A novel cancer-germline gene DAZL promotes progression and cisplatin resistance of non-small cell lung cancer by upregulating JAK2 and MCM8

Background Germline-specic genes are usually activated in cancer cells and drive cancer progression; such genes are called cancer-germline or cancer-testis genes. Deleted in azoospermia-like (DAZL) is a germline-specic gene whose role in gametogenesis is well characterized, but whose expression and function in non-small cell lung cancer (NSCLC) is unknown. Methods RNA-seq data from Genotype Tissue Expression and The Cancer Genome Atlas were used to analyze DAZL mRNA levels. An immunochemical assay was performed to detect DAZL protein levels in a lung tissue microarray. Cell viability, EdU staining, colony formation, wound healing, and invasion assays were performed for in vitro functional analysis. The in vivo effect of DAZL on tumor growth was investigated using a subcutaneous murine xenograft model. Quantitative proteomics was performed to explore the potential targets of DAZL.


Introduction
Lung cancer is the second most commonly diagnosed cancer and the leading cause of cancer death worldwide, with an estimated 2.2 million new cancer cases and 1.8 million deaths in 2020 [1].Non-small cell lung cancer (NSCLC) accounts for approximately 85% of lung cancer cases and includes three subtypes: lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), and large cell lung carcinoma [2].Surgery is the rst choice for the treatment of early-stage NSCLC (stage I and II) [3], while adjuvant chemotherapy has become the standard treatment for patients with advanced NSCLC [4].Platinum-based drugs (such as cisplatin, carboplatin, and oxaliplatin) form the backbone of therapeutic agents [5].Recently, targeted therapy has achieved improvements in the treatment of NSCLC [6].However, only a subset of patients can bene t from targeted therapy, and drug resistance still exists.Therefore, a deeper understanding of the molecular pathology of NSCLC is necessary to nd new therapeutic targets and improve the e cacy of NSCLC treatment.
Cancer cells usually express some germline genes known as cancer/germline (CG) or cancer/testis genes [7].Under normal physiological conditions, CG genes are predominantly expressed in germ cells of adult testes and fetal ovaries, but they are aberrantly expressed in various cancers [8].There is increasing evidence that CG genes are involved in the development and progression of lung cancer and represent potential biomarkers and targets for lung cancer diagnosis, prognosis, and therapy [9].DAZL (deleted in azoospermia-like) has been shown to be exclusively expressed in germ cells of ovaries and testes and plays an important role in gametogenesis [10].DAZL was also found to be overexpressed in glioblastomas and to play an oncogenic role [11].However, its role in other cancers and the underlying mechanism have not yet been investigated.

Data acquisition and bioinformatic analysis
RNA-seq data of 53 normal tissues from 570 healthy donors were downloaded from the Genotype Tissue Expression (GTEx) project (http://commonfund.nih.gov/GTEx) and used to compare DAZL expression between testicular and non-testicular tissues.The mRNA pro ling data and clinical information of NSCLC patients were downloaded from The Cancer Genome Atlas (TCGA) (https://www.cancer.gov/tcga)and used to analyze the differential expression between cancer and adjacent noncancerous tissues and correlation between DAZL expression and clinicopathological factors.The above analyzes were performed using R software (version 4.1.2)(https://www.r-project.org).

Cell lines, antibodies, and chemicals
All cell lines were purchased from ATCC (Manassas, VA, USA) and authenticated using short tandem repeat analysis by Yubo Biological Technology (Shanghai, China).All antibodies used in this study were purchased from ABclonal Technology (Wuhan, China).Fedratinib (TG101348) was purchased from Selleck (Houston, TX, USA).

Plasmid construction and transfection
The coding sequences of the longer and shorter transcriptional variants of the human DAZL gene were cloned into the pCMV-Myc vector (Clontech, Mountain View, CA, USA), resulting in the expression plasmids Myc-DAZL-1 and Myc-DAZL-2, respectively.Overlapping PCRs were performed to generate expression constructs of mutant DAZL1 and DAZL2 (Arg to Gly substitution at residue 115 of DAZL1 isoform and at residue 135 of DAZL2 isoform, respectively): Myc-DAZLR115G and Myc-DAZLR135G.All the primers used in the study are listed in Table S1.Transfection was performed with Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA).

