Exploring the Molecular Mechanisms and Shared Genetic Characteristics of Monoclonal Gammopathy of Undetermined Significance and Multiple Myeloma

doi:10.21203/rs.3.rs-4701045/v1

Download PDF

Article

Exploring the Molecular Mechanisms and Shared Genetic Characteristics of Monoclonal Gammopathy of Undetermined Significance and Multiple Myeloma

https://doi.org/10.21203/rs.3.rs-4701045/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Monoclonal gammopathy of undetermined significance (MGUS) serves as a precursor to multiple myeloma (MM), with a subset of MGUS cases advancing to MM each year. Despite extensive research, the mechanisms driving this progression are not yet fully understood. Leveraging transcriptomic data from the Gene Expression Omnibus (GEO) database, this study conducted an analysis of differentially expressed genes (DEGs) between MGUS and MM. Weighted gene co-expression network analysis (WGCNA) was utilized to pinpoint significant gene modules linked to disease progression. Univariate Cox analysis and LASSO regression were employed to identify genes associated with MM prognosis. Immune scores and immune cell proportions were calculated using ESTIMATE and CIBERSORT tools. A total of 961 DEGs related to MM and 355 DEGs associated with MGUS were identified. Cross-analysis revealed 12 intersecting genes, with four (DAP3, HIST1H1C, MRPL4, and UBE2S) as core genes. The MGUSscore effectively stratified MM patients into high-risk and low-risk groups, with the high-risk group showing significantly shorter overall survival (P < 0.05). Core genes were closely linked to immune cell infiltration. A ceRNA network identified 68 miRNAs and 10 lncRNAs related to core genes. DAP3, HIST1H1C, MRPL4, and UBE2S are potential therapeutic targets for MM.

Biological sciences/Cancer/Haematological cancer

Biological sciences/Cancer/Tumour biomarkers

Biological sciences/Computational biology and bioinformatics/Databases

monoclonal gammopathy of undetermined significance

multiple myeloma

weighted gene co-expression network analysis

immune response

Monoclonal Gammopathy of Undetermined Significance (MGUS) is characterized by the presence of abnormal monoclonal immunoglobulin or its light chain fragments in the blood and/or urine, with clonal plasma cells or lymphoplasmacytic cells constituting less than 10% of bone marrow cells. Predominantly affecting individuals over 50, MGUS has a higher incidence in males^1,2. Although initially asymptomatic, patients with MGUS face a continual risk of progression to more severe plasma cell malignancies, particularly multiple myeloma (MM). Approximately 1% of MGUS cases deteriorate annually, and about 10% of Smoldering Multiple Myeloma (SMM) cases progress to MM within five years of diagnosis. Alarmingly, 18% of patients with MGUS are at an increased risk of progression within 20 years³. The asymptomatic nature of MGUS often leads to delayed diagnosis, complicating prevention and treatment strategies^4,5.

MM is the second most common type of hematologic malignancy, marked by the uncontrolled proliferation of abnormal clonal plasma cells in the bone marrow. This pathological expansion causes severe clinical manifestations, including bone lesions, renal failure, anemia, and hypercalcemia⁶. The transition from MGUS to MM is a complex process involving significant genetic and microenvironmental changes. Key genetic alterations, such as hyperdiploidy and specific gene translocations, play a crucial role in promoting the clonal expansion of malignant plasma cells. Concurrently, the bone marrow microenvironment evolves to support malignant progression, with increased angiogenesis and enhanced adaptability of stromal cells⁷. Understanding these clinical and biological markers is vital for predicting the progression from MGUS to MM, facilitating patient stratification for targeted surveillance and early intervention. Despite these insights, the precise mechanisms driving MGUS to MM transformation remain unclear, necessitating further investigation.

