CXC chemokines generate age-related increases in
neutrophil-mediated brain inflammation and blood–brain barrier breakdown
Daniel Anthony*, Robert Dempster*, Sara Fearn*, John Clements
†
,
Graham Wells
†
, V. Hugh Perry* and Katharine Walker*
‡
Children are at greater risk than adults of permanentbrain damage and mortality following head injury or infection [1–5]. Rodent models have demonstrated a ‘window of susceptibility’ in young animals during which the brain parenchyma is at greater risk of acute
neutrophil-mediated breakdown of the blood–brain barrier [6,7]. The exact mechanism of this age-related susceptibility to brain inflammation has yet to be defined, but animal models have revealed that the potent pro-inflammatory cytokine interleukin-1b (IL-1b) initiates an intense acute neutrophil-mediated
inflammatory response in the brains of young rats and mice that is not seen in adults [6]. Here, we demonstrate the rapid induction of CXC chemokines (which contain a Cys–X–Cys motif), in particular the cytokine-induced neutrophil chemoattractant CINC-1, following the intra-cerebral administration of IL-1b. The CXC chemokines produced a more intense neutrophil response in young rats than in adults. The IL-1b-induced blood–brain barrier breakdown in young rats could be attenuated by an anti-CINC-1 neutralising antibody. These results show that the immature central nervous system (CNS) is dramatically more susceptible to the chemotactic effects of CXC chemokines. Blocking the CXC chemokine activity associated with brain inflammation inhibits neutrophil-mediated blood–brain barrier damage and represents a significant therapeutic possibility.
Addresses: *Department of Pharmacology, University of Oxford, Mansfield Road, Oxford OX1 3QT, UK. †British Biotech Ltd,
Watlington Road, Cowley, Oxford, UK.
Present address: ‡Novartis Institute for Medical Sciences, 5 Gower
Place, London WC1E 6BN, UK. Correspondence: Katharine Walker E-mail: [email protected] Received: 1 May 1998 Revised: 5 June 1998 Accepted: 24 June 1998 Published: 27 July 1998 Current Biology1998, 8:923–926 http://biomednet.com/elecref/0960982200800923 © Current Biology Publications ISSN 0960-9822
Results and discussion
The early expression of CXC chemokines that induce neu-trophil chemotaxis was examined in adult and juvenile rat brain to see whether these chemokines might underlie the
acute pro-inflammatory effects of IL-1β. A competitive reverse transcriptase (RT)–PCR assay [8] was used to determine the level of acute mRNA expression for the rat CXC chemokines CINC-1 and macrophage inhibitory protein-2 (MIP-2) following intra-cerebral administration of rat recombinant (rr) IL-1β [9,10]. After extracting mRNA, the cDNA was reverse transcribed from it and amplified by PCR in the presence of known concentrations of a synthetic standard plasmid containing primer sites for rat β-actin, CINC-1 and MIP-2 that yielded products dis-tinguishable by size from the PCR products copied from the reverse-transcribed mRNA (see Supplementary mater-ial, published with this paper on the internet). The assay revealed a rapid increase in both MIP-2 and CINC-1 mRNAs in the groups of rats treated with IL-1β, whereas these mRNAs were virtually undetectable in saline-treated controls (Figure 1a). CINC-1 mRNA, in particular, showed a 10,000-fold increase in expression in IL-1β-treated brain. Using a rat CINC-1 enzyme-linked immunosorbent assay (ELISA) we found that CINC-1-like immunoreactivity was also increased in adult and juvenile rats 4 hours follow-ing IL-1βadministration (Figure 1b).
