Abstract
Background/Aim: Alpha-kinase 2 (ALPK2), suggested to be a novel tumour-suppressor gene down-regulated by oncogenic KRAS, plays a pivotal role in luminal apoptosis in normal colonic crypts. The aim of this study was to determine the association between ALPK2 germline variants and colorectal cancer. Materials and Methods: Missense single nucleotide variants in the exons of the ALPK2 gene in 2,343 consecutive autopsy cases (1,446 cases with cancer and 897 cases without cancer) were screened using HumanExome BeadChip arrays. To address the functional effect of a missense ALPK2 variant, a 3D floating cell culture was performed using HCT116-derived human colorectal cancer cells stably expressing wild-type (wt) ALPK2 (HCT116-wtALPK2) or amino acid-substituted (sub) ALPK2 (HCT116-subALPK2). Results: We identified that one of the ALPK2 germline variants, rs55674018 (p.Q1853E), was significantly associated with the presence of cancer (adjusted odds ratio(OR)=4.39; 95% confidence interval(CI)=1.31-14.78, p=0.001). The p.Q1853E variant was present in the East Asian population and located in the immunoglobulin-like domain. Notably, the basolateral polarity of actin in the surface of HCT116-wtALPK2 spheroids was more attenuated compared to that of HCT116-subALPK2 spheroids. Furthermore, luminal apoptosis and cell aggregation were promoted by wtALPK2, but not by subALPK2 in 3D culture. Conclusion: The p.Q1853E variant of ALPK2, which had been accumulating in the Japanese population, induced a metastatic phenotype by disrupting ALPK2 function.
- ALPK2
- missense variation
- luminal apoptosis
- basolateral polarity
- intercellular interaction
- three-dimensional floating culture
- colorectal cancer
Human tumorigenesis is associated with multiple genetic alterations. Mutations of Kirsten rat sarcoma viral oncogene homolog (KRAS) are frequently observed during the early stages of colorectal cancer (CRC) development, and even in adenomas (1-3), suggesting that oncogenic KRAS plays several crucial roles in the adenoma–carcinoma sequence (4). We previously investigated the behavior of HKe3 cells, which are HCT116 human CRC cells with a disruption in oncogenic KRAS (5), in three-dimensional (3D) culture and reported that they formed an organized structure resembling a colonic crypt (6). In this model, we found that the alpha-kinase 2 (ALPK2) gene was one of the genes down-regulated by oncogenic KRAS (7).
ALPKs constitute a recently discovered family of protein kinases that have no detectable sequence homology to conventional protein kinases (8). To date, six human ALPKs have been identified, namely eukaryotic elongation factor-2 kinase (9), ALPK1 (lymphocyte ALPK, LAK), ALPK2 (heart ALPK, HAK), and ALPK3 (muscle ALPK, MAK), transient receptor potential cation channel subfamily M member 6 (TRPM6) and TRPM7 (9, 10).
The ALPK2 gene is mapped to 18q21.31, the distal end of a minimal region of loss of heterozygosity frequently observed in colonic adenomas and colon cancer (11-13). We showed that ALPK2 induces luminal apoptosis and cell polarity and up-regulates DNA-repair genes (including TP53) in a 3D-specific manner (6, 7), suggesting ALPK2 to be a novel tumor-suppressor gene. Indeed, Lawrence et al. recently reported ALPK2 to be one of 33 genes whose somatic point mutations were associated with 21 cancer cell lines (14) (http://www.tumorportal.org). Furthermore, a recent study showed that ALPK2 is down-regulated by miRNA-214, which is overexpressed in human non-small cell lung cancer with metastatic potential (15). These studies strongly suggest that ALPK2 has pivotal roles as a tumor-suppressor gene. However, the functional details of ALPK2 action still remain unknown.
In this study, we hypothesized that certain germline variations in the ALPK2 gene are associated with cancer and examined nonsynonymous variants of ALPK2 in 2,343 autopsied individuals, for whom the presence or absence of cancer was pathologically confirmed (16).
Materials and Methods
Study population. The study group comprised 2,343 consecutive autopsy cases, which were collected from the Tokyo Metropolitan Geriatric Hospital between 1995 and 2012 (1,293 men and 1,050 women; the mean age at the time of death was 80.6±8.8 years) (16). The subjects were enrolled in the online database of Japanese single nucleotide polymorphisms for Geriatric Research (JG-SNP) (17). The presence or absence of any disease was determined by a thorough autopsy on approximately 29% of patients who died in the hospital.
