Ankrd45 Is a Novel Ankyrin Repeat Protein Required for Cell Proliferation

Ankyrin repeats, the most common protein–protein interaction motifs in nature, are widely present in proteins of both eukaryotic and prokaryotic cells. Ankyrin repeat-containing proteins play diverse biological functions. Here, we identified the gene ankrd45, which encodes a novel, two ankyrin repeat-containing protein. Zebrafish ankrd45 displayed a tissue specific expression pattern during early development, with high expression in ciliated tissues, including otic vesicles, Kupffer’s vesicles, pronephric ducts, and floor plates. Surprisingly, zebrafish ankrd45 mutants were viable and developed grossly normal cilia. In contrast, mutant larvae developed enlarged livers when induced with liver specific expression of KrasG12V, one of the common mutations of KRAS that leads to cancer in humans. Further, histological analysis suggested that multiple cysts developed in the mutant liver due to cell apoptosis. Similarly, knockdown of ANKRD45 expression with either siRNA or CRISPR/Cas9 methods induced apoptosis in cultured cells, similar to those in zebrafish ankrd45 mutant livers after induction. Using different cell lines, we show that the distribution of ANKRD45 protein was highly dynamic during mitosis. ANKRD45 is preferably localized to the midbody ring during cytokinesis. Together, our results suggest that ANKRD45 is a novel ankyrin repeat protein with a conserved role during cell proliferation in both zebrafish embryos and mammalian cells.


Introduction
The ankyrin repeat (ANK), a motif of 33 amino acid residues, is one of the most common protein motifs that is widely present in all forms of life, including bacteria, archaea, eukaryotes, and viruses. ANK proteins often contain such motifs in tandem arrays with each exhibiting a canonical helix-turn-helix conformation, a well-known motif involved in protein-protein interactions [1,2]. ANK repeats were first identified in the yeast cell cycle regulator Swi6/Cdc10 and the Drosophila cell signaling Notch protein, and later named after the human membrane-associated ankyrin protein, which contains 24 such repeats and regulates the interaction between the cytoskeleton and the plasma membrane [3,4].
Currently, thousands of ANK-containing proteins have been identified, which perform a wide range of functions, including signal transduction, cell cycle regulation, vesicle trafficking, cytoskeletal organization, and transcriptional regulation [2,5]. Dysfunction of ANK proteins is associated with many human disorders. Mutation of p16, a tumor suppressor protein with four ankyrin repeats, is associated with several human cancers due to abnormal cell cycle defects [6,7]. Disruption in the ankyrin repeat domains in Notch proteins leads to neurological disorders in humans [8,9]. The ANK protein IκB, an inhibitor of nuclear factor kappa B (NF-κB), is involved in transcription regulation and mediates metabolism and inflammatory responses [10,11]. IκB may also induce apoptosis in cancer cells as inhibition of IKKa, an IκB kinase leading to its degradation, can switch the effects of estrogens on human breast cancer MCF-7 cells from anti-to pro-apoptotic [12,13]. Inversin (INVS), also known as NPHP2, is a ciliary-localizing protein with multiple ANK domains. Patients harboring mutations in the INVS gene manifest multiple defects, including renal cystic disease and left-right asymmetry defects due to abnormal functioning of cilia [14].
Cilia are tiny organelles protruding from the cell surface and perform diverse biological functions [15]. Dysfunction of cilia may lead to multiple defects during embryonic development and result in a class of genetic disorders collectively termed as ciliopathies [16]. Recently, zebrafish have been used as disease models for ciliopathies [17]. Cilia are present in various organs of developing zebrafish larvae. Particularly, the olfactory pits, pronephric ducts, floor plates, and Kupffer's vesicles are tissues rich in motile cilia, and cilia genes are often expressed at a higher level in these organs [17]. Zebrafish cilia mutants frequently develop curly body axis phenotype due to motile cilia defects in the spinal cord [18].
KRAS, together with HRAS and NRAS, are members of the RAS family of small GTPases and mutations of these RAS genes are associated with one third of human cancers [19]. The KRAS G12V mutation is one of the common mutations that is found in many human cancers [19,20]. The G12V oncogenic mutation renders the KRAS protein more active by diminishing its hydrolysis from the GTP-bound active state to the GDP-bound inactive state. The GTP-bound KRAS G12V proteins chronically bind to and activate multiple downstream signaling pathways, including MAPK or PI3K/AKT signals, which lead to excessive cell proliferation and subsequent carcinogenesis [20].
In this study, we report the functions of a novel ANK protein, ANKRD45. We show that ankrd45 exhibits a tissue specific expression pattern with high enrichment in ciliary tissues during early zebrafish development. Although zebrafish ankrd45 mutants were viable with grossly normal cilia, mutant larvae displayed proliferation defects when induced with a liver specific kras G12V transgene. We further investigated the role of ANKRD45 both in zebrafish and in cell lines. Our data suggests that ANKRD45 is a novel player during cell cycle regulation.

