Lidocaine suppressed the movement of A549 cells by the CXCR4 signal

Background Lidocaine is a local anesthetic that wildly used in surgical treatment and postoperative medical care for lung cancers. We hypothesized that lidocaine at clinical plasma concentration can inhibit CXCL12/CXCR4 axis-regulated cytoskeletal remodeling thereby reduce the migration of Non-small-cell lung cancers (NSCLC) cells. We determined the effect of lidocaine at clinical plasma concentration on CXCL12-induced cell viability, apoptosis, cell death, monolayer cell wound healing rate, individual cell migration indicators, expression of CXCR4, CD44, and ICAM-1, intracellular Ca 2+ level, and lamentous actin level alteration of NSCLC cells A549 and CXCR4-knocked down A549 cells using CCK-8, Bcl-2 ELISA, Cell death ELISA, wound healing assay, chemotaxis assay, western blotting, QPCR, Fura-2-based intracellular Ca 2+ assay, and Fluorescein Phalloidin staining respectively. bers during CXCL12 stimulation. Results showed that lidocaine reduced cytoskeleton remodeling by about 50% and both the pertussis toxin group and CXCR4 knockdown group failed to show any reaction following CXCL12 stimulation. These results suggested that lidocaine inhibited CXCL12-induced intracellular Ca 2+ releasing thereby reduced cytoskeleton remodeling.


Introduction
Lung cancer is one of the most fatal cancer types with the most global death number among males and the second most global death number among females [1]. As the medical care for lung cancer develops, the mortality of lung cancer decreased in the USA and UK has decreased in recent years. However, in many industrialized nations, the emerging social smoking culture has resulted in higher lung cancer rates in these areas [2]. Among all lung cancer cases, more than four-fths of clinical lung cancer diagnosed are non-small cell lung cancer (NSCLC) [3], a lung cancer type that has diverse pathological features and with an undesirable prognosis.
Clinical medical advancements have improved cancer therapy extensively in recent years [4]. As clinical diagnosis is critical for disease treatment [5], in-depth analyses of lung cancer subtypes have been developed for potential targeting therapy and customized treatment according to their genetic and cellular heterogeneity [6]. Surgical treatment is the most applicable intervention for early-stage lung cancer diagnoses and is critical for the further prescription of proper therapeutic options [7]. One of the clinical factors that might affect the outcome of surgical treatments is the anesthesia during the surgery [8,9]. Many local anesthetics have been found to affect multiple cancer cells [10]. A local anesthetic, lidocaine, has been wildly used in surgical treatment [11] and postoperative medical care [12,13].
However, the doses of lidocaine used in most previous studies are much higher than the concentration in plasma during clinical use of lidocaine (we mentioned it as "plasma concentration"). In this study, we focused on a lower concentration range of lidocaine that is correspondent to clinical plasma concentration. Lidocaine at this concentration range has been reported to inhibit cytoskeletal remodeling and migration of breast cancer cell MDA-MB-231 [22]. We proposed that a similar effect of lidocaine on cell migration is also present in NSCLC cells.
The metastatic potential of NSCLC has been found to be strongly associated with the chemokine CXCL12 (C-X-C Motif Chemokine Ligand 12) and the activity of the CXCL12 receptor CXCR4 (C-X-C chemokine receptor type 4) [23]. CXCR4/CXCL12 axis controls the immunity of the body and the survival, invasion, and metastasis of cancer cells [24] [25]. CXCR4 has been found to be differently expressed in cancer cells and might contribute to the motility of cancer cells [26]. The CXCR4/CXCL12 axis has been proposed as a potential drug target for NSCLC [27]. In this study, we hypothesized that lidocaine at clinical plasma concentration can inhibit CXCL12/CXCR4 axis-regulated cytoskeletal remodeling, thereby decrease migration of NSCLC cells. This investigation can provide a better understanding of the potential pharmacological effects of lidocaine on the clinical treatment of NSCLC.

Cell line and cells culture
A549 cells were purchased from Biofeng (China). Cells were cultured using Ham′s F12K + 2mM Glutamine + 10% Foetal Bovine Serum (FBS) in a cell culture incubator at 37•C 5% CO 2 . Cells were cultured in an FBS-free medium for 24 hours to Synchronize before any assay.