Genotyping. Genomic DNA was extracted from the renal cortex using a standard procedure. All samples were genotyped using Infinium HumanExome BeadChip Kit Version 1.1 (Illumina, San Diego, CA, USA) as previously described (16).
Association analysis. In the exome array, 14 probes were available for nonsynonymous variants of the ALPK2 gene: rs7240666 (p.I2157V), rs723499 (p.H1767Y), rs17065127 (p.K1730E), rs33910491 (p.Q1579R), rs3809977 (p.H1174P), rs3809975 (p.S977T), rs4940404 (p.N916K), rs3826593 (p.T891I), rs3809973 (p.K829N), rs3809970 (p.G810S), rs12103986 (p.H719Q), rs9944810 (p.R136S), rs6566987 (p.K2T) and rs55674018 (p.Q1853E). Multiple logistic regression analyses under a dominant model were performed to determine the association between the genotypes and phenotypes as described previously (16). By applying the Bonferroni method for multiple testing correction, the significance threshold was set to 0.00357.
Ethics statement. This study was approved by the Tokyo Medical and Dental University Ethics Committee (approval no. 2009-19-4) and the Tokyo Metropolitan Geriatric Hospital Ethics Committee (approval no. 230405). Written informed consent was obtained from a family member of each participant involved in this study before autopsy.
Cell culture. HCT116 human CRC cells were obtained from the American Type Culture Collection (Frederick, MD, USA). HKe3 cells, HCT116 and HCT116-derived cells were maintained as previously described (5, 6, 18). All cell lines used were confirmed to be mycoplasma-free using the MycoAlert system (Lonza, Verviers, Belgium). Cell morphology was checked regularly to ensure that the cell lines were not cross-contaminated.
Retroviral production and generation of stable cell lines. Dasher Green Fluorescent Protein (DGFP) cDNA from DNA2.0's Cas9 vectors (pD1401-AD) was subcloned into the retroviral pMSCVpuro vector (Clontech,, Palo Alto, CA, USA) at the multi-cloning site (pMSCV-DGFP). Each of the cDNAs for human wtALPK2 (accession number NM052947.3) and subALPK2 (Q1853E) was inserted into the pMSCVpuro and pMSCV-DGFP vectors. Retroviruses were produced by the transfection of the retrovirus vectors as previously described (19). HCT116 cells in 6-well plates were infected with the viruses by being spun for 2 h at 32°C and 2000 × g. Approximately 48 h after infection, the cells were treated with 2 μg/ml puromycin (Sigma-Aldrich, St. Louis, MO, USA) for 1 week to establish HCT116-derived cells stably expressing wtALPK2 or subALPK2. The cells were further maintained in medium containing 2 μg/ml puromycin.
Antibodies and reagents. The antibodies used were monoclonal antibody to ALPK2 (4E5) from Abnova (Taipei, Taiwan), anti-heat-shock protein 90 (HSP90; 68/Hsp90) from BD Biosciences (San Jose, CA, USA), anti-E-cadherin from BD Biosciences, anti-cleaved caspase-3 (5A1) from Cell Signaling Technology (Beverly, MA, USA) and anti--actin (A2066) from Sigma-Aldrich. 4’,6-Diamidino-2-phenylindole (DAPI) was obtained from Sigma-Aldrich.
Subcellular fractionation. Subcellular fractions from mouse thymus were obtained as described previously (20).
3D Floating (3DF) culture. 1×103 cells were seeded in a 96-well plate with an ultra-low attachment surface and round bottom (product number 7007; Corning Inc., Corning, NY, USA). Cells were cultured for 2 to 6 days with or without EtOH in CO2 incubator as previously described (19). Photomicrographs of cells were taken using a BIOREVO BZ9000 microscope (Keyence, Osaka, Japan), while the area of spheroids was measured using a BZ Analyzer (Keyence) as previously described (6, 7, 20, 21).
