British Journal of Anaesthesia, 120 (2): 368e375 (2018)

doi: 10.1016/j.bja.2017.11.066
Advance Access Publication Date: 21 November 2017
Translational Studies

TRANSLATIONAL STUDIES

Differential effects of sevoflurane on the metastatic

potential and chemosensitivity of non-small-cell
lung adenocarcinoma and renal cell carcinoma
in vitro
S. Ciechanowicz1, H. Zhao1, Q. Chen1,2, J. Cui1, E. Mi1, E. Mi1, Q. Lian3 and
D. Ma1,*
1
Department of Surgery and Cancer, Anaesthetics, Pain Medicine and Intensive Care, Faculty of Medicine,
Imperial College London, Chelsea & Westminster Hospital, London, UK, 2Department of Anesthesiology,
Southwest Hospital, Third Military Medical University, Chongqing, China and 3The Second Affiliated
Hospital, Wenzhou Medical University, Wenzhou, China

*Corresponding author. E-mail: d.ma@imperial.ac.uk

Abstract
Background: Increasing evidence suggests that perioperative factors including anaesthetics influence cancer recurrence
and metastasis after surgery. This study investigated the influence of sevoflurane on the response of lung and renal
cancer cells to cisplatin, with focus on transforming growth factor-beta (TGF-b) and osteopontin (OPN) that are both
closely associated with cancer tumorigenesis and metastasis.
Methods: Non-small cell lung adenocarcinoma (A549) and renal cell carcinoma (RCC4) cells were exposed to 3.6% sev-
oflurane for two hrs. Malignant potential represented by cell viability, migration, chemosensitivity to cisplatin was
evaluated. Expression of OPN, TGF-b1, TGF-b receptor type II (TGF-bRII) and the canonical downstream effector Smad3
was assessed. SiRNA knockdown of TGF-b1 and OPN and chemical inhibition of TGF-bRI/II was performed.
Results: Sevoflurane reduced cell viability (0.394) versus control (0.459) (P < 0.01), enhanced chemosensitivity but had no
effect on migration of A549 cells. It enhanced viability (0.467) versus control (0.347) (P < 0.001), chemoresistance and
migration of RCC4. In A549, there was enhanced nuclear Smad3. In RCC4, TGF-bRII and OPN were upregulated, while
TGF-b1 was over- expressed with reduced nuclear Smad3. TGF-bRII inhibition and OPN knockdown abolished
sevoflurane-mediated viability, and migration, respectively, in RCC4.
Conclusions: Sevoflurane promotes the metastatic potential of renal carcinoma, but not of non-small cell lung cancer.
This may be associated with its differential effect on cellular signalling including TGF-b. Our findings indicate that
sevoflurane may have different effects on the metastatic potential and chemosensitivity of different tumour types.

Keywords: anaesthetic drugs; carcinoma; kidney; lung; metastasis

Editorial decision: June 7, 2017; Accepted: June 28, 2017

© 2017 Published by Elsevier Ltd on behalf of British Journal of Anaesthesia.
For Permissions, please email: permissions@elsevier.com

