Background
Acute myeloid leukemia (AML) is a genetically and clinically heterogeneous hematologic malignancy [
1] with a 3.7/100 000 incidence per year [
2]. Nearly 80% of adult acute leukemia are AML [
3], and the 5-year overall survival rate for AML patients older than 60 years is about 25% [
4]. The application of various first-line chemotherapeutic agents, molecular-targeted agents, immunotherapeutic agents and hematopoietic stem cell transplantation (HSCT) has dramatically improved the treatment outcomes of AML patients [
5,
6]. However, the relapse rate of AML patients is 80% [
7], and the mortality rate has been over 50% [
8]. Stratified diagnosis and treatment of AML have become the focus of research in recent years. The classification of AML is based on cytogenetic and mutational profiles. Moreover, some additional pre-disposing features are considered as prognostic factors, which include therapy-related, prior myelodysplastic syndrome (MDS) or MDS/myeloproliferative neoplasm [MPN], and germline predisposition [
9]. Also, the response to initial therapy and assessment of early minimal residual disease (MRD) is also crucial in risk classification[
9]. Based on the existing risk-stratification system, patients in different risk strata accept the corresponding treatment. Even so, there is still a chance of drug resistance and relapse in low- and intermediate-risk patients.
Karyotyping is crucial in the risk stratification of AML. It was reported that 40–50% AML patients have a normal karyotype (NK) [
10]. Single genes, such as
CEBPA [
11],
FLT3-ITD [
12,
13] and
NPM1 [
13], could provide references for predicting the prognosis of NK-AML patients. It used to be considered that patients with NK-AML had a medium prognosis. However, it was found that NK-AML patients with high-risk mutated genes or aberrantly expressed genes had poor prognosis [
14,
15]. This situation suggested that the risk stratification of NK-AML patients required further improvements. Gene expression-based scoring systems have potentials in predicting the prognostic value in AML [
16]. Therefore, novel prognostic biomarkers are highly anticipated to improve the risk stratification for NK-AML.
Based on the ImmuCo database and clinical specimens, we screened seven genes associated with acute leukemia as candidate prognosis biomarkers. In the follow-up verification, we found that the expression level of Small Integral Membrane Protein 3 (
SMIM3, also called Nid67
) in AML was significantly higher than that in normal controls. SMIM families contain multiple members. Small integral membrane protein 1 (
SMIM1), a tail-anchored transmembrane protein [
17], is associated with Vel-negative blood type [
18,
19].
SMIM1 also could influence red blood cell traits [
20]. A study revealed that
SMIM4 is a respiratory chain assembly factor [
21].
SMIM20 is expressed in the adult brain, and may function in fertility and reproduction [
22]. A study suggested that
SMIM20 could be a new target of endometriosis [
23]. Sha Liu et al. found that
Lnc-SMIM20-1 upregulation is associated with poor prognosis in AML [
24]. In addition,
SMIM30 could promote the progression of hepatocellular carcinoma [
25,
26].
The
SMIM3 gene is located on chromosome 5q33.1, coding a single-pass transmembrane protein consisting of 60 amino acids with a total molecular weight of 6593 Da. It is expressed in various tissues, with the highest expression in heart, ovarian and adrenal glands, and the lowest expression in skeletal muscle and cerebellum [
27]. It may play a role in cell channel regulation and be associated with neuronal differentiation [
27]. There are few studies on
SMIM3, and certainly not have been studied in AML. Thus, large gaps remain to be filled in our understanding of the function and mechanism of
SMIM3 in AML
. Currently,
SMIM3-related hematological disease is 5q- syndrome of MDS. A study showed that a gene or genes in the Cd74 to Nid67 interval might be associated with MDS [
28]. However, the mechanism has not been studied in details. In addition,
SMIM3 can be used as a sensitive and specific biomarker of radiation exposure in the radiation emergency department, and patients with expression of
SMIM3 had a poor prognosis [
29]. Weining Wang et al. found that eleven genes including
SMIM3 in NCCS (n = 36) and TCGA (n = 40) databases were associated with poor overall survival rate in oral squamous cell carcinoma patients without nodal metastases [
30]. However, more evidence is needed to prove whether
SMIM3 is a prognostic biomarker in oral squamous cell carcinoma.
