Differences in microRNA Changes of Healthy Rat Liver between Sevoflurane and Propofol Anesthesia
Ishikawa, Masashi M.D.*; Tanaka, Shunsuke M.D.*; Arai, Masae M.D.*; Genda, Yuuki M.D.*; Sakamoto, Atsuhiro M.D., Ph.D.†
Background: In previous studies, the authors showed that anesthetics affect the expression ratios of many genes in rat liver. microRNAs (miRNA) negatively regulate more than 30% of genes in cells, and control cell proliferation, inflammation, and metabolism. The authors hypothesized that anesthetics influence miRNA expression in the liver, and performed miRNA screening tests using TaqMan low-density arrays.
Methods: Rats were randomly assigned to the 2.4% sevoflurane group, the 600 µg·kg−1·min−1 propofol group, and the control group without anesthetics. Rats were allowed to breathe spontaneously under anesthesia for 6 h. The miRNA expression profile of the liver was analyzed, and 15 representative miRNAs were validated by quantitative real-time reverse transcriptase polymerase chain reaction.
Results: TaqMan low-density arrays analysis showed 46 miRNAs that were differentially expressed by anesthetics. After sevoflurane treatment, 16 miRNAs were significantly increased and 11 were significantly decreased compared with controls, whereas after propofol treatment, 31 miRNAs were increased and 8 were decreased. Twenty expressed miRNAs were common to both anesthetics, whereas three miRNAs were differentially expressed. Bland-Altman analysis was performed across the validations to compare the fold changes measured by both methods, and they were equivalent (mean difference = 0.01, 95% CI = −0.26 to 0.27). This showed that the TaqMan low-density arrays results are accurate and can be confirmed using an independent experimental approach.
Conclusion: The results showed that anesthetics cause many miRNA expression changes, and the miRNA expression pattern was particular for each anesthetic. Further studies are needed to determine the functional consequence of miRNA modulation by anesthetics.
What We Already Know about This Topic
* microRNAs mediate posttranscriptional regulation of gene expression and control cell proliferation, metabolism, and inflammation
* The effects of anesthetic agents on microRNAs remains unknown
What This Article Tells Us That Is New
* Anesthetics cause many microRNA expression changes in the rat liver and the microRNA expression pattern differs between sevoflurane and propofol
* Further studies are needed to understand the functional consequences of these changes
THE inhalation anesthetic sevoflurane and the intravenous anesthetic propofol are widely used during hepatobiliary surgery. Anesthetic agents are known to influence hepatic function in the perioperative period. Both anesthetic agents affect blood flow, oxygen consumption, and liver functions.1–3
In previous studies, we showed that anesthetic agents affected the expression ratios of many genes in the rat liver.4
microRNAs (miRNAs) are approximately 22-nt-long, single-stranded RNA molecules that can cause either messenger RNA degradation or translational repression, resulting in reduced protein expression. miRNAs negatively regulate more than 30% of the genes in cells, and control cell proliferation, inflammation, and metabolism. Recent studies have provided evidence that miRNAs influence hepatocyte regeneration after partial hepatectomy,6
and influence metabolism8–10
in the liver.
We assumed that miRNAs have an effect on liver-function changes during perioperative periods, and hypothesized that sevoflurane and propofol anesthesia affect miRNA expression in the liver. To test this, we comprehensively investigated the miRNA expression changes induced by sevoflurane and propofol. Quantitative real-time reverse transcriptase polymerase chain reaction (qRT-PCR) is increasingly being used to detect and quantify miRNAs. Furthermore, TaqMan low-density arrays (TLDA), which enable hundreds of qRT-PCR reactions to be performed simultaneously,11
have recently been introduced as miRNA screens. Therefore, we performed miRNA screening tests using TLDAs, and subsequently measured miRNAs for the validation of TLDA data.
