stable cell line of human sh-sy5y uniformly expressing twik-related acid-sensitive potassium channel...

Stable Cell Line of Human SH-SY5Y UniformlyExpressing TWIK-Related Acid-Sensitive PotassiumChannel and eGFP Fusion

Chaokun Li & Linyu Wei & Hongbo Jiang & Linlin Shan &

Xinjuan Li & Na Lu & Guohong Wang & Dongliang Li

Received: 13 December 2013 /Accepted: 29 January 2014# Springer Science+Business Media New York 2014

Abstract TWIK-related acid-sensitive potassium channels (TASK3) are pharmacologicaltargets of CNS inflammation induced by acidification. They function as molecular switchesbetween survival and death of neurons. In this report, TASK3 cloned from human brain cDNAwas tagged with enhanced green fluorescent protein (eGFP), and the fusion gene wastransiently expressed in human neuroblastoma SH-SY5Y cells. A cell line stably expressingTASK-eGFP fusion proteins was generated from transient expression cells by usingfluorescence-activated cell sorting followed by antibiotic selection. The uniform expressionof TASK3 fusion proteins was further confirmed by flow cytometry. Moreover, the localizationof TASK3 tagged with eGFP was checked by confocal microcopy. TASK3-eGFP fusionproteins are observed on the SH-SY5Y cell membrane. The strategies using eGFP as a fusiontag facilitate the monitoring of the TASK3 expression and enable the successful employmentof FACS for screening and construction of cell lines stably expressing TASK3. The TASK3overexpression cell line will lay a fundamental for the in vitro evaluation of TASK3 functionduring hypoxic/ischemic injury.

Keywords TWIK-related acid-sensitive potassium channels . TASK3 . KCNK9 . Stablecell line . TASK3-eGFP fusions . SH-SY5Y

Two-pore domain potassium channels (K2P, KCNK) play a pivotal role in setting neuronalexcitability by modulating the resting membrane potential, and they are also key players in thecontrol of firing frequency in neurons [1, 2]. In this family, TWIK-related acid-sensitivepotassium channels (TASK1, KCNK3) and TASK3 (KCNK9) are pH-sensitive subunitspredominately expressed in many brain regions such as the cortex, cerebellum, hippocampus,

Appl Biochem BiotechnolDOI 10.1007/s12010-014-0768-7

C. Li (*) : L. Wei : L. Shan : X. Li :N. Lu :G. Wang :D. LiDepartment of Physiology and Neurobiology, Xinxiang Medical University, Xinxiang, Henan 453003,People’s Republic of Chinae-mail: [email protected]

H. JiangThe Third Affiliated Hospital, Xinxiang Medical University, Xinxiang, Henan 453003, People’s Republic ofChina

thalamus, and spinal cord [3–5]. In different types of neurons, TASK channels are composedof TASK3 homomeric channels, TASK3/TASK1 heterodimers, and/or TASK1 homomericchannels [6, 7]. It had been suggested that TASK1 homodimers will be a minority [8].

It is clear that TASK1 channels are predominantly open at alkaline conditions. In physio-logical conditions such as pH 7.2 to 7.4, a half percentage of TASK1 had already closed [9,10]. In contrast, TASK3 channels keep in the open state in physiological conditions and onlyshut under more acid conditions [11]. TASK3 channels are more sensitive to acidification and,therefore, become pharmacological targets of CNS inflammation induced by acidification.Multiple reports had identified that TASK channels were functionally involved in neuronalapoptosis and ischemic brain injury [7, 12, 13]. The opening or leaking of TASK3 channelsshut in the initial of ischemic injury, which dramatically induced both the depolarization ofCNS neurons and disturbances in homeostasis. Hours to days after TASK3 closed, the neuronseventually die. To date, TASK3 channels are deemed as molecular switches between survivaland death of neurons.

