USP21 contributes to the aggressiveness of laryngeal cancer cells by deubiquitinating and stabilizing AURKA

Abstract

Laryngeal cancer is a usual malignant tumor of the head and neck. The role and mechanism of deubiquitinase USP21 in laryngeal cancer are still unclear. We aimed to explore whether USP21 affected laryngeal cancer progress through deubiquitinating AURKA. USP21 and AURKA levels were evaluated by qRT‐PCR and Western blot. Kaplan–Meier analysis was conducted by survival package. MTT was performed to detect cell proliferation. The wound healing assay was applied to evaluate cell migration. Transwell was used to measure cell invasion. Co‐IP and GST‐pull down determined the interaction between USP21 and AURKA. In addition, AURKA ubiquitination levels were analyzed. USP21 was signally elevated in laryngeal cancer tissues and cells. USP21 level in clinical stages III–IV was higher than that in clinical stages I–II, and high levels of USP21 were highly correlated with poor prognosis in laryngeal cancer. USP21 inhibition suppressed AMC‐HN‐8 and TU686 cell proliferation, migration and invasion. Co‐IP and GST‐pull down confirmed the interaction between USP21 and AURKA. Knockdown of USP21 markedly increased the ubiquitination level of AURKA, and USP21 restored AURKA activity through deubiquitination. In addition, overexpression of AURKA reversed the effects of USP21 knockdown on cell growth, migration, and invasion. USP21 stabilized AURKA through deubiquitination to promote laryngeal cancer progression.

Abbreviations

ANOVA

one‐way analysis of variance

AURKA

A aurora kinase A

CHX

cycloheximide

Co‐IP

co‐immunoprecipitation

GST

glutathione‐S‐transferase

qRT‐PCR

quantitative real‐time PCR

USP

ubiquitin‐specific protease

1. INTRODUCTION

Laryngeal cancer accounts for about one‐third of all head and neck cancers. ¹ It is characterized by ineffective conventional treatment and poor prognosis. ² The gold standard for the diagnosis of laryngeal cancer is histopathological image analysis. However, the pathologist's diagnosis is highly subjective and easy to miss and misdiagnose. ³ Current treatments include monomodal treatment of early tumors by surgery or radiotherapy, while advanced tumors are usually treated with multimodal chemotherapy or radiotherapy after (total) laryngectomy. Nevertheless, the recurrence rate of advanced laryngeal cancer is still between 25% and 50%. ⁴ The successful treatment of laryngeal cancer depends on the pretreatment assessment of patients and disease factors to achieve accurate staging and select appropriate treatment methods for patients with this highly affected disease. ⁵ Therefore, introducing molecular biomarkers as predictive factors combined with new treatment modalities may help improve laryngeal cancer patients' survival rate.

Ubiquitination is the basic mechanism of signal transduction, regulating immune response and many other biological processes. It is a reversible process and a key post‐translational modification, which is counter‐regulated by ubiquitinating enzymes and deubiquitinases. ⁶ , ⁷ Ubiquitination regulates protein stability and activity and is involved in homeostatic cellular functions. ⁸ By labeling proteins degraded by the proteasome, ubiquitination has always been considered a key determinant of protein fate. ⁹ Deubiquitinases drive the cleavage of ubiquitin from substrates, which control the stability of most cellular proteins. ¹⁰ Ubiquitin‐specific protease (USP) 21 is a deubiquitinating enzyme involved in the malignant progression of various cancers. ¹¹ USP21 can maintain the stability of related proteins through deubiquitination and promote the progress of cancer. For example, Arceci et al. reported that USP21‐mediated FOXM1 deubiquitination regulated basal breast cancer cell cycle progression and paclitaxel sensitivity. ¹² Chen et al. found that USP21 promoted bladder cancer cell proliferation and metastasis by inhibiting EZH2 ubiquitination. ¹³ However, the function of USP21 in laryngeal cancer has not been reported yet.

