Isoflurane Inhibits Growth but Does Not Cause Cell Death in Hippocampal Neural Precursor Cells Grown in Culture

All animals were cared for according to procedures approved by the Institutional Animal Care and Use Committee of the University of California, San Francisco. NPCs were isolated by following methods previously described with slight modification.17–20Unanesthetized postnatal day 2 Sprague-Dawley rats were separated from the dam and decapitated using a guillotine. Hippocampi were immediately dissected out and placed in 10-ml ice-cold Hanks Basic Salt Solution without calcium. Whole hippocampi pooled from 10 animals were collected by centrifugation for 2 min at 300 relative centrifugal force. Supernatant was removed, and hippocampi were minced with a razor blade then triturated 5 times using a 200-μl pipette. Cells were then resuspended in 10 ml of Hanks Basic Salt Solution for 2 min. After settling for 2 min, the supernatant was transferred to a new tube. Cells were then collected by centrifugation at 300 relative centrifugal force for 5 min. Supernatant was discarded, and cells were resuspended in proliferation medium consisting of 3:1 Dulbecco's Modified Eagle's Medium: Ham's F12 (University of California, San Francisco cell culture facility, San Francisco, CA), 1% penicillin and streptomycin, 1× B-27 supplement (Invitrogen, Grand Isle, NY), 20 ng/ml basic fibroblast growth factor (Chemicon, Temecula, CA), 0.75 units/ml heparin (Abraxis, Schaumburg, IL). Hippocampal precursor cells were then grown in 5% carbon dioxide in air at 37°C and with media changed thrice weekly and cells transferred to new flasks every 5 days so that adherent differentiated cells were left behind and nonadherent proliferating NPCs were moved to the new flask. NPCs had been grown in culture for 2 to 12 weeks at the time of use. To differentiate NPCs, basic fibroblast growth factor containing medium was replaced with differentiation medium consisting of Neurobasal-A (Invitrogen, Grand Isle NY), B27 supplement, 1% penicillin-streptomycin (University of California, San Francisco cell culture facility), l-glutamine (Invitrogen, Carlsbad CA), and 5% fetal bovine serum (University of California, San Francisco cell culture facility) on day one. The next day, medium was replaced with differentiation medium lacking serum.

Cells were plated at a density of 20,000 per well on 8-chamber microscope slides that were precoated with poly-l-ornithine (Sigma-Aldrich, St. Louis, MO) in water overnight and then with laminin (Sigma-Aldrich) in phosphate-buffered saline (PBS), pH7.4 overnight. At the end of the experiment, the slides were fixed with 4% paraformaldehyde (Sigma-Aldrich) in PBS for 15 min at room temperature and blocked with 10% goat serum and 0.03% Triton X-100 (Sigma-Aldrich, St. Louis, MO) in PBS for 2 h at room temperature. Primary antibodies were diluted in PBS and added for overnight incubation at 4°C. Secondary antibodies were diluted 1:1000 in PBS and added for 2 h at room temperature. Nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min at room temperature. Slides were washed with PBS, and coverslips were applied with Aquapolymount (Polysciences, Warring, PA). Table 1gives a summary of the antibodies used.

NPCs in slides or flasks were placed in a Billups-Rothenburg chamber (Billups-Rothenburg, Del Mar, CA). The chamber was flushed with isoflurane in humidified air with 5% carbon dioxied at 37°C for 15 min while monitoring the temperature and concentration of gases inside the chamber. The chamber was then sealed and placed inside a 37°C incubator. The concentrations of oxygen, isofluorane, and carbon dioxide in the chamber were monitored using a Datex Capnomac Ultima gas analyzer (Datex Ohmeda, Helsinki Finland) at 2-h intervals, and the chamber was reflushed if there was any change. Control experiments were performed in the same manner, except no isoflurane was added when flushing the chamber. We chose 3.4% isoflurane for these experiments on the basis of our experience anesthetizing postnatal day 7 animals. Using the method of Merkel and Eger,21we found the inspired minimum alveolar concentration of isoflurane to be 3.6– 4% initially, with a slow decline thereafter.4 

NPCs were grown as neurospheres in suspension as described above and plated in 96-well plates in proliferation medium. The next day, the cells were exposed to 3.4% isoflurane for 4 h as described above (Isoflurane Exposure) and returned to standard incubator growth conditions for 18 h. NPCs were then analyzed for release of lactate dehydrogenase (LDH) using the Cytox-96 assay kit (Promega, Madison, WI) and following the manufacturers protocol. Treatment with Triton X-100 (0.9% volume/volume, provided by manufacturer) for 45 min at 37°C was used as a positive control.

