whether or not NF-κB is activated in peripheral mononuclear cells (PBMC) in neonatal asphyxia by flow cytometry. In addition, we examined the relationship between NF-κB activation in PBMC and the neurological prognosis. Flow cytometry analysis demonstrated that the level of NF-κB activation in CD14⁺ monocytes/macrophages of the patients with asphyxia who had neurological sequelae was significantly higher than in the controls, and in the patients with asphyxia who survived (31·7 ± 7·2%versus 2·5 ± 0·9%, P = 0·008, and versus 1·6 ± 1·4%, P = 0·014, respectively). Our findings suggest that NF-κB activation in peripheral blood CD14⁺ monocytes/macrophages in neonatal asphyxia is important for predicting the subsequent neurological sequelae.

INTRODUCTION

Neonatal asphyxia causes hypoxic–ischaemic encephalopathy, and might result in a poor neurological prognosis. Previous studies on hypoxic–ischaemic brain injury suggested that inflammatory mechanisms, particularly ones involving certain cytokines, are important for the final common biochemical pathway to hypoxic–ischaemic cell death [1–12]. The principal sequence of events comprises activation of microglia in the first hours after the insult, followed by the release of a variety of neurotoxic products, including excitatory amino acid agonists, reactive oxygen species, nitric oxide, proteases and proinflammatory cytokines such as tumour necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) [1–15]. The mRNA expression of the latter two cytokines is increased after brain ischaemia [8,9,13,14]. In addition, TNF-α and IL-1β are able to induce an inflammatory reaction in the central nervous system (CNS), and IL-1β blocking agents in animal models of brain ischaemia have been shown to attenuate the post-ischaemic increase in brain water content that leads to a decrease in brain injury, and improvement of the neurological score [9,15,16].

NF-κB is a pivotal transcription factor for genes that encode proinflammatory cytokines, chemokines and adhesion molecules that mediate inflammation: IL-1, IL-2, IL-6, IL-8, TNF-α, monocyte chemoattractant protein-1, E-selectin, vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 [17–23]. NF-κB is retained in an inactive form in the cytoplasm [24,25]; its prototypic form consists of a heterodimer of p50 and p65 that is normally bound by members of the IκB family. Activation of NF-κB requires degradation of the cytoplasmic inhibitor IκB protein. Phosphorylation of IκB by drugs, cytokines, bacterial products, and viruses leads to IκB degradation, translocation of NF-κB to the nucleus and transcription of proinflammatory cytokine genes. Asphyxia induces NF-κB activation in the CNS in animal models [26–29]. However, the role of NF-κB in neonatal asphyxia is unknown. We examined whether or not NF-κB activation occurs in peripheral blood CD14⁺ monocytes/macrophages or CD3⁺ T cells to clarify the immunological pathophysiology of neonatal asphyxia, and investigated the relationship between NF-κB activation in peripheral blood mononuclear cells (PBMC) and the neurological prognosis.

PATIENTS AND METHODS

Patients with neonatal asphyxia

Nine neonates (seven boys and two girls; mean gestational age, 37·9 weeks; mean birth weight, 2602 g) with neonatal asphyxia (Apgar score at 1 min lower than 8 points) admitted to our NICU between June 1999 and June 2002 were enrolled in this study (Table 1). The parents of the neonates gave informed consent for their participation in the study. All samples were obtained within 24 h of birth.

Table 1.

Data for neonates

No.	Gestational age	Birth weight (g)	Apgar score at 1 min/5 min	Prognosis	Complication
Controls
1	35W6D	2132	9/9	Normal	Normal
2	38W2D	2446	8/9	Normal	Normal
3	39W6D	3416	8/9	Normal	Normal
4	40W0D	3136	8/8	Normal	Normal
5	38W6D	2830	8/9	Normal	Normal
6	37W5D	2984	8/8	Normal	Normal
7	39W3D	2934	8/9	Normal	Normal
Asphyxia group
1	41W0D	3402	6/9	Normal	Normal
2	39W0D	2670	4/5	Normal	Normal
3	37W1D	2100	5/7	Normal	Normal
4	37W1D	1716	6/8	Normal	Normal
5	34W1D	2324	6/7	Normal	RDS
6	37W1D	2526	4/5	CP, Epi	Normal
7	36W2D	2552	4/6	MeR	Normal
8	38W2D	3172	7/7	MeR	Normal
9	40W6D	2952	1/1	CP, Epi	Normal

CP = cerebral palsy; Epi = epilepsy; MeR = mental retardation; RDS = respiratory distress syndrome.

Controls

Seven normal newborns (five boys and two girls; mean gestational age, 37·6 weeks; mean birth weight, 2840 g) without neonatal asphyxia (Apgar score at 1 min higher than 7 points) admitted to our NICU between September and November 2002 were enrolled in this study (Table 1). The parents of the neonates gave informed consent for their participation in the study. All samples were obtained within 24 h of birth.

Cells

Whole blood cells were obtained from peripheral venous blood of the subjects. Some of the whole blood cells of normal newborns were mixed with 1 µg/ml lipopolysaccharide (LPS, from Escherichia coli, 0111:B4; Sigma Chemical Co., St Louis, MO, USA) and incubated at 37°C for 2 h.