Aurora kinase A (AURKA) is overexpressed in 96% of human cancers and regulates cell cycle progression, mitosis, and a series of key oncogenic signaling pathways. AURKA is an independent marker of poor prognosis in tumors. ¹⁴ In laryngeal squamous cell carcinoma, AURKA may induce epithelial–mesenchymal transition by activating the FAK/PI3K pathway to promote the metastasis of laryngeal squamous cell carcinoma. ¹⁵ Moreover, ubiquitination of AURKA plays an important role in different cancers. It was reported that ubiquitination and degradation of AURKA play a vital role in colorectal cancer progress. ¹⁶ However, the relationship between the level of AURKA ubiquitination and the occurrence and development of laryngeal cancer has not been reported yet. Using the HitPredict database, we predicted and found a potential interaction between USP21 and AURKA at the protein level. Nevertheless, it is still unclear whether USP21 regulates the occurrence and development of laryngeal carcinoma by interacting with AURKA.

Based on the above background, we speculated that USP21 could inhibit the ubiquitination degradation of AURKA, thereby promoting the occurrence and development of laryngeal cancer. To this end, we collected clinical samples and conducted in vitro cell experiments. Our study may provide a new molecular biomarker for laryngeal cancer.

2. MATERIALS AND METHODS

2.1. Collection of clinical samples

This study collected 50 samples of laryngeal cancer tissues and para‐carcinoma tissues from newly diagnosed patients with laryngeal cancer. The patients were diagnosed using histology and did not undergo preoperative radiotherapy or chemotherapy. According to the clinical stage, the patients were divided into clinical stages I–II (n = 25) and III–IV (n = 25). The correlation between USP21 expression level and clinical characteristics of laryngeal cancer patients were summarized in Table 1. Tissue samples were stored at −80°C immediately after acquisition. We have obtained the written informed consent of subjects and approval of the First Affiliated Hospital of Jiamusi University before the start of the study.

TABLE 1.

Correlation between USP21 expression and clinical characteristics in 50 laryngeal cancer patients

Characteristics	Number (n)	USP21 expression		P value
		High	Low
Age				0.5709
<50	23	10	13
≥50	27	15	12
Sex				0.5672
Female	21	9	12
Male	29	16	13
Differentiation				0.3961
Well	24	10	14
Poorly/moderately	26	15	11
Tumor stage				0.0095*
T1 + T2	23	8	15
T3 + T4	27	20	7
TNM stage				0.0222**
I + II	25	9	16
III + IV	25	18	7
Survival				0.0120**
Yes	36	11	25
No	14	10	4

Note: TNM staging is referring to the 8th AJCC TNM staging criteria.

^* P < 0.01;

^** P < 0.05.

2.2. Cell culture and treatment

The normal human bronchial epithelial cell line 16HBE, laryngeal cancer cell lines (AMC‐HN‐8, TU686, TU177, M4E, and LSC‐1) and 293T cells were purchased from American type culture collection (ATCC, USA). The cells were cultured in an incubator at 37°C and 5% CO₂. The culture conditions were: Dulbecco's modified Eagle's medium (DMEM) (Hyclone, USA) containing 10% fetal bovine serum (Gibco, USA) and 1% penicillin–streptomycin (Invitrogen, USA).

2.3. Plasmid construction and cell transfection

To knock down USP21, sh‐USP21 was transfected into AMC‐HN‐8 and TU686 cells for 48 h. Moreover, wild‐type USP21 (wt‐USP21) and C221A (USP21 catalytic site mutation) were constructed into pcDNA 3.1 vector and then transfected into the 293T cells. To overexpress AURKA, pcDNA 3.1 AURKA was transfected into AMC‐HN‐8 and TU686 cells, respectively. The above transfection sequences were designed and synthesized at GenePharma (China). Cell transfection was performed according to Lipofectamine 2000 (Invitrogen, USA) kit method.

2.4. MTT assay

Cells were digested and counted and then inoculated into 96 well plates (1 × 10⁴ cells/well), 100 μl per well. Exactly 10 μl Thiazolyl blue (MTT) was added and incubated at 37°C with 5% CO₂ for 4 h. Next, the medium containing MTT was sucked out. Exactly 150 μl dimethyl sulfoxide (DMSO) was added to each well. After shaking slowly for 10 min, a bio‐Tek microplate analyzer (MB‐530, Heales, China) determined the absorbance value (490 nm).

2.5. Wound healing assay

Cells were digested and inoculated on 6‐well culture plates (5 × 10⁵ cells/well). Cells were cultured at 37°C in a 5% CO₂ incubator for 24 h until cells were covered with 6‐well plates. Cells were scratched with the tip of a sterile pipette along a horizontal line at the back of 6‐well plates. PBS was utilized to remove scratched cells. Serum‐free DMEM medium was added. Scratches of 0 and 48 h were taken under an inverted biological microscope (DSZ2000X, Beijing Zhongxian Hengye Instrument, China).

