Genome-Engineered mpkCCDc14 Cells as a New Resource for Studying AQP2

mpkCCDc14 cells, a polarized epithelial cell line derived from mouse kidney cortical collecting ducts, are known to express the vasopressin V2 receptor (V2R) and aquaporin-2 (AQP2) that are responsive to vasopressin. However, a low abundance of the endogenous AQP2 protein in the absence of vasopressin and heterogeneity of AQP2 protein abundance among the cultured cells may limit the further application of the cell line in AQP2 studies. To overcome the limitation, we aimed to establish mpkCCDc14 cells constitutively expressing V2R and AQP2 via CRISPR/Cas9-mediated genome engineering technology (i.e., V2R-AQP2 cells). 3′- and 5′-Junction PCR revealed that the V2R-AQP2 expression cassette with a long insert size (~2.2 kb) was correctly integrated. Immunoblotting revealed the expression of products of integrated Aqp2 genes. Cell proliferation rate and dDAVP-induced cAMP production were not affected by the knock-in of Avpr2 and Aqp2 genes. The AQP2 protein abundance was significantly higher in V2R-AQP2 cells compared with control mpkCCDc14 cells in the absence of dDAVP and the integrated AQP2 was detected. Immunocytochemistry demonstrated that V2R-AQP2 cells exhibited more homogenous and prominent AQP2 labeling intensity in the absence of dDAVP stimulation. Moreover, prominent AQP2 immunolabeling (both AQP2 and pS256-AQP2) in the apical domain of the genome-edited cells was observed in response to dDAVP stimulation, similar to that in the unedited control mpkCCDc14 cells. Taken together, mpkCCDc14 cells constitutively expressing V2R and AQP2 via genome engineering could be exploited for AQP2 studies.


Introduction
Urine concentration and body water homeostasis are controlled by the anti-diuretic hormone, vasopressin, which increases osmotic water permeability and water reabsorption in the kidney collecting duct [1,2]. Aquaporin-2 (AQP2) is the water channel protein localized at the apical plasma membrane and subapical vesicles in the connecting tubule and collecting duct cells, and is regulated by vasopressin on a short-term and a long-term basis for water reabsorption and urine concentration [1,[3][4][5][6]. The short-term regulation of AQP2 is dependent on the intracellular translocation of AQP2-expressing vesicles to the apical plasma membrane, resulting in an increase in the osmotic water permeability of the cells [4,[7][8][9]. The long-term regulation of AQP2 is based on the changes in AQP2 protein abundance [1,8,10,11]. Vasopressin induces Aqp2 gene transcription [8,12] and increases the half-life of the AQP2 protein [13]. In contrast, vasopressin withdrawal induces the 2 of 13 intracellular degradation of AQP2 protein, mediated by ubiquitination and proteasomallysosomal degradation [14][15][16][17]. mpkCCDc14 cells, a polarized epithelial cell line derived from mouse kidney cortical collecting ducts [18], have been used for AQP2 studies since they are known to express both vasopressin V2 receptor (V2R) and AQP2, which are responsive to vasopressin [19][20][21][22]. However, the endogenous AQP2 protein abundance in mpkCCDc14 cells is very low in the absence of vasopressin stimulation. In addition, the heterogeneity of AQP2 protein abundance among the cultured mpkCCDc14 cells was shown in the in vitro culture system [23]. Vasopressin pre-treatment is, therefore, required to induce sufficient AQP2 expression to study the regulation of functional AQP2 proteins in mpkCCDc14 cells. In terms of AQP2 trafficking studies, however, dDAVP pre-treatment may have limitations. Although the induction of endogenous AQP2 expression by the pre-treatment of dDAVP produces detectable endogenous AQP2 proteins, dDAVP withdrawal for 2 or 4 h [24,25] might not fully recover cellular signaling pathways, including kinase activity associated with AQP2 trafficking. Therefore, the pre-treatment of dDAVP for the induction of endogenous AQP2 is likely to cause an artifact in studying cellular processes for AQP2 trafficking. More extended withdrawal may affect total AQP2 abundance, which might not be enough for detection. To overcome these limitations, we aimed: (1) to establish mpkCCDc14 cells stably expressing both V2R and AQP2 via CRISPR/Cas9-mediated genome engineering technology; and (2) to examine whether mpkCCDc14 cells with products from the targeted integration of both Avpr2 gene and Aqp2 gene are responsive to 1-deamino-8-D-arginine vasopressin (dDAVP) stimulation, a V2R agonist, to induce intracellular cAMP level and AQP2 translocation into the apical plasma membrane.