Whole Mount In Situ Hybridization and Immunohistochemistry
Full-length ankrd45 gene was amplified from 24-h post fertilization (hpf) zebrafish cDNA with the following primers: Forward 5 -CACACCACATCACTACTCTTC-3 , Reverse 5 -GTAATGCAGTCCAACAGTTTC-3 . The PCR products were ligated into pEASY-T3 vectors. Probe preparation and hybridization were performed using standard protocols. To analyze its expression in liver, zebrafish larvae were anaesthetized and fixed at 5 days post fertilization (dpf) for cryosectioning. Transverse sections through the liver were collected for fluorescence in situ hybridization analysis. Fluorescent signal amplification using a TSA-plus Fluorescein System (Perkin Elmer Life Sciences, Boston, MA, USA) was carried out according to the manufacturer's protocols. For immunohistochemistry, the anti-monoglycylated Tubulin Antibody (TAP952, Merck-Millipore, Darmstadt, Germany) was used to visualize cilia in wild-type and mutant larvae. The antibody staining procedure was performed as described previously [23].

Pharmaceutical Treatments and mRNA Overexpression
To evaluate liver development, we generated wild-type and ankrd45 homozygous mutants carrying the Tg (fabp10:rtTA2s-M2;TRE2:EGFP-kras G12V ) transgene. In our experience, we have found that the homozygous EGFP-kras G12V transgene developed larger livers compared with the heterozygote transgene in wild-type larvae. To exclude this situation, we crossed ankrd45 homozygous mutants carrying the EGFP-kras G12V transgene with ankrd45 homozygous mutants, and compared them with wild-type embryos carrying the heterozygote transgene. The collected embryos were treated with 60 µg/mL doxycycline hydrochloride (Sangon Biotech, Shanghai, China) or DMSO starting from 60 hpf. Images of the liver with green fluorescence were captured using a Leica M165FC microscope at different time points after treatment. The size of liver was evaluated using Image J software according to the area of green fluorescent protein (GFP) expression.
For rescue experiments, full length ankrd45 cDNA was subcloned into pCS2+ expression vectors. After linearization, mRNA was transcribed using the mMessage mMachine Sp6 transcription kit (Invitrogen-Ambion, Austin, TX, USA). Microinjections were performed at the one-cell stage at a concentration of 100 ng/µL. Injected embryos were maintained for drug treatments at later stages to evaluate liver development.

Histological Analysis and TUNEL Assay
Zebrafish larvae carrying the EGFP-kras G12V transgene were fixed at three days after doxycycline treatment in 4% paraformaldehyde (PFA) (w/v) in PBST overnight at 4 • C. After gradually dehydrating in ethanol, the fixed embryos were finally embedded in JB4 embedding medium (Polysciences Inc., Warrington, PA, USA). Transverse sections through the liver were collected using Leica RM2235 microtome and stained with hematoxylin and eosin. Images were taken using a Leica DM2500 microscope. For TUNEL analysis, the doxycycline-treated embryos were fixed in 4% PFA overnight at 4 • C, infiltrated in 30% sucrose overnight at 4 • C, then embedded in frozen medium for cryosectioning using LEICA cryostat (CM1860). The sections were collected using adhesive slides and air dried for one hour at room temperature (RT). After that, sections were incubated in 70% ethanol/30% glacial acetic acid at −20 • C for 5 min, washed 5 times with 50 mM Tris-HCl, then blocked for 30 min at RT with blocking solution (1× PBS/10% normal goat serum/0.5% TrionX-100). The TUNEL assay was performed using the in situ cell death detection kit (Roche, Indianapolis, IN, USA) according to standard protocols from the manufacturer.