368
 Differential effects of sevoflurane on metastatic potential and chemosensitivity
Editor’s key points
 It is unknown how sevoflurane affects the sensitivity of
lung or renal carcinoma metastases to cisplatin.
 Sevoflurane promoted the metastatic potential and
chemoresistance of renal cell carcinoma, but not of
lung carcinoma.
 The distinct effect of sevoflurane on both cell lines was
related to differential effects on TGF-b and osteopontin
signaling.
Surgical excision is the primary treatment for most solid ma-
lignancies. Despite extensive surgical resection, the existence
of minimal residual disease can lead to a high incidence of
cancer recurrence and metastasis postoperatively, and is the
leading cause of death in cancer patients.1 Survival and pro-
gression of these residual cancer cells depends on multiple
synergistic perioperative factors, which may include anaes-
thetics and analgesics.2
Volatile anaesthetics are used extensively in surgical
oncology. In vitro study has indicated a differential effect on
tumours, with sevoflurane inhibiting ADC3 and colon cancer;4
but potentiating breast5 and ovarian cancer.6 However, the
molecular mechanisms underlying these effects are unknown.
Investigation of direct effects of volatile agents on malignant
molecular processes has thus far largely focused on receptor
tyrosine kinase-dependent pathways. However, transforming
growth factor-b (TGF-b) signaling could be implicated.7 This
signal transduction network is involved in regulation of a
number of developmental processes including growth and
cellular differentiation, proliferation, migration, adhesion and
apoptosis. TGF-b binding to the cell surface TGF-b type I/II re-
ceptor complex initiates a phosphorylation cascade of intra-
cytoplasmic effectors. The TGF-b family share a canonical
signaling pathway, ’similar to Mothers against Decapentaplegic’
(Smad). Receptor-regulated Smad proteins (Smad2/3) combine
with common-mediator Smad4 to regulate transcription of
target genes. Non-canonical or ’non-Smad’ signaling also oc-
curs, alongside cross-talk with other pathways, resulting in wide
ranging effects of TGF-b signaling.8 Osteopontin (OPN) is a
secreted phosphoprotein normally expressed in bone and
epithelial cells including the lung bronchi and renal tubules.
Overexpression is observed in many cancers,9 and it is associ-
ated with invasion and metastasis, including in lung cancer.10
Cisplatin is a cytotoxic chemotherapeutic agent routinely
used in the treatment of solid malignancies, including renal
cell carcinoma (RCC) and non-small-cell lung adenocarcinoma
(ADC). These aggressive carcinomas commonly develop che-
moresistance clinically and were therefore selected for this
study.
The aim of this study was to investigate the effect of sev-
oflurane on cell viability, migration and chemoresistance to
cisplatin of ADC and RCC cell lines, and the role of TGF-b and
OPN signaling in the mechanism of sevoflurane-mediated
effects.
Methods
Cell culture
Human non-small cell lung adenocarcinoma A549 and renal
cell carcinoma RCC4 cell lines were cultured in 75 cm2 tissue
culture flasks (VWR, Leicestershire, UK) as a monolayer in
- 369
RPMI-1640 medium (GIBCO, Invitrogen, Paisley, UK), supple-
mented with 10% heatinactivated fetal bovine serum (Thermo
Scientific, Epsom, UK), with 2mM L-glutamine and 100Uml-1
penicillin-streptomycin (Invitrogen). Cells were maintained at
37
C in humidified air with 5% CO2.
Sevoflurane exposure
Before gas exposure, cells were split and seeded at equal cell
density in 30-mm2 Petri dishes (VWR, Leicestershire, UK) and 6,
24 and 96-well plates, and incubated at 37
C and 5% CO2. When
cultures reached 70-80% confluence, cell media was replen-
ished and plates were placed in a 1.5 L purpose-built airtight,
temperature-controlled gas chamber with inlet and outlet
valves and an internal electric fan to ensure mixture of gases.
For sevoflurane delivery, the chamber inlet was connected to a
closed gas delivery system consisting of a calibrated oxygen
flow meter and an inline sevoflurane vaporizer (Abbott Labo-
ratories, Maidenhead, UK). Outlet gas was analysed (Datex gas
monitor, Helsinki, Finland) until a sevoflurane concentration of
3.6% (equivalent MAC value to 2% isoflurane)6 11 was reached
from 2L/min flow air carrier gas (oxygen 21%) balanced with 5%
carbon dioxide (CO2), with gas flow delivery continuing for a
minimum of 5 min to ensure equilibration. Cells were incu-


bated in the sealed chamber for two h at 37 C before they were
returned to the standard cell culture incubator at maintenance
conditions. Cells in the control group were placed in an iden-
tical chamber with 21% oxygen and 5% CO2. Cells in the naı̈ve
group remained in the standard cell culture incubator
throughout. An exposure duration of two h was used, in
accordance with previous studies,6 11 as this is a reasonable
minimum duration for abdominal and thoracic cancer surgery.
Treatment with chemotherapeutic agent
Cisplatin [cis-diammine (cyclobutane-1,1-dicarboxylato) plat-
inum(II)] (DDP) is a well-established cytotoxic agent for solid
malignancies. 1mM stock solution (Calbiochem, USA, Cat
232120) was diluted with RPMI media to make a clinically
relevant concentration of 80 mM DDP-supplemented media.
Just before gas exposure, cells were treated with or without
DDP- supplemented media and 24 h later reverted back to
standard RPMI media.
Cell viability assay
Cells were split into 96-well plates and incubated overnight. Cells
were treated with or without 80 mM DDP-supplemented media,
and exposed to sevoflurane gas for two h. At 18 h post gas
exposure, 10% Alamar Blue™ was added to each well and incu-