Targeted Integration of Avpr2 and Aqp2 Genes into the Rosa26 Locus in mpkCCDc14 Cells
The Rosa26 locus on chromosome 6 is frequently exploited for the integration of transgene constructs. To create CRISPR/Cas9-mediated knock-in of rat Avpr2 and Aqp2 genes into Rosa26 locus, we selected sgRNA target sequences ( Figure 1A). The sgRosa26, together with a Cas9 mRNA that includes a plasmid coded poly A tail, was selected. Rosa26 homology-directed repair (HDR) donor vector included homology arms (HAs) of 0.8 kb flanking a splicing acceptor (SA)-2A-puromycin element and the V2R-P2A-AQP2 expression cassette ( Figure 1A).
To confirm the targeted integration of Avpr2 and Aqp2 genes in mpkCCDc14 cells (i.e., V2R-AQP2 cells), junction PCR was performed in two different ways ( Figure 1B). The first one was 3 -junction PCR, where the forward primer had homology with integrated gene and the reverse primer had homology with 3 -mpkCCDc14 cells chromosome of Rosa26 locus ( Figure 1A). When the gene was correctly integrated,~1.3 kb of PCR product could appear ( Figure 1B). The second way was 5 -junction PCR, in which its forward primer had homology with 5 -mpkCCDc14 cells chromosome of Rosa26 locus and reverse primer had homology with integrated gene ( Figure 1A). In this case, the PCR product could appear at~1.5 kb ( Figure 1B). 3 -and 5 -Junction PCR revealed that the V2R-AQP2 expression cassette with a long insert size (~2.2 kb) was correctly integrated ( Figure 1B).
In addition, the PCR products (V2R and AQP2) from the V2R-AQP2 cells were shown at the predicted sizes, demonstrating that V2R-AQP2 expression cassette was correctly integrated ( Figure 1C). PCR experiments in control mpkCCDc14 cells and V2R-AQP2 cells demonstrated that AQP2 and V2R expression was much stronger in the V2R-AQP2 cells ( Figure 1C). Immunoblotting revealed that the protein expression of AQP2 (unglycosylated band) was strongly observed in V2R-AQP2 cells in the absence of vasopressin stimulation ( Figure 1D). The results indicated that Avpr2 and Aqp2 genes were correctly integrated, and the gene product was observed constitutively, even in the absence of vasopressin stimulation. tively-expressed AQP2 (integrated) as well as vasopressin-responsive AQP2 endogenously expressed in mpkCCDc14 cells. In particular, we inserted the rat Aqp2 gene as an integrated Aqp2 gene to detect two AQP2 forms at the transcript level, not affecting the function of AQP2 proteins. As shown in RT-qPCR experiment (Figures S1 and S2), the expression of two distinct forms can be clearly validated using species-specific primers for transcripts. Figure 1. Integration of V2R-AQP2 expression cassette into the Rosa26 locus using CRISPR/Cas9 genome-engineering technology in mpkCCDc14 cells. (A) Schematics of integration into the Rosa26 locus in mpkCCDc14 cells. Protospacer targeting the Rosa26 locus and the PAM element were indicated. The Rosa26 homology-directed repair (HDR) donor vector included homology arms (HAs) of 0.8 kb flanking a SA-2A-puromycin element and the V2R-P2A-AQP2 expression cassette. HA-left (L) was indicated in green and HA-right (R) was in blue. Two different junction PCR sites (3′-and 5′-junction) were depicted. (B) Junction PCR analysis in V2R-AQP2 cells to identify the targeted integration of Avpr2 and Aqp2 genes. PCR with 5′-F/5′-R and 3′-F/3′-R primers amplified the 5′-junction PCR product (1477 bp) and 3′-junction PCR product (1271 bp) of the V2R-AQP2 cells. (C) PCR showing the expression of AQP2 and V2R in control mpkCCDc14 cells and V2R-AQP2 cells. The predicted PCR product size of AQP2 and V2R were 245 bp and 216 bp, respectively. (D) Immunoblotting of AQP2 in control mpkCCDc14 cells and V2R-AQP2 cells. PAM, protospacer adjacent motif; HA-L, homologous arm-left; HA-R, homologous arm-right; pA, poly A; Puro, puromycin resistance gene; CMV, cytomegalovirus promoter; HDR, homology-directed repair; 5′-JCT, 5′-junction; 3′-JCT, 3′-junction; Con, control mpkCCDc14 cells; V2R-AQP2, mpkCCDc14 cells constitutively expressing V2R and AQP2 via CRISPR/Cas9-mediated genome engineering technology. The engineered V2R-AQP2 cell model was designed to express both the constitutivelyexpressed AQP2 (integrated) as well as vasopressin-responsive AQP2 endogenously expressed in mpkCCDc14 cells. In particular, we inserted the rat Aqp2 gene as an integrated Aqp2 gene to detect two AQP2 forms at the transcript level, not affecting the function of Int. J. Mol. Sci. 2023, 24, 1684 4 of 13 AQP2 proteins. As shown in RT-qPCR experiment ( Figures S1 and S2), the expression of two distinct forms can be clearly validated using species-specific primers for transcripts.