Immunoblotting. Cells grown in two-dimensional culture (2D) were lysed in RIPA buffer (50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS and protease inhibitor cocktail; Roche, Basel, Switzerland) and subjected to immunoblotting as previously described (20, 22). For 3DF culture, cells were seeded in 96-well plates and cultured. At day six, by inverting the 96-well plate, the resultant spheroids were collected into plastic dishes. The collected spheroids from the plastic dishes were transferred to 50 ml tubes and centrifuged for 3 min at 200 × g. The spheroid pellets were then lysed in RIPA buffer for immunoblotting. Quantitative analysis of immunoblotting was performed using ImageJ software (National Institute of Health, Bethesda, MD, USA).
Immunofluorescence labelling and confocal microscopy. Immunofluorescence experiments were performed as described previously (6, 7). To examine 3D structures, a TCS-SP5 Laser Scanning Confocal Microscope (Leica, Wetzlar, Germany) was used.
Statistical analyses in cell culture experiments. Data are presented as the mean±standard deviation. Statistical analyses were performed using unpaired two-tailed Student's t-test. All p-values less than 0.05 were considered to be statistically significant.
Results
The p.Q1853E variant is associated with the presence of cancer at the time of death. In this study, we analyzed nonsynonymous germline variants of the ALPK2 gene in a study population previously described (16): the 2,343 subjects included 1,446 cases with and 897 without cancer, respectively. Regardless of the cause of death, 61% of the patients had at least one cancer at the time of death. The top three cancer types were lung (18.9%, n=274), gastric (18.1%, n=262) and colonic (12.0%, n=173) cancer.
Among the 14 nonsynonymous variants in the ALPK2 gene, there was a significant association between the presence of cancer at the time of death and the p.Q1853E variant (p=0.001; Table I). For individual cancer types, p.Q1853E was associated with several types of cancer, including those of the prostate, lung, biliary tract and rectum (data not shown).
p.Q1853E is a rare and potentially rare functional variant present in the East Asian population. We then examined the association between the ALPK2 variants, 14 of which were analyzed here, and four different ethnic populations using the Exome Aggregation Consortium (ExAC) Browser (http://www.http://exac.broadinstitute.org/). As shown in Table II, the ALPK2 locus contains many missense variants with a high variant-allele frequency. Notably, the p.Q1853E variant was only found in the Asian populations.
Many of the rare missense variants are potentially functional and may be crucial for yielding insights into the genetic basis of human disease (23). Amino acid residue 1853 of APLK2 is located in the immunoglobulin-like domain, which is known to be involved in intercellular interactions (24). Therefore, there is the possibility that the accumulation of p.Q1853E variant in the East Asian population is functional. We examined the potential effect of the p.Q1853E variant on APLK2 protein function using the SIFT browser (http://sift.jcvi.org.), which is a sequence homology-based tool that sorts-out intolerant amino acid substitutions from tolerant ones and predicts the extent of their phenotypic effects (25). The amino acid substitution of p.Q1853E variant was predicted as damaging as shown in Table II, suggesting that the p.Q1853E variant was functionally important.
The p.Q1853E variant affects luminal apoptosis of HCT116 spheroids. In order to address the effects of the p.Q1853E variant on cell proliferation in 3D culture, we established HCT116 cells stably expressing wtALPK2 or subALPK2. Expression levels of HCT116-wtALPK2 and -subALPK2 cells were similar to those of endogenous ALPK2 in HKe3 cells (Figure 1A). The areas of HCT116-wtALPK2 and HCT116-subALPK2 spheroids were compared. Although no significant differences in the growth rates were observed between HCT116-wtALPK2 and HCT116-subALPK2 spheroids (Figure 1B), the signal for cleaved caspase-3 was higher in HCT116-wtALPK2 than in HCT116-subALPK2 spheroids, suggesting that the p.Q1853E variant affects luminal apoptosis in the 3D microenvironment, as previously reported (7).
Distribution of p.Q1853E variants in autopsy cases with and without cancer.
