| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and Critical Care Medicine, University of Heidelberg, Mannheim, Germany; and Department of Physiology and Pathophysiology, University of Heidelberg, Heidelberg, Germany.
Address correspondence and reprints requests to Dr. Armin Kalenka, Department of Anesthesiology and Critical Care Medicine, University of Heidelberg, Theodor Kutzer Ufer 1-3, 68167 Mannheim, Germany. Address e-mail to armin.kalenka{at}urz.uni-hd.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
METHODS: Rats were anesthetized by spontaneous inhalation of 2.4% sevoflurane in air for 3 h. Animals (n = 6 for each group) were killed either directly, 72 h, or 28 days after anesthesia. Brains were removed and subjected to global protein expression profiling based on two-dimensional gel electrophoresis and mass spectrometry. Expression factors were compared to results from untreated conscious animals at each time point. Data were statistically analyzed by ANOVA (P < 0.01) and a cut of more than two-fold change in the expression factor.
RESULTS: We found 11 protein spots differentially regulated directly after anesthesia. Seventeen proteins were differentially expressed 72 h after the anesthesia. Only one spot was differentially regulated 28 days after anesthesia. The plausible targets of these differentially regulated proteins can be attributed to synaptic vesicle handling and cell–cell communication.
CONCLUSIONS: Sevoflurane induced relevant changes in protein expression profiles directly and 72 h after an anesthesia with 1 MAC. Twenty-eight days after the anesthesia, all proteins except one had returned to baseline levels of abundance.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Volatile anesthetics affect cerebral bloodflow and metabolism (4,5), and signaling via 
-aminobutyric acid receptors (6), N-methyl-d-aspartate receptors (7), and voltage-gated sodium channels (8). These effects occur during anesthesia, but not for later periods after withdrawal of anesthetic gases. These and other effects suggest that volatile anesthetics can have profound effects on cellular function, as well as altering expression at the message (9–13) and protein levels (14,15).
In the present study we used a high throughput proteomic analysis to determine changes in protein abundance after sevoflurane anesthesia. After anesthesia for 3 h with one minimum alveolar concentration (1 MAC) sevoflurane in air (16,17) proteome analysis revealed changes in abundance of 11 proteins directly after anesthesia, a change of 17 proteins 72 h and only 1 protein 28 days after anesthesia.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Experimental Protocol
The rats taken for proteome analysis were randomly assigned to one of six groups, each group consisting of six animals. For induction of anesthesia, the animals were placed in a box flooded by 1.0 MAC of sevoflurane (2.4% v/v) in air (16,17). When anesthesia had been induced, rats were taken out of the box and allowed to spontaneously breathe the anesthetic gas via a nose cone. Gas flow was set to 3 L/min and end-tidal gas analysis (concentrations of oxygen, carbon dioxide, and the volatile anesthetic) was performed. During anesthesia, we monitored oxygen saturation transdermally and kept the body temperature constant at 37°C. In this state, the animals appeared motionless and without visible skeletal muscle tone. The control rats (3 groups of 6 rats each) were placed in a box and allowed to breathe air without anesthetics for 3 h. Animals in the 3 h group were decapitated under anesthesia at the end of the sevoflurane inhalation. Rats in the 72 h and 28 day groups were returned to their cages, and they were euthanized by decapitation at the appropriate time. Each brain was removed immediately and tissue samples for protein extraction were stored at –80°C until further processing.
Physiological Variables
In additional two separate groups not taken for brain proteome analysis (n = 5 each), we measured physiological variables during 1 MAC of sevoflurane, or lack of anesthesia, respectively. In contrast to rats taken for proteome analysis, we cannulated the femoral artery to draw blood samples for measuring arterial Po2, arterial Pco2, arterial blood pH, heart rate, and mean arterial blood pressure. After 3 h of anesthesia (one group) or under control conditions without anesthesia (one group), we allowed these animals to recover in a rat restrainer (Braintree Scientific, Braintree, MA, USA) for an additional 2 h to detect potential alterations after the end of anesthesia. Blood samples were taken 180 min after induction of anesthesia and 2 h later.
Two-Dimensional Gel Electrophoresis
We performed two-dimensional gel electrophoresis essentially as described previously (15). Briefly, we mechanically homogenized approximately 2000 mg brain in lysis buffer (1 µL/mg tissue) consisting of 40 mM Tris-HCl, 7 mol/L urea, 2 mol/L thiourea, 4% (w/v) 3-[(3-Cholamidopropyl)dimethylamino]-1-propanesulphonate, 10 mmol/L 1,4-dithiothreitol, 1 mmol/L EDTA, and a protease inhibitor (Complete Protease Inhibitor, Roche, Mannheim, Germany). The samples were centrifuged for 10 min at 1000g to remove nuclei and undissolved material. The supernatant was centrifuged for 1 h at 100,000g and protein content in the supernatant was determined by the Bradford method (18) and stored at –80°C.
