MINOCYCLINE ATTENUATES LIPOPOLYSACCHARIDE-INDUCED WHITE MATTER INJURY IN THE NEONATAL RAT BRAIN L.-W. FAN, Y. PANG, S. LIN, P. G. RHODES AND Z. CAI* Department of Pediatrics, Division of Newborn Medicine, University of Mississippi Medical Center, Jackson, MS 39216-4505, USA Abstract—Our previous studies have shown that intracere- bral administration of endotoxin, lipopolysaccharide (LPS), induces selective white matter injury and hypomyelination in the neonatal rat brain and that the LPS-induced brain injury is associated with activation of microglia. To test the hypothe- sis that inhibition of microglial activation may protect against LPS-induced white matter injury, we examined roles of mino- cycline, a putative suppressor of microglial activation, on LPS-induced brain injury in the neonatal rat. A stereotactic intracerebral injection of LPS (1 mg/kg) was performed in postnatal day 5 Sprague–Dawley rats and control rats were injected with sterile saline. Minocycline (45 mg/kg) was ad- ministered intraperitoneally 12 h before and immediately after LPS injection and then every 24 h for 3 days. Inflammatory responses, activation of microglia and brain injury were ex- amined 1 and 3 days after LPS injection. LPS injection re- sulted in brain injury in selective brain areas, including bilat- eral ventricular enlargement, cell death at the sub- and periventricular areas, loss of O4 and O1 oligodendrocyte (OL) immunoreactivity and hypomyelination, as indicated by decreased myelin basic protein immunostaining, in the neo- natal rat brain. Minocycline administration significantly atten- uated LPS-induced brain injury in these rat brains. The pro- tective effect of minocycline was associated with suppressed microglial activation as indicated by the decreased number of activated microglial cells following LPS stimulation and with consequently decreased elevation of interleukin 1 and tu- mor necrosis factor- concentrations induced by LPS and a reduced number of inducible nitric oxide synthase express- ing cells. Protection of minocycline was also linked with the reduction in LPS-induced oxidative stress, as indicated by 4-hydroxynonenal positive OLs. The overall results suggest that reduction in microglial activation may protect the neo- natal brain from LPS-induced white matter injury and inhibi- tion of microglial activation might be an effective approach for the therapeutic treatment of infection-induced white mat- ter injury. © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: minocycline, lipopolysaccharide, white matter injury, oligodendrocyte, microglial activation, oxidative stress. Periventricular leukomalacia (PVL), a form of white matter disease closely associated with cerebral palsy, remains a major problem in premature infants. Increasing evidence has indicated that in addition to ischemia/reperfusion in- jury, cytokine-induced brain injury associated with mater- nal or fetal infection may also play an important role in the pathogenesis of PVL (Hagberg et al., 2002; Volpe, 2003). Recent studies have also demonstrated that late oligoden- drocyte (OL) progenitors (O4+/O1-), which are the pre- dominant OL lineage during the peak period of PVL (i.e. 24 –32 gestation weeks), are the major target of brain injury in infant brain with PVL (Back et al., 2001, 2002). In our previous studies, we have developed a neonatal rat model to mimic the scenario of infection through intra- cerebral injection of lipopolysaccharide (LPS) in postnatal day 5 (P5) rat brain (Pang et al., 2003; Cai et al., 2003). LPS, a component of the cell wall of gram-negative bac- teria, is responsible for most of the inflammatory effects of infection from gram-negative bacteria. LPS has been de- tected in the amniotic fluid (Romero et al., 1987). It is possible that LPS may reach the fetal brain during mater- nal infection. We have used a neonatal rat model, instead of a maternal–fetal model, because the rat brain white matter develops postnatally. Late OL progenitor cells, the major target in cerebral white mater injury in human infants (Back et al., 2002), represent the predominant OL lineage stage in the rat cerebral hemispheres between P2 and P7 (Back et al., 2002). Using this rat model, we found that LPS resulted in selective white matter injury in the neonatal rat brain and that inflammatory cytokines and microglial acti- vation play important roles in LPS-induced brain injury (Pang et al., 2003; Cai et al., 2003). Microglia, which are the only nonneuronal cell type that expresses the Toll-like receptor 4 (TLR4), have been identified as the major LPS- responsive cell in the CNS (Lehnardt et al., 2002). Activa- tion of innate immunity through the TLR4 plays a critical role in LPS-induced OL injury (Lehnardt et al., 2002) and neuronal and axonal injury (Lehnardt et al., 2003). There- fore, we hypothesized that reduction in activation of micro- glia might provide protection against LPS-induced brain injury. Minocycline is a semi-synthetic second-generation de- rivative of tetracycline. Tetracyclines are bacteriostatic agents with broad-spectrum anti-microbial activity. Mino- cycline has a superior tissue penetration into the brain and is absorbed rapidly and completely (Aronson, 1980). Re- *Corresponding author. Tel: +1-601-984-2786; fax: +1-601-815-3666. E-mail address: zcai@ped.umsmed.edu (Z. Cai). Abbreviations: DAPI, 4=,6-diamidino-2-phenylindole; GFAP, glial fibrillary acidic protein; H&E, hematoxylin and eosin; 4-HNE, 4- hydroxynonenal; IL1-, interleukin 1-; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; MBP, myelin basic protein; NO, nitric oxide; OL, oligodendrocyte; P, postnatal day; PBS, phosphate- buffered saline; PN, polymorphonuclear; PVL, periventricular leu- komalacia; ROS, reactive oxygen species; RT, room temperature; TLR4, Toll-like receptor 4; TNF, tumor necrosis factor-; TUNEL, terminal deoxynucleotidyl transferase (TdT)-mediated uridine 5=- triphosphate-biotin nick end labeling. Neuroscience 133 (2005) 159 –168 0306-4522/05$30.00+0.00 © 2005 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2005.02.016 159