Knockdown and overexpression
Page 4/24 CXCR4 knockdown was achieved by transfecting the CXCR4 shRNA (TRCN0000256866) plasmid into A549 cells to silence the expression of CXCR4. The plasmid was purchased from Sigma-Aldrich (USA). The Sequences of shRNA oligonucleotides are as follows: 5'-TCCTGTCCTGCTATTGCATTA-3'. CXCR4 overexpression was achieved by transfecting the CXCR4 expressing (hCXCR4-mTFP1) plasmid into A549 cells to overexpress CXCR4. The plasmid was purchased from Addgene (USA). The transfection method was described previously [28]. The knowdown and overexpression of CXCR4 were validated by western blotting experiments.

CCK-8 assay
The CCK-8 assay is used to determine cell overall viability and cytotoxicity. The CCK-8 assay was conducted as previously described [29]. Brie y, cells were seeded in 96 well plates. After the incubation with testing drugs, 10 μl of the CCK-8 labeling reagent was added to each well. The cells were further incubated for 4 hours. The absorbance at 450 nm was measured using a microplate reader.

Apoptosis and Cell death detection
Apoptosis and cell death detection were determined using Bcl-2 ELISA kit (Abcam, UK) and Cell Death Detection ELISA plus (Roche, USA) respectively. The method of ELISA was described previously [30,31].
The positive control was induced by high-temperature culture (55 •C for 30 min) as a previous study [32].

Western blotting
Membrane protein expression was determined using the western blotting assay. Membrane proteins of A549 were extracted using Mem-PER™ Plus Membrane Protein Extraction Kit (Thermo Fisher Scienti c, USA). The extraction was conducted as previously described [33,34]. Subsequently, the western blotting assay was conducted as in previous studies [35,36]. Brie y, SDS-PAGE gel electrophoresis was used to separate proteins and semi-dry protein transfer was conducted as described [37]. The membrane was then incubated with primary and secondary antibodies according to the recommended experimental condition. Na+/K+ ATPase protein was used as an internal reference gene to normalized the data. Antibodies used including CXCR4 Antibody (PA3-305), Anti-Alpha 1 Na+/K+ ATPase Antibody (#ANP-001), Recombinant Anti-CD44 antibody [EPR1013Y] (ab51037), Anti-ICAM1 antibody [EP1442Y] (ab53013), Goat Anti-Mouse IgG H&L (HRP) (ab6789), and Goat Anti-Rabbit IgG H&L (HRP) (ab6721). ECL reagent (Thermo Scienti c, USA) was used to visualize the target protein.

QPCR
The CXCR4 mRNA level was determined using a QPCR assay. The extraction was conducted as previously described [38,39]. Brie y, Trizol buffer (Sigma-Aldrich, USA) was used to isolate cellular RNA. An Agilent Bioanalyzer (Agilent 2100; Agilent Technologies, Inc, USA) was used to assess the concentration and integrity of RNA extracted. QuantiTect Reverse Transcription Kit (QIAGEN, USA) and miRCURY LNA SYBR Green PCR Kit (QIAGEN, USA) were used to conduct RT-PCR. The protocol of PCR reaction was described previously [40]. The GAPDH gene was used as an internal reference gene to normalize the data. CXCR4 Human qPCR Primer Pair (NM_003467) and GAPDH Human qPCR Primer Pair (NM_002046) were purchased from OriGene Technologies (USA).

Wound healing assay
Cells migration was determined using the wound healing assay. The method was described previously [41,42]. Brie y, cells were grown in six-well plates at over 90% con uency. A scratch wound was created in the cell monolayer using a 200-µl pipette tip. The cells were cultured with testing reagents and images were recorded every 3 or 6 hours. The wound distance was analyzed using Fiji16 software.

Chemotaxis assay
Real-time cell migration of individual A549 cells was recorded using the m-Slide chemotaxis system (ibidi, Germany). The method was described previously [43,44]. Brie y, A549 cells were cultured on the central channel of the chemotaxis slide at 10% con uency for 8 h to allow adherence. Testing drugs were applied when the assay started and the images of cells were recorded for 15 hours with a time-lapse Micro-Imager. Single-cell tracking was analyzed using the ImageJ software. Spider plots representing the aggregated trajectories of cells. Forward migration indexes and cell velocity were analyzed using the Ibidi software.