bated in the dark at 37 C for six h. Absorbance was measured at
540 nm. Recordings were blanked using media and media-DDP
only wells and negative controls provided by wells with media/
media-DDP plus Alamar Blue™ but no cells. For TGF-bRI/II in-
hibition, cells were incubated in media containing 5-20 mM
LY2109761 (LY21) (Santa Cruz Biotechnology, Inc Cat SC- 396262),
a dual inhibitor of TGF-b receptor I/II kinases (IC50¼69 nM),
concentration optimised to cell line, for two h pregas exposure.
Wound healing (migration) assay
Cell migration was assessed using a method as previously
described.12 After cultured cells reached full confluence in six or
24-well plates, an artificial gap was created by scratching with a
sterile plastic pipette tip (1000 mL) to form a ’cross’ devoid of
 370 - Ciechanowicz et al.

adherent cells. Gap closure or healing was measured from im-
ages taken by a phase-contrast microscope and digital camera
(Olympus CK30, Tokyo, Japan). The same four scratch areas from
each well were analysed before and 24 h after treatment using
ImageJ 1.35 (NIH, Bethesda, MD, USA) software. Wound healing
was quantified as the mean percentage of the remaining gap area
compared with the area of the initial gaps at t ¼ 0.13 For siRNA-
mediated TGF-b1 and OPN knockdown, A549 and RCC4 cells at
70-80% confluence were transfected with 20 mM human-specific
TGF-b1 siRNA: sense strand 5’-AGGUUAUU UCCGUGGGAUATT-
3’, antisense strand 5’-UAUCCCACGGAAAU AACCUAG-3’; or
OPN siRNA: sense strand 5’-GGUUGUCCAGC AAUUAAUATT-3’,
antisense strand 5’-UAUUAAUUGCUGGA CAACCGT-3’ (Flex-
iTube siRNA, Qiagen, Sussex, UK). A scrambled non-sense siRNA
without specific gene-silencing activity was used as a negative
control (Qiagen, Sussex, UK). Transfection was achieved using

Fig 1. Sevoflurane inhibited cell viability of A549 cells. (A) Alamar Blue reduction of naı̈ve, control and sevoflurane-treated A549 cells at 24 h
post gas exposure (n¼3). (B) Alamar Blue absorbance after inhibition of TGF-bRI/II with LY2109761 (LY21) at 24h post gas exposure (n¼3). (C)
Wound healing at 24 h post sevoflurane exposure in cultures with and without 24 h of 40 mM DDP treatment. Images are X4 magnification.
Dashed-line indicates initial wound at t ¼ 0h. (D) Wound/gap closure expressed as percentage remaining wound area at 24h post gas
exposure. 24h data naive (n¼4), control (n¼4), sevo (n¼4), naive þ DDP (n¼3), control þ DDP (n¼3), sevo þ DDP (n¼3). N¼number of in-
dependent experiments. (E) Wound healing after siRNA knockdown of TGF-b1 and OPN at 24h post gas exposure (n¼6). SSi¼scramble RNA
negative control, TSi¼TGF-b1 SiRNA-treated, OSi¼OPN SiRNA-treated. Data are expressed as mean (SEM) and analysed using one-way
ANOVA with Tukey’s post hoc test. *P<0.05, **P<0.01, ***P<0.001.
 Differential effects of sevoflurane on metastatic potential and chemosensitivity - 371

20nM lipofectamine RNAi MAX (Invitrogen, Paisley, UK). Cells

were then incubated for 24h at 37
C and 5% CO2, after which cells
were washed with PBS (Sigma-Aldrich, Gillingham, UK) and RPMI
medium was replaced before gas exposure.