The Functional Properties of V2R-AQP2 Knock-In mpkCCDc14 Cells
Cell proliferation assay was performed to examine whether the targeted integration of Avpr2 and Aqp2 genes in mpkCCDc14 affects cell proliferation (Figure 2A). Cell Counting Kit-8 (CCK-8) analysis showed that the cell proliferation rate of the V2R-AQP2 cells (92 ± 3%) was not different compared to the control mpkCCDc14 cells (100 ± 3%, not significant (n.s.), Figure 2A), indicating that the targeted integration of Avpr2 and Aqp2 genes into the Rosa26 locus did not affect cell proliferation in mpkCCDc14 cells.

The Functional Properties of V2R-AQP2 Knock-In mpkCCDc14 Cells
Cell proliferation assay was performed to examine whether the targeted integration of Avpr2 and Aqp2 genes in mpkCCDc14 affects cell proliferation (Figure 2A). Cell Counting Kit-8 (CCK-8) analysis showed that the cell proliferation rate of the V2R-AQP2 cells (92 ± 3%) was not different compared to the control mpkCCDc14 cells (100 ± 3%, not significant (n.s.), Figure 2A), indicating that the targeted integration of Avpr2 and Aqp2 genes into the Rosa26 locus did not affect cell proliferation in mpkCCDc14 cells.
Since cAMP/PKA signaling is a main regulatory pathway for vasopressin-mediated AQP2 expression and trafficking in the renal collecting duct, intracellular cAMP levels were measured in control mpkCCDc14 cells and V2R-AQP2 cells ( Figure 2B). After the pre-treatment of 3-isobutyl-1-methylxanthine (IBMX, 1 mM) for 15 min, cells were treated with two different concentrations of dDAVP (10 −11 M or 10 −10 M) for 15 min in the presence of IBMX. dDAVP (10 −10 M) treatment significantly increased intracellular cAMP levels in both cell types to a similar level, demonstrating that vasopressin-induced cAMP production was not affected by the targeted integration of Avpr2 and Aqp2 genes.