We applied 100 µg pooled protein, with an equivalent amount of protein from each animal, from each experimental group on gradient gel strips with an immobilized nonlinear pH gradient ranging from pH 3 to 10 (Immobiline DryStrips pH 3–10 NL, 18 cm, Amersham Biosciences, Uppsala, Sweden). After 12 h of reswelling at 30 V, voltages were increased to 200, 500, and 1000 V for 1 h each, and kept constant at 8000 V for 12 h, resulting in a total of 100,313 V/h.
For the second dimension, we used 12.5% linear polyacrylamide gels in the presence of 10% sodium dodecyl sulfate. The gels were run at 30 mA for 30 min and 100 mA for about 4 h in a water-cooled electrophoresis apparatus. Each sample was run four times.
For image analysis, gels were silver stained. For spot identification, gels with 250 µg protein content were stained with colloidal Coomassie blue and matched to a silver stained reference gel. Spots of interest in the Coomassie blue gel were excised, digested in the gel by trypsin, and subjected to matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF-MS) at the Center for Molecular Medicine (University of Cologne, Cologne, Germany). MA spectra were searched in the Mascot database (http://www.matrixscience.com). A Mascot score more than 63 were considered statistically significant (P < 0.05) (19).
Image and Statistical Analysis
We performed image and statistical analysis using published procedures (20). Silver-stained gels were digitized, and the images were analyzed using the Phoretix 2D Expression software (Nonlinear Dynamics, Newcastle-upon-Tyne, UK). Each single spot was matched to a corresponding spot in the reference gel. The reference gel was created as a virtual PC-generated averaged gel, based on the gels from the control group.
A normalized spot volume was created, calculated from the spot volume, defined as the integral of the optical density over the spot area, in relation to the total volume of all detected spots. We then compared normalized spot volumes between the control group and experimental group at each respective time point using a statistical package for analysis of variance. Statistical significance cut off was set to P < 0.01.
Expression factors in relation to each specific control group were calculated by dividing the mean normalized spot volumes in the experimental group by the mean normalized spot volumes of the control group. A value of 2.5 therefore indicates an increase of 2.5-fold, a value of 0.5 a decrease of 2-fold. Biological relevant differences were defined as more than two-fold changes of the expression factor if analysis of variance revealed a P < 0.01.
An ANOVA test was used to compared the physiological variables between experimental and control groups. P < 0.01 was considered to be significant. Data are given as mean ± sd.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
Two-Dimensional Gel Electrophoresis
We were able to discriminate more than 1650 spots in each gel. In the experimental groups, we could match 1051 ± 99 spots (n = 12, range 884–1205) to the reference gel. We found 33 spots differentially expressed in all experimental groups. Two spots (spots 79 and 1128) were differentially expressed at two time points. These resulted in 31 protein spots which were sent for mass spectrometry. Of these, 29 spots could be identified. The remaining two mass spectra produced nonsignificant Mascot scores (P > 0.05).
Time-Course Analysis of Proteome Changes
Directly after anesthesia, 11 spots were differentially expressed. Nine spots were down-regulated and two spots were up-regulated in their expression. Seventy-two hours after the anesthesia, 17 spots were differentially expressed, 15 down-regulated and 2 spots up-regulated. Twenty-eight days after the experiments, only one spot (spot 1128, Septin 8) was differentially expressed. This specific spot was up-regulated 72 h after anesthesia (EF 3.34) and kept up-regulated 28 days after anesthesia (EF 11.99). No other spots were differentially expressed among the groups after 28 days. The expression factors in all experimental groups ranged from 0.10 (10-fold down-regulation) to 11.99 (nearly a 12-fold up-regulation) (Table 2).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The present study has three main results. First, sevoflurane at a 1 MAC concentration induced significant changes in abundance of proteins directly after 3 h exposure. These alterations were persistent 72 h after the anesthesia. Second, the differentially regulated proteins are mainly associated with synaptic vesicle handling and cell–cell communication. Third, 28 days after the anesthesia all proteins except one had returned to baseline levels of abundance.
Proteome Analysis
Volatile anesthetics modulate gene expression in the brain (9–13). However, the study of messenger RNA expression, obviously a potent factor in the control of protein production, cannot predict the structure and dynamics of the respective proteins. The regulatory processes take place mainly at the protein level place (21,22).