In vivo studies using gene knockout animals had verified the function of TASK3 channelsin stroke or traumatic brain injury. However, in vitro study on hypoxic/ischemic injury is still alimitation due to the weak expression of TASK3 in most culture cell lines. In the current study,human brain TASK3 (KCNK9) gene was cloned and its carboxyl terminal was fused withenhanced green fluorescent protein (eGFP). The fusion gene was expressed by a mammalianexpression vector. The recombination plasmid was transfected into human neuroblastoma SH-SY5Y cells, and the expression of TASK3-eGFP fusion was confirmed by fluorescencemicroscopy. For stable TASK3-eGFP expressing cell line construction, the TASK3-eGFPtransient expression cells were sorted by flow cytometry and followed by antibiotic selection.The aims of this study were to construct a SH-SY5Y cell line uniformly expressing TASK3-eGFP fusions and also to evaluate whether the fusion strategy is suitable for the localization ofTASK3 fusions on cell surface.

Materials and Methods

Cells and Chemicals

Human neuroblastoma SH-SY5Y cell line was obtained from ATCC. The cell culture mediumand fetal bovine serum were Gibco products. Transfection agent (x-fect) was purchased fromClontech (Takara Bio Company, Japan). PCR kits and enzymes were purchased from NewEngland Biolabs (Ipswich, MA, USA). All reagents including TASK3 antibody were pur-chased from Sigma-Aldrich (St. Louis, MO, USA).

Construction of Plasmids

Human brain TASK3 (KCNK9) cDNAwas obtained from Thermo Scientific (Beijing, China).PCR-based cloning methods were performed following FastCloning methods [14]. Thesequences of TASK3 subcloning primers were as follows: TASK3-FWD, 5′-CGTCAGATCCGCTAGCGCCACCATGAAGAGGCAGAACGTGCGGAC-3′; TASK3-REV, 5′-CGCCCTTGCTCACCATAACGGACTTCCGGCGTTTCA-3′. The underlined data indicated thesequences of TASK3 open reading frame without the stop codon, and the remaining sequenceswere the two 16 overlapping sequences with pEGFP-N1 plasmid. The primers for amplifyingthe pEGFP-N1 vector were as follows: N1-start-REV, 5′-GCTAGCGGATCTGACGGTTCAC-3′; N1-end-FWD, 5′-ATGGTGAGCAAGGGCGAGGA-3′. PCR reaction was performed

Appl Biochem Biotechnol

with Phusion polymerase (NEB) as follows: 1 min at 98 °C followed by 18 cycles of 30 s at98 °C, 30 s at 56 °C, and 90 s at72 °C, and a final step of 5 min at 72 °C. PCR products werechecked on agarose gel and then 1 μl DpnI was added to the PCR product. The PCR productswere incubated at 37 °C for at least 3 h. Equal volumes of PCR products were mixed andtransformed into Escherichia coli DH5a.

Colony PCR was used with sequencing primers for positive clone validation. The sequenc-ing primer was as follows: N1-seq-FW, 5′-TGGGAGTTTGTTTTGGCAC-3′; N1-seq-RV, 5′-GGACACGCTGAACTTGTGG-3′. The plasmid from the positive clone was purified withEndo-free plasmid kits (Takara Bio Company, Japan). The final plasmid was renamed aspTASK3-eGFP. The plasmid was further confirmed by DNA sequencing.

Cell Culture and Plasmid Transfection

Human neuroblastoma SH-SY5Y cells were cultivated in a six-well plate with Dulbecco’smodified Eagle’s medium (DMEM, Gibco) containing 10 % fetal bovine serum (FBS,Hyclone). When the cell density reached 50 %, 5 μg plasmid DNA was used for celltransfection. Plasmid transfection was mediated with x-fect kits following standard instructions(Clontech, Japan). Forty-eight hours after transfection, the plate was placed under a fluores-cence microscopy and a 488 nm/512 nm filter was selected for eGFP fluorescence detection.

Fluorescence-Activated Cell Sorting

Seventy-two hours after transient expression of TASK3-eGFP fusions, SH-SY5Y cells wereharvested by using 0.25 % trypsin-EDTA (Gibco) and cells were checked by a flow cytometer.A 488 nm/512 nm filter was selected for detecting TASK3-eGFP fluorescence. The controlSH-SY5Y cells were used for gate determination. Around 10,000 cells were sorted out andrecultured into a six-well plate with DMEM (10 % FBS) containing 600 μg/ml G418 sulfateantibiotic.