Fura-2-based intracellular Ca 2+ assay
Fura-2-based uorescence was used to determine the intracellular Ca 2+ level. Fura-2 has been wildly used in the indication of intracellular Ca 2+ [45,46]. A549 cells were cultured on poly-D-lysine coated TC-treated Cell Culture Dishes (Falcon) with lidocaine or vehicle for 24 hours. Cells of the Gai blocker group were pretreated with 1 mg/ml pertussis toxin for 2 hours before the Fura-2-loading. Cells were loaded with Fura-2 loading buffer (1 mM Fura-2, 140 mM NaCl, 5 mM KCl, 20 mM HEPES, and 1 mM CaCl 2 , pH 7.4.) for 20 min at 37 •C. Then the loaded cells were gently washed with a washing buffer (140 mM NaCl, 5 mM KCl, 20 mM HEPES, and 1 mM CaCl 2 , pH 7.4.). Images were recorded real-timely on an Axiovert 200 inverted microscope with an excitation at 340 and 380 nm. CXCL12 was perfused 50 s after the recording start and the monitoring continued until 150 s. The F340/380 ratio was calculated and plotted against time to indicate the changes in intracellular Ca 2+ level.

Actin polymerization detection
The cytoskeleton remodeling was indicated by the determination of actin polymerization using Fluorescein Phalloidin staining. The method was described previously [47]. Brie y, cells were xed in 4% paraformaldehyde for 10 min at 4 •C, then were permeabilized with 0.01% Triton-X-100 for 5 min. Filamentous actin was stained using Fluorescein Phalloidin (F432, Thermo Fisher Scienti c, USA) for 40 min. Images of uorescence were recorded and the uorescence intensity was analyzed using the Fiji software.

Plotting and statistical analysis
Means and standard deviations are displayed in the gures. A T-test or ANOVA was used to analyze the signi cance of the difference (p < 0.01). Dunnett's post hoc tests were used to test the difference between groups. GraphPad Prism (version 8) was used to plot the data and calculate statistics. The anti-arrhythmia plasma concentration and approximately equipotent nerve block concentrations of lidocaine are around 10 µM [48]. We rst evaluated the toxicity of lidocaine at the clinical plasma concentration range (1-100 µM) on NSCLCA549 cells. Results showed that at these concentrations, lidocaine did not affect cell viability, apoptosis, and cell death (Fig. 1ABC). Therefore, we suggested that Lidocaine at clinical plasma concentrations did not inhibit proliferation or induce apoptosis or necrosis of A549. In addition, CXCL12 at 100 nM also failed to affect A549 viability (Fig. 1D). In this study, we performed a knockdown experiment of CXCR4 in A549. Results showed that CXCR4 knockdown did not affect the sensitivity of A549 to lidocaine at the clinical plasma concentration range in terms of cell viability (Fig. 1E). As we focused on the effect of lidocaine at clinical plasma concentration on the chemokinesis or chemotaxis of A549 cells, 100 µM lidocaine was used in the subsequent study.

lidocaine decreased CXCR4 expression.
In this study, we focused on the effect of the CXCR4/CXCL12 signal on cancer cell migration. CXCR4/CXCL12 signal has been wildly accepted as a critical signal for the migration of lung cancer cells [49][50][51] and CXCR4 has been thought to be a drug target for the inhibition of lung cancer migration [27,52]. We extracted the membrane expression of CXCR4 in A549 and observed the effect of lidocaine on the surface expression of CXCR4 in A549 cells. Results showed that lidocaine signi cantly decreased the surface expression of CXCR4 and the total expression of CXCR4 mRNA (Fig. 2). In the CXCR4 knockdown experiments, we knocked down over 80% CXCR4 mRNA (Fig. 2C) and the surface CXCR4 expression was very low that was almost not detectable in the western blotting (Fig. 2AB). After CXCR4 knockdown, the effect of lidocaine on CXCR4 expression was eliminated (Fig. 2). In the CXCR4 overexpressing expression, we increased the mRNA of CXCR4 by about 8 times, and the CXCR4 protein expression was increased by about 2 times. At the presence of lidocaine, both the CXCR4 mRNA and protein levels were signi cantly decreased compared to the control (Fig. 2). Thus, we suggested that lidocaine down-regulated CXCR4 expression when it was highly expressed, but did not affect CXCR4 when it was at a low expression level.

lidocaine inhibited CXCL12 induced migration.
CXCL12 is the agonist of the CXCR4 and has been wildly used for studying chemokinesis or chemotaxis of cells [53]. To tested if lidocaine affected CXCL12 induced migration of A549 cells, we performed both the wound healing assay to observed the migration of monolayer A549 cells as a group and in vitro chemotaxis assay to observe the migration of individual A549 cells. Results showed that CXCL12 signi cantly increased the A549 wound healing rate. In addition, CXCL12 also increased the migration index and velocity of individual cells. As shown in the aggregated trajectories of individual A549 cells, A549 migrated a longer distance at the presence of CXCL12. At the presence of lidocaine, the A549 wound healing rate was the same as control, but the wound healing rate decreased compared with CXCL12-induced cells (Fig. 3). These results revealed that lidocaine did not affect A549 cell migration but blocked the stimulation of CXCL12.