BrdU Incorporation Assay
At 68 h after doxycycline inducement, zebrafish larvae were incubated in 10 mM BrdU (Sigma, St. Louis, MO, USA) in 1% DMSO for four hours at 28.5 • C, then anesthetized in tricaine and fixed with 4% paraformaldehyde for two hours at RT. The fixed embryos were then sent for cryosectioning. Sections were treated with 2 N HCl for 1 h at RT, blocked 1 h with blocking solution, then incubated with mouse anti-BrdU antibody (Invitrogen) overnight at 4 • C. The next day, after a 1-h wash with PBST, the sections were further incubated with secondary goat anti-mouse IgG antibody. For cell culture analysis, Hep3B cells were incubated with BrdU at a final concentration of 20 µM for 2 h. After a PBS wash, cells were fixed with 4% PFA for 5 min, permeabilized with acetone:methanol (1:1) for 3 min on ice, then rehydrated in PBS for 5 min. The following denaturing and antibody incubation steps were similar to those of the staining of the cryosections.
For immunofluorescence, cells were fixed in 4% PFA at RT for 10 min, then treated with cold methanol at −20°C for 5 min. After two washes with PBS, the slides were permeabilized using 0.25% Triton-X-100 for 5 min, blocked with 3% BSA in PBS for 45 min, then incubated with anti-ANKRD45 and anti-α-tubulin antibodies in 3% BSA in PBS overnight at 4°C. After washing with PBS, the cells were incubated with secondary antibodies for 4 h at RT. All the images were captured using a Lecia Sp8 confocal microscope. We used two antibodies (Abcam, ab97878 and Invitrogen, PA5-21796) to detect ANKRD45 localization in cultured cells. Both of these two antibodies exhibited a similar staining pattern to that of ANKRD45 proteins. We only showed the immunofluorescence images in the paper with anti-ANKRD45 antibodies from Abcam (Cambridge, MA, USA).

Knockdown of ANKRD45 with the CRISPR/Cas9 System
Two sgRNAs that target the human ANKRD45 gene were designed with the following sequences: sgRNA-1: 5 -CTTAGAAGAAGACATCGTT-3 and sgRNA-2: 5 -TTGTAAAAGGGGGTTCTTA-3 . The target sequences were cloned into the sgRNA expression vector via BbsI restriction enzyme digestion. Each gRNA expression construct was transfected together with the pSpCas9 expression vector into Hep3B cells. The empty sgRNA expression vector was transfected together with pSpCas9 as a negative control. After 48 h of transfection, the expression level of ANKRD45 was examined by immunoblot analysis.

Time-Lapse Assay
HeLa cells were first transfected with control or ANKRD45 siRNA for 48 h as described above. After that, cells were plated on 20 mm glass-bottomed dishes. Time-lapse images were recorded at 5 min intervals for 12 h using a Nikon A1 laser scanning confocal microscope equipped with 20× objective lens. The movies were generated using the Image J software (NIH Image, Bethesda, MD, USA).

CCK8 Assay
Cell proliferation was assessed using the CCK-8 assays kit (Sigma), which uses a water-soluble tetrazolium salt to quantify the number of live cells by producing an orange formazan dye upon bio-reduction in the presence of an electron carrier. Briefly, Hep3B cells were transfected with 50 µM control or ANKRD45 siRNA and seeded onto 96-well plates at an initial density of 1 × 10 3 cells/well. At 1, 2, 3, and 4 days, 10 µL of CCK-8 solution was added to each well of the plate. After a one-hour incubation at 37 • C, spectrophotometric readings (A450 nm) were obtained on a microplate reader (Bio-Rad Laboratories, Hercules, CA, USA). Each experiment was repeated at least three times, and the data presented represent the mean of all measurements.