NPCs (1 × 10⁴) were plated in proliferation medium and grown as neurospheres in suspension in opaque 96-well plates. The next day, plates were exposed to 3.4% isoflurane, 500 ng/ml fentanyl citrate (0.95 μm), or control for 4 h, followed by standard incubator conditions for 2 h for caspase 3/7 or as indicated in the figure for caspase 9. Fentanyl has previously been shown to activate opioid receptors on cultured differentiated neurons at concentrations of 0.1 to 2.0 μm22–24and did not induce caspase activation in the brain when used as a sole anesthetic.25Fentanyl was used as a control treatment because it does not interact with GABA^Aor N -methyl-d-aspartate receptors and does not cause cell death in vivo  but is commonly used for clinical anesthesia. Cells were then lysed and analyzed for caspase 3/7 or caspase 9 activity (Promega, Caspase-Glo 3/7 or Glo 9 kit, Promega). Nuclear cleaved caspase 3 was quantified using immunocytochemistry methods described above and performed on NPCs plated in 8-chamber glass slides coated with poly-l-ornithine and laminin. The total (DAPI-labeled) and cleaved caspase 3 (immunofluorescence-labeled) nuclei were counted in multiple fields of view to determine the fraction of nuclei positive for cleaved caspase 3.

After triturating to a single cell suspension, equal numbers of NPCs were plated into 6- or 12-well plates in proliferation medium and grown floating in culture. The next day, cultures were exposed to 3.4% isoflurane or control for 6 h before returning to the incubator for growth under standard conditions. Forty-eight hours later, the contents of each well were collected by centrifugation, triturated to a single cell suspension, and then resuspended in 500 μl of medium before counting the cells by trypan blue exclusion using a hemocytometer (VWR, Batavia, IL).

NPCs (0.5–1 × 10⁵) grown as neurospheres in suspension in T-25 flasks (Corning supplied by Fisher, Pittsburg, PA) were exposed to isoflurane for 4 h. At specified times, cells were collected by centrifugation, and RNA was isolated using Trizol (Invitrogen, Carlsbad, CA) by following the manufacturer's directions 6, 12, and 24 h after treatment. RNA (1 μg) was reverse-transcribed to make copy DNA using the Quigen Omniscript kit (Quiagen, Valencia, CA) by following manufactures instructions and using a mix of poly d-thymidine and random hexamers. Quantitative polymerase chain reaction was performed using Quiagen Quantitect SYBR Green kit (Quiagen) on an Stratagene MX 3000P thermocycler (Agilent Technologies, La Jolla, CA). Changes in gene expression were determined using the delta-delta cT method.26α-Actin was used as a reference gene for loading control, and expression of the gene of interest was defined relative to untreated cultures, which were arbitrarily assigned a value of 100%. Primer sequences for target genes are listed in table 2.

NPCs were plated in coated eight-chamber microscope slides in proliferation medium. The next morning, NPCs were exposed to isoflurane as described above (isoflurane exposure). 5-Bromodeoxyuridine (BrdU), a thymidine analog, was added to the medium for 2 h at the end of the isoflurane exposure or 12 h later. Slides were fixed and stained as described above and then photographed with 40× or 100× objective lenses on an E400 (Nikon, Melville, NY) fluorescence microscope equipped with filters at 385, 490, and 570 nm. Six to twelve sets of images were acquired at different locations and were subsequently merged using National Institutes of Health Image J software (Bethesda, MD), and the total number of cells (DAPI) as well as the number of BrdU-positive cells was determined.