Tissue array (TMA) and immunohistochemistry (IHC)
This retrospective study complied with the Declaration of Helsinki and relevant ethical guidelines in China and was approved by the Ethics Committee of Hunan Normal University.NSCLC TMA (catalog number BC041115c) was purchased from Alenabio (Xian, China).DAZL protein expression in the TMA was detected using a streptavidin-biotin-peroxidase IHC kit (Maxim Biotech, Fuzhou, China).Slides were digitally scanned using the Aperio ScanScope XT (Aperio Technologies, Inc., Vista, CA, USA).Depending on the number of positive cells, IHC scores were classi ed as 0 (no staining), 1 (less than 25% positive cells), 2 (more than 25% and less than 50% positive cells), and 3 (more than 50% positive cells).

Identi cation of transcript variants of DAZL
Total RNA was extracted from cell lines and formalin-xed and para n-embedded testicular tissue (Alenabio) using TRIzol reagent and a PureLink FFPE RNA kit (Invitrogen, Carlsbad, CA, USA), respectively, and then reverse-transcribed into cDNA using M-MLV RTase and random primers (GeneCopoeia, Guangzhou, China).These cDNAs were ampli ed by PCR with a variant-speci c primer pair or an ACTB primer pair (positive control).The forward primers speci c to DAZL1 and DAZL2 are located in exon 1 of DAZL1 and DAZL2, respectively; the reverse primer is located in exon 6, which is common to both variants.

In vitro functional assays
MTT assays, colony formation assays, wound healing assays, and transwell invasion assays were performed as described previously [12].

Subcutaneous murine xenograft model
Male BALB/c nude mice (4-6 weeks old) were purchased from SLACCAS Jingda (Changsha, China).A total of 5 × 10 7 Bease-2B cells stably overexpressing DAZL-1 or empty vector were injected into the dorsal ank of the nude mice.At 56 days after injection, the mice were sacri ced to obtain tumor samples.All animal experiments were performed in accordance with ARRIVE guidelines 2.0 and the National Research Council Guide for the Care and Use of Laboratory Animals.
2.10 Quantitative proteomics analysis based on tandem mass tag (TMT) A549 cells stably expressing scramble shRNA or shDAZL were lysed, digested, labeled, and subjected to LC-MS/MS analysis as previously described [14].Differentially expressed proteins (fold change ≥ 2 and p value < 0.05) were subjected to enrichment analysis of Gene Ontology (GO) terms using the Database for Annotation, Visualization and Integrated Discovery (DAVID) (https://david.ncifcrf.gov/).

RNA binding protein immunoprecipitation (RIP) assay
The RIP assay was performed using the RIP kit (Guangzhou BersinBio Co. Ltd.) as described previously [15].RNA precipitated with IgG or anti-DAZL antibodies was reverse-transcribed into cDNA and ampli ed by PCR.PCR products were separated by electrophoresis on a 1.5% agarose gel and visualized with ethidium bromide.

Western blotting
Protein extraction and Western blotting were carried out as previously described [15].

Statistical analysis
R software (version 4.1.2) was used for statistical analysis.The Wilcoxon test or Student's t test were used to compare the difference in expression between two groups.The log-rank test was used to compare two survival curves.The value is shown as the mean ± SD.A p value < 0.05 was considered statistically signi cant.The level of statistical signi cance is expressed as the p value; *, p value ≤ 0.05 and **, p value ≤ 0.01.

DAZL expression is upregulated in NSCLC
Consistent with a previous report [10], mRNA expression of DAZL was 1611-fold (log2-fold change = 10.56) higher in testes than in non-testicular tissues based on the GTEx database (Fig. 1A).Analysis of the TCGA database showed that DAZL mRNA levels were signi cantly upregulated in LUAD and LUSC (Fig. 1B, C).DAZL mRNA levels were signi cantly associated with tumor grade and tumor stage in LUAD but not in LUSC (Table S2).IHC assays showed that DAZL protein was undetectable in normal lung tissue but overexpressed in lung cancer tissue (Table S3, Fig. 1D) and that DAZL protein expression was positively associated with tumor grade in both LUAD and LUSC (Table S4).Among the lung cell lines, the normal lung epithelial cell line Beas-2G had the lowest DAZL protein levels (Fig. 1E, F).