Immunocytochemistry
Cells were fixed in 4% paraformaldehyde then rinsed with PBS
(Sigma-Aldrich). Permeabilisation and blocking was achieved
with 0.1% Triton X-100 in PBS (PBST) containing 10% normal
donkey serum (Millipore) for one h. Cells were incubated in the
dark overnight at 4
C in PBST with one of the following anti-
bodies: rabbit anti-TGF-b (Abcam, Cambridge, UK), rabbit anti-
Osteopontin, (Abcam), rabbit anti-Smad3 (Abcam), or mouse
anti-TGF-bRII (Abcam). Cells were then washed and incubated
for one h in PBST with one of the following secondary anti-
bodies: Rhodamine-conjugated donkey anti-rabbit IgG (Milli-
pore, Oxford, UK), Fluorescein isothiocyanate (FITC)-
Fig 2. Effect of sevoflurane on TGF-b1 expression in A549 cells.
conjugated donkey anti-rabbit IgG (Millipore) or FITC-
TGF-b1 expression in A549 cells detected by immunofluores-
conjugated donkey antimouse IgG (Millipore), followed by cence staining. Mean fluorescent intensity is displayed as a box
washing and sealing with 4,6-diamidino-2-phenylindole and whisker plot with mean plus or minus upper and lower
(DAPI) (Vector Laboratories). Images of 10 high-power fields limits (n¼6). Data are expressed as mean (SEM) and analysed
at x20 magnification were generated from each cover slip us- using Kruskal-Wallis with Dunn’s post hoc test.
ing an AxioCam digital camera (Zeiss, Welwyn Garden City,
UK) mounted on an Olympus BX60 microscope (Olympus,
Middlesex, UK) with Zeiss KS-300 software (Zeiss). Images for (P¼0.0124) (Fig. 1E) and similar results were observed for
each marker were obtained using identical exposure times and control cells, which indicates that OPN is important for cell
fluorescent intensity measured from five random cells and five migration in A549.
background areas using ImageJ 1.35 software. Net mean in-
tensity was calculated as mean cell minus background in-
tensity for each of the 10 images obtained from the coverslip. Sevoflurane upregulated expression of nuclear Smad3
in DDP-treated A549 cells
Statistical analysis TGF-b1 expression was unchanged in cells exposed to sevo-
flurane at 24 h (Fig. 2). Nuclear Smad3 expression was elevated
Data was expressed as mean (SEM). One-way ANOVA was
performed with Tukey’s post hoc multiple comparisons test
using GraphPad Prism (GraphPad Software, La Jolla, CA, USA),
except for immunocytochemistry data which was assessed
nonparametrically using Kruskal-Wallis with Dunn’s post hoc
test. A P-value of < 0.05 was considered to be statistically sig-
nificant. Power analysis revealed a group size of n¼6 to detect a
difference in means of 0.5 to achieve a power of 0.8 (1-b) and
a¼ 0.05.

Results
Sevoflurane inhibited cell viability but not migration of
A549 cells
Sevoflurane exposed A549 cells had reduced cell viability
compared with control at 24 h post gas exposure (0.394 vs
0.459) (P<0.01) and in the presence of DDP (0.490 vs 0.544)
(P<0.001) (Fig. 1A). Inhibition of TGF-bRI/II signalling with
LY2109761 did not reverse the inhibitory effect of sevo-
flurane on A549 cell survival (P¼0.2028) (Fig. 1B). Wound
healing/migration was not inhibited by sevoflurane at 24 h
[-1.6 (4.8%) vs 9(2.2%)] (P¼0.0506) (Fig. 1C and D). Treatment
Fig 3. Sevoflurane exposure enhanced nuclear Smad3 expres-
of control cells with DDP inhibited cell migration and sub- sion in A549 cells. (A) Nuclear Smad3 expression in DDP-treated
sequent gap closure (P¼0.0001). To investigate possible ad- A549 cells as detected by immunofluorescence staining at 24 h
ditive inhibitory effects, gap expansion of DDP-treated cells post sevoflurane exposure. Fluorescent intensity of Smad3 dis-
with sevoflurane co-treatment was measured but this did played as a box-and-whisker plot with mean upper and lower
not achieve statistical significance [-39 (22.6%) vs -9.3 (4.9%)], limits. Naive þ DDP (n¼3), Control þ DDP (n¼3) Sevo þ DDP
(P¼0.21) (Fig. 1C and D). OPN knockdown resulted in signif- (n¼4). Data were analysed with Kruskal-Wallis with Dunn’s post
icant inhibition of migration compared with SSi control in hoc test. ***P<0.001.
sevoflurane treated cells [-9.1% (8.8) vs 18.3% (1.8)]
 372 - Ciechanowicz et al.