Recent advancements in bioinformatics and the use of gene arrays have become indispensable in the biomedical field^8,9. In-depth studies of differentially expressed genes (DEGs) and their protein levels in patients with MM can uncover the molecular mechanisms driving the progression from MGUS to MM. This can lead to new discoveries and targets for early diagnosis and effective treatment. In this study, bioinformatics analysis methods were employed to identify common pathways, core shared genes, and pivotal genes involved in the pathogenesis of both MGUS and MM. The objective is to explore the shared mechanisms and therapeutic targets for both conditions, enhancing the management of patients with MGUS and providing early diagnosis and treatment options for MM.

Data sources

Bulk Transcriptome Data

Transcriptome sequencing datasets related to MGUS and MM were sourced from the Gene Expression Omnibus (GEO) database¹⁰. The dataset GSE6477¹¹ includes bone marrow transcriptome data from 15 healthy individuals, 22 patients with MGUS, and 73 patients with newly diagnosed MM, annotated under the GPL96 platform. Additionally, the GSE5900¹² Series Matrix File, annotated under GPL570, was utilized, comprising data from 22 healthy individuals and 44 patients with MGUS. For validation, the GSE136337¹³ dataset annotated under GPL2714 and the GSE57317¹⁴ dataset annotated under GPL570 were accessed. These datasets provide overall survival (OS) rate information for patients with newly diagnosed MM, with GSE136337 also offering details on International Staging System (ISS) staging, progression-free survival (PFS), and chromosomal variation.

Single-Cell Transcriptome Data

Single-cell transcriptome data were retrieved from the GEO database, specifically the GSE124310¹⁵ dataset. This dataset includes data from 9 healthy individuals and 7 patients with MM, enabling single-cell correlation analysis.

Identification of DEGs among the MGUS group, the MM group, and the healthy group

Firstly, the 'limma' package¹⁶ was utilized to identify DEGs in the MGUS and MM groups compared to the control samples. The screening criteria for DEGs were set as AveExpr > 5, |logFC| > 1, and P-value < 0.05.

WGCNA for coexpression network construction

Advancements in bioinformatics have led to the development of sophisticated and comprehensive algorithms. Weighted gene co-expression network analysis (WGCNA) is particularly notable for its ability to process large datasets, enabling the identification of gene clusters most relevant to disease progression through clustering and modularization¹⁷. In our study, the 'WGCNA'¹⁷ R package was employed to construct gene co-expression networks for MGUS and MM. To estimate network connectivity, the weighted adjacency matrix was converted to a topological overlap matrix (TOM), and its clustering tree structure was constructed using a hierarchical clustering method. The branches of the clustering tree represent different gene modules, and colors denote the various modules. Genes are classified according to their expression patterns, with those having similar functions grouped into one module. Thousands of genes are classified into modules based on their weighted correlation coefficients derived from their expression patterns.

Venn diagram plotting

To integrate transcriptome data with gene co-expression network analysis and identify core genes closely associated with the progression of MM and MGUS, set operations were performed on four gene sets: MM-related DEGs, MGUS-related DEGs, MM-related WGCNA module genes, and MGUS-related WGCNA module genes. Intersections were calculated between MM-related DEGs and MM-related WGCNA module genes (set A) and between MGUS-related DEGs and MGUS-related WGCNA module genes (set B). The final core gene set (set C) was derived by intersecting sets A and B. This integrative approach was visualized with a Venn diagram, where distinct closed curves represented different gene sets, and intersection areas highlighted the core genes of interest.

Clinical significance of DEGs

To corroborate the findings, receiver operating characteristic (ROC) curve analysis was conducted to assess the diagnostic potential of key DEGs. Additionally, associations between survival status, disease staging, chromosomal variations, and differential gene expression were examined (Table 1).