Despite previous demonstrations of more intense IL-1β -induced neutrophil recruitment and blood–brain barrier breakdown in juvenile rats [6], the experiments described above indicate that the expression of CXC chemokines, particularly CINC-1, occurs rapidly and to much the same level in the brains of adults and juveniles. We therefore examined whether, like IL-1β, the intra-cerebral adminis-tration of CXC chemokines would precipitate a different pattern of responses in adult and juvenile brains. Striking age-related differences were observed in the effects of both rrCINC-1 and rrMIP-2 (Figure 2). Whereas in control rats the microinjection of vehicle produced no neutrophil recruitment or blood–brain barrier breakdown measured as horseradish peroxidase (HRP) extravasation, microinjection of CINC-1 induced profound HRP extravasation and neutrophil recruitment as early as 4 hours following its administration to juvenile rat brain (Figures 2a,3a,3c). Adult rats showed significantly less neutrophil recruitment and HRP extravasation at the 4 hour time-point (Figures 2b,3a,3c). CINC-1-induced neutrophil recruitment was observed in adult rats, particu-larly at 12 hours after CINC-1 administration (Figure 3a); the overall neutrophil response was lower and delayed (analysis of variance, p < 0.05), however, compared with juveniles. HRP extravasation following CINC-1 adminis-tration was rapid and extensive in juveniles, showing a
gradual decline over 24 hours (Figure 3c). In adults, however, HRP extravasation was almost absent at the early time-points and never reached the maximal response displayed in juvenile brain (Figure 3c). MIP-2 also induced a more intense response in juvenile rats than in adults, although in comparison with CINC-1 the onset of neutrophil recruitment and HRP extravasation induced by MIP-2 was delayed (Figures 2c,3b,3c). In contrast to the rapid and widespread effects of CINC-1, MIP-2 induced intense, but focal, neutrophil recruitment to the striatum from 12 to 24 hours following its administration (Figure 3b). MIP-2 induced less HRP extravasation than CINC-1 (Figure 3c). In keeping with the putative neu-trophil-selective chemotactic effects of these CXC chemokines, almost all the leukocytes that were recruited within the 24 hours following the administration of CINC-1 and MIP-2 were neutrophils (Figure 3a,b). The HRP extravasation induced by both CINC-1 and MIP-2 was also neutrophil-dependent, because it was inhibited following the depletion of circulating neutrophils with an anti-neutrophil antiserum (Figure 2d).
Given these results, we decided to test whether the intense inflammatory response and blood–brain barrier breakdown induced by IL-1β in juvenile rats could be attenuated by the administration of an anti-rat CINC-1 neutralising antibody. Juvenile rats treated with IL-1β plus a control non-specific rabbit immunoglobulin G (IgG) displayed a brain inflammatory response 4 hours following treatment that was comparable to that previ-ously described for IL-1β[6] (Figures 2e,4). The IL-1β -treated brains showed extensive blood–brain barrier breakdown with concomitant intraparenchymal neu-trophil recruitment. Injection of the non-specific rabbit IgG or the anti-CINC-1 neutralising antibody alone did not produce an inflammatory response (Figure 4a). In marked contrast, animals treated with IL-1β plus the 924 Current Biology, Vol 8 No 16
Figure 1
Microinjection of rrIL-1β(1 ng) into adult or juvenile rat striatum induces expression of CXC chemokines. (a) Quantitation of rat CINC-1 and MIP-2 mRNA (measured via quantitation of cDNA) in rat striatum 4 h following the injection of rrIL-1βor saline. The ratio of standard cDNA to cellular cDNA was plotted on a log–log scale against the dilution of standard. The dilution value when the ratio of standard to cellular cDNA was equal to 1 was calculated [8]. Log mean ± SEM (n = 3 rats per group) cDNA levels are expressed as a percentage of β-actin cDNA levels for each sample. An asterisk denotes a significant difference (p < 0.01) between the IL-1-treated and saline-treated groups. (b) Expression of rat CINC-1-like immunoreactivity 4 h following IL-1βor saline administration to adult and
juvenile rat striatum (mean ± SEM; n = 4 rats per group). An ELISA (rat CINC-1 ELISA kit, IBL [11]) of each sample was carried out in
triplicate and quantitation of CINC-1-like immunoreactivity was determined using a standard curve of rrCINC-1.
(a) Current Biology (b) * * * 30 25 20 15 CINC-1-like immunoreactivity (ng) per g tissue
cDNA levels compared
to β -actin (%) 10 5 0 10–4 Adult IL-1 Juvenile IL-1 Adult Saline CINC-1
MIP-2 JuvenileAdult
Juvenile Saline
IL-1 Saline Naive 10–3
10–2 10–1
*
Figure 2
Blood–brain barrier breakdown induced by CXC chemokines in adult and juvenile rat brain measured by HRP extravasation. (a) Juvenile (3-week-old) and (b) adult (3-month-old) rat brains 4 h following an intra-striatal injection of 0.5µg rrCINC-1 (n = 3 rats per group). The extent of HRP extravasation in juvenile rat brain 12 h following CINC-1 administration was similar to that at 4 h (see Figure 3c). (c) Juvenile rat brain 12 h following an intra-striatal injection of 0.5µg rrMIP-2. (d)Inhibition of CINC-1-induced (0.5µg) HRP extravasation in juvenile rat brain by depletion of circulating neutrophils following systemic administration of a rat neutrophil-depleting antiserum. (e) HRP extravasation in juvenile rat brain 4 h following intra-striatal injection of rrIL-1β(1 ng) plus control IgG (1µg). (f) This IL-1β-induced HRP extravasation was inhibited by co-administration of rrIL-1βplus an anti-CINC-1 neutralising antibody (1µg).