by sevoflurane treatment at 24h [16.5 (2.1) vs 5.3 (0.4)] (P<0.001)
(Fig. 3).
Sevoflurane enhanced cell viability of RCC4 cells
Exposed cultures had greater cell viability compared with con-
trol at 24 h post sevoflurane exposure (0.467 vs 0.347) (P<0.001)
and when in combination with 80 mM DDP (0.428 vs 0.355)
(P<0.05) (Fig. 4A). In the non-DDP treated cultures growth was
promoted by 14.1% (3.3%) with sevoflurane compared with un-
treated control, and in DDP-treated cells, growth was promoted
by 12.2% (5.0%) when sevoflurane was added compared with
DDP-control. Inhibition of TGF-bRI/II with 20 mM LY21 reverses
the growth promoting effects of sevoflurane (P<0.05) (Fig. 4B).

Fig 4. Sevoflurane promoted cell viability, chemoresistance and migration of RCC4 cells. (A) Alamar Blue reduction of control and
sevoflurane-treated RCC4 cells with and without treatment with DDP at 24 h post gas exposure. (B) Alamar Blue absorbance after LY21
treatment at 24 h post gas exposure, (C) Wound healing at 24h post sevoflurane exposure in cultures with and without 24h of 40 mM DDP
treatment. Images are X4 magnification. Dashed-line indicates initial wound at t ¼ 0h. (D) Wound/gap closure expressed as percentage
remaining wound area at 24 h post gas exposure. (E) Wound healing after siRNA knockdown of TGF-b1 and OPN at 24h post gas exposure
(n¼3). SSi¼scramble RNA, TSi¼TGF-b1 siRNA-treated, OSi¼OPN siRNA-treated. Data expressed as mean (SEM) and analysed using ANOVA
with Tukey’s post hoc test. *P<0.05, **P<0.01, ***P<0.001.
 Differential effects of sevoflurane on metastatic potential and chemosensitivity
Sevoflurane enhanced migration of RCC4 cells
Migration was markedly enhanced in sevoflurane-treated cells
compared with control at 24h after exposure (74.9% vs 52.1%)
(P¼0.03) (Fig. 4C and D). DDP-treated cells did not demonstrate
a difference in migration, but cell confluence appears more
preserved in cultures exposed to sevoflurane at 24 h post-
treatment (Fig. 4C).
OPN is involved in sevoflurane-mediated effects on
migration of RCC4 cells
OPN knockdown significantly reversed the positive effect of
sevoflurane on migration in RCC4 at 24 h post gas exposure
(42% vs 72.9%) (P¼0.017). Knockdown of TGF-b1 also appeared
to reverse sevoflurane-mediated effects on RCC4 migration,
although this did not achieve significance (P>0.05) (Fig. 4E).
Knockdown of both TGF-b1 and OPN had no effect on the
migration of control cells (Fig. 4E).
Fig 5. Sevoflurane increased TGF-b1 and OPN expression in
RCC4 cells. (A) TGF-b1 expression in RCC4 cells was detected by
immunofluorescence staining at 24h post sevoflurane exposure.
Fluorescent intensity of TGF-b1 displayed as a box-and-whisker
plot with mean plus or minus upper and lower limits. Naive
(n¼3), Control (n¼4), Sevo (n¼4). (B) OPN in RCC4 cells was
detected by immunofluorescence staining at 24 h post sevo-
flurane exposure. Fluorescent intensity of OPN displayed as a
box-and-whisker plot with mean plus or minus upper and lower
limits (n¼4). Data were analysed with Kruskal-Wallis with
Dunn’s post hoc test. ***P<0.001.
- 373
Sevoflurane upregulated expression of TGF-b1, TGF-
bRII and OPN but reduces nuclear Smad3 in DDP-
treated RCC4 cells
TGF-b1 expression was enhanced by sevoflurane exposure
[22.4 (1.5) vs 11.7 (0.6)] (P<0.001) (Fig. 5A). Levels of OPN were
also markedly raised at 24 h post sevoflurane exposure [18.1
(1.4) vs 9.1 (0.5)] (P<0.001) (Fig. 5B). Correspondingly, TGF- bRII
was elevated in DDP-treated cells after exposure to sevo-
flurane [13.1 (0.5) vs 4.7 (0.4)] (P<0.001) (Fig. 6A). Nuclear Smad3
expression was reduced in sevoflurane-treated cells [12.4
(0.99) vs 22.3 (1.4)] (P<0.001) (Fig. 6B).
Fig 6. Sevoflurane enhanced TGF-bRII and reduced nuclear
Smad3 expression in RCC4 cells. (A) TGF-bRII expression in DPP-
treated RCC4 cells was detected by immunofluorescence stain-
ing at 24 h post sevoflurane exposure. Fluorescent intensity of
TGF-bRII displayed as box-and-whisker plot with mean plus or
minus upper and lower limits. Naive þ DDP (n¼3), Control þ
DDP (n¼4), Sevo þ DDP (n¼4). (B) Nuclear Smad3 expression in
DDP-treated RCC4 cells was detected by immunofluorescence
staining at 24 h post sevoflurane exposure. Fluorescent intensity
of Smad3 displayed as a box-and-whisker plot with mean plus
or minus upper and lower limits. Naive þ DDP (n¼4), Control þ
DDP (n¼4), Sevo þ DDP (n¼3). Data were analysed using Kruskal-
Wallis with post hoc Dunn’s test. **P<0.01, ***P<0.001.
 374 - Ciechanowicz et al.