Materials and Methods
The Animal Research Committee of Nippon Medical School, Tokyo, Japan, approved this study (approval number 22-147). Seven-week-old male Wistar rats (Saitama Experimental Animals Supply, Saitama, Japan) weighing 250 ± 10 g were maintained under a 12-h light/12-h dark cycle starting at 06:00 AM and 07:00 PM, respectively, for 1 week before performing the experiments. After the rats were anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg) for the surgical procedure, we cannulated the left femoral artery and vein. Normal saline was administered via the venous catheter at a rate of 1 ml/h. Each rat was allowed to breathe spontaneously and housed in an anesthesia box (Sanplatec, Osaka, Japan) supplied with an air-oxygen mixture (fraction of inspired oxygen = 0.4) at a rate of 6 l/min, with body temperature maintained at approximately 37°C using a heat lamp. During the experiment, we continuously monitored rectal temperature, mean arterial blood pressure, and heart rate, and also measured arterial blood pH, arterial PO2, and arterial PCO2 at 3 h and 6 h after starting the experiments.
The rats were randomly assigned to three groups, each group consisting of six rats. Rats undergoing sevoflurane anesthesia were supplied with 2.4% sevoflurane (Maruishi Pharmaceutical, Osaka, Japan)12
and the propofol anesthesia group was infused with 1% propofol (AstraZeneca, Osaka, Japan) at 600 µg·kg−1
from 9:00 AM to 3:00 PM. The control group rats received no anesthesia. All rats were decapitated at 3:00 PM. The left lateral lobe of the liver was obtained from each rat within 3 min after death. Liver samples were washed twice with cold phosphate buffered saline and immediately stored at −20°C in RNAlater solution (Applied Biosystems, Foster City, CA). After storage for 1 day, the RNAlater solution was rapidly separated from the samples by centrifugation (10,000g
, 4°C, 5 min). Total RNA was extracted using a mirVana miRNA Isolation kit (Applied Biosystems). RNA quantity and quality were assessed with a NanoDrop ND-1000 (Thermo Fisher Scientific, Waltham, MA). A260/280
nm of 1.8 or more was qualified for quantitative analysis; then, the total RNA sample containing miRNA was used for qRT-PCR.
miRNA Screening Test; TaqMan Low-density Arrays and Standard qRT-PCR of Individual miRNAs
The miRNA expression profile was analyzed using TLDA Rodent MicroRNA Cards v.3 A and B (Applied Biosystems). Cards contain 373 preloaded rat miRNA targets, all cataloged in the miRBase database, and three endogenous controls: Mamm U6, U87, and Y1. All procedures were performed according to a previous report.14
In brief, TLDAs were performed by a two-step process. During the first step, 1200 ng total RNA per sample was reverse-transcribed using Megaplex RT primer Pool A and B, up to 381 stem-looped primers per pool and a TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems). In the second step, each of the resulting complementary DNA pools was diluted, mixed with TaqMan Universal PCR master mix (Applied Biosystems) and deionized distilled water (Wako, Tokyo, Japan), and loaded into one of the eight fill ports on the TLDA microfluidic card. The card was briefly centrifuged for 1 min at 1,600 rpm to distribute samples to the multiple wells connected to the fill ports, and then sealed to prevent well-to-well contamination. Finally, the cards were processed and analyzed using a 7900HT Fast Real Time PCR System (Applied Biosystems). The data analysis was performed using DataAssist software v2.0 (Applied Biosystems). The resulting data were represented as the threshold cycle (Ct) values, where Ct represents a unitless value defined as the fractional cycle number at which the sample fluorescence signal passes a fixed threshold above baseline. Relative amounts of all miRNAs were calculated by the comparative Ct method. ΔCt was the difference in the Ct values derived from the experimental samples and the control, and ΔΔCt represented the difference between the paired samples, as calculated by the formula: ΔΔCt = ΔCt of sample after anesthesia − ΔCt of the control group. The expression ratio shows the relative quantity of the target gene (Xtarget) to the control gene (Xcontrol). The fold change was computed by the formula Xtarget / Xcontrol = 2−ΔΔCt
. Graphic displays were visualized as heat map results of hierarchical clustering. Distances between samples and assays were calculated for hierarchical clustering based on ΔCt values, using Euclidean Distance.