G418 Antibiotic Mediated Stable Cell Line Screening

DMEM medium was replaced with fresh medium containing 600 μg/ml G418 every48 h. The G418 antibiotic was used for 15 days, and fluorescence was monitoredevery time after DMEM medium was changed. After 15 days, the surviving cellswere digested by 0.25 % trypsin-EDTA. The number of cells was counted and thena limiting dilution assay was used. Each individual cell was placed into a single wellof a 96-well plate. Around 300 cells were placed into three 96-well plates and thencultured with DMEM (10 % FBS) containing 600 μg/ml G418. The DMEM mediumwas changed every 3 days until the cells have grown into a colony. After 2 weeks,cell fluorescence was observed using microscopy, and the best colony in which allcells had high fluorescence emission was labeled and expanded gradually using a24-well plate, a six-well plate, and a 25 cm dish. The seeds were kept in liquidnitrogen.

Flow Cytometry-Based Positive Cell Colony Confirmation

The labeled stable cell line was cultured for 3 days, and the cells were harvested and placedinto a flow cytometer. Around 10,000 cells were checked for the uniform expressing TASK3fusions. Intact SH-SY5Y was used as negative control.


Confocal Microscopy for TASK3-eGFP Cell Localization

The labeled stable cell line was cultivated in a live cell imaging culture dish (LivefocusCompany). Forty-eight hours later, the dish was plated under a confocal microscope. A layerscanning assay was used to detect the localization of TASK3-eGFP in SH-SY5Y cells.

Western Blot Assay

Cells were harvested and resuspended in a cell lysis solution containing 1 mM PMSF (pH 7.4)on ice for 30 min. After sonication, the cells were centrifuged at 12,000 rpm at 4 °C for 5 min.The soluble protein was stored at −80 °C. For sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE), a 40 μg sample was mixed with protein loading buffer, boiledfor 10 min, and resolved by 10 % (w/v) SDS-PAGE. For western blot analysis, proteins on gelwere transferred onto nitrocellulose membranes (Millipore) with a tank transfer system (Bio-Rad). Membranes were then placed in a blocking solution of 3 % BSA in TBST buffer(100 mM Tris-HCl, 0.9 % (w/v) NaCl, 0.1 % (v/v) Tween 20) for 30 min. Forimmunodetection, primary TASK3 polyclonal antibody (Sigma) was diluted 1:1,000 in TBSTbuffer and incubated with membrane overnight at 4 °C. The second antibody was an alkalinephosphatase-conjugated goat anti-rabbit IgG diluted 1:1,000 (Beyotime Biotechnology, Chi-na). After 1 h of incubation and washing with TBST, the reaction of alkaline phosphatase wasfurther developed by a solution containing 5-bromo-4-chloro-3-indolyl phosphate di-sodiumsalt (BCIP)/Nitro blue tetrazolium chloride (NBT).

Results

Amplification of TASK3 and Construction of pTASK3-eGFP

Polymerase chain amplifications of TASK3 and pEGFP-N1 are shown in Fig. 1. A1.2 kb band of TASK3 PCR product and a 4.7 kb pEGFP-N1 backbone were clearlyshown on the 1 % agarose gel (Fig. 1a). Two PCR products were then mixed with a1:1 ratio and transformed to E. coli DH5a competent cells. Colony PCR wasemployed for the positive cell validation. Six clones were selected randomly onthe LB (Kan+) plate. As shown by the PCR results in Fig. 1b, five clones werepositive cells harboring recombination plasmids. The recombination plasmids werepurified and renamed as pTASK3-eGFP. When the DNA sequencing result wascompared with the original TASK3 sequence obtained from GenBank, the alignmentis around 100 % and the above results indicate the successful construction ofpTASK3-eGFP.

Transient Expression of TASK3-eGFP Fusions and Fluorescence-Activated Cell Sorting

To monitor the transient expression of TASK3-eGFP in SH-SY5Y cell after plasmid transfec-tion, fluorescence microscopy was employed for the evaluation of cell fluorescence at 48 h. Asshown in Fig. 2a, SH-SY5Y cells grow well during 48 h of transfection. We observed thatsome cells had green light emission under fluorescence microscopy, and some cells were stillnegative (Fig. 2b).

