The changes of oligodendrocytes induced by anesthesia during brain development

With the advent of modern techniques, drugs, and monitoring, general anesthesia has come to be considered an unlikely cause of harm, particularly for healthy patients. While this is largely true, newly emerging clinical and laboratory studies have suggested that exposure to anesthetic agents during early childhood may have long-lasting adverse eff ects on cognitive function. This concern has been the focus of intense study in the fi eld of anesthesia research. A recent high-profi le review by Rappaport et al. (2015) concluded that while many questions remain unanswered, there is strong evidence from laboratory studies that commonly used anesthetics interfere with brain development and that clinical studies suggest a correlation between early childhood exposure to these agents and subsequent eff ects on learning and cognition. The issue is of suffi cient public health importance that a public-private partnership known as Smar-Tots (Strategies for Mitigating Anesthesia-Related Neurotoxicity in Tots) was developed by the FDA to study pediatric anesthetic neurotoxicity. The mechanism of injury underlying this phenomenon has yet to be fully elucidated, and there is evidence to suggest that anesthetics may have direct cytotoxic eff ects on neurons leading to cell death or suppressed neurogenesis (Stratmann et al., 2010) and that they may interfere with key processes in neuronal growth and development that underlie brain circuit development (Wagner et al., 2014).

Nearly all of the focus of basic science investigations of mechanisms of anesthetic toxicity has fallen on the eff ects of anesthetics on neurons, and strikingly little attention has been paid to possible mechanisms of injury mediated through eff ects on glial cells. Both the development and mature function of neurons depends heavily on the two principal species of glia: astrocytes, which support neuronal function via a variety of biochemical and structural roles and oligodendrocytes, which provide myelin to electrically insulate neuronal transmission. This article highlights evidence that anesthetic exposure during development may have toxic eff ects on astrocytes and oligodendrocytes, including recent evidence that these eff ects could lead to impairments in the function of the neuronal circuitry that underlies cognition.

Currently, only a few studies exist which have examined the effects of anesthetic exposure on astrocytes in the context of brain development. Todorovic and co-workers fi rst opened this line of investigation by describing injury to developing astrocytes that resulted from a 4 hour exposure to 3% isofl urane in a primary culture model. The authors describe a substantial disruption of the actin cytoskeleton, particularly evident in examination of the stress fi bers, that is accompanied by a reduction in paxillin, a marker of focal adhesions, and that correlates with reduced levels of RhoAGTPase, a signaling molecule that is critical for the development of normal glial morphology (Lunardi et al., 2011). Another study of the eff ects of isofl urane on astrocytes in vitro that used a lower dose, a 4 hour exposure to 1.4% isofl urane, revealed no changes in, phalloidin binding, a gross measure of F-actin levels, but instead showed a decrease in glial fi brillary acid protein expression and a decrease in alpha tubulin levels (Culley et al., 2013). These data provide strong evidence that isofl urane, the benchmark potent volatile anesthetic, impairs the cytoskeleton in developing astrocytes. The cytoskeletal eff ects do not appear to be lethal for astrocytes, as assays of astrocyte survival after anesthesia exposure show no signifi cant change from controls. Given the fundamental nature of the cytoskeleton for cellular function, it is likely that such a disruption would have some effect on glial activity, but neither of these manuscripts uncovered a functional consequence related to brain development.

In a study by Ryu et al., (2014), we asked whether isofl urane exposure in astrocytes impaired the ability of astrocytes to support neuronal growth, one of their key roles in early development. Culley et al. (2013) had previously addressed this question by asking whether neurons grown in conditioned media from astrocytes treated with isofl urane manifested a change in synapse number compared to neurons growth in control conditioned media. They found no change, and thus concluded that isofl urane treatment did not alter the ability of astrocytes to support neuronal growth. We took a diff erent approach, in which we measured axonal growth in naïve neurons co-cultured with astrocytes that had been previously exposed to isofl urane or control conditions (Ryu et al., 2014). We found that axon length was reduced in the isofl urane treated co-culture group in a dose dependent manner over a range from 2.4% to 3.6% of 5 hour treatment with isofl urane, and furthermore that isofl urane treated astrocytes remained less capable of supporting axonal growth for up to 24 hours. The cues that guide targeted axonal growth are only transiently expressed, and thus axons must grow at the correct speed in order to encounter the appropriate cues for the duration of their extension towards synaptic targets.