A longer transcript variant of the DAZL gene is overexpressed in NSCLC cells and promotes tumor growth
DAZL has two transcript variants: DAZL-1 (GenBank accession number: NM _001190811) and DAZL-2 (GenBank accession number: NM _001351) in the GenBank database.A search of the human genome data using the Basic Local Alignment Search Tool (BLAST) revealed that DAZL-1 and DAZL-2 are transcribed from different promoters of the DAZL gene and have a unique exon 1 but ten exons in common (Fig. 2A).The DAZL-1 transcript encodes a longer isoform of 315 amino acids, whereas the shorter isoform encoded by the DAZL-2 transcript contains 295 amino acids, lacking the N-terminal 20 amino acids of the DAZL-1 isoform.To con rm the expression of the two variants, we designed primer pairs for speci c detection of each variant and performed RT-PCR.The results showed that DAZL-1 and DAZL-2 were expressed in A549 cells and testes, respectively, but there was no DAZL-1 or DAZL-2 expression in the normal lung cell line BEAS-2B (Fig. 2B).Therefore, only DAZL-1 was examined in the following experiments (unless otherwise indicated, DAZL-1 will be referred to as DAZL in the following sections).Xenograft experiments in mice showed that the mice bearing Beas-2B-DAZL cells had larger and heavier tumors than control mice bearing Beas-2B-vector cells (Fig. 2C, D), suggesting that DAZL promotes tumor growth.

Proteomics analysis screens the targets of DAZL in NSCLC cells
DAZL has been shown to regulate mRNA translation [16].To explore the targets of DAZL, TMT-based quantitative proteomic analysis was performed to analyze the differential expression of proteins between A549-shDAZL and A549-shNC cells.A total of 89 proteins were identi ed as signi cantly differentially expressed (|log2 (shDAZL/shNC) | ≥ 1 and p value < 0.05), 46 proteins were downregulated in A549-shDAZL cells, and 43 proteins were upregulated in A549-shDAZL cells compared with A549-shNC cells (Fig. 4A, Table S5).Western blot experiments were then performed to con rm the results of proteomics analysis.Consistent with the results of proteomics analysis, knockdown of DAZL decreased the protein levels of EGFR, FGF2, AKT3, EPHA2, ITGB4, and MET (Fig. 4B).Enrichment analysis of the differentially expressed genes revealed that the upregulated genes were signi cantly enriched in cytoskeleton organization, whereas the downregulated genes were signi cantly enriched in signal transduction, inner ear receptor cell development, regulation of JAK-STAT cascade, cell cycle, negative regulation of neuron apoptotic process and sensory perception of sound (Fig. 4C).

DAZL promotes the translation of JAK2 and MCM8
We selected two genes (JAK2 and MCM8) for further investigation.JAK2 and MCM8 belong to the JAK-STAT and cell cycle pathways, respectively.The above proteomics analysis results showed that they were signi cantly downregulated in A549-shDAZL cells (Fig. 4A).Western blotting showed that knockdown of DAZL in A549 and H1437 cells decreased, but overexpression of DAZL in Beas-2B and H446 cells increased, the expression of JAK2 and MCM8 (Fig. 4D, E).To investigate whether the regulatory role of DAZL on JAK2 and MCM8 depends on its RNA-binding activity, we constructed DAZL1-R135G and DAZL2-R115G mutants.It has been shown that the R115G mutant of DAZL-2 can impair RNA-binding activity [17].Transfection of wild-type DAZL or DAZL2 signi cantly increased the levels of JAK2 and MCM2, but transfection of mutant DAZL1 or DAZL2 had no signi cant effects in Hek293 cells (Fig. 4F, G).In addition, RIP assays showed that DAZL could bind JAK2 and MCM8 mRNA in A549 and H1437 cells (Fig. 4H).These results suggest that DAZL promotes the translation of JAK2 and MCM8 by binding to their mRNA.

DAZL promotes cell viability, colony formation, migration, and invasion of NSCLC cells by upregulating JAK2
The above results showed that DAZL increased JAK2 expression.JAK2/Stat3 signaling is frequently activated in solid tumors and plays a crucial role in oncogenesis [18].Therefore, the present study investigated whether DAZL plays an oncogenic role via JAK2.We overexpressed DAZL or an empty vector in Beas-2B and H446 cells and then treated them with the JAK2 inhibitor fedratinib.Western blotting showed that overexpression of DAZL increased, but treatment with fedratinib decreased, the levels of phosphorylated JAK2 and phosphorylated STAT3 (Fig. 5A).In vitro functional assays demonstrated that overexpression of DAZL increased the viability, colony formation, migration, and invasion of NSCLC cells, but treatment with fedratinib decreased these parameters (Fig. 5B-H).These results suggest that DAZL plays an oncogenic role by upregulating JAK2.