Table 1

Clinical significance of GSE136337
Database	GSE136337
Platform	GPL27143
New cases of MM	426
Age
≤ 60 years	213 (50%)
>60 years	213 (50%)
Gender
female	165 (38.7%)
male	261 (62.3%)
Median OS (years)	7.58 [0-14.5]
Median PFS (years)	4 [0-12.92]
Chromosomal variations.
del 1p	90(23.2%)
del 13q	77(18.1%)
del 1p32	85(19.6%)
ISS stage
I	168(39.4%)
II	135(31.7%)
III	121(28.4%)
NA	2(0.5%)

Development and validation of the prognostic model

First, univariate Cox analysis was performed to identify genes associated with MM prognosis. The LASSO algorithm was then applied to address redundancy issues caused by collinearity among MGUS-related DEGs in the MM data. By shrinking the regression coefficients, an optimal subset of DEGs related to MM prognosis was identified. A prognostic model based on MGUS was constructed, consisting of four hub genes (DAP3, HIST1H1C, MRPL4, UBE2S). The GSE136337 dataset was divided into high-risk and low-risk groups based on the risk score, and OS between the two subgroups was compared using Kaplan-Meier analysis. Multivariable Cox proportional hazards models were utilized to analyze the discrimination ability of clinical risk score variables. Time-dependent ROC analysis with the area under the curve (AUC) was used to evaluate the prognostic performance of the risk score. Additionally, the GSE57317 dataset served as a validation set to verify the predictive value of the risk score. A predicted nomogram based on the results of multivariate Cox analysis was developed using the 'rms' R package to assess the prediction of 1-, 3-, and 5-year OS probability.

Functional enrichment analysis

Following the identification of the hub genes, gene enrichment analyses were conducted to investigate the underlying biological processes and pathways. Gene Ontology (GO) enrichment analysis was employed to annotate the biological processes and functions of these hub genes, examining them according to the three sub-ontologies: molecular functions (MF), cellular components (CC), and biological processes (BP). Additionally, the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was used to identify potential signaling pathways associated with these genes. The 'clusterProfiler'¹⁸ package in R facilitated extensive GO/KEGG enrichment analyses of the shared genes, potentially elucidating mechanisms of disease onset and progression. Patients with MM were stratified into low and high expression groups based on the median expression level of the MGUSscore. Subsequent Gene Set Enrichment Analysis (GSEA) of these groups provided further insights into the biological implications of gene expression differences.

Immune-related analysis

The significant influence of the immune response on disease progression has become increasingly evident with advancements in disease research. The ESTIMATE algorithm calculated the immune score, stromal score, and ESTIMATE score for high and low-risk groups. The 'CIBERSORT'¹⁹ tool determined the proportion of immune cell infiltration. Linear regression was then employed to elucidate the association between immune cell infiltration and risk score.

Regulatory network analysis of hub genes

This study utilized the R package 'RcisTarget'²⁰ for predicting transcription factors, focusing on specific motifs for all computations. The normalized enrichment score (NES) for each motif was determined based on the total number of motifs in the database. To assess the overexpression of each motif within the gene set, the area under the curve (AUC) of the recovery curve was initially calculated. This curve ranks gene sets and motifs, thereby defining the AUC for each motif-gene set pair. The distribution of AUCs across all motifs in the gene set was then used to calculate the NES for each motif. The 'rcitarget.hg19.motifdb.cisbpont.500 bp' database was employed as the gene-motif ranking reference.

Construction of ceRNA network

The competing endogenous RNA (ceRNA) network illustrates complex regulatory interactions between non-coding RNAs and coding RNAs within cells, mediated by microRNA (miRNAs)²¹. These small RNA molecules, approximately 20–25 nucleotides long, regulate gene expression by promoting mRNA degradation or inhibiting mRNA translation²². Long non-coding RNAs (lncRNAs), exceeding 200 nucleotides, modulate miRNA activity by binding to them, thereby influencing miRNA-mediated regulation of target genes ²³. miRNAs associated with hub genes were identified from the miRTarBase database²⁴, while lncRNAs targeting these miRNAs were predicted using the StarBase database²⁵. The ceRNA network was visualized with Cytoscape²⁶.