Expression of Functional AQP2 in Control mpkCCDc14 Cells and V2R-AQP2 Cells
AQP2 expression was very low in control mpkCCDc14 cells cultured in vitro system without vasopressin stimulation. We examined whether the targeted integration of Avpr2 and Aqp2 genes in mpkCCDc14 cells (V2R-AQP2 cells) enables to exhibit detectable AQP2 protein abundance, even in the absence of vasopressin stimulation. To address the question, two different protocols were used.
In Protocol 1, control mpkCCDc14 cells and V2R-AQP2 cells were treated with dDAVP (10 −9 M) applied to the basolateral side for 0 h, 3 h, or 6 h after starvation for 24 h ( Figure 3A). Despite the absence of dDAVP stimulation (at 0 h), immunoblotting revealed that the abundance of glycosylated AQP2 (~35-50 kDa) was significantly higher in V2R-AQP2 cells, compared with control mpkCCDc14 cells (309 ± 31% of control at 0 h, p < 0.05, Figure 3B,C). Moreover, integrated AQP2 protein (bottom of the lower three AQP2 bands) was detected in V2R-AQP2 cells in the absence of dDAVP stimulation ( Figure 3B,E), indicating that basal AQP2 protein abundance was more abundant in V2R-AQP2 cells. Shortterm dDAVP treatment for 3 h or 6 h further increased both glycosylated and integrated AQP2 abundance in V2R-AQP2 cells (glycosylated AQP2: 532 ± 28% of control at 3 h, p < 0.05; 506 ± 15% of control at 6 h, p < 0.05, Figure 3B,C and integrated AQP2: 170 ± 5% at 3 h, p < 0.05, 168 ± 11% at 6 h, p < 0.05, Figure 3B,E), possibly resulted from the prolonged Since cAMP/PKA signaling is a main regulatory pathway for vasopressin-mediated AQP2 expression and trafficking in the renal collecting duct, intracellular cAMP levels were measured in control mpkCCDc14 cells and V2R-AQP2 cells ( Figure 2B). After the pre-treatment of 3-isobutyl-1-methylxanthine (IBMX, 1 mM) for 15 min, cells were treated with two different concentrations of dDAVP (10 −11 M or 10 −10 M) for 15 min in the presence of IBMX. dDAVP (10 −10 M) treatment significantly increased intracellular cAMP levels in both cell types to a similar level, demonstrating that vasopressin-induced cAMP production was not affected by the targeted integration of Avpr2 and Aqp2 genes.