Flow cytometric analysis

Flow cytometoric analysis was performed according to the previously published procedure [30,31]. Whole blood cells were labelled with phycoerythrin (PE)-conjugated anti-CD14 monoclonal antibodies and peridinin chlorophyll protein (PerCP)-conjugated anti-CD3 monoclonal antibodies, and then permeabilized with 4% paraformaldehyde in phosphate-buffered saline (PBS), pH 7·2, containing 0·1% saponin and 10 mm HEPES. The cells were then labelled with mouse anti-NF-κB (nuclear-localized signal) antibodies (IgG3; Boehringer Mannheim, Mannheim, Germany). The mouse anti-NF-κB (nuclear-localized signal) antibodies recognize an epitope overlapping the nuclear location signal of NF-κB-p65 and therefore selectively recognize the activated form of NF-κB. The cells were then labelled with fluorescein isothiocyanate (FITC)-conjugated rat antimouse IgG3 monoclonal antibodies (Pharmingen, San Diego, CA, USA). After washing, the cells were fixed with 1% paraformaldehyde in PBS and then stored at 4°C until flow cytometric analysis. Immunofluororescence staining was analysed with a FACScan flow cytometer equipped with CellQuest software (Becton-Dickinson Biosciences, San Jose, CA, USA). We analysed 5000 cells for each subject in the flow cytometric studies.

Statistical analysis

The differences in the results between groups were analysed by means of the Mann–Whitney U-test.

RESULTS

The percentages of cells exhibiting NF-κB activation among CD3⁺ T cells and CD14⁺ monocytes/macrophages on flow cytometry analysis are presented in Tables 2 and 3. The level of NF-κB activation in CD14⁺ monocytes/macrophages of the patients with asphyxia who had neurological sequelae was significantly higher than in the controls, and in the patients with asphyxia who did not have neurological sequelae (31·7 ± 7·2%versus 2·5 ± 0·9%, P = 0·008, and versus 1·6 ± 1·4%, P = 0·014, respectively). There were no significant differences among the three groups in the level of NF-κB activation in CD3⁺ T cells. The percentage of cells exhibiting NF-κB activation among LPS-stimulated CD14⁺ monocytes/macrophages of controls was 36·4 ± 11·6%.

Table 2.

Flow cytometric analysis data for the controls

	Percentage of cells exhibiting NF-xB activation(%)
No.	CD3+ cells	CD14+ cells	CD14+ cells stimulated by LPS
Controls
1	4·8	2·8	58·7
2	1·1	1·0	23·4
3	1·1	2·3	25·8
4	3·4	3·6	36·8
5	2·0	3·2	41·0
6	3·3	2·8	34·1
7	2·7	1·7	34·8
Mean ± s.d.	2·6 ± 1·4	2·5 ± 0·9	36·4 ± 11·6

Table 3.

Flow cytometric analysis data for the asphyxia group

	Percentage of cells exhibiting NF-kB
No.	CD3+ cells	CD14+ cells
Asphyxia group without sequelae
1	2·4	0·8
2	4·8	3·9
3	1·1	0·4
4	1·8	1·6
5	7·0	1·3
Mean ± s.d.	3·4 ± 2·4	1·6 ± 1·4
Asphyxia group with sequelae
6	7·2	26·6
7	8·0	33·9
8	19·7	23·0
9	0·0	39·1
Mean ± s.d.	8·7 ± 8·2	31·7 ± 7·2 *,**

^**P = 0·008, compared with controls;

^*P = 0·014, compared with asphyxia group with sequelae.

DISCUSSION

Previous studies have demonstrated that oxidative stress induces NF-κB activation [32,33]. In experimental animal models, NF-κB activation was found to be induced in the brain after ischaemia [26–29]. Moreover, a recent study demonstrated that NF-κB activation is induced in the brains of rats with perinatal asphyxia [34]. Proinflammatory cytokines, the inducible form of nitric oxide synthase (iNOS) which generates NO, and inducible cyclooxygenase (COX-2) which generates prostanoids, are induced via NF-κB activation [35]. Inflammation induced by NF-κB activation occurs in the ischaemic brain. In NF-κB subunit p50 knockout mice, ischaemic damage is reduced significantly [36]. Therefore, inflammation mediated by NF-κB activation is important as to the pathogenesis of hypoxic–ischaemic encephalopathy.

Our data revealed that NF-κB was activated in PBMC, especially CD14⁺ monocytes/macrophages, of the asphyxic neonates with neurological sequelae. The levels of NF-κB activation were the same as those in LPS-stimulated CD14⁺ monocytes/macrophages. How does NF-κB activation in the CNS affect PBMC? Proinflammatory cytokines, such as TNF-α and IL-1β, produced in the CNS may enter the peripheral blood and activate NF-κB in PBMC. Alternatively, peripheral blood T cells and monocytes/macrophages may enter the CNS by means of chemokines induced in the ischaemic brain and be activated, and then return to the peripheral blood. The level of NF-κB activation in PBMC may reflect the severity of hypoxic–ischaemic encephalopathy caused by neonatal asphyxia. The present study demonstrated that NF-κB activation in CD14⁺ monocytes/macrophages is related to neurological sequelae, but not that in CD3⁺ T cells. CD14⁺ monocytes/macrophages may be more sensitive than CD3⁺ T cells to hypoxic–ischaemic encephalopathy. Previous studies have revealed that activated monocytes/macrophages play an important pathogenic role in hypoxic and ischaemic brains [37–39]. CD14⁺ cells increase in both the perivascular space and the brain parenchyma in ischaemic brain lesions [37]. The mean infarct volume is significantly smaller in MCP-1^−/− mice in an ischaemic brain model, and immunostaining revealed a reduction of phagocytic macrophage accumulation in MCP-1^−/− mice [38]. Macrophage migration inhibitory factor, which is a cytokine that activates macrophages, is up-regulated in secondary brain damage after ischaemia and reperfusion stress [39].

Taking the role of monocytes/macrophages in hypoxic–ischaemic encephalopathy into consideration, our data suggest that NF-κB activation in peripheral blood CD14⁺ monocytes/macrophages in neonatal asphyxia is important for predicting the subsequent neurological sequelae.