Our fi ndings could not be explained by reduced numbers of astrocytes since we did not observe any cell death resulting from isofl urane exposure, so we examined other putative mechanisms. We found that levels of brain derived neurotrophic factor (BDNF) were reduced in isoflurane treated co-culture media and that axon outgrowth in the isofl urane treated group could be rescued to normal levels by adding exogenous BDNF to the co-culture (Ryu et al., 2014). BDNF is also critical for synaptogenesis, and thus it is likely that the absence of an eff ect on synapses seen previously was due to the diff erence in paradigm. Our eff ect was only evident at 2.4%, and not at a lower dose of 1.2% that is more consistent with the 1.4% dose used by Culley and co-workers. Alternately, and perhaps equally likely, the eff ect may be due to the relationship between BDNF and astrocytes. In conditions of injury and infl ammation, astrocytes can produce high levels of BDNF, but in normal development, the predominance of BDNF is secreted by neurons, and the role of astrocytes is thought to be in uptake and recycling of this growth factor (Bergami et al., 2008), a process that would not have been observed in a conditioned media model, but which would have been evident in a co-culture paradigm. As astrocytes co-exist with neurons in the developing brain in vivo, these data suggest that isofl urane may inhibit neuronal growth indirectly via impairment of astrocyte eff ects on BDNF levels. The mechanism by which isofl urane might act on astrocytes to impair BDNF recycling is unclear, and we speculate that the structural defi cits discussed above may contribute to an inability of the astrocytes to project processes into appropriate proximity with neurons to serve this function. Alternately, both the cytoskeletal disruption and their reduced capacity to support neuronal growth may both be consequences of metabolic impairment caused by isofl urane, which has been shown to cause mitochondrial injury during development (Sanchez et al., 2011).

The effects of developmental exposure to anesthetics on oligodendrocytes appear to refl ect the potential cytotoxicity properties of anesthetics. Seminal work in pediatric anesthetic

toxicity by Jevtovic-Todorovin et al. (2013) showed that commonly used anesthetics cause an increase in neuronal apoptosis in rodents. Depending on the dose and timing of exposure, anesthetics have been shown to cause cell death, but the literature on this topic is complex and highly dependent on context and cell type. None of the studies of the eff ects of anesthetics on astrocytes in vitro discussed above revealed any cytotoxicity, and similar work in vivo support this conclusion. Brambrink and co-workers exposed postnatal day 6 rhesus macaques to fi ve hoursof isofl urane administered between 0.7% and 1.5% to ablate movement and physiologic response to painful stimulus, using physiologic monitoring and support consistent with standards applied in the clinical practice of anesthesia (Brambrink et al., 2012). Brains were harvested 3 hours after the termination of anesthesia exposure, and a substantial increase in apoptosis relative to unexposed controls was appreciated as measured by quantitative microscopy for activated caspase and Hoechst labeling in serial sections throughout the brain. Unlike astrocytes and microglia, the oligodendrocyte population showed a marked increase in apoptosis after isofl urane treatment (Brambrink et al., 2012). Using histological analysis, they estimated that this eff ect amounted to a loss of 6.3% of total population of myelinating oligodendrocytes. These degenerative changes were not confi ned to any single anatomical location, but rather were diff usely present throughout the brain. Interestingly, the same group of researchers reported a similar eff ect with propofol, an intravenous anesthetic commonly used in clinical practice (Creeley et al., 2013). In a study of postnatal day 6 and fetal gestational day 120 rhesus macaques using the same methodology discussed above, a substantial increase in oligodendrocyte apoptosis was detected in isofl urane exposed brains. Deletion of myelinating oligodendrocytes could cause pathological changes in axonal structure or function. In addition, given that one oligodendroglial cell will contribute myelin to several axons, a loss of myelinating oligodendrocytes could be expected to broadly disrupt myelination and nerve conduction for multiple axons within the fi eld of a single oligodendrocyte. However, these studies were conducted only at a single, early time-point, without any long term follow-up. Thus, it is possible that a recovery from injury might occur, particularly as the authors note that oligodendrocyte precursors in the isofl urane exposed groups did not undergo increased apoptosis, but rather appeared to be engaged in reactive proliferation.