anti-CINC-1 neutralising antibody showed a significant reduction in neutrophil recruitment and almost complete absence of HRP extravasation (Figures 2f,4).
These results demonstrate that the intense neutrophil recruitment and concomitant blood–brain barrier break-down induced by IL-1β in the brains of juvenile rats is largely accounted for by an increased susceptibility to the neutrophil chemotactic effects of CXC chemokines. An anti-rat CINC-1 neutralising antibody significantly inhib-ited both neutrophil recruitment to the juvenile brain parenchyma and IL-1β-induced HRP extravasation. The latter finding is consistent with the previous demonstra-tion that IL-1β-induced HRP extravasation in juvenile rats is neutrophil dependent [6]. Yamasaki et al. [11] have also demonstrated that systemic administration of an anti-CINC-1 neutralising antibody can reduce increases in
brain water content following ischemic reperfusion injury in adult rats. The increased neutralising effect of the anti-CINC-1 antibody seen in this study compared with that of Yamasaki et al. [11] is likely to be due to IL-1β-induced chemokine synthesis and presentation by CNS cells, such as astrocytes [12].
The induction of cytokines by CNS cells has been reported in conjunction with many different neuropatholo-gies [13,14]. Nevertheless, experiments in rats and mice have shown that IL-1β-induced leukocyte recruitment and plasma extravasation in the adult brain parenchyma is slower and attenuated compared with that seen in periph-eral tissues [15]. This inherent resistance to brain inflam-mation offers a direct advantage to the host in terms of protecting neural tissue from permanent damage. Compar-ison of the dose–response relationships for IL-1β-induced Figure 3
Adult ANS+ cells Juvenile ANS+ cells Adult ED1+ cells Juvenile ED1+ cells
Adult ANS+ cells Juvenile ANS+ cells Adult ED1+ cells Juvenile ED1+ cells
CINC-1-treated adults CINC-1-treated juveniles MIP-2-treated adults MIP-2-treated juveniles 12 4 0 500 * * * * * * * 1000 1500 2000 (a)
Positively stained cells per mm
2
Time following CINC-1 administration (h) Time following MIP-2 administration (h) Time following chemokine administration (h)
Current Biology 0 500 1000 1500 2000 (b)
Positively stained cells per mm
2 0 2 6 4 8 10 (c) Area of HRP extravasation (mm 2) 24 4 12 24 4 12 24
Inflammation and blood–brain barrier breakdown induced by CXC chemokines in adult and juvenile rat brain. Leukocyte recruitment to adult and juvenile striatum, measured at the intervals shown, following an injection of (a) rrCINC-1 (0.5µg) or (b) rrMIP-2 (0.5µg). Polymorphonuclear neutrophils were identified using a rat polyclonal anti-neutrophil antibody (ANS) [18]. Macrophages were identified
using the monoclonal antibody ED1 (donated by the Dunn School of Pathology). (c) HRP extravasation in adult and juvenile brain 4 h following an injection of rrCINC-1 (0.5µg) or rrMIP-2 (0.5µg). An asterisk denotes a significant difference between the respective adult and juvenile groups (mean ± SEM; n = 3 rats per group; t test, p < 0.05).
Figure 4
Inhibitory effects of an anti-CINC-1 neutralising antibody on inflammation and blood–brain barrier breakdown in juvenile rats. (a)Neutrophil recruitment to juvenile brain parenchyma 4 h following co-administration of either rrIL-1β(1 ng) plus control IgG (1µg; positive control), rrIL-1βplus anti-CINC-1 neutralising antibody (1µg), or anti-CINC-1 neutralising antibody (1µg) alone
(mean ± SEM; n = 4 rats per group). Because polyclonal antibodies had already been administered, neutrophils were identified using the monoclonal antibodies OX1 and OX30 (combined in a 1:1 ratio; donated by the Dunn School of Pathology) against leukocyte common antigen (LCA) and by their
characteristic polylobar nuclei. (b) HRP extravasation in the brains of the same animals shown in (a). An asterisk denotes a significant
difference between the indicated bars,
p < 0.01 (mean ± SEM; n = 4 rats per group).