Discussion to TGF-b1. Sevoflurane could be changing the balance of signal

transduction to favour the Smad pathway. This is supported
Our study provides evidence that sevoflurane promotes the
by a recent studies showing inhibition of the non-Smad
metastatic potential of RCC cells, with associated upregulation
pathway components p38 MAPK,21 Akt,23 and ERK24 by sevo-
of TGF-b and OPN signaling. Sevoflurane conferred cytopro-
flurane in lung. Gordian and colleagues25 found exogenous
tection to RCC cells in the presence of DDP at 24h and cellular
TGF-b1 enhances the chemosensitivity of A549. However other
confluence was more preserved at 48 h. This demonstrates an
studies have found that TGF-b/Smad promotes cell invasion
“antagonistic” action of sevoflurane on DDP cytotoxicity with
and epithelial-mesenchymal transition in NSCLC,26 and Smad
enhanced chemoresistance, in keeping with findings by Bro-
activation correlates with poor prognosis in ADC.27 An expla-
zovic and colleagues14 Inhibition of TGF-b RI/II mitigated the
nation for these contradictory findings may be that the
promoting effects of sevoflurane, supporting the role of TGF-b
inhibitory effects of sevoflurane on ADC are downstream to
in sevoflurane-mediated metastatic potentiation. TGF-b1 has
Smad3 activation, or potentiated by an alternative pathway,
been recognized to promote chemoresistance to DDP.11
which should be investigated further.
This study found a positive effect of sevoflurane on
It is possible that the differences in behaviour to sevo-
migration of RCC cells at 24 h, which agrees with effects found
flurane exposure of A549 and RCC4 originate from differences
in breast cancer.5 From a clinical perspective, this could indi-
in the stage of carcinogenesis, which is related to TGF-b sig-
cate that induced phenotypic changes persist in residual
nalling alterations as a result of mutations in TGF-bRI/II and
cancer cells into the postoperative period per se.
Smad proteins, and the presence of other transcription factors
Sevoflurane treatment resulted in increased TGF-b1 and
causing cross-talk differences. This is an assumption and
TGF-bRII and reduced nuclear Smad3 expression. This could
warrants further study.
illustrate a direct or indirect pro-tumorigenic and chemo-
In summary, our data indicate a potential role of TGF-b and
resistance action of sevoflurane on RCC cells, involving
OPN signalling in sevoflurane-mediated metastatic potentia-
uncoupling of the TGF-b1/Smad pathway, resulting in auto-
tion of RCC, suggesting a possible increased risk of post-
induction of TGF-b1,7 thus promoting tumour cell ’aggres-
operative recurrence after sevoflurane anaesthesia, and
siveness’. Knockdown of TGF-b1 may mitigate the pro-
conversely, minor inhibitory effects of sevoflurane on ADC.
migratory effects of sevoflurane on RCC, supporting this as a
Our findings indicate that sevoflurane may have different ef-
mechanism. This effect concurs with our previous study on
fects on different tumour types. In vivo studies that model the
cell growth and migration of RCC4 cells after exposure to iso-
tumour microenvironment and clinical trials are urgently
flurane,15 and taken together, implies a class-effect of volatile
needed to evaluate this further.
agents on RCC.
It has been reported that increased TGF-b1 and TGF-bRI/II
expression and reduced Smad2/3 are possible biomarkers of Authors’ contributions
tumorigenesis in RCC,16 and studies have shown negative cor-
Study design/planning: S.C., H.Z., D.M.
relation of Smad3 to tumour grade and patient survival,17 and