NPCs growing in suspension in proliferation medium were collected, switched to differentiation medium, and plated on coated 8-chamber microscope slides. The slides were immediately exposed to isoflurane as described above under isoflurane exposure and then returned to standard incubator conditions for 4 days. Media was replaced on day 2 with differentiation medium lacking serum. Slides were fixed and stained as described above (immunocytochemistry) and then photographed with 40× or 100× objective lenses on a Nikon E400 fluorescence microscope with filters at 385, 490, and 570 nm. Six to twelve sets of images were acquired at random locations and were subsequently merged using NIH Image J software. The total number of cells (DAPI) and the ratio of cells positive for neuron-specific beta 3 tubulin (Tuj1), to mark neurons, or glial fibrillary acidic protein (GFAP), to mark astrocytes, was determined.

All statistical analyses were performed and all graphs were produced using Prism 4 (GraphPad Software, San Diego, CA). Data are expressed as mean ± SE and analyzed using parametric test except in cases for which assumptions required for parametric analyses (homogeneity of variance, normality) were violated, when data were expressed as medians ± interquartile ranges and analyzed using nonparametric tests. Caspase 3/7 and 9 enzyme activity was evaluated by analysis of variance (ANOVA) and Bonferroni post test to correct for multiple comparisons. BrdU incorporation immediately after isoflurane exposure was analyzed by Kruskal-Wallis with Dunn's post test. Nuclear localization of cleaved caspase 3, BrdU incorporation at 12 h, and expression of Tuj1 or GFAP by differentiated NPCs were analyzed by Mann–Whitney U test. Growth experiments were analyzed by Wilcoxon signed rank test, with fold increases from control values being compared to the theoretical value of 1, assigned to the control group. Subsequent comparisons of fold increase in untreated versus  isofluorane-treated cultures were done with Wilcoxon matched pairs test. Changes in gene expression in quantitative polymerase chain reaction experiments were analyzed by ANOVA with Dunnett post test to compare all time points to control untreated cultures.

Two to four days after isolation, NPCs in proliferation medium began to grow in characteristic neurospheres as has been described previously (fig. 1A).17–20,27,28The same NPCs plated on slides with poly-l-ornithine and laminin become adherent and put out processes (fig. 1B); however, they remain largely undifferentiated as evidenced by expression of stem cell markers nestin and Sox2 (fig. 1C) but little expression of neuronal marker Tuj1 or the astrocyte marker GFAP (fig. 1D). After 4 days in differentiation medium, cells positive for GFAP (fig. 1E, red) and Tuj1 (fig. 1E, green) were observed, confirming that the precursor cells had been multipotent before differentiation.

The activity of the apoptotic proteins caspase 3/7 and caspase 9 was determined (fig. 2). Although 1 μm staurosporine increased caspase 3/7 activity, neither 4 h of 3.4% isoflurane nor 4 h of 500 ng/ml fentanyl citrate changed the activity of caspase 3/7 2 h after exposure (n = 3; ANOVA P = 0.0005; Bonferroni post test P > 0.05 for control vs.  isoflurane and control vs.  fentany; fig. 2A). Similarly, when cells were grown adherently on glass slides, no increase was seen in nuclear-cleaved caspase 3 staining 2 h after a 4-h isoflurane exposure (control n = 3; isoflurane n = 4; staurosporine n = 2; Mann–Whitney U P > 0.05; fig. 2C). Staurosporine (1 μm) did induce nuclear localization of cleaved caspase 3 (fig. 2, C and D). Caspase 9 activity was increased in the presence of 1 μm staurosporine but was unchanged by fentanyl and decreased after 2 or 4 h of isoflurane or 4 h of isoflurane + 2 h standard incubator conditions (n = 3; except staurosporine n = 5; ANOVA P < 0.0001; Bonferroni post test P < 0.001 control vs.  any isoflurane group, P < 0.05 control vs.  staurosporine, and P > 0.05 control vs.  fentanyl; fig. 2B).

To determine if nonapoptotic cell death occurred, NPCs were exposed to isoflurane for 4 h, and the amount of LDH released into the media was determined. No difference was seen between control and isofluorane-exposed groups, but addition of detergent (0.9% Triton X-100) to the medium did increase LDH release (n = 8; isoflurane vs.  control t  test P = 0.80; fig. 2E).