In this study, we examined the expression and prognostic value of SMIM3 in AML. We further demonstrated the effect of SMIM3 on cellular and biological behavior both in vitro and in vivo. Meanwhile, we investigated the mechanism of SMIM3 regulation. We also correlated the SMIM3 expression with targeted therapy responsiveness.
Methods
Subjects
The 236 bone marrow samples from newly diagnosed AML patients and 23 samples from healthy donors were enrolled at the First Affiliated Hospital of Zhengzhou University between February 2017 and March 2022. The exclusion criteria were patients who didn’t treat or treat elsewhere and patients with acute promyelocytic leukemia. The inclusion criteria was patients who accepted at least one course of treatment. We referred to this cohort as the ZZU cohort. Clinical information acquired from patient medical records mainly included gender, age, white blood cell count (WBC), hemoglobin (HGB), platelet (PLT), peripheral blasts (PB), bone marrow (BM) blasts at diagnosis, fusion gene, gene mutations and chromosomal karyotype, risk stratification, treatment regimens, transplant, and survival status. The induction therapy contained IA and DA regimens: standard-dose cytarabine (Ara-C) 100–200 mg·m
− 2·d
− 1 × 7 d combined with idarubicin 10–12 mg·m
− 2·d
− 1 × 3d or daunorubicin (DNR) 60 mg·m
− 2·d
− 1 × 3d. After remission, patients accepted high-dose Ara-C 3 g/m
2, every 12 h × 3d. Patients without HSCT accepted four courses. Patients with HSCT accepted two courses, and then they accept HSCT. Patients over 60 years old or patients who cannot tolerate intensive chemotherapy accepted chemotherapy with demethylating drugs ± CAG ± venetoclax regimen until progression. Subjects were followed up until death, loss to follow-up or March 2022. The diagnosis of AML, complete remission (CR), relapse, risk stratification and overall survival (OS) were defined according to NCCN guideline for acute myeloid leukemia Version 2.2021[
33]. The study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University and informed consent was obtained according to the Declaration of Helsinki.
Next-generation sequencing
Next-Generation sequencing was applied to assess the mutational hotspots of genes. Based on an Illumina MiSeq System (Illumina, San Diego, CA) high-throughput sequencing platform, a Rightongene AML/MDS/MPN Sequencing Panel (Rightongene, Shanghai, China) was applied to finish the detection. Details of the variant calling, filtering, and annotation are shown in the published reports [
34].
Cytogenetics and fusion genes analysis
Based on the International System for Human Cytogenetic Nomenclature, chromosomal banding analyses were performed by G-banding techniques. Real-time quantitative polymerase chain reaction (RT-qPCR) was performed to detect fusion genes with Multiplex RT-qPCR Fusion Gene Kits (Rightongene, Shanghai, China).
Cell lines and reagents
The human AML cell line Kasumi-1 was purchased from American Type Culture Collection (Manassas, VA, USA). The human AML cell line THP-1 was purchased from the cell bank of the Chinese Academy of Sciences (Shanghai, China). The cell lines were cultured in 90% Roswell Park Memorial Institute (RPMI) 1640 supplemented with 10% Fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S) (all from Gibco, Billings, MT, USA). The culture conditions were 37 °C, 5% CO2, and 95% humidity. SC79 (Beyotime, Shanghai, China) was used as an AKT activator.
Lentiviral transduction
Lentiviral shRNA transduction was performed in Kasumi-1 and THP-1 cells with human SMIM3 shRNA lentiviral particles (Genechem, Shanghai, China) or empty control lentiviral particles (Genechem, Shanghai, China). All infections were done at a multiplicity of infection (MOI) of 100. At 12 h post-transfection, media containing lentiviral particles was replaced with fresh complete medium. Stably transfected Kasumi-1 and THP-1 cells were selected with 2 mg/ml puromycin dihydrochloride (Genechem, Shanghai, China) at 72 h post-infection. Stable SMIM3-knockdown cells and control cells were acquired 4 weeks after antibiotic selection. The expression level of SMIM3 was confirmed by RT-qPCR and the immune fluorescence (IF) technique.