* IL-1 + IgG 0 200 400 Neutrophils per mm 2 600 800 (a) IL-1 + anti-CINC-1
Anti-CINC-1 IL-1 + IgG 0 2 4 6 8 10 Area of HRP extravasation (mm 2) 12 (b) IL-1 + anti-CINC-1 Anti-CINC-1 Current Biology *
inflammation in the brains of young and adult rats, however, indicates that the brains of young rats are more susceptible to the inflammatory effects of IL-1β [6,7]. Neither the postnatal development of the blood–brain barrier nor the differential expression of adhesion mole-cules mediate these age-related differences [6,7,16]. The inhibition of blood–brain barrier breakdown by the deple-tion of circulating neutrophils or by an anti-CINC-1 neu-tralising antibody clearly shows, however, that it is the neutrophil chemotactic effects of CXC chemokines that are most destructive to the juvenile rat brain.
Once in the brain, neutrophils are potent contributors to the acute inflammatory response and can promote neural injury. This has been clearly demonstrated in transgenic mice expressing the CXC chemokine N51/KC in brain oligodendrocytes [17]. Neutrophil-mediated blood–brain barrier breakdown is a mechanism by which CNS tissue can be damaged. Neurological studies have demonstrated that alterations in blood–brain barrier breakdown perme-ability, and cerebral oedema in children and infants with bacterial meningitis, are directly associated with poor prognosis [5]. Several clinical studies have noted the increased risk of permanent brain damage and poor prog-nosis following head injury or cerebral infection in young children [4,5]. Mortality rates following head injury are much higher in children [3] and acute infections, such as meningitis and malaria, are not only more frequent but also more severe in young children than in adults [1,2]. Clearly, if the childhood risks associated with brain injury and infection are due to an increased susceptibility to neu-trophil-mediated brain inflammation then the prevention of neutrophil recruitment to the brain by antagonising the effects of CXC chemokines would represent a significant therapeutic possibility.
Materials and methods
Wistar rats (Charles River) received intra-striatal microinjections (maximum volume, 1µl) as described [6]. Rat recombinant IL-1β (National Institute for Biological Standards and Controls) was dis-solved in a solution of 0.1% bovine serum albumin in PBS, or mixed with an anti-rat CINC-1 neutralising antibody (Peprotech) or rabbit IgG (Serotec). Rat recombinant MIP-2 and CINC-1 (Peprotech) were freshly made up from lyophilised protein immediately before use, and dissolved in sterile filtered 0.9% saline. Assessment of the blood–brain barrier with type II HRP (Sigma), 104U/kg, as a
blood-borne tracer was carried out as described [6]. Quantitation of the area of HRP extravasation from brain sections was calculated using Optilab 2.1 for Macintosh software (Graphtec). Immunohistochemistry was used to confirm the presence and distribution of specific cell popula-tions [6]. For PCRs and ELISAs, rats were killed 4 h following IL-1β administration, brains were removed and the striatums from the injected hemisphere were dissected and quickly frozen in liquid nitro-gen. A neutrophil-depleting antiserum (Acurate) was used to deplete circulating neutrophils in 3-week-old rats, as described [6], and resulted in profound neutrophil depletion.
Supplementary material
A table showing primer sequences and amplicon lengths for the com-petitive RT–PCR assay is published with this paper on the internet.
Acknowledgements
This study was funded by a Wellcome Trust Fellowship (number 045972/Z/95/Z) to K.W.
References
1. Jakobsen PH, Bate CAW, Taverne J, Playfair JHL: Malaria — toxins,
cytokines and disease. Parasite Immunol 1995, 17:223-231. 2. Rosenfeld EA, Rowley AH: Infectious intracranial complications of
sinusitis, other than meningitis, in children: 12-year review. Clin
Infect Dis 1994, 18:750-754.
3. Sharples PM, Stuart AG, Matthews DSF, Aynsleygreen A, Eyre JA:
Cerebral blood-flow and metabolism in children with severe head-injury. 1. Relation to age, Glasgow Coma Score, outcome, intracranial-pressure, and time after injury.J Neurol Neurosurg Psychiatry 1995, 58:145-152.