Expression of Functional AQP2 in Control mpkCCDc14 Cells and V2R-AQP2 Cells
AQP2 expression was very low in control mpkCCDc14 cells cultured in vitro system without vasopressin stimulation. We examined whether the targeted integration of Avpr2 and Aqp2 genes in mpkCCDc14 cells (V2R-AQP2 cells) enables to exhibit detectable AQP2 protein abundance, even in the absence of vasopressin stimulation. To address the question, two different protocols were used.
In Protocol 1, control mpkCCDc14 cells and V2R-AQP2 cells were treated with dDAVP (10 −9 M) applied to the basolateral side for 0 h, 3 h, or 6 h after starvation for 24 h ( Figure 3A). Despite the absence of dDAVP stimulation (at 0 h), immunoblotting revealed that the abundance of glycosylated AQP2 (~35-50 kDa) was significantly higher in V2R-AQP2 cells, compared with control mpkCCDc14 cells (309 ± 31% of control at 0 h, p < 0.05, Figure 3B,C). Moreover, integrated AQP2 protein (bottom of the lower three AQP2 bands) was detected in V2R-AQP2 cells in the absence of dDAVP stimulation ( Figure 3B,E), indicating that basal AQP2 protein abundance was more abundant in V2R-AQP2 cells. Short-term dDAVP treatment for 3 h or 6 h further increased both glycosylated and integrated AQP2 abundance in V2R-AQP2 cells (glycosylated AQP2: 532 ± 28% of control at 3 h, p < 0.05; 506 ± 15% of control at 6 h, p < 0.05, Figure 3B,C and integrated AQP2: 170 ± 5% at 3 h, p < 0.05, 168 ± 11% at 6 h, p < 0.05, Figure 3B,E), possibly resulted from the prolonged half-life of AQP2 protein. In contrast, the glycosylated AQP2 protein abundance in control mpkCCDc14 cells was not changed after short-term dDAVP stimulation for 3 h or 6 h ( Figure 3B,C). A dDAVP treatment for 6 h induced the expression of endogenous AQP2 (middle of the lower three AQP2 bands) to similar level in both cell types (97 ± 6% of control at 6 h, not significant (n.s.), Figure 3B,D).   We performed real-time quantitative PCR (RT-qPCR) using mouse-or rat-specific Aqp2 primers to examine the constitutive expression of integrated rat Aqp2 gene in V2R-AQP2 cells and to determine the dDAVP-induced upregulation of Aqp2 mRNA. Firstly, to verify the specificity of mouse or rat Aqp2 primer, we prepared mouse kidney IMCD tubule suspension treated with 10 −9 M dDAVP for 2 h. Mouse Aqp2 primers not only amplified mouse Aqp2 mRNA, but also showed an increase in Aqp2 mRNA expression in response to dDAVP stimulation for 2 h (mouse Aqp2 primer 1: 174 ± 20%, p < 0.05; mouse Aqp2 primer 2: 157 ± 49%, not significant (n.s.), Figure S1B). In contrast, rat Aqp2 primers did not amplify mouse Aqp2 mRNA in mouse kidney IMCD tubule suspension ( Figure  S1C), indicating the specificity the mouse-or rat-AQP2 primers.
Next, the validated mouse and rat Aqp2 primers were applied to both control mpkC-CDc14 cells and V2R-AQP2 cells treated with dDAVP as performed in Protocol 1. RT-qPCR with mouse Aqp2 primer 1 showed that Aqp2 mRNA was amplified ( Figure S2A,B) and significantly increased by dDAVP stimulation for 6 h in both cell types (fold change: 89-fold increase in the control mpkCCDc14 cells at 6 h, p < 0.05; 79-fold increase in the V2R-AQP2 cells at 6 h, p < 0.05, Figure 3F). In contrast, rat Aqp2 primer 1 amplified Aqp2 mRNA only in the V2R-AQP2 cells, in which rat Aqp2 was inserted ( Figure S2B), but not Aqp2 mRNA in control mpkCCDc14 cells ( Figure S2A), indicating that rat Aqp2 was inserted as an integrated Aqp2 gene in V2R-AQP2 cells. Moreover, Aqp2 mRNA expression in V2R-AQP2 cells was increased by dDAVP treatment for 3 h (fold change: 4.86-fold increase at 3 h, p < 0.05, Figure 3G). The results indicate that the targeted integration of Avpr2 and Aqp2 genes did not affect the dDAVP-induced upregulation of endogenous AQP2 in genome-engineered mpkCCDc14 cells (V2R-AQP2 cells).