Cell Cycle and Annexin V Apoptosis Assay
Cell cycle distributions were determined by flow cytometry through staining of the cells with propidium iodide (PI). Briefly, Hep3B cells were harvested at 48 h or 72 h after siRNA treatment, then washed with 1× PBS and resuspended in 75% cold ethanol at 4 • C overnight. After a further wash with 1× PBS, fixed cells were treated with RNase for 30 min, and incubated with 30 µg/mL PI for one hour in the dark. The cells were analyzed with a Beckman Coulter Cytomics FC500MPC Flow Cytometry Analyzer.
The percentage of apoptotic cells was assessed with an eBioscience™ Annexin V-FITC Apoptosis Detection Kit according to the manufacturer's instructions (Invitrogen, 88-8005-72). Hep3B cells transfected with siRNA or sgRNA constructs were harvested and washed with cold PBS, then resuspended in 100 µL binding buffer containing 5 µL Annexin V-FITC. After 15 min incubation in the dark, cells were centrifuged at 1000 rpm for 5 min and resuspended in 500 µL binding buffer containing 5 µL of PI for 15 min in the dark at RT. The apoptosis was analyzed by a Beckman Coulter Cytomics FC500MPC Flow Cytometry Analyzer.

Expression of Ankrd45 Is Enriched in Ciliated Organs
When searching for genes involved in ciliogenesis, we identified ankrd45 as a candidate gene using whole-mount in situ-hybridization gene expression screening. The staining of ankrd45 was first weakly detected in the dorsal forerunner cells at 80% epiboly to bud stages ( Figure 1A). At the 10-somite stage, strong staining was detected in the otic placodes, Kupffer's vesicles, and floor plates ( Figure 1B). Later, at 24 h post-fertilization (hpf), ankrd45 was mainly expressed in those organs rich in motile cilia, including otic vesicles, pronephric ducts, and floor plates ( Figure 1C) [17]. At later stages, ankrd45 displayed a ubiquitous expression pattern. Real-time PCR results showed that the expression of ankrd45 persists into the adult stage with higher expression in the brain, liver, and testis ( Figure 1D).

Ankrd45 Is Dispensable for Zebrafish Development
Zebrafish Ankrd45 contains 224 amino acids with two ankyrin repeats, which are conserved between humans and zebrafish ( Figure 1E). The expression pattern suggests that Ankrd45 might play a role during ciliogenesis. To elucidate this, we generated zebrafish ankrd45 mutants using the CRISPR/Cas9 system and recovered a mutant allele containing a 5 bp deletion in the target area ( Figure 1E). This deletion introduced a new restriction site in the target region, and led to a premature termination of translation that encodes only the N-terminal 90 amino acids of the wild-type protein ( Figure S1A,B). Further in situ hybridization results suggested that mutant mRNA was degraded due to nonsense-mediated mRNA decay, a common phenomenon in zebrafish mutants ( Figure S1C,D) [24,25].
Next, we compared cilia in several organs of mutant and wild-type larvae, including olfactory pits, pronephric ducts, spinal canals, otic vesicles, and neuromasts, by immunofluorescence with anti-glycylated tubulin antibodies. Surprisingly, cilia developed normally in these tissues, and the length of mutant cilia was comparable to those in wild type ( Figure 1F,J). Furthermore, ankrd45 mutants were viable and fertile. These results suggest that Ankrd45 is dispensable for ciliogenesis, as well as zebrafish development.

Liver Specific Expression of Kras G12V Induced Larger Livers in Ankrd45 Mutants
Considering that many ANK proteins play a role during cell cycle regulation, we investigated whether loss of function of Ankrd45 could lead to cell proliferation defects in zebrafish larvae. Both real-time PCR and fluorescence in situ hybridization results suggested that ankrd45 displayed a higher expression level in the liver ( Figure 1D, Figure 2A-B"). We further generated ankrd45 mutants carrying a doxycycline-inducible transgenic line, Tg (fabp10:rtTA2s-M2; TRE2:EGFP-kras G12V ), which contained a liver-specific double transgene to induce the expression of EGFP-Kras G12V specifically in the liver [21,22]. KRAS is one of the three RAS isoforms that is most frequently mutated in human cancers. The KRAS G12V mutation reduces the ability of the RAS small GTPase to hydrolyze to GDP, rendering it constitutively active for downstream signaling [22,26]. Upon induction, as reported previously, enlarged and hyperplastic livers developed in both wild-type and ankrd45 mutant larvae ( Figure 2C,D). We compared the size of the liver between mutant and wild-type larvae from day 1 to day 5 after doxycycline treatment. Unexpectedly, ankrd45 mutants developed a larger liver than wild-type larvae after treatment ( Figure 2C,D). The average liver size was almost double in ankrd45 mutants compared to that in control larvae ( Figure 2C). We further injected wild-type ankrd45 mRNA into the mutant larvae and found that the formation of the enlarged liver was diminished in the mutants ( Figure S2A). These data suggested that Ankrd45 protein is involved in the Kras G12V -induced liver tumor growth.
Genes 2019, 10, x FOR PEER REVIEW 6 of 14 day 5 after doxycycline treatment. Unexpectedly, ankrd45 mutants developed a larger liver than wildtype larvae after treatment ( Figure 2C,D). The average liver size was almost double in ankrd45 mutants compared to that in control larvae ( Figure 2C). We further injected wild-type ankrd45 mRNA into the mutant larvae and found that the formation of the enlarged liver was diminished in the mutants ( Figure S2A). These data suggested that Ankrd45 protein is involved in the Kras G12V -induced liver tumor growth.