RNA extraction and RT-qPCR
Bone marrow samples were collected into Ethylene Diamine Tetraacetic Acid (EDTA) anticoagulant–containing tubes. Mononuclear cells were obtained via density gradient centrifugation. Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). The cDNA was synthesized using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) [
35].
SMIM3 transcript levels were detected by the Taqman method using RT-qPCR as previously described [
35]. Serial dilutions of plasmids expressing
SMIM3 and
ABL1 (Genechem, Shanghai, China) were amplified to construct standard quantification curves. The
SMIM3 and
ABL1 copy numbers were calculated from standard curves using Ct values. The
SMIM3 transcript level was calculated as the ratio of the
SMIM3 copy number/
ABL1 copy number as previously described [
36].The primers and probe sequences of
SMIM3 and
ABL1 are shown in Additional file
1: Table S1.
Immune fluorescence analysis
The cells in T25 flasks were collected and washed several times with phosphate-buffered saline (PBS). Then the cells were fixed using 4% paraformaldehyde (PFA) for 20 min at room temperature (RT). Further, Triton-X-100 (Beyotime, Shanghai, China) was applied to permeabilized cells for 10 min, and nonspecific binding was blocked with 5% BSA (Solaibao Biotechnology, Beijing, China) for 30 min at RT. Followed by washing, cells were incubated overnight at 4 °C with diluted (1:100) primary anti-SMIM3 Polyclonal Antibody (Thermo Fisher Scientific, Waltham, MA, USA). The cells were then incubated with diluted (1:200) secondary antibody Cy3 conjugated Goat Anti-Rabbit IgG (H + L) (Servicebio, Wuhan, China) for 1 h at RT, followed by washing in PBS and staining with DAPI (Solaibao Biotechnology, Beijing, China). Analysis was conducted under a confocal laser scan microscope (Zeiss, Oberkochen, Germany).
Western blot analyses
RIPA lysis buffer (Beyotime, Shanghai, China) supplemented with Protein phosphatase inhibitor (Biomed, Beijing, China) and phenylmethylsulfonylfluoride (PMSF, Biomed) was used for protein extraction. Lysates were run on 10% polyacrylamide gel electrophoresis (PAGE) gels, and protein bands were transferred to 0.45 μm polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA, USA), then blocked with 5% skim milk at room temperature for 2 h. The membrane was incubated with primary antibodies (GAPDH, cleaved-PARP, cleaved caspase3, p27 Kip1, Cyclin D1, CDK4, p-AKT, AKT, PI3K, Cell Signaling Technology [CST], MA, USA, 1:1000; p-PI3K, Affinity Biosciences LTD, Jiangsu, China, 1:1000) overnight at 4 °C and probed with secondary antibodies (goat anti-rabbit IgG horseradish peroxidase (HRP), Zhongshan Golden Bridge Biotechnology, Beijing, China, 1:2000) at room temperature for 1 h. The immunoreactive bands were detected using Super ECL Prime (US EVERBRIGHT, Suzhou, China) according to the manufacturer’s protocol.
Cell proliferation
Cell proliferation was measured through Cell Counting Kit-8 (CCK8, Dojin Laboratories, Kumamoto, Japan) assay. CCK8 assay was carried out according to the standard protocol by seeding cells in a 96-well plate at a density of 1 × 105 cells/well. Then 10 μl of the kit reagent was added into each well after 0, 24, 48, 72, 96 h. After incubation for 3 h, the absorbance was measured at 450 nm spectrophotometrically. The experiments were performed in triplicate.
To analyze the colony formation, we seeded cells in 35 mm dishes at 6 × 103 cells per well in methylcellulose-based MethoCult medium (STEMCELLTM TECHNOLOGIES, Vancouver, British Columbia, Canada). The surviving colonies (≥ 30 cells per colony) were counted under an inverted microscope after 10 days of growing in a humid incubator. All experiments were performed 3 times independently.