The evidence strongly suggests that glia are a possible target of anesthetic toxicity. Astrocytes clearly undergo cytoskeletal disruption as a result of early anesthetic exposure and their capacity to support neuronal growth is transiently impaired. These results have been obtained using dissociated culture models, and must still be translated to the context of the intact brain. Anesthetics induce apoptotic cell death in a substantial population of oligodendrocytes in the developing brain, but the signifi cance of this fi nding will be greatly enhanced if it is shown to be persistent. While clear eff ects of anesthetics on astrocytes and oligodendrocytes have been demonstrated, a possible role for microglia has not been explored to date. Recent evidence has shown that microglia play an important role in normal brain development (Michell-Robinson et al., 2015), and anesthetics have been shown to modulate microglial cytokine release in an in vitro model of infl ammation (Ye et al., 2013). This fi nding suggests the possibility that anesthetics might alter brain development via eff ects on microglia. The currently available literature on the eff ects of anesthetics on glia in the developing brain is fairly sparse compared to studies of developmental anesthetic toxicity in neurons, and we conclude that this is a promising area for further study in the fi eld of pediatric anesthetic neurotoxicity.

This study was supported by NIH 1K08GM104329 to CDM.

Danye Jiang, Sanghee Lim, Minhye Kwak, Yun Kyoung Ryu, C. David Mintz*

Department of Anesthesiology and Critical Care Medicine, Johns Hopkins Medical Institutes, Baltimore, MD, USA

*Correspondence to: C. David Mintz, M.D., Ph.D.,

cmintz2@jhmi.edu.

Accepted: 2015-06-29

Bergami M, Santi S, Formaggio E, Cagnoli C, Verderio C, Blum R, Berninger B, Matteoli M, Canossa M (2008) Uptake and recycling of pro-BDNF for transmitter-induced secretion by cortical astrocytes. J Cell Biol 183:213-221.

Brambrink AM, Back SA, Riddle A, Gong X, Moravec MD, Dissen GA, Creeley CE, Dikranian KT, Olney JW (2012) Isofl urane-induced apoptosis of oligodendrocytes in the neonatal primate brain. Ann Neurol 72:525-535.

Creeley C, Dikranian K, Dissen G, Martin L, Olney J, Brambrink A (2013) Propofol-induced apoptosis of neurones and oligodendrocytes in fetal and neonatal rhesus macaque brain. Br J Anaesth 110 Suppl 1:i29-38.

Culley DJ, Cotran EK, Karlsson E, Palanisamy A, Boyd JD, Xie Z, Crosby G (2013) Isofl urane aff ects the cytoskeleton but not survival, proliferation, or synaptogenic properties of rat astrocytes in vitro. Br J Anaesth 110 Suppl 1:i19-28.

Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, Olney JW, Wozniak DF (2003) Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning defi cits. J Neurosci 23:876-882.

Lunardi N, Hucklenbruch C, Latham JR, Scarpa J, Jevtovic-Todorovic V (2011) Isofl urane impairs immature astroglia development in vitro: the role of actin cytoskeleton. J Neuropathol Exp Neurol 70:281-291.

Michell-Robinson MA, Touil H, Healy LM, Owen DR, Durafourt BA, Bar-Or A, Antel JP, Moore CS (2015) Roles of microglia in brain development, tissue maintenance and repair. Brain 138:1138-1159.

Rappaport BA, Suresh S, Hertz S, Evers AS, Orser BA (2015) Anesthetic neurotoxicity--clinical implications of animal models. N Engl J Med 372:796-797.

Ryu YK, Khan S, Smith SC, Mintz CD (2014) Isofl urane impairs the capacity of astrocytes to support neuronal development in a mouse dissociated coculture model. J Neurosurg Anesthesiol 26:363-368.

Sanchez V, Feinstein SD, Lunardi N, Joksovic PM, Boscolo A, Todorovic SM, Jevtovic-Todorovic V (2011) General anesthesia causes long-term impairment of mitochondrial morphogenesis and synaptic transmission in developing rat brain. Anesthesiology 115:992-1002.

Stratmann G, Sall JW, May LD, Loepke AW, Lee MT (2010) Beyond anesthetic properties: the eff ects of isofl urane on brain cell death, neurogenesis, and long-term neurocognitive function. Anesth Analg 110:431-437.

Wagner M, Ryu YK, Smith SC, Mintz CD (2014) Review: eff ects of anesthetics on brain circuit formation. J Neurosurg Anesthesiol 26:358-362.

Ye X, Lian Q, Eckenhoff MF, Eckenhoff RG, Pan JZ (2013) Differential general anesthetic eff ects on microglial cytokine expression. PLoS One 8:e52887.

10.4103/1673-5374.165244 http://www.nrronline.org/

Jiang D, Lim S, Kwak M, Ryu YK, Mintz CD (2015) The changes of oligodendrocytes induced by anesthesia during brain development. Neural Regen Res 10(9):1386-1387.