Enlarged Cystic Liver Is Due to Cell Degeneration in Kras G12V -Induced Ankrd45 Mutant
To better understand the mechanisms underlying the enlarged liver development in the mutant, we further investigated the liver morphology using hematoxylin and eosin (H&E) staining and oil red O staining on the histological sections. The oil red O staining showed no difference in lipid formation between mutant and wild-type embryos (data not shown). In wild-type embryos, H&E staining results showed a hyperplastic liver due to Kras G12V -induced cell proliferation. In contrast, multiple cyst-like areas were present throughout the mutant liver ( Figure 2E,F). To further explain

Enlarged Cystic Liver Is Due to Cell Degeneration in Kras G12V -Induced Ankrd45 Mutant
To better understand the mechanisms underlying the enlarged liver development in the mutant, we further investigated the liver morphology using hematoxylin and eosin (H&E) staining and oil red O staining on the histological sections. The oil red O staining showed no difference in lipid formation between mutant and wild-type embryos (data not shown). In wild-type embryos, H&E staining results showed a hyperplastic liver due to Kras G12V -induced cell proliferation. In contrast, multiple cyst-like areas were present throughout the mutant liver ( Figure 2E,F). To further explain the reason for cyst formation, we compared cell apoptosis between mutant and wild-type larvae at different time points after induction. TUNEL assay results showed that the number of apoptotic cells increased significantly (p < 0.001) in the mutant liver at 3, 4, and 5 days post treatment (dpt) ( Figure 2G-I, Figure S2B,D). In contrast, BrdU labeling results showed that the number of proliferating cells decreased substantially in the mutant ( Figure 2J-L, Figure S2C,E). Noticeably, we observed significant loss of proliferating cells at five-days post treatment, compared with those at 3 and 4 dpt ( Figure S2E). These data suggest that loss of ankrd45 results in a large number of apoptotic cells at the expense of proliferating cells in the Kras G12V -induced liver tumor, which leads to the formation of large areas of cystic degeneration and represents the main reason for enlarged livers in the mutants. the reason for cyst formation, we compared cell apoptosis between mutant and wild-type larvae at different time points after induction. TUNEL assay results showed that the number of apoptotic cells increased significantly (p < 0.001) in the mutant liver at 3, 4, and 5 days post treatment (dpt) ( Figure  2G-I,S2B,D). In contrast, BrdU labeling results showed that the number of proliferating cells decreased substantially in the mutant ( Figure 2J-L,S2C,E). Noticeably, we observed significant loss of proliferating cells at five-days post treatment, compared with those at 3 and 4 dpt ( Figure S2E). These data suggest that loss of ankrd45 results in a large number of apoptotic cells at the expense of proliferating cells in the Kras G12V -induced liver tumor, which leads to the formation of large areas of cystic degeneration and represents the main reason for enlarged livers in the mutants.