Single‑cell sequencing analysis

Single-cell RNA-seq data analysis utilized the 'Seurat'²⁷ R package, with normalization of the expression matrix of all cells performed via the 'LogNormalize' method. Principal component analysis (PCA) reduced data dimensionality, selecting the top 15 principal components for clustering cells based on marker genes identified by the elbow plot. Uniform manifold approximation and projection (UMAP) visualized the cellular landscapes, while cell clusters were annotated using the 'SingleR'²⁸ software package. The expression levels of DAP3, HIST1H1C, MRPL4, and UBE2S across various cell types were examined.

Statistical analysis

Statistical analysis was performed using R language (version 4.3.3), with P < 0.05 indicating statistical significance.

Identification of Hub Genes from GSE6477 and GSE5900

Analysis of the GSE6477 dataset identified 961 DEGs between patients with MM and healthy controls (Fig. 1A). Conversely, the GSE5900 dataset revealed 355 DEGs between MGUS and normal controls (Fig. 1B). Heatmaps in Figs. 1C and 1D depict the 20 DEGs ++. Using WGCNA on the GSE6477 dataset, 12 gene modules significantly associated with MM were identified, each uniquely color-coded (Figs. 2C, 2D). The green, yellow, blue, and black modules demonstrated strong correlations with MM, with correlation coefficients and statistical significances as follows: green (r = -0.61, P = 4e-10), yellow (r = -0.86, P = 2e-27), blue (r = 0.67, P = 1e-12), and black (r = -0.46, P = 7e-06) (Figs. 2A, 2E). Analysis of the MGUS dataset GSE5900 also identified 12 modules, with the green module showing a significant positive correlation with MGUS (r = 0.56, P = 1e-06) (Figs. 2B, 2F). Focusing on the blue module linked to MM and the green module associated with MGUS, these modules were intersected with the previously identified DEGs, resulting in the identification of 12 hub genes (Fig. 2G).

ROC curves for the 12 intersection genes

ROC curves for the 12 intersection genes in the GSE6477 dataset were plotted to evaluate their diagnostic value. The AUC reflects the diagnostic accuracy; values approaching 1.0 indicate high accuracy, while values near 0.5 indicate low accuracy. An AUC of 0.5 signifies no diagnostic value. As shown in Fig. 3A, all 12 genes have AUC values greater than 0.8, demonstrating significant diagnostic efficacy for MM.

Construction and verification of the prognostic model

The GSE136337 dataset, containing OS and status information of patients with newly diagnosed MM, was first downloaded for analysis. Univariate analysis of 12 genes revealed that CD320, DAP3, EXOSC5, HIST1H1C, MRPL4, PSMD8, and UBE2S are prognostically relevant, indicating poor prognosis (Fig. 3B). Subsequently, Lasso regression analysis was used to refine these genes, ultimately selecting four valuable predictive genes: DAP3, HIST1H1C, MRPL4, and UBE2S (Fig. 3C). Based on these four genes, an MGUS-related risk score (MGUSscore) was calculated. Patients with MM were stratified into two subgroups according to the MGUSscore, with the high-MGUSscore group correlating with higher expression levels of the four prognostic molecules and showing significantly worse OS compared to the low-MGUSscore group (Fig. 3D). ROC curves indicated that the MGUSscore could serve as a predictive marker for MM patient OS (1-year AUC: 0.532, 3-year AUC: 0.644, 5-year AUC: 0.650) (Fig. 3F). Validation of the MGUSscore in the GSE57317 dataset confirmed that the high MGUSscore group had significantly worse OS, with statistically significant results (SFigure 1A). The ROC curves based on the risk score showed a 1-year AUC of 0.878 and a 3-year AUC of 0.936 (SFigure 1B). Furthermore, multivariate Cox regression analysis demonstrated that the MGUSscore is an independent risk factor for patients with MM (HR 2.494, 95% CI 1.794–3.467, P < 0.001) (Fig. 3E). A nomogram for patients with MM was established by integrating the risk score, age, and ISS stage, performing well in predicting 3-, and 5-year OS (Fig. 3G). Internal validation using the Bootstrap method (1000 samples) confirmed the model's effectiveness in distinguishing event occurrence. The corrected model demonstrated high predictive accuracy, with some bias. Overall, the model showed good stability and reliably predicted survival outcomes. Figure H and figure I presents calibration curves assessing the model's predictive accuracy at 3, and 5 years, comparing predicted (X-axis) and observed (Y-axis) survival probabilities. Points represent bootstrap estimates with error bars, and curves near the diagonal line indicate high accuracy. The red lines show the fit between predictions and actual values, demonstrating the model's performance closely matches observed outcomes.