The p.Q1853E variant alters the localization of filamentous-actin (F-actin) in the surface cells of spheroids. The intracellular localization of ALPK2 is not well-characterized. Staining of HKe3 cells with phalloidin and antibody to ALPK2 revealed that the endogenous ALPK2 proteins co-localized with F-actin, which is recognized by phalloidin, and localized in the cytoplasm and membrane in 2D culture (Figure 2A). To further characterize the subcellular localization pattern of ALPK2, cells were lysed with RIPA buffer, and the lysate was separated into the supernatant containing the soluble fraction, including actin and the pellet containing the insoluble fraction, including E-cadherin. The immunoblotting detected the ALPK2 protein predominantly in the supernatant fraction (Figure 2B). Therefore, consistent with the results of the confocal microscopic analysis (Figure 2A), our results suggest that the majority of the ALPK2 protein is localized in the cytoplasm, while a little is localized in the membrane (Figure 2B). To examine whether the p.Q1853E variant changes the subcellular localization of ALPK2, we prepared HCT116-wtALPK2-DGFP and HCT116-subALPK2-DGFP cells grown in 3DF culture and stained the surface of the spheroids using DAPI and phalloidin. ALPK2 colocalized with F-actin in HCT116-wtALPK2-DGFP and HCT116-subALPK2-DGFP cells. Notably, although ALPK2 was localized near the basolateral membrane (upper panels) in HCT116-wtALPK2 cells, that in HCT116-subALPK2 cells was thoroughly localized in the cytoplasmic region (lower panels), suggesting that ALPK2 is involved in the basolateral polarity of the surface cells of spheroids (Figure 2C).
Allelic frequencies of alpha-kinase-2 (ALPK2) variants in different ethnic populations.
The p.Q1853E variant induces the disassembly of HCT116 spheroids. The p.Q1853E variant is located in the immunoglobulin-like domain, which is known to be involved in intercellular interactions (24). In order to further understand the function of the p.Q1853E variant in vitro, we developed a 3DF culture system for examining cellular aggregation. Ethanol is known to cause redistribution and intracellular mislocalization of tight junction proteins, including zonula occludens-1, and occluding without affecting the levels of protein expression (26). To examine whether ALPK2 was associated with spheroid assembly, HCT116 cells overexpressing wt ALPK2 or variant ALPK2 were used. The addition of ethanol partially prevented the formation of HCT116-subALPK2 spheroids in a dose-dependent manner, but did not affect that of HCT116-wtALPK2 spheroids (Figure 3). These results suggest that the p.Q1853E variant attenuates intercellular interaction.
The p.Q1853E variant of alpha-kinase 2 (ALPK2) reduces caspase-3 cleavage in HCT116 spheroids. A: Western blot analyses of ALPK2 expression in HKe3, parental (Par) HCT116 and HCT116 cells stably expressing wild-type (wt)ALPK2 (HCT116-wtALPK2) or amino acid-substituted (sub) ALPK2 (HCT116-subALPK2) grown in 2D culture. B: The area of spheroids formed by HCT116-derived cells at day 8 in 3D floating (3DF) culture. One thousand each of HCT116-wtALPK2 and -subALPK2 cells were seeded for 3DF culture. The areas of spheroids were measured at day 8. n.s.: Not significantly different. C: Expression of cleaved caspase-3 in HKe3, HCT116, HCT116-wtALPK2 and HCT116-subALPK2 cells at day 6 in a 3DF culture. β-Actin was used as a loading control. The upper panel shows western blotting results and the lower panel shows quantitative analyses of the relative expression levels. Bars represent the relative signal intensities of cells normalized to the signal of HCT116 cells. Significantly different at *p<0.05 and **p<0.005.
The p.Q1853E variant of alpha-kinase-2 (ALPK2) alters the localization of ALPK2 and filamentous-actin (F-actin) in the surface cells of spheroids. A: Signals for antibody to ALPK2 (green), phalloidin-labelled F-actin (red) and 4’,6-diamidino-2-phenylindole (DAPI)-labelled nuclei (blue) in HKe3 cells grown in 2D culture. Scale bar=10 μm. B: Western blot analyses for the expression of ALPK2 (left panel), E-cadherin and β-actin (right panel) in different fractions of HKe3 cells. C: The signals for Dasher Green Fluorescent Protein (DGFP) (green), phalloidin-labelled F-actin (red) and DAPI-labelled nuclei (blue) in HCT116 cells stably expressing wild-type ALPK2-DGFP (upper panels) or amino acid-substituted (sub) ALPK2-DGFP (lower panels) cells at day six in 3D floating culture. Scale bar=20 μm.
The p.Q1853E variant of alpha-kinase 2 (ALPK2) attenuates cellular aggregation. Representative images of HCT116 spheroids stably expressing wild-type (wt) ALPK2 or amino acid-substituted (sub) ALPK2 grown in the presence or absence of ethanol (EtOH, 0.2% or 1%) at day two in 3D floating culture. Scale bar=100 μm.