To sort the positive cell from the whole cell colony, fluorescence-activated cellsorting was used. A control SH-SY5Y population (without transfection) was checked


on a flow cytometer for the sorting gate determination (Fig. 2c). As clearly shown inFig. 2d, after 72 h of induction, around 50 % of the whole population was success-fully expressing the TASK3 fusion proteins. About 10,000 cells in gate P2 wereisolated for further experiment.

Construction of Stable Cell Lines Expressing the TASK3-eGFP Fusions

Cell lines stably expressing the TASK3-eGFP proteins were further developed fromthe FACS-sorted cells and further incubated with DMEM medium containing 600 μg/ml G418 in the next 2 weeks. The colonies were checked under fluorescencemicroscopy, and then each individual cell was inoculated into a 96-well plate bylimiting dilution assay. After cultured in G418 medium for 1 month, the population inwhich all cells had strong fluorescence emission was selected. As illustrated in Fig. 3,the selected clone was uniformly expressing TASK3-eGFP and all cells in the dishhad eGFP fluorescence (Fig. 3a, b). The SH-SY5Y negative control cells wereevaluated at the same time (Fig. 3c, d).

Uniformity Determination of the Cell Line Stably Expressing TASK3 Fusions by FlowCytometry

To confirm the selected SH-SY5Y population uniformly expressing TASK3-eGFP fusionproteins, flow cytometry assay was employed. As shown by the data in Fig. 4a, the negativecontrol cells peaked up only in the M1 area and almost all cells were kept in this region. Incontrast, the fluorescent peak of the cells stably expressing TASK3-eGFP was shifted to M2gate (Fig. 4b) and around 96 % was kept in the M2 region, indicating that the selected cell linewas uniformly expressing TASK3-eGFP.

Surface Localization of TASK3-eGFP Fusions

TASK3 channels are membrane proteins expressed on the cell membrane. After taggedwith eGFP in its C-terminal, the localization of TASK3-eGFP fusions in SH-SY5Ycell line was determined by confocal microscopy using cell layer scanning. The SH-

Fig. 1 Construction of TASK3-eGFP mammalian expression vector. a Electrophoresis of PCR-amplifiedTASK3 and pEGFP-N1 fragments. Lane 1, TASK3 fragment. Lane 2, DNA ladder (bands from top to bottomare 2,000, 1,000, 750, 500, 250, and 100 bp). Lane 3, pEGFP-N1 fragment. b Electrophoresis of colony PCR forclone validation. Lane 1, DNA ladder. Lanes 2 to 7, six clones selected randomly from the agar plate


SY5Y negative control was detected and no fluorescence was observed under laserexcitation (Fig. 5a, b). As the selected cells stably expressing TASK3-eGFP werechecked by confocal microscopy, only the cell membrane was labeled with TASK3-eGFP fluorescence (Fig. 5d), and both the cytoplasm and nucleolus were eGFPnegative, suggesting the membrane expression of TASK3-eGFP proteins.

Western Blot Analysis for TASK3-eGFP Expression

To compare the expression level between the TASK3-eGFP stably expressing cellsand intact SH-SY5Y cells, western blot analyses using the soluble protein werecarried out by the rabbit anti-TASK3 antibody, and the antibody was subsequentlydetected with AP-conjugated goat anti-rabbit IgG. As shown by the western blot resultin Fig. 6, weak bands corresponding to the 45 kDa TASK3 intact proteins weredetected in both SH-SY5Y control cells and stably expressing cells, suggesting that

Fig. 2 Transient expression of TASK3-eGFP in SH-SY5Y cells. a Bright-light images of plasmid-transfectedSH-SY5Y cells after 48 h of induction (original magnification×100). b Fluorescent images of the same view fieldof transfected SH-SY5Y cells. c Flow cytometry assay of SH-SY5Y control cells. d Flow cytometry assay oftransient expression of TASK3-eGFP in SH-SY5Y cells. The gate for fluorescence-activated cell sorting waslabeled as P2


the TASK3 protein was weakly expressed in SH-SY5Y cells. A prominent bandcorresponding to around 73 kDa, which is the molecular mass of TASK3-eGFP fusion