Cell cycle and apoptosis analyses
The cells were inoculated in six-well plates with a density of 1 × 105 cells per ml. Then they were synchronized via serum starvation (grown in RPMI 1640 without FBS). After 24 h, the medium was replaced with complete medium for an additional 72 h. The Cell Cycle Staining Kit (Lianke Biotechnology, Hangzhou, China.) was used for cell cycle analyses and the Annexin V-APC/PI Apoptosis Kit (US EVERBRIGHT) was applied for the apoptosis assay. The cell cycle and apoptosis were examined by BD FACSCelesta™ flow cytometry (BD Biosciences, California, USA).
Xenograft tumor mouse model
Xenograft model experiments were conducted using 6-week-old male BALB/c nude mice (Beijing HFK Bioscience Co., Ltd.; Beijing, China). All mice were divided into three groups (CTRL, KD1 and KD2), each group consisted of 3 mice. After 2 days of cyclophosphamide intraperitoneal injection (100 mg/kg/d × 2d), the transfected cells were syringed in the mice’s right flank. Tumor volume was calculated every other day for 14 or 18 days using the following formula: volume (mm3) = (L*I2)/2, where L and I are the lengthiest and shortest diameters, respectively. Subsequently, all mice were euthanized and the xenograft tumors were harvested, weighed, and photographed. This study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University.
H&E staining
The tissue was fixed in 10% formaldehyde. Firstly, the paraffin sections were deparaffinized. Then the sections were stained with Hematoxylin and Eosin staining solutions. After that, the sections were dehydrated with gradient ethanol, transparentized using xylene. Finally, neutral resin was applied to seal the sections.
Immunohistochemical assays
Tissue was also fixed in 10% formaldehyde. According to the standard protocol, the paraffin sections were deparaffinized. After repairing antigen and blocking endogenous peroxidase act, tissue sections were blocked in 3% BSA. Tissue sections were incubated with primary antibodies (Ki67, cleaved caspase3, cleaved PARP1, Servicebio, Wuhan, China) at 4 °C overnight, followed by conjugated secondary antibodies (Servicebio, Wuhan, China) and diaminobenzidine (DAB, Servicebio, Wuhan, China). Then nuclei counterstaining was performed with Mayers hematoxylin (Servicebio, Wuhan, China). Finally, the tissue sections were dehydrated and sealed.
Statistical analyses
Pearson Chi-square test or Fisher exact analysis was applied for categorical data.
Student’s t-test or
Mann–Whitney U-test was applied for continuous variables. Survival was estimated using the
Kaplan–Meier method and
Log-Rank test. In the TCGA-LAML, GEO 12,417 and ZZU cohort, patients were classified into the high expression group and the low expression group according to cutoff value of
SMIM3. A
Cox proportional hazard regression model was used to determine associations between
SMIM3 transcript levels and OS. Variables with
P < 0.2 in the single variable analysis were included in the model.
P <
0.05 (two-sided) was considered significant (*,
P <
0.05; **,
P < 0.01; ***,
P < 0.001). The receiver operating characteristic (ROC) curve was generated using the R package “timeROC” [
37] and “survival” (
https://CRAN.R-project.org/package=survival) to evaluate the diagnostic value. The hazard ratio (
HR) and corresponding 95% confidence interval (
CI) were also calculated. Data analysis was performed with Graphpad Prism
™ 8.01 (San Diego, California, USA) and R (version 4.1.1, Auckland, NZ, United States,
http://www.r-project.org/).
Discussion
This study investigated the prognostic value and molecular mechanism of SMIM3 in adult AML. Based on the bioinformatics analyses and our clinical data, we found that adult AML patients showed a significant increase in the expression level of SMIM3 compared with normal controls. In NK-AML, the high SMIM3 expression was independently associated with a poor prognosis. To highlight the function of SMIM3 in AML, we further explored the critical effects of SMIM3 on cell behaviors through both in vivo and in vitro experiments. Our results suggested that the knockdown of SMIM3 could inhibit cell proliferation and colony formation, arrest cell cycle progression and promote apoptosis. The effect on cell proliferation may occur through downregulating PI3K-AKT signaling pathway.
We performed a pan‐cancer expression analysis of SMIM3 and showed that SMIM3 was highly upregulated in 8 cancers and commonly downregulated in 16 cancers. This suggests that SMIM3 has complex regulatory roles, and can act as either a potential oncogene or a tumor suppressor gene in different cancer types. The high expression of SMIM3 in AML was also further validated in GEO database, ZZU cohort and cell lines.