Discussion
To our knowledge, this is the first report to focus on the association between the germline genetic variants of ALPK2 and the presence of cancer in Japanese autopsy cases. The genotype data available in the ExAC Browser clearly demonstrate that the p.Q1853E variant is specific to the Asian population: its frequencies in East and South Asian populations are 0.48% and 0.02% but 0% in the other ethnic populations (Table II). Surprisingly, the frequency of the p.Q1853E variant in individuals with cancer in our study population was 1.5% (Table I). The East Asian population is a population showing rapid growth. Rare variants accumulating in such a population tend to be functional due to weak purifying selection (23). The other ALPK2 missense variants, whose total allelic frequency is less than 0.1% (Table II), are also accumulating in the East Asian population and localized in the functional domains of ALPK2, suggesting that these variants cooperatively disrupt ALPK2 function.
We previously showed that siRNA inhibition of ALPK2 expression prevents luminal apoptosis (7). In this study, we demonstrated that overexpression of wt ALPK2 protein also induces luminal apoptosis of HCT116 cells in a 3DF culture, while the overexpression of variant ALPK2 did not (Figure 1). These results indicate that that the 1853rd glutamine residue of ALPK2 protein is essential for luminal apoptosis.
Wild-type and variant ALPK2 both co-localized with F-actin in 2D and 3D cultures, indicating the physical association between ALPK2 and the cytoskeleton. Notably, although wt ALPK2 was localized near the basolateral membrane, variant ALPK2 was thoroughly localized in the cytoplasmic region. These results show that ALPK2 is involved in the basolateral polarity of the surface cells of spheroids (Figure 2C). A recent study of invertebrate ALPK2 family proteins, such as myosin heavy chain kinase (MHCK) A, MHCK-B and MHCK-C, suggested that these ALPK family members target proteins involved in filament disassembly, including myosin II (10). Ivanov et al. reported that myosin II regulates the spherical shape of epithelial cysts by controlling actin polymerization at the cyst surface (27). Furthermore, in smooth muscle cells, the inhibition of myosin light chain kinase induces apoptosis in vitro and in vivo (28). A recent study also showed that the inhibition of non-muscle myosin IIA (NMIIA) reduces the nuclear localization of wt TP53 (29), suggesting the increased induction of apoptosis. Interestingly, we previously reported that siRNA inhibition of ALPK2 expression leads to the down-regulation of DNA repair-related genes that are regulated by TP53 (7), suggesting the functional association between ALPK2 and TP53 via NMIIA. Collectively, our results and other reports suggest that the p.Q1853E variant affects ALPK2 kinase activity and induces F-actin depolarization and luminal apoptosis.
We also demonstrated that the p.Q1853E variant prevents cell aggregation under conditions where tight junction protein function is attenuated (Figure 3), suggesting that p.Q1853E is involved in intercellular interaction through the immunoglobulin-like domain.
Taken together, the intact intercellular interaction, which includes apical and basolateral membranes, induces luminal apoptosis (6). These results strongly suggest that p.Q1853E plays an essential role in disrupting normal cyst formation. Further elucidation of ALPK2 function and its substrates, as well as the validation of the association between ALPK2 variants and cancer progression, will help in establishing diagnosis, therapeutics and prognosis for patients with cancer having rare ALPK2 variants or the deletion of ALPK2 gene.
Acknowledgements
The Authors thank Takami Danno, Shiori Yamano and Yumiko Hirose for their technical assistance.
Footnotes
↵* These Authors contributed equally to this study.
Funding
This work was supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan; by GMEXT/JSPS KAKENHI Grants (Numbers: A-25242062, A-22240072, B-21390459, C-26670481, C-21590411 and CER-24650414 to MT); by Grants-in-Aid for Research on Intractable Diseases Mitochondrial Disorders (grant numbers 23-016, 23-116 and 24-005 to MT) from the Ministry of Health, Labor, and Welfare of Japan; by the Practical Research Project for Rare/Intractable Diseases from the Japan Agency for Medical Research and Development, AMED (15ek0109088h0001 and 15ek0109088s0401 to MT); by the Takeda Science Foundation (to MT); by the Smoking Research Foundation (to TA, and MS); and by the Joint Usage/Research Program of the Medical Research Institute, Tokyo Medical and Dental University (to MM).
- Received May 2, 2017.
- Revision received May 25, 2017.
- Accepted May 29, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.