To verify the accuracy of our TLDA data, we performed single qRT-PCR experiments for representative miRNAs (mmu-let-7e, mmu-miRNA-101a, mmu-miRNA-125a-5p, mmu-miRNA-139-5p, mmu-miRNA-145, mmu-miRNA-146a, mmu-miRNA-150, mmu-miRNA-186, mmu-miRNA-191, mmu-miRNA-26a, mmu-miRNA-27b, mmu-miRNA-301b, mmu-miRNA-335-5p, mmu-miRNA-449a, and rno-miRNA-450a) with TaqMan MicroRNA Assays (Applied Biosystems) in accordance with the manufacturer’s instructions.15
Single TaqMan assays rely on the same technology as TLDA. Normalization was performed with U87, the same internal control as TLDA. Reactions were run in triplicate in 96-well plates on a 7900HT Fast RealTime PCR System. Relative amounts of all miRNAs were represented as fold change.
Values are expressed as means ± SD. ANOVA followed by Tukey test was used to compare the physiological data between the control and anesthesia groups. P
value less than 0.05 (two-tailed) was considered to be significant. We applied Tukey test to identify genes and miRNAs that were significantly differentially expressed upon exposure to anesthetic agents. Furthermore, to control for potential false-positive results due to multiple testing, we applied the false discovery rate following the method described by Storey and Tibshirani.16 P
values (two-tailed) were considered significant after correction for multiple testing by adjusted false discovery rate (q-value) at the 0.05 level. We used Bland-Altman analysis between miRNA expression levels of TLDA and those of standard qRT-PCR to establish the reliability of measuring miRNA expressions by these methods. ANOVAs followed by Tukey test were performed using Kyplot 5.0 (KyensLab Incorporated, Tokyo, Japan). Bland-Altman analysis was performed using GraphPad Prism 5.0 (GraphPad Software, Inc., San Diego, CA), and the false discovery rate procedure was performed with the program QVALUE 1.0.†
All rats survived until sacrifice and the data for all animals were used. In this study, we administered one minimum alveolar concentration (2.4%) sevoflurane and effective dose 50 (600 µg·kg−1
) propofol for the comparison between sevoflurane and propofol anesthesia, because these doses were equally effective and mirrored clinical doses.12
We have used this protocol extensively in our previous genomic, proteomic, and metabolomics studies.4
Physiological data from the anesthesia groups were compared with those from the control group (table 1
). We observed slightly low mean arterial pressures and high PCO2
after the administration of sevoflurane or propofol; however, there were no significant differences in the anesthesia groups and the control group in physiological data. Hypoxia, hyper/hypocapnia, hypotension, or hypothermia did not occur in any group.
Careful manual inspection of all amplification plots excluded miRNAs that did not amplify in all samples, had very high variations, or had Ct values above 35, indicating that their expression was too low for an accurate analysis. To identify suitable endogenous controls, three different candidate miRNAs were analyzed for variance in gene expression with DataAssist software v2.0. The statistical method ranked the candidate endogenous control genes with an excellent correlation of raw stability values (data not shown). The most stably expressed U87 was chosen as the endogenous control, and relative miRNA expression was normalized against U87.