A proteomic approach revealed that a 3-h 1 MAC desflurane anesthesia altered rat brain proteome directly after the end of the anesthesia and as long as 72 h after the anesthesia (15). To minimize the effects on brain protein expression, which are not related to the volatile anesthetic exposure, animals in the present study were kept completely undisturbed during sevoflurane exposure. Physiological variables were analyzed in separate groups not taken for proteome analysis but handled in a similar manner.
Proteome analysis is a semiquantitative method to separate a complex protein mixture, to detect hundreds of proteins and their alterations to a given control. We used lysates from whole brains in our study. Using analysis of substructures (e.g., hippocampus), another proteome profile may have been revealed. Further studies are needed to analyze the effects of volatile anesthetics on such substructures. The protocol we used for tissue preparation analyzes hydrophilic proteins, mainly from the cytosol, whereas hydrophobic proteins, e.g., membrane proteins are not well represented (23). To ensure biological significance we stated that only proteins with a P < 0.01 and a more than two-fold change in expression were significant. It may possible that if we used a more liberal approach more proteins would have been differentially expressed.
A well known problem in proteome analysis is intergel variability. We used a specific standard operating procedure for all gel preparation steps done by one person and ran each sample four times to overcome this variability (24,25).
With MALDI-TOF-MS we were able to identify 27 of the differentially expressed spots. Several proteins were identified in more than one spot, further confirmation of their alteration in response to sevoflurane.
Functional Groups of Proteins Differentially Expressed Directly and 72 h After the End of Anesthesia
The findings of a previous study (15) using 1 MAC desflurane anesthesia in rats suggested that volatile anesthetics interact with brain proteins. These differentially expressed proteins can be thought of as belonging to functional groups. In this study we found that differentially expressed proteins belong mainly to the metabolism, and stress response groups and a group consisting of proteins involved in vesicle formation, transport, exocytosis, cytoskeletal rearrangement, and synaptic transmission.
A first group of proteins altered after sevoflurane anesthesia include enzymes well known for their metabolic functions. Pyruvate kinase (spot 1119), transketolase (spots 971 and 974) and aconitase (spot 744) were down-regulated. As sevoflurane decreases local cerebral glucose use (4), it is likely that these changes reflect a down-regulation of energy-dependent cellular pathways, similar to that described in other tissues exposed to volatile anesthetics (26).
In addition, cellular stress responsive proteins, such as heat shock proteins, molecular chaperones and ubiquitin carboxy-terminal hydrolase L1 (UCH L1), were altered in their expression. UCH L1 (spot 2018) was highly up-regulated directly at the end of the 3 h of sevoflurane anesthesia. In the brain, UCH L1 serves as a multifunctional protein involved in protein proteolysis, apoptosis, synaptic function and morphology and is localized exclusively to neurons (27). Recently, evidence for a relationship between UCH L1 expression and neurotransmitter receptors was reported (28).
The 90-kDa heat shock protein, HSP90, consisting of two isoforms, 
and ß (spots 635 and 637), is an abundant molecular chaperone associated with a number of signaling proteins. The association of HSP90 with these signaling proteins is essential for their stability, correct intracellular location, and biological activity (29). HSP90 plays a critical and indispensable role in regulating cell growth through modulation of several signal transduction pathways. Like other chaperones (e.g., spot 1066: Cytosolic chaperonin containing T-complex polypeptide subunit zeta), HSP90 is also involved in the folding of cytoskeletal proteins, such as tubulins (spots 1207, 1267, 1808), actin and centractin (30). In contrast to other molecular chaperones, HSP90 has an independent function in neurotransmitter release and is necessary for the efficient release of neurotransmitters at the presynaptic terminal (31).
A third functional group of differentially regulated proteins was a group associated with vesicle formation, transport, exocytosis, cytoskeletal rearrangement, and synaptic transmission.
Dihydropyrimidinase-like proteins (spots 941, 1039, and 1091), also called collapsin response mediator proteins (CRMP), were down-regulated at different time points. These proteins are involved in neuronal differentiation and axonal outgrowth (32). The interaction of CRMP-2 with tubulin (spots 1207, 1267, 1808) dimers promotes microtubule assembly for axon outgrowth (33).
We also found proteins involved in the cytoskeletal arrangement (Cofilin, Tubulin) down-regulated. Cofilin (spot 2200), the highest down-regulated protein in the present study, is an actin-binding protein involved in cytoskeletal arrangement and plays a central role in actin turnover (34).
Several proteins (Dynamin 1, NSF, Synpasin 1, Synapsin 2) involved in synaptic vesicle traffic were down-regulated. Synapsins (spots 830 and 1137) are anchor proteins regulating the synaptic vesicle pool near the presynaptic membrane and the active pool for exocytotic release (35). Down-regulation of synapsins results in synaptic fatigue as a result of their being fewer synaptic vesicles distal to the active zone (36).