Constitutive and Homogenous Expression of the Integrated AQP2 in V2R-AQP2 Cells
Immunocytochemistry of AQP2 was performed to examine the homogeneity of AQP2 expression as well as AQP2 translocation to the plasma membrane in the absence or the presence of dDAVP treatment. dDAVP (10 −9 M) was applied to the basolateral side for 30 min without a period of dDAVP-mediated AQP2 induction. The relative intensity of AQP2 labeling was more prominent in V2R-AQP2 cells than in control mpkCCDc14 cells in the absence of dDAVP ( Figure 4C vs. Figure 4A), indicating the constitutive expression of integrated AQP2 in V2R-AQP2 cells, as demonstrated in Figure 3B. Moreover, AQP2 immunolabeling was more homogenous in V2R-AQP2 cells compared with control mpkC-CDc14 cells ( Figure 4C vs. Figure 4A), indicating the homogenous expression of integrated AQP2 in V2R-AQP2 cells. Furthermore, prominent AQP2 immunolabeling in the apical domain of the genome-edited cells was observed in response to dDAVP stimulation, similar to that in the unedited control mpkCCDc14 cells ( Figure 4D vs. Figure 4B). Consistent with AQP2 immunolabeling, the relative intensity of pS256-AQP2 labeling was more prominent in V2R-AQP2 cells than in control mpkCCDc14 cells in the absence of dDAVP ( Figure 4E vs. Figure 4G). dDAVP stimulation induced prominent of pS256-AQP2 immunolabeling in the plasma membrane, similar level in both cell types, as observed in the x-z images ( Figure 4F vs. Figure 4H and x-z image). The finding indicated that mpkCCDc14 cells constitutively expressing V2R and AQP2 via genome engineering could be exploited for AQP2 trafficking studies.

Constitutive and Homogenous Expression of the Integrated AQP2 in V2R-AQP2 Cells
Immunocytochemistry of AQP2 was performed to examine the homogeneity AQP2 expression as well as AQP2 translocation to the plasma membrane in the absen or the presence of dDAVP treatment. dDAVP (10 −9 M) was applied to the basolateral s for 30 min without a period of dDAVP-mediated AQP2 induction. The relative intens of AQP2 labeling was more prominent in V2R-AQP2 cells than in control mpkCCD cells in the absence of dDAVP ( Figure 4C vs. 4A), indicating the constitutive express of integrated AQP2 in V2R-AQP2 cells, as demonstrated in Figure 3B. Moreover, AQ immunolabeling was more homogenous in V2R-AQP2 cells compared with cont mpkCCDc14 cells ( Figure 4C vs. 4A), indicating the homogenous expression of integra AQP2 in V2R-AQP2 cells. Furthermore, prominent AQP2 immunolabeling in the api domain of the genome-edited cells was observed in response to dDAVP stimulation, s ilar to that in the unedited control mpkCCDc14 cells ( Figure 4D vs. 4B). Consistent w AQP2 immunolabeling, the relative intensity of pS256-AQP2 labeling was more prom nent in V2R-AQP2 cells than in control mpkCCDc14 cells in the absence of dDAVP ( Fig  4E vs. 4G). dDAVP stimulation induced prominent of pS256-AQP2 immunolabeling the plasma membrane, similar level in both cell types, as observed in the x-z images (F ure 4F vs. 4H and x-z image). The finding indicated that mpkCCDc14 cells constitutiv expressing V2R and AQP2 via genome engineering could be exploited for AQP2 traffi ing studies.  Figure 4. Immunofluorescence of total AQP2 and pS256-AQP2 in control mpkCCDc14 cells and V2R-AQP2 cells. Immunofluorescence staining of total AQP2 (A-D) and pS256-AQP2 (E-H) in control mpkCCDc14 cells and V2R-AQP2 cells treated with dDAVP (10 −9 M) to the basolateral side for 30 min without an AQP2 induction period. Con, control mpkCCDc14 cells; V2R-AQP2, mpkCCDc14 cells constitutively expressing V2R and AQP2 via CRISPR/Cas9-mediated genome engineering technology. Scale bar is 20 µm.