Study conduct: S.C.
negative association of Smad expression with migration.18 Our
Data analysis: S.C., H.Z.
findings suggest that this malignant phenotype is exaggerated
Writing paper: S.C., H.Z., Q.C., J.C., E.M., E.M., Q.L., D.M.
by sevoflurane exposure. In contrast, renal ischaemiar-
Revising paper: All authors
eperfusion injury studies show that sevoflurane is cytopro-
tective to renal proximal tubule epithelium via phosphorylation
of cytoprotective kinases ERK and Akt, and that this is mediated
Declaration of interest
via upstream TGF-b signalling,19 specifically the TGF-b1/Smad3
pathway. Our data do not support the hypothesis that this could D.M. is a board member of BJA and other authors have no
also be a mechanism of protumorigenesis, and instead find the competing interests.
opposite effect, of reduced TGF-b1/Smad3 activation. An
explanation for this is that the differential response to TGF-b1 of
normal and malignant cells could confer different pathways of
Funding
cytoprotection from sevoflurane in normal renal epithelium This work was supported by BOC Chair grant of Royal College
and in RCC. This warrants further study. of Anaesthetists, London, UK and a collaborative grant be-
OPN is a metastasis-associated-protein and is implicated in tween Q.L. and D.M.
chemoresistance in RCC.10 RCC OPN expression was elevated
with sevoflurane exposure consistent with findings in previ-
ous studies of normal renal epithelium.20 Knockdown References
reversed the pro-migration effect of sevoflurane, suggesting
1. Tohme S, Simmons RL, Tsung A. Surgery for Cancer: A
OPN has a major role in the action of sevoflurane in RCC. In
Trigger for Metastases. Cancer Res 2017; 77: 1548e52
contrast, sevoflurane caused minor inhibition of cell viability
2. Buggy DJ, Hemmings HC. Special issue on anaesthesia and
of ADC, with enhanced chemosensitivity, but no significant
cancer. Br J Anaesth 2014; 113(Suppl 1): i1e3
change to cell migration. This inhibition was evident at 24 h
3. Liang H, Gu MN, Yang CX, Wang HB, Wen XJ, Zhou QL.
but can persist to 48 h, as demonstrated by Liang and col-
Sevoflurane inhibits proliferation, induces apoptosis, and
leagues,3 21 which is likely because of its induction of
blocks cell cycle progression of lung carcinoma cells. Asian
apoptosis.3 22
PacJ Cancer Prev 2011; 12: 3415e20
Our data is the first to demonstrate upregulation of Smad3
4. Kawaraguchi Y. Sevoflurane enhances proliferation of
signalling in ADC by a volatile agent. Unchanged levels of TGF-
human colon cancer HCT116 cells via adenosine
b1 could indicate normal homeostatic regulation, with this triphosphate-sensitive potassium channels (1052.8).
pattern suggesting a possible functional ’sensitisation’ of cells
FASEB J 2014: 28
 Differential effects of sevoflurane on metastatic potential and chemosensitivity
5. Ecimovic P, McHugh B, Murray D, Doran P, Buggy DJ. Ef-
fects of sevoflurane on breast cancer cell function. in vitro.
Anticancer Res 2013; 33: 4255e60
6. Iwasaki M, Zhao H, Jaffer T, et al. Volatile anaesthetics
enhance the metastasis related cellular signalling
including CXCR2 of ovarian cancer cells. Oncotarget 2016;
7: 26042e56
7. Zhang Q, Yu N, Lee C. Vicious cycle of TGF-beta signaling
in tumor progression and metastasis. Am J Clin Exp Urol
2014; 2: 149e55
8. Massague J. TGFbeta signalling in context. Nat Rev Mol Cell
Biol 2012; 13: 616e30
9. Han J, Zhang F, Xie J, Linhardt RJ, Hiebert LM. Changes in
cultured endothelial cell glycosaminoglycans under hy-
perglycemic conditions and the effect of insulin and