CXCR4 knockdown block lidocaine effect in migration
To investigate whether lidocaine affects migration through CXCR4, we also performed migration assays in CXCR4 knockdown A549 cells. Results showed that knockdown of CXCR4 eliminated the effects of CXCL12 or lidocaine on A549 migration. The exposure of CXCL12 or lidocaine had no signi cant effect on the wound healing speed and the migrate index and velocity of individual cells (Fig. 4). This suggested that CXCR4 was essential for the effects of CXCL12 and lidocaine.

Lidocaine inhibited migration caused by CXCR4 overexpression
To further investigate the effect of lidocaine on CXCR4 mediated migration, we also performed migration assays in CXCR4 overexpressing A549 cells. Results showed that CXCL12 signi cantly increased CXCR4 overexpressing the A549 wound healing rate. In addition, CXCL12 also increased the migration index and velocity of individual CXCR4 overexpressing A549 cells. As shown in the aggregated trajectories of individual CXCR4 overexpressing A549 cells, these cells migrated a longer distance at the presence of CXCL12. At the presence of lidocaine, CXCR4 overexpressing the A549 wound healing rate was the same as control, but the wound healing rate decreased compared with CXCL12-induced cells (Fig. 5). These results revealed that lidocaine did not affect CXCR4 overexpressing A549 cell migration but blocked the stimulation of CXCL12. (Fig. 6). These results demonstrated that the effect of lidocaine was mediated by CXCR4/CXCL12 signal.
To further explore the potential mechanisms for lidocaine action on A549 migration, we investigated two critical adhesion molecules on the epithelial cell membrane, CD44 and ICAM-1. CD44 has been wildly used as a key migration-related biomarker for lung cancers [54,55]. Another migration-related extracellular molecule, ICAM-1, is also thought to be critical in lung cancer cell migration [56][57][58]. Both CD44 and ICAM-1 levels were signi cantly increased by CXCL12. Lidocaine increased CD44 in both wildtype A549 and CXCR4 knockdown A549 cells in the absence or presence of CXCL12, but it did not affect ICAM-1 (Fig. 6). These results suggested that lidocaine can up-regulate CD44 but not ICAM-1. However, the regulation of lidocaine on CD44 was not essential for CXCL12 induced migration. 3.7. Lidocaine inhibited CXCL12-induced intracellular Ca 2+ releasing and cytoskeleton remodeling.
Another potential mechanism we hypothesized underlying lidocaine's effect on migration is the regulation of the cytoskeleton remodeling. To test this hypothesis, we monitored the effect of lidocaine on the intracellular Ca 2+ level alteration following the addition of CXCL12. The pertussis toxin was used as a control to block the Gai subunit of CXCR4 [47]. The CXCR4 knockdown A549 cells were used as another control. Results showed that lidocaine decreased the peak of the Ca 2+ curve by about 50%. CXCR4 knockdown A549 cells reacted slightly (less than 5% of the peak of Ca 2+ curve in the control group) to CXCL12 and pertussis toxin blocked almost all the effects of CXCL12 (Fig. 7). To further study the CXCL12-induced cytoskeleton remodeling of A549 at the presence of lidocaine, we stained lamentous actin with phalloidin coupled to uorescein isothiocyanate to observe the rearrangement of cortical actin bers during CXCL12 stimulation. Results showed that lidocaine reduced cytoskeleton remodeling by about 50% and both the pertussis toxin group and CXCR4 knockdown group failed to show any reaction following CXCL12 stimulation. These results suggested that lidocaine inhibited CXCL12-induced intracellular Ca 2+ releasing thereby reduced cytoskeleton remodeling.