4 Tuomanen E: Susceptibility to infection and age-related-changes
in the blood-brain-barrier.Neurobiol Aging 1994, 15:757-758.
5. Scott BL, Jankovic J: Delayed-onset progressive movement
disorders after static brain lesions.Neurology 1996, 46:68-74.
6. Anthony DC, Bolton SJ, Fearn S, Perry VH: Age-related effects of
interleukin-1 beta on polymorphonuclear neutrophil-dependent increases in blood-brain barrier permeability in rats.Brain 1997,
120:435-444.
7. Lawson LJ, Perry VH: The unique characteristics of inflammatory
responses in mouse brain are acquired during postnatal development.Eur J Neurosci 1995, 7:1584-1595.
8. Wells GMA, Catlin G, Cossins JA, Mangan M, Ward GA, Miller KM, et
al.: Quantitation of matrix metalloproteinases in cultured rat
astrocytes using the polymerase chain reaction with a multi-competitor cDNA standard.Glia 1996, 18:332-340.
9. Driscoll KE, Hassenbein DG, Howard BW, Isfort RJ, Cody D, Tindal MH, et al.: Cloning, expression, and functional characterization of
rat MIP-2: a neutrophil chemoattractant and epithelial cell mitogen.J Leukoc Biol 1995, 58:359-364.
10. Suzuki H, Suematsu M, Miura S, Liu YY, Watanabe K, Miyasaka M, et
al.: Rat CINC/Gro - a novel mediator for locomotive and
secretagogue activation of neutrophils in vivo. J Leukoc Biol 1994,
55:652-657.
11. Yamasaki Y, Matsuo Y, Zagorski J, Matsuura N, Onodera H, Itoyama Y, et al.: New therapeutic possibility of blocking cytokine-induced
neutrophil chemoattractant on transient ischemic brain damage in rats.Brain Res 1997, 759:103-111.
12. Ransohoff RM, Tani M: Do chemokines mediate leukocyte
recruitment in post-traumatic CNS inflammation?Trends Neurosci
1998, 21:154-159.
13. Matsumoto T, Yokoi K, Mukaida N, Harada A, Yamashita J, Watanabe Y, et al.: Pivotal role of interleukin-8 in the acute respiratory
distress syndrome and cerebral reperfusion injury.J Leukoc Biol
1997, 62:581-587.
14. Rothwell NJ, Luheshi G, Toulmond S: Cytokines and their receptors
in the central-nervous-system - physiology, pharmacology, and pathology.Pharmacol Ther 1996, 69:85-95.
15. Andersson PB, Perry VH, Gordon S: Intracerebral injection of
proinflammatory cytokines or leukocyte chemotaxins induces minimal myelomonocytic cell recruitment to the parenchyma of the central nervous system.J Exp Med 1992, 176:255-259.
16. Bell MD, Perry VH: Adhesion molecule expression on murine cerebral
endothelium following the injection of a proinflammagen or during acute neuronal degeneration.J Neurocytol 1995, 24:695-710.
17. Tani M, Fuentes ME, Peterson JW, Trapp BD, Durham SK, Loy JK, et
al.: Neutrophil infiltration, glial reaction, and neurological disease
in transgenic mice expressing the chemokine N51/KC in oligodendrocytes. J Clin Invest 1996, 98:529-539. 18. Anthony DC, Miller KM, Fearn S, Townsend MJ, Wells GMA,
Clements JM, et al.: Matrix metalloproteinase expression in an
experimentally-induced DTH model of multiple sclerosis in the rat
CNS.J Neuroimmunol 1998, 87:62-73.
CXC chemokines generate age-related increases in
neutrophil-mediated brain inflammation and blood–brain barrier breakdown
Daniel Anthony, Robert Dempster, Sara Fearn, John Clements, Graham Wells,
V. Hugh Perry and Katharine Walker
Current Biology
27 July 1998, 8:923–926
Table S1
Primer sequences and amplicon lengths for the CXC chemokine competitive RT–PCR assay.
EMBL accession Forward primer Reverse primer Standard amplicon cDNA amplicon
number sequence (5′→3′) sequence (5′→3′) length (bp) length (bp)
Rat β-actin L00981 TCCTTCCGCCTATGGAATC ACTCATCGTACTCCTGCTTG 205 300
Rat CINC-1 M86536 GCTCGCTTCTCTGTGCAGC CCATCGGTGCAATCTATCTTC 204 298