Dynamic Subcellular Localization of ANKRD45 During Mitosis
Considering that high levels of Kras G12V expression can lead to hepatocellular carcinoma (HCC) in zebrafish and that KRAS G12V is also the most popular mutation driving cancer cell proliferation [19,22], we further analyzed the role of ANKRD45 using several human cancer cells. First, we analyzed subcellular localization of ANKRD45 using human hepatoma Hep3B cells. Interestingly, the subcellular localization of ANKRD45 was cell-cycle dependent. Although we did not detect any signals during interphase, ANKRD45 localized diffusely in the entire cytoplasm during metaphase ( Figure 3A-A"). During anaphase and telophase, ANKRD45 was detected to be highly concentrated at the cleavage furrow region ( Figure 3B-C"). ANKRD45 appeared to accumulate rapidly during cleavage-furrow formation and was enriched at the midbody ring at the later cytokinesis ( Figure 3D,D"). Of note, ANKRD45 was distributed in a dotted manner at early stages of cell division and was later enriched in the apical region beneath the cell membrane, similar to those proteins that are essential for midbody ring formation [27,28]. We observed a similar localization pattern of ANKRD45 in several other cell lines, suggesting the conserved role of this protein during cell division ( Figure S3A-D). Furthermore, we pre-incubated the antibody with recombinant human ANKRD45 peptide before antibody staining, such treatment completely blocked the fluorescence signal in the midbody ( Figure S3E,F).

Loss of ANKRD45 Results in Cell Proliferation Defects in Cultured Cells
To further examine the role of ANKRD45 during cell division, we designed four ANKRD45 siRNAs to silence ANKRD45's expression. siRNA#2 showed more than 50% knockdown efficiency compared to other siRNAs in both Hep3B cells and HeLa cells ( Figure 3E, Figure S3G,H). When treated with this siRNA, the fluorescence signal of ANKRD45 in the midbody ring decreased significantly (p < 0.001) ( Figure 3F-H), which further confirmed the localization of ANKRD45.
Surprisingly, we observed robust, round-shaped, detached cells floating in the medium in siRNA-treated cells, indicating that these were apoptotic cells. To better describe this observation, we performed time-lapse microscopy to compare cell division between control and ANKRD45 siRNA-treated HeLa cells. In control siRNA-treated cells, cell cycle progressed through mitosis and cytokinesis normally within 2-3 h ( Figure 4A, Supplementary Movie S1). In contrast, silencing of ANKRD45 resulted in cell division defects and cell death characterized by rounding up of the cell body and eventual detachment from the plate ( Figure 4B, Supplementary Movie S2). Consistent with this, the cleaved form of PARP-1, a hallmark of cell apoptosis, was also increased in both HeLa and Hep3B cells ( Figure 3E, Figure S3G). We stained these cells with Annexin V and performed flow cytometry analysis, which further confirmed cell apoptosis after siRNA treatment ( Figure 4C,D).  Next, we performed a cell proliferation assay using a cell counting kit-8 (CCK-8) to evaluate cell viability in ANKRD45 siRNA-treated cells. Compared with the control group, knockdown of ANKRD45 expression dramatically affected cell viability ( Figure 4E). Flow cytometry analysis results suggested that the percentages of cells in G0/G1 stages were greatly increased at both 48 h and 72 h after treatment, suggesting that knockdown of ANKRD45 expression delayed cell cycle progression ( Figure 4F,G). Finally, BrdU incorporation and TUNEL assay further suggested that cell proliferation was diminished and cell apoptosis was induced after siRNA treatment ( Figure 4H-J).
Next, we performed a cell proliferation assay using a cell counting kit-8 (CCK-8) to evaluate cell viability in ANKRD45 siRNA-treated cells. Compared with the control group, knockdown of ANKRD45 expression dramatically affected cell viability ( Figure 4E). Flow cytometry analysis results suggested that the percentages of cells in G0/G1 stages were greatly increased at both 48 hours and 72 hours after treatment, suggesting that knockdown of ANKRD45 expression delayed cell cycle progression ( Figure 4F,G). Finally, BrdU incorporation and TUNEL assay further suggested that cell proliferation was diminished and cell apoptosis was induced after siRNA treatment ( Figure 4H-J). Considering that siRNA-mediated knockdown may cause off-target effects, we further designed two single-guide RNAs (sgRNAs) to transiently knockdown ANKRD45 expression using the CRISPR/Cas9 system ( Figure 5A). The efficiency of these two sgRNAs was confirmed by T7EI assay ( Figure 5B). Western blot with ANKRD45 antibody showed that the expression of ANKRD45 was successfully knocked down in both sgRNA transfected cells ( Figure 5C). Again, we observed a large number of dead cells at 72 h after sgRNA transfection ( Figure 5D-F). We stained the transfected cells with Annexin V, which further confirmed the increase of apoptotic cells ( Figure 5G-H). Finally, the expression level of the cleaved form of PARP-1 was also upregulated when ANKRD45 was knocked down ( Figure 5C).
Together, both the siRNA-and sgRNA-mediated knockdown results suggest that ANKRD45 plays an important role during cell division, and ANKRD45 deficiency can lead to cell proliferation defects and finally result in cell apoptosis, similar to results seen in zebrafish ankrd45 mutants. Considering that siRNA-mediated knockdown may cause off-target effects, we further designed two single-guide RNAs (sgRNAs) to transiently knockdown ANKRD45 expression using the CRISPR/Cas9 system ( Figure 5A). The efficiency of these two sgRNAs was confirmed by T7EI assay ( Figure 5B). Western blot with ANKRD45 antibody showed that the expression of ANKRD45 was successfully knocked down in both sgRNA transfected cells ( Figure 5C). Again, we observed a large number of dead cells at 72 h after sgRNA transfection ( Figure 5D-F). We stained the transfected cells with Annexin V, which further confirmed the increase of apoptotic cells ( Figure 5G,H). Finally, the expression level of the cleaved form of PARP-1 was also upregulated when ANKRD45 was knocked down ( Figure 5C).
Together, both the siRNA-and sgRNA-mediated knockdown results suggest that ANKRD45 plays an important role during cell division, and ANKRD45 deficiency can lead to cell proliferation defects and finally result in cell apoptosis, similar to results seen in zebrafish ankrd45 mutants.