2.6. Transwell assay

Cells were digested, and the cell suspension was prepared with the serum‐free basal medium. Cell density was adjusted to about 1 × 10⁶/ml. Exactly 100 μl cell suspension was absorbed and added into the upper of the Transwell chamber (#3412, Corning, USA). A complete medium containing 10% FBS was added into the lower chamber and cultured for 48 h. Cells were wiped on the upper ventricle with a wet cotton swab. Four percentage of paraformaldehyde was fixed for 10 min. Exactly 0.5% crystal violet was stained. Cells on the upper outdoor surface were observed under the microscope (Olympus, Japan).

2.7. Quantitative real‐time PCR (qRT‐PCR)

In simple terms, Trizol (Invitrogen, USA) extracted total RNA. RNA was reverse transcribed into cDNAs by a cDNA reverse transcription kit (#4368814, Invitrogen, USA). SYBR Green PCR Mastermix (SR1110, SolarBio, China) tested relative gene expression on ABI 7900 system. Using GAPDH as an internal reference, gene levels were calculated by the 2^−ΔΔCt method. Primer sequences used in this study are as follows:

• USP21‐F: 5′‐GCAGGATGCCCAAGAGTT‐3′,

• USP21‐R: 5′‐GCAGGGACAGGTCACAAAA‐3′,

• AURKA‐F: 5′‐GGATATCTCAGTGGCGGACG‐3′,

• AURKA‐R: 5′‐GCAATGGAGTGAGACCCTCT‐3′,

• GAPDH‐F: 5′‐GAAGGTGAAGGTCGGAGTC‐3′,

• GAPDH‐R: 5′‐GAAGATGGTGATGGGATTTC‐3′.

2.8. Western blot

RIPA lysis (P0013B, Beyotime, China) extracted total protein. Protein was quantified by BCA protein kit, and protein was adsorbed on PVDF membrane by gel electrophoresis and mixed with SDS‐PAGE loading buffer (P0015L, Beyotime, China). Then protein was sealed with a 5% skim milk solution for 90 min. USP21 (17856‐1‐AP, 1:600, Proteintech, USA), AURKA (ab52973, 1:5000, Abcam, UK), and GAPDH (ab9485, 1:2500, Abcam, UK) were incubated overnight at 4°C. Then, HRP antibodies were incubated. ECL color exposure was performed to detect protein bands.

2.9. Co‐immunoprecipitation (Co‐IP)

For immunoprecipitation experiments, 293T/AMC‐HN‐8/TU686 cells were lysed in BC100 buffer after being transfected with Flag/Myc labeled plasmids for 24 h. Then the cell lysates were incubated with FLAG‐M2 magnetic beads (M8823, Sigma, USA) at 4°C overnight. After elution and dissolution, protein A/G Beads (88803, Thermo Fisher Scientific, USA) and indicated antibodies were added to cell lysates and incubated overnight. After being washed with PBS three times, the beads were re‐suspended in a ×1 SDS‐PAGE loading buffer. Bound proteins were eluted in an SDS sample buffer for Western blot analysis.

2.10. Glutathione‐S‐transferase‐pull down

The fusion protein was prepared according to previous reports. ¹⁷ About 100 μg glutathione‐S‐transferase (GST) (ab89494, Abcam, UK) and the GST‐USP21 fusion protein were fixed in 50 μl glutathione agarose. After equilibration, they were incubated at 4°C with gentle shaking for 60 min. His‐AURKA fusion protein was added to immobilized GST‐USP21 and GST. Next, ~100 μg of the fusion protein was added. Two fusion proteins were incubated at 4°C with gentle shaking overnight. Bound protein was eluted with elution buffer (10 mM glutathione in PBS, pH 8.0) and examined by immunoblotting.

2.11. Analysis of AURKA ubiquitination level

The sh‐NC and sh‐USP21 were transfected into AMC‐HN‐8 and TU686 cells. Cells were lysed by lysis buffer after 48 h. Cell lysates were immunoprecipitated with anti‐AURKA or anti‐ubiquitin antibodies (#3936, CST, USA) at 4°C for 24 h. Then mix the samples with Protein A‐Sepharose beads and incubate at 4°C for 2 h. They were then washed with lysis buffer and sample buffer to obtain target proteins. Proteins were analyzed by Western blot.