TLDA analysis showed 177 expressed miRNAs (47%) out of 373 miRNAs, and the expressed miRNAs were the same in each group. There were 46 miRNAs (26%) out of 177 expressed miRNAs that were differentially expressed in response to anesthetic agents. After sevoflurane, 16 miRNAs were significantly increased and 11 were significantly decreased compared with controls, whereas after propofol, 31 miRNAs were increased and eight were decreased (tables 2
). TLDA identified 20 expressed miRNAs that were common to both anesthetics, whereas three miRNAs were significantly differentially expressed between both anesthetics (table 4
A clustergram of the samples and the significant differentially expressed miRNAs is shown in figure 1
as a heat map generated using ΔCt. Heat maps are commonly used for visualization of high-dimensional data in a two-dimensional image with colors representing the intensity values. It is typically used in gene-expression analysis to represent the level of expression of many genes across a number of comparable samples. After individually clustering columns (samples) and rows (miRNAs), the heat map simultaneously displays the separate samples and miRNA clusterings in one graphic. The magnitude of miRNA expression changes is denoted by color, and the numbers that correspond to these colors are the ΔCt. The dendrogram on the right of the heat map (fig. 1A
) orders miRNAs into groups based on the divergence of miRNA expression values among the samples. The dendrogram (fig. 1B
) indicates the relatedness of the samples based on overall miRNA expression values, and separates three main branches, the control group, the sevoflurane anesthesia group, and the propofol anesthesia group.
To confirm the results from the TLDA experiments, standard qRT-PCR in triplicate was performed. Fifteen miRNAs were tested, including 10 miRNAs that were predicted by TLDA to be significantly differentially expressed in response to both anesthetics compared with controls, and five miRNAs that were not significantly differentially expressed. Bland-Altman analysis was performed across the validations to compare the fold changes measured by TLDA with those measured by standard qRT-PCR (fig. 2
). Bland-Altman analysis demonstrated that the mean difference and the 95%CIs for the limits of agreement between both methods were 0.01 and −0.26 to 0.27, respectively. There was agreement in 30 of 30 pairs of samples (100%). The Bland-Altman limits of agreement were narrow and the mean difference was low, which suggest that the differential expression identified by these two techniques was concordant.
The detailed mechanisms of the influence of anesthetic agents on the liver are still unknown. It is now well established that miRNAs are critical regulators of carcinogenesis, inflammation, and metabolism, and that the reactions of drugs are mediated via
Our study provides the first evidence that sevoflurane and propofol anesthesia affect miRNA expression in healthy rat livers. By analyzing the expression profiles obtained from rat livers under anesthesia, the results show that these anesthetic agents caused many miRNA expression changes. We used hierarchical clustering with heat map presentation to visualize both semantic similarities and expression levels of the miRNAs, to determine the similarities between the samples and the miRNAs, and to clarify whether a group-specific expression pattern could be revealed. With similar expression patterns, samples derived from the same subtypes were clustered together. The results show that hierarchical clustering separated the samples into three distinct branches, representing the sevoflurane anesthesia group, the propofol anesthesia group, and the control group. This means that the miRNA expression pattern was particular for each anesthetic agent. Our study suggests that the anesthetic agents prescribed during animal sampling will affect miRNA expression in studies by other groups. Investigators should thus pay attention to the choice and influences of anesthetic agents when conducting studies in which miRNAs play a role. It has been calculated that each miRNA binds to 100 different target messenger RNAs on average, and the same target gene may be regulated by a given miRNA under different situations, allowing for enormous complexity and flexibility of their regulatory potential.20
In our previous studies, we detected that 99 transcripts of 10,000 genes were influenced by sevoflurane anesthesia by performing microarray analysis,4
and that propofol anesthesia also affected some of them by qRT-PCR in the liver.5
The findings in the present study are in keeping with our previous reports.
There are some methods to comprehensively measure miRNA expression. Previous reports have compared these profiling platforms and concluded that the results of TLDA based on qRT-PCR are the most stable.21
A robust experimental system should not only be technically replicable, but should also yield similar results to alternative experimental methodologies. Accordingly, we directly compared TLDA with standard qRT-PCR. The fold changes predicted for these 15 miRNAs by the two methods were equivalent. Hence, these analyses show that the TLDA results are accurate and can be confirmed using an independent experimental approach. In addition to our study, Hui et al
showed that the technical reproducibility of TLDAs was high (intrasample correlations more than 0.9, accuracy 92.8%). Mees et al
also demonstrated that the miRNA expression validation rate of TLDA is 100% against standard qRT-PCR.