Discussion
Ankyrin repeat-containing proteins are one of the most common protein families present in eukaryotic cells and they play a role in various biological functions. Here, we reported the function of a novel ANK protein, ANKRD45, which contains two ankyrin repeats. The expression pattern of ankrd45 in early zebrafish embryos was tissue specific with high enrichment in ciliated organs, suggesting that Ankrd45 possibly plays a role during cilia development. Indeed, ANKRD45 was also identified recently from proteomics analysis by comparison of ciliary proteins in different species [29]. Unexpectedly, homozygous ankrd45 mutants were viable and showed no apparent defects. All the cilia detected were grossly normal in ankrd45 mutants. Although we cannot rule out subtle defects, our results suggest that Ankrd45 is not essential for ciliogenesis in zebrafish. Interestingly, ankrd45 mutants displayed abnormal reaction during stress response, such as Kras G12V -induced tumor formation. Overexpression of the oncogenic Kras G12V could promote tumor formation when strongly expressed in the liver in wild-type zebrafish [22]. The induction of enlarged livers seems to be more prominent in ankrd45 mutant larvae. However, our histological results suggest that the enlarged livers were indeed due to the presence of a large amount of cystic degeneration due to cell apoptosis. These data suggest that Ankrd45 participates in the regulation of Kras G12V -induced cell proliferation. Similarly, ANKRD45 is also important for cell proliferation in several cancer cells, including human hepatoma Hep3B cells, which are relatively similar to the Kras G12V -induced liver cells.
Although deficiency of Ankrd45 proteins resulted in cell division defects both in Kras G12V -induced liver cells and cultured cancer cells, we did not observe any cell proliferation defects in ankrd45 mutants. One possible explanation is that the embryos are complex, multicellular organisms with elaborate gene regulation systems, which can trigger multiple signaling pathways to compensate for the loss of Ankrd45 in the mutant larvae [30]. Under stress conditions, such as rapid cell division in tumor cells, which might require high ANKRD45 levels, the gene regulation systems fail to fully compensate for the loss of ANKRD45, which eventually leads to cell apoptosis. The high requirement of ANKRD45 during cancer cell proliferation may be worthy of further attention to evaluate the potential target effects during therapeutic trials.
The dynamic subcellular localization of ANKRD45 during cell cycle progression is consistent with its role in cell proliferation. Time-lapse microscopy showed that cell division was arrested during the cell cycle, resulting in cell death at interphase and/or metaphase, which suggested that ANKRD45 is required for cell division both in early and late mitotic stages. Currently, the mechanism by which ANKRD45 regulates mitosis is still unclear. ANKRD45 was slightly enriched beneath the plasmid membrane during the cleavage furrow formation and later in the midbody ring ( Figure 3A-D"). Such a distribution was similar to that of some actin binding proteins, such as ezrin/radixin/moesin (ERM) proteins, which are involved in the regulation of membrane morphology [28,31]. Future studies will be necessary to decipher the mechanisms by which ANKRD45 regulates mitosis.