NPCs were plated in 6- or 12-well plates, grown in nonadherent proliferative conditions, and exposed to 3.4% isoflurane for 6 h. Both untreated and isoflurane-treated cultures proliferated relative to time 0 (Wilcoxon signed rank test relative to theoretical value of 1 at time 0; control P < 0.01; isoflurane P < 0.05,). However, the fold increase in the number of cells 48 h later was smaller in isoflurane-treated than in untreated cultures (n = 9; Wilcoxon matched pairs t  test P < 0.05; fig. 3).

NPCs grown in proliferative conditions were exposed to 3.4% isoflurane for 4 h, and expression of the cell cycle regulator Ki-67 and the stem cell marker Sox2 was evaluated 6, 12, and 24 h later by using quantitative polymerase chain reaction. Both Ki-67 and Sox2 messenger RNA were decreased maximally at 12 h (34 and 45% lower) and began to return to baseline by 24 h (Ki-67: n = 6, ANOVA P > 0.05 Dunnett post test 12 h P < 0.05, fig. 4A; Sox2: n = 5, ANOVA P < 0.05, Dunnett post test 12 h P < 0.05, fig. 4B).

To determine if the isoflurane-mediated decrease in proliferation was the result of a change in cell cycle, progenitors were exposed to 4 h of 3.4% isoflurane followed immediately by 2 h of BrdU. Control cultures showed a median of 21.5% of cells incorporated BrdU compared to only 13.1% of isoflurane-exposed cells. Fentanyl citrate (500 ng/ml) given for 6 h did not change BrdU incorporation (Kruskal-Wallis P = 0.029; Dunn post test P < 0.05, control vs.  isofluorane; fig. 5A). To assess the duration of the effect BrdU was added 12 h after completion of the isoflurane treatment leading to a median 18.6% BrdU-positive cells in the isoflurane group compared to 23.3% in control (Mann–Whitney P < 0.05; fig. 5B).

We next evaluated the effect of isoflurane on fate selection and differentiation by exposing NPCs to isoflurane for 4 h at the time of plating in differentiation medium. Cells were fixed 4 days later and stained for Tuj1 (neurons), GFAP (astrocytes), and DAPI (all nuclei). NPCs exposed to isoflurane were more likely to express a neuronal marker 4 days later (n = 6; P < 0.01, Mann–Whitney U; fig. 6). No difference was seen in the rate of expression of the astrocyte marker GFAP (n = 6; P = 0.94, Mann–Whitney U; fig. 6).

Multiple anesthetics have previously been reported to lead to caspase activation and neuronal cell death throughout the brain of postnatal day 7 animals.1We report in an accompanying article in this issue that isoflurane decreases NPC proliferation (BrdU uptake) in vivo  in adults and neonates.4This could be the result of isofluorane-mediated decreases in neuronal firing in neonatal and adult animals29or of direct toxicity in neonates where cell death has been observed.3Here we have attempted to determine whether this effect is independent of the actions of isoflurane on the surrounding brain by growing NPCs in an in vitro  environment. Cultures of hippocampal precursor cells were isolated from postnatal day 2 rats to maximize the contribution of cells from the DG and minimize the contribution from the subventricular zone and other brain regions. Embryonic animals are often used for NPC cultures; however, the hippocampus is poorly defined or absent until birth in rats, making isolation nearly impossible. We are most interested in NPCs specifically from the dentate to correlate with in vivo  studies of this region; therefore, we obtained our cultures from postnatal day 2 animals.

Translating anesthetic studies from animals (let alone cell culture) to humans is at least flawed if not impossible. To minimize differences between in vitro  and in vivo  models of anesthesia, a comparable dose must be used. One can use an equal dose such as 1% isoflurane that is not clinically equivalent across species or one can choose to measure a clinical endpoint such as minimum alveolar concentration and use a dose that produces an equivalent endpoint between species. We have chosen to base our in vivo  studies on minimum alveolar concentration and therefore use the same dose for the in vitro  studies presented here. Fentanyl was used as a control in these studies because both GABA^Aand N -methyl-d-aspartate receptors mediate signals that are known to influence proliferation and differentiation of NPCs,13–16and most anesthetic drugs act at one or both of these receptors.29Opiates do not act at GABA^Aor N -methyl-d-aspartate receptors, and fentanyl is a commonly used opiate and was chosen as a control for that reason.