Fig. 3 Generation of a stable cell line expressing TASK3-eGFP fusions. Top row, SH-SY5Y cells stablyexpressing TASK3 tagged with eGFP in bright-light field (a) and fluorescent field (b). Bottom row, controlSH-SY5Y cells in bright-light field (c) and fluorescence field (d). Original magnification×200

Fig. 4 Analysis of the uniformity of the stable cell line expressing TASK3 fusions by flow cytometry. a SH-SY5Y control cells. b SH-SY5Y cells stably expressing TASK3 tagged with eGFP


Fig. 5 Surface localization of TASK3-eGFP fusion by confocal microscopy. Top row, control SH-SY5Y cells inbright-light field (a) and fluorescent field (b) in confocal microscopy. Bottom row, SH-SY5Y cells stablyexpressing TASK3 tagged with eGFP in bright-light field (c) and fluorescence-mediated layer scanning (d).The arrows showed that the TASK3-eGFP fusions were expressed on the cell membrane. Originalmagnification×400

Fig. 6 Comparison of TASK3 expression levels between TASK3 fusion uniformly expressing cells and SH-SY5Y control cells by western blot assay. Lanes 1 and 2, soluble proteins purified from SH-SY5Y cells stablyexpressing TASK3-eGFP fusions and two bands (TASK3 and TASK3-eGFP) shown on the membrane. Lanes 3and 4, soluble proteins from intact SH-SY5Y control cells. Lane 5, protein marker


proteins, was observed in the portion of the cells infected with TASK3 fusions(Fig. 6, lane 1 and lane 2), demonstrating that both the TASK3 part and the eGFPpart of the fusion protein were successfully expressed in the constructed cell line.

Discussion

In mammalian central nervous system, the two-pore domain potassium channels playan essential role in maintaining the rest potential and duration of neuronal firingfrequency and then directly affect the responsiveness to synaptic inputs [2, 13]. Theacid-sensitive subunit TASK channels are the main sources for CNS homeostasis andfunction as internal sensors for numerous stimuli such as fluctuation in temperatureor intracellular pH, osmolarity, oxygen tension, and membrane stretch [15, 16]. It isclear that TASK1 and TASK3 channels are the pharmacological targets for manydrugs including volatile general anesthetics [17, 18]. In the recent decade, TASK3channels have attracted considerable research attention due to their actions asmembrane switch between neuron survival and death in CNS diseases affected byacidification [12, 13, 19]. They may serve as the main potassium channels kept inan active or alive status in the acidification of neurons subjected to ischemia/hypoxiastress.

Currently, the successful development of TASK knockout animals makes it tech-nologically possible for evaluating the role of TASK channels involved in neuroninjury caused by stroke acidification or such diseases. However, the weak expressionlevels of TASK3 in most of the cultivated cell lines formed a barrier for the in vitroevaluation of the TASK3 function involved in ischemic/hypoxia injury. This barriercan be overcome by developing a genetically engineered overexpression cell line. Inthis report, human brain TASK3 was tagged with eGFP, and the TASK3-eGFP fusionproteins were transiently expressed in human neuroblastoma SH-SY5Y cells. Cellsexpressing TASK-eGFP fusion proteins were obtained from a transient expressionpopulation with fluorescence-activated cell sorting. A stable cell line was furtherdeveloped under G418 antibiotic selection. The uniformity of the selected cell lineexpressing TASK3 fusions was confirmed by flow cytometry. The localization ofTASK3 fusions was verified with layer scanning assay in confocal microcopy andprotein expression levels were evaluated by western blot assay.