DAZL decreases cisplatin sensitivity by upregulating MCM8
It has been shown that the MCM8/9 complex is able to reduce sensitivity to DNA-damaging agents (such as cisplatin, hydroxyurea, olaparib) by maintaining replication fork integrity upon DNA damage [19][20][21].To investigate whether DAZL exerts its effect by upregulating MCM8.We used the lentivirus-mediated system to overexpress DAZL and/or silence the expression of MCM8 in Beas-2B and H446 cells (Fig. 6A).DAZL overexpression increased, but MCM8 knockdown decreased, cell viability.Moreover, MCM8 knockdown counteracted the effects of DAZL overexpression (Fig. 6B).DNA synthesis was detected based on the incorporation of EdU.As shown in Fig. 6C-D, DDP treatment decreased, but DAZL increased, DNA synthesis; MCM8 knockdown suppressed DAZL-induced DNA synthesis.Accordingly, MCM8 knockdown decreased, whereas DAZL overexpression increased, the IC50 value of DDP; MCM8 knockdown may counteract the effects of DAZL on cisplatin sensitivity (Fig. 6E-F).

Discussion
The role of DAZL in gametogenesis is well characterized, but its expression and function in tumorigenesis have not been fully elucidated.Analysis of the public transcriptome and immunohistochemical experiments showed that DAZL was exclusively expressed in the testis among normal human tissues but was ectopically expressed in NSCLC tissues.DAZL had two transcript variants.The longer and shorter transcripts were expressed in cancer cells and testicular cells, respectively.Overexpression of the longer transcript promoted tumor growth in a xenograft mouse model.However, knockdown of DAZL suppressed cell viability, colony formation, migration, and invasion in NSCLC cells.These results indicate that DAZL is a CG gene.
DAZL is an RNA-binding protein that can regulate mRNA translation, mainly as an activator of translation [22,23].TMT-based quantitative proteomic analysis and Western blotting showed that DAZL increased the levels of JAK2 and MCM8.RIP experiments showed that DAZL can bind to the mRNA of JAK2 and MCM8.These results suggest that DAZL enhances the translation of JAK2 and MCM8 by binding to their mRNAs.
JAK2 is a member of the Janus kinase family that phosphorylates STAT3[18] Phosphorylated STAT3 monomers form homodimers and translocate to the nucleus to modulate gene expression [18].Aberrant activation of JAK2/STAT3 is frequently observed in cancers, including NSCLC, and plays an important role in cancer initiation and development.JAK2 is mainly activated by cytokine-mediated phosphorylation or gain-of-function mutations in cancers [18].JAK2 p. V617F mutation is frequently detected in myeloproliferative disorders, resulting in constitutive activation of JAK2 kinase and the JAK2/STAT3 pathway [24].However, it rarely occurs in NSCLC [25].In this study, we demonstrated that DAZL can activate JAK2/STAT3 by promoting JAK2 translation.Treatment with the JAK2 inhibitor fedratinib can counteract the effects of DAZL overexpression on cell viability, colony formation, migration, and invasion.
These results suggest that DAZL plays an oncogenic role by activating JAK2.
MCM8 is a minichromosome maintenance (MCM) protein that is a DNA helicase.MCM2-7 forms a heterohexameric helicase complex that is required for the initiation of DNA replication [26].MCM8 and MCM9 are not essential for normal DNA replication but are rapidly recruited to DNA damage sites to restart DNA synthesis when the replication fork is stalled by cellular stresses such as DNA lesions and MCM2 depletion [27].Inhibition of the MCM8-9 complex reduces the synthesis of damaged DNA and sensitizes cancer cells to DNA-damaging agents (such as cisplatin, hydroxyurea, and olaparib) [19,20].Consistent with this nding, we found that knockdown of MCM8 decreased, but overexpression of DAZL increased, cisplatin-damaged DNA synthesis and IC50 values of cisplatin in NSCLC cells.Moreover, silencing of MCM8 attenuated the effects of DAZL overexpression.These results suggest that DAZL may reduce sensitivity to cisplatin by upregulating MCM8. 5

Conclusions
In summary, this study identi ed DAZL as a CG gene in NSCLC and demonstrated that it promotes NSCLC progression and cisplatin resistance by upregulating JAK2 and MCM8, respectively.This suggests that DAZL is a novel therapeutic target for the treatment of NSCLC.

Figure 2 A
Figure 2

F
-H The statistical analysis results of triplicate colony information assay (F), transwell invasion assay (G), and wound healing assay (H).