The clinical significance of 4 hub genes

Validation of the differential expression of DAP3, HIST1H1C, MRPL4, and UBE2S genes in the GSE6477 dataset revealed higher expression levels in patients with MM compared to patients with MGUS and healthy controls (Fig. 4A). Clinical analyses in the GSE136337 dataset demonstrated that high expression of these four genes is associated with shorter OS and PFS (Figs. 4B and 4C). Additionally, DAP3, HIST1H1C, and UBE2S correlate with the ISS of MM (Fig. 5A). Furthermore, high expression of HIST1H1C, MRPL4, and UBE2S is associated with chromosomal regions 1p, 13q, and 1p32, while DAP3 is linked only to 13q (Figs. 5B-5D). These findings suggest that DAP3, HIST1H1C, MRPL4, and UBE2S may play a role in the malignant progression of MM.

Pathway enrichment analysis of dysregulation of MGUSscore

To understand the impact of changes in the four core genes on the biological functions of MM, GO and KEGG analyses were conducted. The dataset was divided into two groups based on the median MGUSscore. After filtering for upregulated and downregulated genes, GSEA was performed. KEGG results indicated that these four genes are involved in the ribosome and ubiquitin-mediated proteolysis pathways (Fig. 6A). GO analysis revealed their participation in pathways regulating ubiquitin-protein ligase activity, ubiquitin-protein transferase regulator activity, ubiquitin cycle, mitochondrial gene expression, mitochondrial ribosomal subunit, cytosolic ribosomal subunit, mitochondrial inner membrane, mitochondrial matrix, mitochondrial respiratory chain complex, ribosome, ribosomal subunit, and mitochondrial ribosome (Fig. 6B). The group with high MGUSscores showed enrichment in pathways such as the tricarboxylic acid (TCA) cycle, DNA replication, nucleotide excision repair, oxidative phosphorylation, proteasome, and ribosome. The figure displays the top 10 enriched pathways in high-scoring patients with MM identified by GSEA (Fig. 6C). The study reveals the critical roles of the four core genes in the pathology of MM, particularly in mitochondrial function, metabolism, and protein homeostasis. Existing research indicates that the ribosome pathway is essential for MM cell survival. Specific genes involved in ribosome biogenesis, such as BRD9, are crucial for MM cell proliferation. Dysregulation of ribosomal proteins affects protein synthesis and cell growth, leading to the progression and malignant transformation of MM²⁹.

Tumor immune microenvironment and four hub genes

Spearman correlation analysis identified a significant association between the four core genes and infiltrating immune cells in the tumor microenvironment (TME), particularly plasma cells (Fig. 7A-D). Further analysis revealed that a high MGUSscore significantly correlated with increased expression of plasma cells, resting dendritic cells, gamma delta T cells, M1 macrophages, and memory CD4 resting T cells (Fig. 7E). The high-MGUSscore group displayed lower Stromal, Immune, and ESTIMATE scores, indicating higher tumor purity and poorer prognosis (Fig. 7F).