Our study has some limitations. (1) This animal study was conducted in rats; hence, the study results cannot be directly extrapolated to humans; (2) Our report examined and analyzed only miRNA expression changes. Although changes in gene expression are likely to affect subsequent protein levels, there are many other dynamically regulated processes involved, very often resulting in a lack of correlation between gene and protein expression. More importantly, the final functional entities in cells are not messenger RNAs and miRNAs, but proteins that undergo many posttranslational modifications controlling protein activity, localization, and interactions with other proteins and molecules;23
(3) We used extracts of whole liver as samples. Because miRNAs are induced not only in hepatocytes but also in other cells of the liver, it is possible that our observed result reflects a summation of both suppression and activation in different cells. However, we think this study is a useful preliminary investigation into the relationship between anesthetics and miRNA expression changes. For further analyses, localization of gene-expression changes should be investigated; (4) We tested a very large number of miRNAs and thus there is the risk that some of the results are false positives. We dealt with this point as follows. We used careful manual inspection and applied a false discovery rate control to reduce the risk of false positives. Although the raw P
values of our results are low, the risk of false positives remains; (5) Further research using antagonist or specific animal models are needed to demonstrate that each miRNA function is induced by anesthetic agents; and (6) In this study, experimental times under anesthesia were only 6 h. There is a need to verify time-dependent changes of miRNA expression.
In conclusion, the results obtained in our study revealed that different miRNA expression patterns are induced by sevoflurane and propofol anesthesia. The differences in miRNA expression patterns are anticipated to play an important role in distinctive gene-expression changes of each anesthetic agent. Further studies are needed to determine the functional consequences of miRNA modulation by anesthetics.
† QVALUE was written by Alan Dabney and John Storey. http://genomics.princeton.edu/storeylab/qvalue/
; version 1.0. Accessed April 20, 2012. Cited Here...
1. Frink EJ Jr, Morgan SE, Coetzee A, Conzen PF, Brown BR Jr. The effects of sevoflurane, halothane, enflurane, and isoflurane on hepatic blood flow and oxygenation in chronically instrumented greyhound dogs. ANESTHESIOLOGY. 1992;117:1245–1252
2. Carmichael FJ, Crawford MW, Khayyam N, Saldivia V. Effect of propofol infusion on splanchnic hemodynamics and liver oxygen consumption in the rat. A dose-response study. ANESTHESIOLOGY. 1993;79:1051–60
3. Crawford MW, Lerman J, Saldivia V, Carmichael FJ. Hemodynamic and organ blood flow responses to halothane and sevoflurane anesthesia during spontaneous ventilation. Anesth Analg. 1992;75:1000–6
4. Sakamoto A, Imai J, Nishikawa A, Honma R, Ito E, Yanagisawa Y, Kawamura M, Ogawa R, Watanabe S. Influence of inhalation anesthesia assessed by comprehensive gene expression profiling. Gene. 2005;356:39–48
5. Nakazato K, Yoshida Y, Takemori K, Kobayashi K, Sakamoto A. Expressions of genes encoding drug-metabolizing enzymes are altered after sevoflurane, isoflurane, propofol or dexmedetomidine anesthesia. Biomed Res. 2009;30:17–24