Discussion
The therapeutic effect of surgery in metastatic NSCLC has been controversial [59,60]. During the surgery, many clinical factors might contribute to the metastasis and recurrence of lung cancer surgery. The application of lidocaine during the surgery or preoperational treatment results in a plasma lidocaine micro-environment for lung cancer cell migration and survival. Hence, the potential impact of lidocaine on metastasis and recurrence of lung cancer surgery should be further studied. However, so far, the effect of lidocaine at plasma concentration on cancers was less studied. The doses of lidocaine used in most previous studies are much higher than the clinical plasma concentration. Therefore, although many previous studies demonstrated the effect of lidocaine on lung cancer cells, they fail to convince clinical surgeons that lidocaine exerts a considerable impact on the surgery outcome. Lidocaine is known as a sodium channel blocker. Although the major target of lidocaine, the voltage-gated sodium channels, has been found to play a role in cancer developments [61], lidocaine might also affect cancer independent of sodium channel blockade.
Although a previous study suggested that lidocaine at the "mM" concentration range inhibited proliferation [15] and induced apoptosis of A549 [62], our result showed that, at plasma concentration, lidocaine had almost no effect on cell proliferation, apoptosis, and cell death. The previous study also suggested that the migration of A549 was inhibited by lidocaine at 8 mM, but as shown by our data, the effect of lidocaine on migration was not signi cant when the doses of lidocaine decreased to 100 µM.
Lidocaine at clinical plasma concentrations only reduced CXCL12-induced migration in A549. The western blotting assay showed that lidocaine decreased the expression of CXCR4 especially the membrane expression of CXCR4. We suggested that the reduction of CXCR4 on the membrane surface impaired the sensitivity of cells to CXCL12 stimulations. A previous study has suggested that lidocaine can regulate CXCR4 sensitivity to CXCL12 in breast cancer cells MDA-MB-231 [22], which was similar to A549 according to our results.
Different membrane surface proteins expressed on lung cancer cells as adhesion molecules can be critical in the migration of cells [63]. CD44 has been reported to play roles in the metastasis of NSCLC cells [54]. In this study, the expression of CD44 was promoted by CXCL12 stimulation. Our results also showed that the surface expression of CD44 on A549 was up-regulated by lidocaine. However, the lidocaine did not further increase the CD44 expression at the presence of CXCL12 and the increase of CD44 in the lidocaine alone group did not affect cell migration. Thus, we suggested that CD44 was directly up-regulated by lidocaine bypassing the CXCL12/CXCR4 axis but the increase of CD44 was not essential for the migration of A549. In addition, we also determined another critical adhesion molecule for migration, the ICAM-1. The expression of ICAM-1 has been associated with lung cancer progression and prognosis [64]. A previous study reported that lidocaine affects the migration of a lung cancer cell line H838 by reducing ICAM-1 [65]. However, in this study, the lidocaine at plasma concentration did not affect ICAM-1 expression. The ICAM-1 expression was up-regulated by CXCR4 activation by CXCL12, hence, we suggested ICAM-1 might be a potential downstream target of lidocaine/CXCR regulation (Fig. 8). Further validation is required in the future.
Another activity that might impact the migration of A549 was the remodeling of the cytoskeleton. The activation of CXCR4 by CXCL12 can trigger intracellular Ca 2+ releasing [66]. Subsequently, the released intracellular Ca 2+ facilitated many Ca 2+ -dependent activities that are involved in cytoskeleton remodeling [67,68]. Results showed that lidocaine signi cantly reduced the intracellular Ca 2+ releasing triggered by CXCL12 stimulated CXCR4 activation. We further determined the level of lamentous actin to observe the rearrangement of cortical actin bers during CXCL12 stimulation. Actin is an important part of the cytoskeleton in most eukaryotic cells [69]. There are two types of actin: globular G-actin and lamentous F-actin [69]. The G-actin can polymerize into the actin laments, the F-actin polymer laments that form the cytoskeleton [69]. Thus, the level of F-actin can indicate the rate of cytoskeleton rearrangement. Our results showed that lidocaine reduced cytoskeleton remodeling by about 50%. Therefore, we suggested that lidocaine inhibited cytoskeleton remodeling via suppressing CXCL12-induced intracellular Ca 2+ releasing (Fig. 8). However, lidocaine, as a multiple channel blocker, might have other effects on intracellular Ca 2+ levels. Many cancer-related ion channels, such as TRP [70] and TPCs [71], might contribute to lidocaine effects on cancer cells and need further investigation.

Conclusion
This study demonstrated that lidocaine at clinical plasma concentrations showed a signi cant inhibition effect on CXCL12-induced CXCR4 activation, thereby reduced the intracellular Ca 2+ dependent cytoskeleton remodeling, resulting in slower migration of A549 cells. This investigation can provide a better understanding of the pharmacological effects of lidocaine at clinical plasma concentrations on the migration of NSCLCcells and can optimize the clinical application of lidocaine in lung cancer surgery.

Funding
This study received funding from the Changzhi Medical College A liated Heping hospital (No. is not available).

Availability of data and materials
The raw data of this study are provided from the corresponding author with a reasonable request.