Discussion
In the present study, mpkCCDc14 cells were knocked-in with the rat Avpr2 gene and Aqp2 gene (i.e., V2R-AQP2 cells) by exploiting CRISPR/Cas9-mediated genome engineering technology. We demonstrated that the V2R-AQP2 gene cassette with a long insert size (~2.2 kb) was correctly integrated in the Rosa26 locus on chromosome 6 in mpkCCDc14 cells by 3 -and 5 -junction PCR. Moreover, PCR and immunoblotting revealed the expression of products of integrated Avpr2 and Aqp2 genes. Cell proliferation rate and dDAVPinduced cAMP production were not affected by the knock-in of Avpr2 and Aqp2 genes in mpkCCDc14 cells. The basal expression level of AQP2 without dDAVP stimulation was significantly higher in V2R-AQP2 cells compared with control mpkCCDc14 cells. Immunocytochemistry demonstrated that V2R-AQP2 cells exhibited more homogenous and prominent AQP2 labeling intensity in the absence of dDAVP stimulation. Moreover, prominent AQP2 immunolabeling (both AQP2 and pS256-AQP2) in the apical domain of the genome-edited cells (x-z images) was observed in response to dDAVP stimulation, similar to that in the unedited control mpkCCDc14 cells.
The mpkCCD cells originally developed by Vandewalle and colleagues were derived from isolated cortical collecting duct (CCD) microdissected from the kidney of SV-PK/Tag transgenic male mice [26]. mpkCCDc14 cells have been used for AQP2 studies, however, their endogenous AQP2 protein abundance is very low without vasopressin stimulation. Moreover, AQP2 expression is not homogenous among cultured cells. Therefore, several limitations lie in the AQP2 study, including trafficking and protein abundance, using mpkCCD cells. Firstly, cells must be treated with vasopressin in advance to induce sufficient AQP2 expression. This makes it difficult to study the shuttling or internalization of AQP2expressing vesicles in response to vasopressin stimulation or withdrawal. Secondly, there is the heterogeneity of AQP2 protein expression among cells. Thus, quantitative measurement and comparison in the AQP2 protein abundance are not simple. Moreover, although we did not study the expression levels of V2R, V2R expression may also be different among cells, which make the interpretation of vasopressin-induced trafficking or expression of AQP2 difficult. These limitations, therefore, could be overcome by the targeted integration of Avpr2 and Aqp2 genes via genome-editing technology in mpkCCDc14 cells.
Several previous attempts were made to obtain the cells highly expressing endogenous AQP2 protein. For instance, the mpkCCDc11 cells were recloned by selecting from the mp-kCCDc14 cells expressing different levels of AQP2 protein in the presence of dDAVP [23,24]. The mpkCCDc11 cells were more responsive to dDAVP with changes in AQP2 expression, phosphorylation, apical trafficking, and water permeability. LLC-PK1 cells, an epithelial cell line derived from pig kidney, express the V2R and increase the intracellular cAMP by vasopressin stimulation. Katsura et al. [27] established the LLC-PK1 cell line stably expressing AQP2 (LLC-AQP2 cells). The AQP2 protein in LLC-AQP2 cells was recruited to the plasma membrane in response to dDAVP [28,29] and the intracellular cAMP level was increased by forskolin [27]. Nevertheless, due to the various limitations of the cells listed above (e.g., vasopressin pre-treatment, heterogeneity of AQP2 or V2R expression), we aimed to generate V2R-AQP2 knock-in mpkCCDc14 cells.
In this study, we successfully established V2R-AQP2 knock-in mpkCCDc14 cells using CRISPR/Cas9 genome-engineering technology. The engineered V2R-AQP2 cells expressing both the constitutively-expressed AQP2 (integrated) and vasopressin-responsive AQP2 endogenously expressed in mpkCCDc14 cells could be exploited for AQP2 studies.