The topology structure of TASK3 channel has displayed that the channels arecomposed of four transmembrane domains, and both amino and carboxy-termini aredisplayed in the intracellular domains [15, 20]. Therefore, the strategy of fusing thecarboxyl terminal of TASK3 with fluorophore may not affect the exposure or poreforming of TASK3 channels. Our data had shown that when the carboxyl terminal ofTASK3 fused with eGFP, localization of TASK3 fusions is still on the SH-SY5Y cellmembrane. The strategy of using eGFP as a tag offers an option for using FACS,which eventually narrows down the chaos between transfected and untransfectedpopulation and also significantly increases the opportunity to obtain the cell linestably expressing TASK3 fusions. The methods may be suitable for other membraneproteins when their carboxyl terminal localized inside of the cell. We eventuallyobtained a stable cell line uniformly expressing TASK3 fusion proteins which madea fundamental for the in vitro evaluation of the modulation of TASK3 during hypoxic/ischemic injury as revealed by experiments where cells suffered oxygen and glucosedeprivation. These experiments are currently under investigation.


Acknowledgments This work was supported by the National Natural Science Foundation of China(grant no. 81100912) and Xinxiang Medical University Foundation to CL and partially supported byNSFC (81271376).

Conflict of Interest The authors declare that there are no conflicts of interest.

References

1. Goldstein, S. A., Bockenhauer, D., O’Kelly, I., & Zilberberg, N. (2001). Nature Reviews Neuroscience, 2,175–184.

2. Kim, D. (2005). Current Pharmaceutical Design, 11, 2717–2736.3. Kindler, C. H., Pietruck, C., Yost, C. S., Sampson, E. R., & Gray, A. T. (2000). Molecular Brain Research,

80, 99–108.4. Callahan, R., Labunskiy, D. A., Logvinova, A., Abdallah, M., Liu, C., Cotton, J. F., & Yost, C. S. (2004).

Biochemical and Biophysical Research Communications, 319, 525–530.5. Torborg, C. L., Berg, A. P., Jeffries, B. W., Bayliss, D. A., & McBain, C. J. (2006). Journal of Neuroscience,

26, 7362–7367.6. Aller, M. I., Veale, E. L., Linden, A. M., Sandu, C., Schwaninger, M., Evans, L. J., Korpi, E. R., Mathie, A.,

Wisden, W., & Brickley, S. G. (2005). Journal of Neuroscience, 25, 11455–11467.7. Berg, A. P., Talley, E. M., Manger, J. P., & Bayliss, D. A. (2004). Journal of Neuroscience, 24, 6693–6702.8. Aller, M. I., & Wisden, W. (2008). Neuroscience, 151, 1154–1172.9. Duprat, F., Lesage, F., Fink, M., Reyes, R., Heurteaux, C., & Lazdunski, M. (1997). EMBO Journal, 16,

5464–5471.10. Czirjak, G., & Enyedi, P. (2002). Journal of Biological Chemistry, 277, 5426–5432.11. Rajan, S., Wischmeyer, E., Xin, L., Preisig-Muller, R., Daut, J., Karschin, A., & Derst, C. (2000). Journal of

Biological Chemistry, 275, 16650–16657.12. Lauritzen, I., Zanzouri, M., Honore, E., Duprat, F., Ehrengruber, M. U., Lazdunski, M., & Patel, A. J. (2003).

Journal of Biological Chemistry, 278, 32068–32076.13. Meuth, S. G., Kleinschnitz, C., Broicher, T., Austinat, M., Braeuninger, S., Bittner, S., Fischer, S., Bayliss, D.

A., Budde, T., Stoll, G., & Wiendl, H. (2009). Neurobiology of Disease, 33, 1–11.14. Li, C., Wen, A., Shen, B., Lu, J., Huang, Y., & Chang, Y. (2011). BMC Biotechnology, 11, 92.15. Patel, A. J., & Honore, E. (2001). European Respiratory Journal, 118, 221–227.16. Enyedi, P., & Czirják, G. (2010). Physiological Reviews, 90, 559–605.17. Patel, A. J., & Honore, E. (2003). Advances in Experimental Medicine and Biology, 536, 9–23.18. Clarkson, A. N. (2007). Life Sciences, 80, 1157–1175.19. Sun, H. S., Feng, Z. P., Barber, P. A., Buchan, A. M., & French, R. J. (2007). Neuroscience, 144, 1509–1515.20. Miller, A. N., & Long, S. B. (2012). Science, 335, 432–436.


stable cell line of human sh-sy5y uniformly expressing twik-related acid-sensitive potassium channel...

Documents