Hub genes and associated transcription factors

Analysis of the four genes and their transcription factors revealed that these genes are regulated by the same transcription factors and common regulatory processes. Enrichment analysis of the cumulative recovery curves of these transcription factors, motif-TF annotations, and significant gene selection results was conducted (Fig. 8A). The results indicated that the motif with the highest normalized enrichment score (NES: 6.51) was taipale_tf_pairs__MYBL1_MAX_NCACGTGNNYAACSGNN_CAP_rep (Fig. 8B, 8C). This motif is more prevalent in the HIST1H1C and UBE2S genes.

Transcriptional regulation analysis

Using the miRTarBase dataframe, reverse prediction of the four core genes identified 62 miRNAs and a total of 108 mRNA-miRNA relationships. The StarBase database then predicted 10 lncRNAs associated with these miRNAs. A ceRNA network of mRNA, miRNA, and lncRNA was constructed using Cytoscape (Fig. 8D). The analysis revealed that HAGLR, FGD5-AS1, and AC087477.2 regulate UBE2S by competitively binding to miR-17-5p, while HCG18 regulates UBE2S by binding to miR-34a-5p. Additionally, LIMD1-AS1, LINC01184, AC135050.6, and AC092295.2 regulate MRPL4 by binding to miR-193b-3p. Furthermore, AC087477.2 and LINC02035 regulate DAP3 by binding to miR-106b-5p and miR-16-5p, respectively.

Overview of hub gene expression in single cells

Single-cell data from GSE124310 were analyzed using the Seurat package. Cells were clustered with the tSNE algorithm, annotated using HumanPrimaryCellAtlasData, and further refined with the SingleR package in R. Cells were categorized into ten groups: B cell, Pro-B cell CD34+, T cells, Monocyte, Erythroblast, Common Myeloid Progenitor (CMP), Granulocyte-Monocyte Progenitors (GMP), Bone Marrow (BM), Megakaryocyte-Erythroid Progenitors (MEP) (Fig. 9A). B cells were subdivided into plasma cells, naive B cells, and Germinal Center B cells (GC B cells) to observe gene expression³⁰(Fig. 9B). In normal samples, DAP3, HIST1H1C, MRPL4, and UBE2S showed baseline expression in monocytes. In MM samples, these genes exhibited significantly increased expression in monocytes, MM plasma cells, and precursor B cells (Figs. 9C, 9D). Validation using the Human Protein Atlas database³¹ confirmed significant expression of DAP3, HIST1H1C, MRPL4, and UBE2S in plasma cells. Notably, HIST1H1C, MRPL4, and UBE2S showed higher expression levels in plasma cells compared to other cell types (Figs. 9E-9H), corroborating our previous findings.

MM is an incurable malignant plasma cell tumor that leads to irreversible organ damage. Typically, MM progresses through three stages: MGUS, SMM, and Symptomatic or active multiple myeloma (AMM)³². In individuals over 50, MGUS prevalence exceeds 3%, with an annual progression rate to MM of 1%. SMM advances to MM at approximately 10% per year during the first five years post-diagnosis^33,34. Early intervention and proactive prevention are crucial for slowing MM progression, yet reliable biomarkers for predicting the transition from MGUS to MM remain elusive. Understanding the molecular basis of this progression is essential for identifying high-risk patients with MGUS and developing new therapeutic targets.

This study conducted bioinformatics analyses on MGUS (GSE5900) and MM (GSE6477) to identify common DEGs in MGUS, MM, and control samples. The WGCNA algorithm, known for its reliability, facilitated co-expression clustering analysis in MGUS and MM. Disease-relevant modules were selected and intersected with DEGs, resulting in the identification of 12 intersecting genes.

Univariate analysis and LASSO regression pinpointed four core genes: DAP3, HIST1H1C, MRPL4, and UBE2S. These genes formed the basis of a new risk score for managing patients with MM. Multivariate analysis demonstrated that this risk score is an independent prognostic factor for MM, with the high-risk group exhibiting significantly shorter OS compared to the low-risk group. The ROC curve further confirmed the risk score's predictive accuracy. Additionally, a nomogram was developed to predict MM patient prognosis.