Transfection
To establish V2R-AQP2 knock-in cells (V2R-AQP2 cells), mpkCCDc14 cells were transfected using the Neon electroporation system (MPK5000, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Cells were washed with DPBS (14190144, Thermo Fisher Scientific, Waltham, MA, USA) and resuspended in Buffer R (Thermo Fisher Scientific, Waltham, MA, USA) at a concentration of 1 × 10 5 cells/mL. For each electroporation reaction, 1 × 10 5 cells in 10 µL of Buffer R were mixed with gRNA, Cas9 mRNA, and donor DNA (1:1:1 µg). The mixture was taken up into a 10 µL Neon Pipette Tip and electroporated using the following settings: 1500 V, 30 ms, 1 pulse. Electroporated cells were transferred to a medium supplemented with 10% FBS without antibiotics. After 3-5 days, V2R-AQP2 knock-in mpkCCDc14 cells were selected by treatment with 1 mg/mL of puromycin. Protocols for the dDAVP treatment: In Protocol 1 (cell culture without a period of dDAVP-mediated AQP2 induction), cells were cultured for 5 days and starved in serum-and hormone-free media for another 24 h. Then, cells were treated with dDAVP (10 −9 M, V1005, Sigma, St. Louis, MO, USA), which was applied to the basolateral side of the cells for 3 or 6 h. In Protocol 2 (cell culture having a period of dDAVP-stimulated AQP2 induction), cells were cultured for 6 days in total and lysed. Before lysis, dDAVP (10 −9 M) was applied to the basolateral side of the cells for 1 day or 2 days. On day 6, dDAVP was withdrawn for 2 h and then cells were treated with dDAVP for 30 min again [24]. The passage numbers of used mpkCCDc14 cells were 29-38. Cells were incubated in a humidified incubator at 37°C with 5% CO 2 .

Genomic DNA Extraction and Junction PCR Analysis of Genomic Integration
Genomic DNA from cultured cells was isolated with DirectPCR Lysis Reagent (Tail) (102-T, Viagen Biotech, Los Angeles, CA, USA). Cells in the 12-well plates were washed with DPBS and suspended in 70 µL of DirectPCR Lysis Reagent containing 0.2 mg/mL Proteinase K (25530049, Thermo Fisher Scientific, Waltham, MA, USA). The samples were incubated for 3-5 h at 55°C until the cells were totally dissolved. To inactivate proteinase K, crude lysates were incubated for 45 min at 85°C. For each junction PCR reaction, 5 µL of lysates and AccuPower PCR PreMix (K-2012, Bioneer, Seoul, Korea) were used. Primers used for the detection of HDR-mediated genomic integration had the following sequences: 5 -junction F: CGGAACTCTGCCCTCTAACG, 5 -junction R: TGAGGAAGAGTTCTTGCAGCT, 3junction F: CAGTGGCAGCCAGGTTAGC, 3 -junction R: CCTGGGATACCCCGAAGAGT.

Cell Counting Kit (CCK)-8 Assay
Cells were seeded into 96-well culture plates at a density of 5 × 10 4 cells per well. Following incubation for 24 h at 37°C, the CCK-8 assay was performed according to the manufacturer's instructions. Briefly, 10 µL of CCK-8 solution (CK04, Dojindo Molecular Technologies, Inc., Rockville, MD, USA) was added to each well and the cells were incubated for 2 h at 37°C. The absorbance was measured at 450 nm wavelength using a microplate reader.

cAMP Assay
Cells were cultured in a 6-well Transwell system (0.4 µm pore size, Transwell Permeable Supports, 3450, Corning, NY, USA) for 5 days and starved for the next 24 h. All measurements were performed in the presence of 1 mM 3-isobutyl-1-methylxanthine (IBMX; I5879, Sigma, St. Louis, MO, USA) to inhibit cyclic nucleotide phosphodiesterase. After pretreatment of 1 mM IBMX for 15 min, vehicle or dDAVP (10 −11 M or 10 −10 M) was added to the basolateral side of the cells for 15 min in the presence of IBMX. Intracellular cAMP levels were measured using a competitive enzyme immunoassay (501040, Cayman Chemical, Ann Arbor, MI, USA) according to the manufacturer's instructions. The results were presented in picomoles per milliliter of cell lysate.

Stastical Analysis
Quantitative data were presented as means ± SEM. Comparisons between two groups were made by the unpaired t-test. A comparison of multiple groups was made by one-way ANOVA followed by Tukey's multiple comparisons test. Multiple comparison tests were only applied when a significant difference was determined in ANOVA. GraphPad Prism software (GraphPad Software, La Jolla, CA, USA) was used for all statistical analysis. p values of <0.05 were considered statistically significant.