In the experiments presented here, with isoflurane as the sole anesthetic, we were unable to induce cell death, caspase activation, or LDH release in cultured NPCs. This suggests that NPCs do not undergo apoptosis or necrosis after isoflurane exposure, consistent with limited cell death being reported in the DG of the hippocampus (the source of these cultured cells) compared with other regions of the brain.3Decreased proliferation 48 h after isoflurane exposure combined with a decrease in the number of cells in S-phase and loss of the cell cycle regulator Ki-67 suggests that the cells are either in growth phase arrest or have exited the cell cycle as would occur with differentiation. Cell cycle exit and differentiation is further suggested by the loss of the stem cell gene Sox2 and increased neuronal fate selection.

Culturing cells in different states can change the characteristics of those cells and how they respond to different signals or stresses and represents a potential limitation of this study. These experiments were performed on cells in two different states: floating as neurospheres or adherent to glass slides but undifferentiated. The results we observed, however, are similar in both cases. Growth inhibition, loss of Ki-67, and lack of caspase activation in floating neurospheres and decreased ratio of cells in S-phase and lack of nuclear-cleaved caspase 3 in adherent cells are all consistent with a lack of cell death and a decrease in proliferation. Similar effects on proliferation in vivo  are also reported in this issue.4 

Isoflurane facilitates opening of the GABA^Areceptor and has previously been shown to raise the intracellular calcium concentration in neurons and slice cultures by liberating it from endoplasmic reticulum.30–33GABA^Areceptor opening and increased intracellular calcium decrease proliferation of precursor cells in subventricular zone13,34and increase differentiation and selection of a neuronal fate in hippocampal progenitors of the DG.14,16,35Isoflurane may lead to cell cycle exit and differentiation of hippocampal NPCs by facilitating GABA^Areceptor opening and increasing intracellular calcium. Future studies are necessary to determine the exact mechanism of isofluorane-mediated growth inhibition and increased neuronal differentiation we have observed in NPCs, but GABA^Areceptor activation and changes in calcium concentration are likely targets.

Decreased proliferation and increased neuronal differentiation caused by isoflurane could lead to cognitive dysfunction in neonates by permanently disrupting the architecture of the hippocampus during a critical period of development or by depleting the pool of precursor cells present for the duration of the animals life. The same mechanism in an adult might not cause any cognitive dysfunction when the hippocampus is already fully developed and connected and can more easily accept an increased number of cells choosing to become neurons at any given time. The adult hippocampus routinely integrates new neurons into its circuitry with learning.10,36 

Elsewhere in this issue, we report isoflurane-mediated changes in NPC proliferation in vivo ,⁴and the current experiments demonstrate an effect of isoflurane on hippocampal NPCs grown in culture, demonstrating that this effect is on the cells themselves rather than being mediated by the surrounding tissue of the neonatal brain or specifically the DG. In addition to suggesting a possible mechanism for isoflurane-induced hippocampal dysfunction, the role of anesthesia in the field of neural stem cells and neural stem cell transplantation remains largely un-explored. The results reported here show that isoflurane may have a direct impact on the proliferation and fate selection of transplanted stem cells. Future studies on the effect of volatile and nonvolatile anesthetics on NPCs are important for our understanding of how these drugs effect the biology of NPCs and eventually for choosing the appropriate anesthetic for trials of stem cell transplantation in animals and someday humans.

This model system can be used to efficiently screen many anesthetics that have been previously reported to cause neurodegeneration to determine if they have an effect on NPC proliferation and differentiation. This system may also be used to address questions of mechanism that are difficult to answer with in vivo  studies. Understanding how anesthetics interact with the complex system of neurogenesis in the neonatal and adult brain may provide some insight into normal development and will guide our choice of drugs for future studies related to both anesthetic-mediated cognitive dysfunction and transplantation of NPCs.