2.12. Statistical analysis

Graphpad Prism8.0 software was used for statistical analysis. Measurement data were expressed as mean ± standard deviation. Student's t‐test or one‐way analysis of variance (ANOVA) was performed for two or inter‐group comparisons. Kaplan–Meier analyses were conducted by survival package. The correlation between USP21 expression and clinicopathological characteristics of laryngeal cancer patients was assessed by χ ² test. P < 0.05 indicated that the difference was statistically significant.

3. RESULTS

3.1. USP21 was elevated in human laryngeal cancer

To study the role of USP21 in laryngeal cancer, qRT‐PCR was used to evaluate USP21 expression in clinical samples. It could be seen that USP21 mRNA was notably promoted in laryngeal cancer tissues, and the USP21 level in clinical stages III–IV was higher than that in clinical stages I–II (Figure 1A,B). Western blot further verified that USP21 protein expression was observably increased in laryngeal cancer tissues (Figure 1C). Kaplan–Meier survival curve revealed that a high USP21 level was related to a poor prognosis of laryngeal cancer (Figure 1D). Moreover, TCGA database predicted that USP21 was positively correlated with AURKA in head and neck squamous cell carcinoma tissues (Figure 1E). Finally, we tested the level of USP21 in vitro. Compared with 16HBE cells, USP21 levels were prominently increased in various laryngeal cancer cell lines (Figure 1F,G). Among them, USP21 was most notably promoted in AMC‐HN‐8 and TU686 cells. Therefore, we chose these two cells for subsequent experiments. In short, the high level of USP21 was related to the laryngeal cancer process, and its underlying mechanism might be related to the regulation of AURKA.

FIGURE 1.

USP21 was elevated in human laryngeal cancer. (A) qRT‐PCR determined USP21 level in laryngeal cancer tissues and para‐carcinoma tissues. (B) qRT‐PCR detected the level of USP21 in clinical phases I–II and III–IV tissues. (C) Western blot measured the level of USP21 protein in laryngeal cancer tissues and adjacent tissues. N, normal; T, tumor. (D) Kaplan–Meier survival curve. (E) The correlation of USP21 and AURKA in head and neck squamous cell carcinoma tissues predicted by TCGA database. (F,G) qRT‐PCR and Western blot evaluated the mRNA and protein levels of USP21 in different laryngeal cancer cells. *P < 0.05, **P < 0.01, and ***P < 0.001.

3.2. Knockdown of USP21 repressed laryngeal cancer cell proliferation, migration, and invasion

To explore the role of USP21 in laryngeal cancer cells, we knocked down USP21 in AMC‐HN‐8 and TU686 cells. As shown in Figure 2A,B, USP21 expression in the sh‐USP21 group was prominently reduced. It was illustrated that we knocked down USP21 successfully. MTT results further showed that the down‐regulation of USP21 remarkably inhibited the proliferation of AMC‐HN‐8 and TU686 cells (Figure 2C). Furthermore, knockdown of USP21 inhibited AMC‐HN‐8 and TU686 cell migration and invasion (Figure 2D,E). This indicated that USP21 could promote the progression of laryngeal cancer in vitro.

FIGURE 2.

Knockdown of USP21 repressed laryngeal cancer cell proliferation, migration, and invasion. (A,B) qRT‐PCR and Western blot measured USP21 knockdown effect. (C) MTT detected cell proliferation. (D) Cell migration was evaluated by wound healing assay. (E) Transwell detected cell invasion. *P < 0.05, **P < 0.01 and ***P < 0.001.

3.3. USP21 interacted with AURKA

To study the related proteins binding to USP21, the HitPredict database (http://www.hitpredict.org/) predicted that the USP21 protein and AURKA protein had the interaction potential. Then we constructed a USP21 mutant to 293T cells, in which the highly conserved catalytic sites were mutated. We co‐expressed wt‐USP21/C221A, and Flag/Myc labeled vectors in 293T cells and performed Co‐IP analysis. The results indicated that both wild‐type and conservative site mutant of USP21 in 293T cells interacted with AURKA (Figure 3A,B). Furthermore, Co‐IP detected the interaction between USP21 and AURKA in AMC‐HN‐8 and TU686 cells. The results illuminated that USP21 interacted with AURKA in AMC‐HN‐8 and TU686 cells (Figure 3C,D). Finally, GST‐pull down also confirmed the interaction between USP21 and AURKA (Figure 3E).