The N-ethylmaleimide-sensitive fusion protein, attachment protein ß (NSF, spot 1859) is involved in neuronal communication, and seems to be essential for the docking and fusion of synaptic vesicles (37). Being involved in the process of membrane fission and fusion, Dynamin 1 (spot 79)37 was also down-regulated.
Taken together, most proteins differentially expressed after a 3 h sevoflurane anesthesia are involved in the context of synaptic vesicle and cell–cell communication, with the exception of proteins with a function in metabolism. Besides UCH L1, all proteins were significantly down-regulated. None of the proteins identified in our study plays a direct role in cognitive dysfunction or other symptoms after volatile anesthesia. However, these proteins suggest that further studies investigating a possible link between anesthetic effects at the cellular level and their clinical consequences are needed. Interestingly, 28 days after the anesthesia all proteins except one had returned to baseline levels of abundance.
Comparison with Studies of Gene and Protein Expression after Volatile Anesthetics
Comparing the proteomic expression profiles after sevoflurane (this study) and desflurane anesthesia (15), we found only one protein to be similarly down-regulated in both studies. Even though only dynamin (spot 79) was differentially expressed in the same manner, all other identified proteins in both studies belong to functional groups related to cellular metabolism, vesicle traffic, and cellular communication.
Three things are of particular interest in comparing these two brain proteome studies. First, in both studies, relative few proteins in the brain were changed in their abundance. This is in accordance with gene expression studies of whole brain after volatile anesthetics which found only six genes differentially expressed after an in vivo exposure to 0.8 MAC halothane and, interestingly, no gene alteration after a 0.8 MAC isoflurane exposure (13). Another study found, after isoflurane exposure, differentially expressed genes related to DNA transcription, protein synthesis, metabolism, signaling cascades, and cytoskeletal structural proteins (11). These are exactly the same functional groups of proteins which were differentially expressed in our study. Another study (10) using 4.5 vol% sevoflurane found only four genes, encoding transcription factors, down-regulated. This is of special interest, as nearly 2 MAC sevoflurane in this study altered several organs, particularly the liver, in many gene products, whereas in the primary target organ, the brain, only in a small amount should be affected.
Second, the differentially expressed proteins are not a specific target for volatile anesthetics. It therefore may be speculated that the different protein expression profiles do not indicate a direct effect of the anesthetic on these proteins. Instead, the expression changes may indicate compensatory efforts to the anesthetic. This is again in strong agreement with gene expression studies (10,11,13) in which the altered gene also was not associated with known anesthetic targets in the brain.
Third, the alterations of protein expression in both studies last as long as 72 h after the anesthesia. In our previous study (15) we also found long lasting effects on brain proteins. It therefore may be speculated that these compensatory effects outlast the duration of anesthesia.
Surprisingly, volatile anesthetics have little influence on gene and protein expression in the brain. In this context, it should be noted that the changes in gene or protein expression observed in this study may also have been altered by the long time courses analyzed. On the other hand, the lack of sustainability in changes does not mean that there is no connection to clinical changes.
Septin 8 Protein Expression Up-Regulated 28 Days After Sevoflurane Anesthesia
The septins are a family of GTP-binding proteins involved in diverse processes, including vesicle trafficking, apoptosis, remodelling of the cytoskeleton, infection, neurodegeneration, and neoplasia. Ten different mammalian septin isoforms have been identified (38). Although the exact role of Septin 8 has yet to be defined, there is growing evidence that Septin 8 and Septin 5 organize into macromolecular complexes around cytosolic vesicle-like organelles (39). Septin 8 is enriched in platelets and neurons. Further studies, e.g., knockout or transgenic over-expression models are needed to examine the exact role of Septin 8 as a target of volatile anesthetics.
Clinical Considerations
Although general anesthesia is considered a safe clinical procedure, it must be kept in mind that even a short anesthetic protocol of 3 h, as in this study, may result in long-lasting changes in gene and protein expression in the brain. This holds true for several volatile anesthetics, such as desflurane (15) and sevoflurane (present study). Changes in postanesthetic gene and protein expression may contribute not only to adverse short-term effects (dizziness, nausea), but, more importantly, to long-term effects such as memory alteration, cerebral bloodflow and metabolism, and neurocognitive reactions. Currently, we cannot predict whether these effects are adverse or favorable. It also remains unclear whether repeated anesthesia will increase the risk of side effects.
|
CONCLUSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Accepted January 19, 2007.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
