Erythropoietin Attenuates Lipopolysaccharide-Induced White Matter Injury in the Neonatal Rat Brain

Periventricular leukomalacia (PVL), a common neonatal brain white matter (WM) lesion, is frequently associated with cerebral palsy. Growing evidence has indicated that in addition to ischemia/reperfusion injury, cytokine-induced brain injury associated with maternal or fetal infection may also play an important role in the pathogenesis of PVL. Recent studies have shown that administration of lipopolysaccharide (LPS) to pregnant rats causes enhanced expression of the cytokines, i.e., IL-1β, TNF-α, and IL-6, in fetal brains. In recent years, it has been shown that erythropoietin (EPO) has a critical role in the development, maintenance, protection and repair of the nervous system. In the present study we investigated the effect of EPO on LPS-induced WM injury in Sprague-Dawley rats. LPS (500 µg/kg) suspension in pyrogen-free saline was administered intraperitoneally to pregnant rats at 18 and 19 days of gestation. The control group was treated with pyrogen-free saline. They were given 5,000 U/kg recombinant human EPO. Seven-day-old Sprague-Dawley rat pups were divided into four groups: control group, LPS-treated group, prenatal maternal EPO-treated group (5,000 U/kg, intraperitoneally given to pregnant rats at 18 and 19 days of gestation), and postnatal EPO-treated group (5,000 U/kg, intraperitoneally given to 1-day-old rat pups). Cytokine induction in the postnatal 7-day-old (P7) rat brain after maternal administration of LPS was determined by the ELISA method. The proinflammatory cytokine levels (IL-1β, TNF-α, and IL-6) in P7 rat pup brains were significantly increased in the LPS-treated group as compared with the control group. Prenatal maternal EPO treatment significantly reduced the concentration of TNF-α and IL-6 in the newborn rat brain following LPS injection. The concentration of IL-1β was decreased in the intrauterine EPO treatment group. Postnatal EPO treatment significantly decreased only the IL-6 concentration in the newborn rat brain following LPS injection. The concentration of cytokines, IL-1β and TNF-α, was reduced in the postnatal EPO treatment group. We demonstrated here that LPS administration in pregnant rats at gestational day 18 and 19 induced WM injury in P7 progeny characterized by apoptosis. Prenatal maternal and postnatal EPO treatment significantly reduced the number of apoptotic cells in the periventricular WM. Using immunohistochemistry techniques, we investigated the effects of maternal administration of LPS on myelin basic protein (MBP) staining, as a marker of myelination in the periventricular area in the neonatal rat brain. MBP staining was significantly less and weaker in the brains of the LPS-treated group as compared with the prenatal maternal EPO-treated group. However, the postnatal EPO treatment did not prevent LPS-stimulated loss of MBP-positive staining. In conclusion, especially prenatal maternal EPO treatment attenuates LPS-induced injury by reducing the expression of inflammatory cytokines and sparing MBP in the neonatal rat brain. While the postnatal EPO treatment prevented LPS-induced brain injury this effect was partial. To our knowledge, this is the first study that demonstrates a protective effect of EPO on LPS-induced WM injury in the developing brain. Regarding the wide use of EPO in premature newborns, this agent may be potentially beneficial in treating LPS-induced brain injury in the perinatal period.