FIGURE 3.

USP21 interacted with AURKA. (A,B) Co‐IP verified the interaction between USP21 (wt‐USP21/conserved site mutant C221A) and AURKA in 293T cells. (C,D) Co‐IP verified USP21 and AURKA interaction in AMC‐HN‐8 and TU686 cells. (E) GST‐pull down detected the interaction between USP21 and AURKA.

3.4. USP21 maintained AURKA level through deubiquitination

Then we further confirmed the regulatory mechanism of USP21 in laryngeal cancer. Firstly, we tested the level of AURKA in laryngeal cancer cell lines after knocking down USP21. Figure 4A,B illuminated that knocking down USP21 did not affect AURKA mRNA levels but significantly down‐regulated AURKA protein levels. In addition, both wt‐USP21 and C221A significantly up‐regulated the level of USP21 protein, and wt‐USP21 significantly up‐regulated the level of AURKA protein. At the same time, mutation of the catalytic site C221A could not increase the protein level of AURKA (Figure 4C). From the above, USP21 might be involved in the post‐translational modification regulation of AURKA. To further verify the regulatory mechanism of USP21, the protein synthesis inhibitor cycloheximide (CHX, 25 μg/ml) was used. It could be seen that the protein level of AURKA in the sh‐USP21 group showed a more rapid decline compared with the sh‐NC group after CHX treatment (Figure 4D). This indicated that USP21 silencing reduced the stability of AURKA and increased its degradation. Then we treated the laryngeal cancer cells with proteasome inhibitor MG132 (10 μM) and found that USP21 stabilized AURKA (Figure 4E). Furthermore, AURKA ubiquitination level analysis showed that knocking down USP21 markedly increased the AURKA ubiquitination level, indicating that USP21 mediated the deubiquitination of AURKA (Figure 4F). Collectively, USP21 could inhibit AURKA degradation through ubiquitination.

FIGURE 4.

USP21 maintained AURKA level through deubiquitination. (A,B) qRT‐PCR and Western blot measured AURKA level. (C) Western blot detected USP21 and AURKA protein levels. (D) Western blot tested AURKA protein level after treatment with the protein synthesis inhibitor CHX. (E) AURKA protein expression was examined by Western blot after treating with proteasome inhibitor MG132. (F) Analysis of AURKA ubiquitination level. **P < 0.01 and ***P < 0.001.

3.5. Overexpression of AURKA reversed the effect of knockdown of USP21 on cells proliferation, migration, and invasion

To explore the role of AURKA regulated by USP21 on laryngeal cancer progression, we then overexpressed AURKA in AMC‐HN‐8 and TU686 cells. Figure 5A,B showed that pcDNA 3.1 AURKA memorably up‐regulated AURKA levels. In addition, we co‐transfected sh‐USP21 and pcDNA 3.1 AURKA in AMC‐HN‐8 and TU686 cells. Figure 5C showed that knockdown of USP21 significantly down‐regulated the level of AURKA protein, and overexpression of AURKA reversed the down‐regulation effect of USP21 inhibition. As shown in Figure 5D–F, knockdown of USP21 inhibited cell proliferation, migration, and invasion. However, overexpression of AURKA reversed the inhibitory effects of USP21 inhibition. Taken together, USP21 promoted proliferation, migration, and invasion of laryngeal cancer cells through stabilizing AURKA.

FIGURE 5.

Overexpression of AURKA reversed the effect of knockdown of USP21 on cells proliferation, migration, and invasion. (A,B) qRT‐PCR and Western blot evaluated the effect of AURKA overexpression. (C) Western blot detected AURKA protein level. (D) Cell proliferation was assessed by MTT. (E) Wound healing assay assessed cell migration. (F) Cell invasion was determined by transwell assay. *P < 0.05, **P < 0.01, and ***P < 0.001.