Current studies found multiple connections between
SMIM3 and various diseases, including pheochromocytoma [
27], 5q- syndrome of MDS [
28] and radiation exposure [
29]. There are few studies on the biological functions of
SMIM3 in different cancers. A study revealed that
SMIM3 may be associated with poor prognosis in oral squamous cell carcinoma [
30], which needs to be further verified. Interestingly, membrane proteins with a similar structure to
SMIM3 (such as the minK family [
39], the γ subunit of the Na, K-ATPase [
40,
41] and phospholamban [
42]) are thought to play a regulatory role in ion channel subunits, suggesting that
SMIM3 may have a similar role [
28]. However, there has not been any in-depth study in the function of
SMIM3 yet. Until now, the upstream and downstream mechanisms of
SMIM3 expression remain unclear. Based on our current results, we would like to further investigate molecular mechanisms in future studies. What’s more, no SMIM family gene has been studied in AML. But several genes do play a role in other cancers. This suggests that SMIM family may play a role in cancers. Further studies should be carried out to define the function of these genes in AML.
The wide application of the existing risk stratification diagnosis and treatment has improved the prognosis of AML patients. Even though the risk stratification system was associated with adult AML prognosis in our cohort, no significant survival difference was found in NK-AML. This indicated the need to improve the risk stratification for NK-AML patients. Our study revealed that high expression of SMIM3 was associated with worse OS in NK-AML. This result was confirmed in several databases. Moreover, multivariate survival analysis indicated that high SMIM3 expression and transplant were independent prognostic factors for unfavorable OS in ZZU NK-AML cohort. NK-AML patients without transplant had a poor prognosis in H- SMIM3 group, but there was no significant difference in the allo‐HSCT subgroup. This result suggested that allo‐HSCT may be an effective way to overcome the adverse impact of SMIM3. Collectively, these results suggested that SMIM3 might be a novel prognostic marker for NK-AML patients.
The abnormally high expression and poor prognosis suggested the biological functions of
SMIM3 in AML. Our study provided further evidence that
SMIM3 affected the proliferation of AML cells through apoptosis and cell cycle regulation both in vitro and in vivo. The results of IF revealed that
SMIM3 was mainly localized in the nucleus. The knockdown of
SMIM3 caused G0/G1 cell cycle arrest via the p27/Cyclin D1-CDK4 pathway. The induction of p27, a cyclin dependent kinase inhibitor, caused cell cycle progression at the G0/G1 phase [
43‐
46]. Also, G0/G1 phase is regulated by CDK4 and Cyclin D1 [
47‐
49]. The upregulation of p27 could inhibit the activity of Cyclin D1-CDK4 [
45]. To further understand the mechanism of the biological functions, we studied the changes in critical signaling pathways related to proliferation, apoptosis and metabolism based on KEGG analysis in AML cell lines. The PI3K-AKT signaling pathway plays a central role in metabolism. It regulates crucial functions including proliferation, differentiation, and survival [
50]. The activation of this pathway was suggested to be associated with adverse prognosis [
51]. Our results showed that the phosphorylated PI3K-AKT was reduced in the
SMIM3-KD cells. These suggested that
SMIM3 could modulate the growth and survival of AML cells by regulating the PI3K-AKT signaling pathway. Based on a previous database, we found that the H-
SMIM3 group was more sensitive to PI3K-AKT-targeted drugs. These were consistent with KEGG, in vitro and in vivo results. Additionally, H-
SMIM3 group was also more sensitive to various first-line and novel drugs, including Imatinib, Selinexor, Sorafenib and Bortezomib. These results provided the basis for the application of targeted drugs, which could reduce the chance of relapses and drug resistance. In addition, GO and KEGG pathway enrichment analyses also found that cation channel complex, ion channel complex and potassium channel complex were associated with high
SMIM3.
There are still some drawbacks in our study. Firstly, our study had the inherent limitations of any retrospective study. Secondly, our results need to be further verified in multicenter large sample prospective cohort studies.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.