heparin. Cardiovasc Diabetol 2009; 8: 46
10. Shevde LA, Samant RS. Role of osteopontin in the patho-
physiology of cancer. Matrix Biol 2014; 37: 131e41
11. Li J, Liu H, Yu J, Yu H. Chemoresistance to doxorubicin
induces epithelial-mesenchymal transition via upre-
gulation of transforming growth factor beta signaling
in HCT116 colon cancer cells. Mol Med Rep 2015; 12:
192e8
12. Yoo KY, Jeong CW, Kim SJ, et al. Sevoflurane concentra-
tions required to block autonomic hyperreflexia during
transurethral litholapaxy in patients with complete spinal
cord injury. Anesthesiology 2008; 108: 858e63
13. Lee HT, Chen SW, Doetschman TC, Deng C, D’agati VD,
Kim M. Sevoflurane protects against renal ischemia and
reperfusion injury in mice via the transforming growth
factor-beta1 pathway. Am J Physiol Renal Physiol 2008; 295:
F128e36
14. Brozovic G, Orsolic N, Knezevic F, et al. Genotoxicity and
cytotoxicity of cisplatin treatment combined with anaes-
thetics on EAT cells. in vivo. Onkologie 2009; 32: 337e43
15. Benzonana LL, Perry NJ, Watts HR, et al. Isoflurane, a
commonly used volatile anesthetic, enhances renal can-
cer growth and malignant potential via the hypoxia-
inducible factor cellular signaling pathway. in vitro. Anes-
thesiology 2013; 119: 593e605
16. Cardillo MR, Lazzereschi D, Gandini O, Di Silverio F,
Colletta G. Transforming growth factor-beta pathway in
- 375
human renal cell carcinoma and surrounding normal-
appearing renal parenchyma. Anal Quant Cytol Histol
2001; 23: 109e17
17. Park JH, Lee C, Suh JH, Chae JY, Moon KC. Nuclear
expression of Smad proteins and its prognostic signifi-
cance in clear cell renal cell carcinoma. Hum Pathol 2013;
44: 2047e54
18. Jingushi K, Ueda Y, Kitae K, et al. miR-629 Targets TRIM33
to Promote TGFbeta/Smad Signaling and Metastatic Phe-
notypes in ccRCC. Mol Cancer Res 2015; 13: 565e74
19. Lee HT, Kim M, Song JH, Chen SW, Gubitosa G, Emala CW.
Sevoflurane-mediated TGF-beta1 signaling in renal prox-
imal tubule cells. Am J Physiol Renal Physiol 2008; 294: F371e8
20. Kharasch ED, Schroeder JL, Bammler T, Beyer R,
Srinouanprachanh S. Gene expression profiling of neph-
rotoxicity from the sevoflurane degradation product
fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether
(“compound A”) in rats. Toxicol Sci 2006; 90: 419e31
21. Liang H, Gu M, Yang C, Wang H, Wen X, Zhou Q. Sevo-
flurane inhibits invasion and migration of lung cancer
cells by inactivating the p38 MAPK signaling pathway.
J Anesth 2012; 26: 381e92
22. Wei GH, Zhang J, Liao DQ, et al. The common anesthetic,
sevoflurane, induces apoptosis in A549 lung alveolar
epithelial cells. Mol Med Rep 2014; 9: 197e203
23. Zhang L, Zhang J, Dong Y, Swain CA, Zhang Y, Xie Z. The
potential dual effects of sevoflurane on AKT/GSK3beta
signaling pathway. Med Gas Res 2014; 4: 5
24. Kim SH, Li M, Pyeon TH, So KY, Kwak SH. The volatile
anesthetic sevoflurane attenuates ventilator-induced lung
injury through inhibition of ERK1/2 and Akt signal trans-
duction. Korean J Anesthesiol 2015; 68: 62e9
25. Gordian E, Li J, Pevzner Y, et al. Transforming growth
factor beta signaling overcomes dasatinib resistance in
lung cancer. PLoS One 2014; 9: e114131
26. Yang H, Zhan L, Yang T, et al. Ski prevents TGF-beta-
induced EMT and cell invasion by repressing SMAD-
dependent signaling in non-small cell lung cancer. Oncol
Rep 2015; 34: 87e94
27. Yu JR, Tai Y, Jin Y, et al. TGF-beta/Smad signaling through
DOCK4 facilitates lung adenocarcinoma metastasis. Genes
Dev 2015; 29: 250e61
Handling editor: C. Boer