4. DISCUSSION

As the most important type of post‐translational modification, ubiquitination is a multistep enzymatic process that participates in various cellular biological activities. The proteasome is a complex molecular machine degrades proteins after they are bound to ubiquitin. ¹⁸ Ubiquitin can be degraded by proteasome via its conjugated substrate when extended with a C‐terminal tail and as a monomer. ¹⁹ Furthermore, disorders of ubiquitination and deubiquitination can lead to various diseases. ²⁰ Masclef et al. revealed the function of BAP1 deubiquitinase in tumor inhibition and regulation of ubiquitin signaling. ²¹ In pancreatic ductal adenocarcinoma, USP21 deubiquitinated and stabilized TCF/LEF transcription factor TCF7, promoting stem cell viability. ²² Laryngeal cancer is the most usual malignant cancers in head and neck. ²³ Wang et al., demonstrated ubiquitin protein ligase E3 component n‐recognin 5 could be involved in regulating laryngeal cancer cell proliferation and radiotherapy sensitivity via P38/MAPK pathway. ²⁴ However, the mechanism of ubiquitination involved in laryngeal cancer needs to be further explored. In this research, we reported the regulating effect of deubiquitinating enzyme USP21 in laryngeal cancer for the first time.

The USP family is the largest family of cysteine proteases. ²² USP2b, USP3, USP18, USP25, UL36USP, and HAUSP provide anti‐virus functions. USP4, USP13, USP15, and USP17 negatively modulate the antiviral immune response. ²⁵ Notably, USP21 has been involved in several types of cancer. ²⁶ It was reported that the USP21 level was signally up‐regulated in different tumors, such as cervical cancer and hepatocellular carcinoma. ²⁷ , ²⁸ Consistently, we found that USP21 was highly expressed in laryngeal cancer tissues and cells. In terms of molecular mechanism, Xu et al. reported that USP21 promoted non‐small cell lung cancer cell proliferation, migration and invasion by stabilizing a well‐known oncogene Yin Yang‐1 and mediating its deubiquitination. ²⁹ In cervical cancer, USP21 regulated Hippo signaling pathway to promote radioresistance through deubiquitination of FOXM1. ²⁷ In the present, we found knockdown of USP21 reduced laryngeal cancer cell proliferation, migration, and invasion capabilities.

AURKA is a serine/threonine kinase that plays a central role in mitosis. ³⁰ AURKA expression was signally elevated in different cancer types, pointing out the role of AURKA as a potential oncogene in tumorigenesis. ³¹ AURKA can regulate tumorigenesis through different pathways. For example, AURKA boosted esophageal squamous cell carcinoma development via EGFR/PI3K/Akt pathway. ³² Furthermore, inhibition of AURKA reduced the proliferation and survival of KRAS‐activated gastrointestinal cancer cells by preventing the activation of ribosomal protein S6 kinase B1. ³³ In glioblastoma, inhibition of AURKA reversed the Warburg effect, triggering the unique metabolic vulnerability of glioblastoma. ³⁴ AURKA ubiquitination plays an important role in cancer. For instance, it was found that the tumorigenic feature of esophageal squamous cell carcinoma was in part mediated by USP3‐facilitated deubiquitination of AURKA. ³⁵ Kim et al. showed that ovarian tumor deubiquitinase 6A was an AURKA‐specific deubiquitinase, which might serve as a therapeutic target in human cancers. ³⁶ However, the level of AURKA ubiquitination in laryngeal cancer has not been explored. Here, we confirmed the interaction between USP21 and AURKA protein through Co‐IP and GST‐pull down. We firstly reported that USP21 affected laryngeal cancer progression by regulating the AURKA protein level.

In conclusion, this study proved that USP21 stabilized AURKA through deubiquitination to aggravate the proliferation, migration, and invasion of laryngeal cancer cells. However, our research still has some shortcomings. We only conducted in vitro studies on the mechanism of USP21 stabilizing AURKA in laryngeal cancer, lacking animal experimental results for further proof. Additionally, the downstream signaling pathways need to be further explored. Our research hopes to provide a novel therapeutic strategy for laryngeal cancer therapy.

CONFLICT OF INTEREST

All authors declare no conflict of interest.

Wang Q‐D, Shi T, Xu Y, Liu Y, Zhang M‐J. USP21 contributes to the aggressiveness of laryngeal cancer cells by deubiquitinating and stabilizing AURKA . Kaohsiung J Med Sci. 2023;39(4):354–363. 10.1002/kjm2.12649