6. Chen X, Murad M, Cui YY, Yao LJ, Venugopal SK, Dawson K, Wu J. miRNA regulation of liver growth after 50% partial hepatectomy and small size grafts in rats. Transplantation. 2011;91:293–9
7. Raschzok N, Werner W, Sallmon H, Billecke N, Dame C, Neuhaus P, Sauer IM. Temporal expression profiles indicate a primary function for microRNA during the peak of DNA replication after rat partial hepatectomy. Am J Physiol Regul Integr Comp Physiol. 2011;300:R1363–72
8. Esau C, Davis S, Murray SF, Yu XX, Pandey SK, Pear M, Watts L, Booten SL, Graham M, McKay R, Subramaniam A, Propp S, Lollo BA, Freier S, Bennett CF, Bhanot S, Monia BP. miR-122 regulation of lipid metabolism revealed by in vivo antisense targeting. Cell Metab. 2006;3:87–98
9. Krützfeldt J, Stoffel M. MicroRNAs: A new class of regulatory genes affecting metabolism. Cell Metab. 2006;4:9–12
10. Zhang B, Farwell MA. microRNAs: A new emerging class of players for disease diagnostics and gene therapy. J Cell Mol Med. 2008;12:3–21
11. Hui AB, Shi W, Boutros PC, Miller N, Pintilie M, Fyles T, McCready D, Wong D, Gerster K, Waldron L, Jurisica I, Penn LZ, Liu FF. Robust global micro-RNA profiling with formalin-fixed paraffin-embedded breast cancer tissues. Lab Invest. 2009;89:597–606
12. Obal D, Preckel B, Scharbatke H, Müllenheim J, Höterkes F, Thämer V, Schlack W. One MAC of sevoflurane provides protection against reperfusion injury in the rat heart in vivo. Br J Anaesth. 2001;87:905–11
13. Orth M, Barter L, Dominguez C, Atherley R, Carstens E, Antognini JF. Halothane and propofol differentially affect electroencephalographic responses to noxious stimulation. Br J Anaesth. 2005;95:477–84
14. Wang B, Howel P, Bruheim S, Ju J, Owen LB, Fodstad O, Xi Y. Systematic evaluation of three microRNA profiling platforms: Microarray, beads array, and quantitative real-time PCR array. PLoS ONE. 2011;6:e17167
15. Schonrock N, Ke YD, Humphreys D, Staufenbiel M, Ittner LM, Preiss T, Götz J. Neuronal microRNA deregulation in response to Alzheimer’s disease amyloid-beta. PLoS ONE. 2010;5:e11070
16. Storey JD, Tibshirani R. Statistical significance for genomewide studies. Proc Natl Acad Sci USA. 2003;100:9440–5
17. Tsuboko Y, Sakamoto A. Propofol anaesthesia alters the cerebral proteome differently from sevoflurane anaesthesia. Biomed Res. 2011;32:55–65
18. Kawaguchi H, Hirakawa K, Miyauchi K, Koike K, Ohno Y, Sakamoto A. Pattern recognition analysis of proton nuclear magnetic resonance spectra of brain tissue extracts from rats anesthetized with propofol or isoflurane. PLoS ONE. 2010;5:e11172
19. Klaassen CD, Lu H, Cui JY. Epigenetic regulation of drug processing genes. Toxicol Mech Methods. 2011;21:312–24
20. Brennecke J, Stark A, Russell RB, Cohen SM. Principles of microRNA-target recognition. PLoS Biol. 2005;3:e85
21. Aldrich BT, Frakes EP, Kasuya J, Hammond DL, Kitamoto T. Changes in expression of sensory organ-specific microRNAs in rat dorsal root ganglia in association with mechanical hypersensitivity induced by spinal nerve ligation. Neuroscience. 2009;164:711–23
22. Mees ST, Schleicher C, Mardin WA, Senninger N, Colombo-Benkmann M, Haier J. Analyzing miRNAs in ductal adenocarcinomas of the pancreas. J Surg Res. 2011;169:241–6
23. Skalnikova H, Vodicka P, Gadher SJ, Kovarova H. Proteomics of neural stem cells. Expert Rev Proteomics. 2008;5:175–86
© 2012 American Society of Anesthesiologists, Inc.
Publication of an advertisement in Anesthesiology Online does not constitute endorsement by the American Society of Anesthesiologists, Inc. or Lippincott Williams & Wilkins, Inc. of the product or service being advertised.