Clinical correlation analysis revealed significantly higher expression levels of the four core genes in MM compared to MGUS and healthy controls. These genes are not only associated with OS and PFS but also correlated with ISS staging, with higher stages exhibiting elevated gene expression levels. Interestingly, HIST1H1C, MRPL4, and UBE2S showed higher expression in groups with 1p, 13q, and 1p32 deletions compared to normal subgroups with intact 1p, 13q, and 1p32, while DAP3 was specifically associated with 13q deletions. Studies indicate that the frequency of 1p deletion in MM is approximately 30%, compared to only 6% in MGUS. This deletion often co-occurs with amplification of the 1q21 region, observed in 40% of MM and 25% of MGUS cases, and is linked to a higher risk of MGUS progressing to MM and poorer prognosis in patients with MM³⁵. Additionally, del13 is associated with progression from MGUS to MM³⁶. These findings suggest that these four core genes may play a role in the malignant transformation from MGUS to MM.

DAP3, also known as S29mt, MRPS29, and bMRP-10, is an apoptosis-associated protein³⁷. Recent research indicates that DAP3 is closely related to tumor progression³⁸ and chemotherapy resistance³⁹. As a mitochondrial ribosomal component located primarily in the mitochondrial matrix, DAP3 is crucial for regulating mitochondrial function. Silencing DAP3 sensitizes cells to mitochondria-mediated intrinsic death pathways⁴⁰. Additionally, DAP3 is implicated in cancers as a splicing-regulatory RNA-binding protein (RBP), with erroneous splicing events regulated by DAP3 occurring in various cancers⁴¹. HIST1H1C, a member of the histone H1 family, regulates higher-order chromatin structure⁴². Mutations in HIST1H1C occur in ≥ 5% of diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), or Burkitt's lymphoma (BL) cases, contributing to the malignant progression of lymphomas^43,44. Mitochondrial ribosomal proteins (MRPs) aggregate in ribosomes to synthesize proteins encoded by mtDNA, crucial for mitochondrial bioenergetics and metabolic processes⁴⁵. MRPL4, a member of the MRPs, is reported as a high-risk factor in prostate cancer and a potential diagnostic biomarker⁴⁶. UBE2S, or Ubiquitin-conjugating enzyme E2S, plays a crucial role in the pathogenesis and progression of various malignant tumors by promoting cell invasion, migration, and proliferation through the regulation of DNA repair, DNA damage, and cell cycle processes^47–49. This study supports these findings. KEGG and GO analyses of the four genes, as well as GSEA conducted after grouping based on risk scores calculated from these genes, indicate that these genes are involved in critical biological pathways related to cell metabolism, growth, and DNA maintenance.

MGUS and MM are both plasma cell-related disorders. Increasing evidence suggests that the development of MGUS and MM is closely associated with the TME. Our study confirms that the four core genes are associated with various immune cells, particularly B cells and plasma cells. Single-cell studies have confirmed that these four genes are primarily expressed at the level of B cells and plasma cells. The ESTIMATE algorithm confirmed that the immune scores were lower in the high MGUSscore group, indicating higher tumor purity and poorer prognosis. The high MGUSscore group was also associated with higher levels of plasma cells and memory B cells.

MGUS and MM are plasma cell-related disorders, with growing evidence linking their development to the TME. This study confirms that the four core genes are associated with various immune cells, particularly B cells and plasma cells. Single-cell studies further validate that these genes are predominantly expressed in B cells and plasma cells. The ESTIMATE algorithm indicated lower immune scores in the high MGUSscore group, suggesting higher tumor purity and poorer prognosis. Additionally, this group exhibited higher levels of plasma cells and memory B cells.

In this study, RcisTarget was used to identify key binding motifs of hub genes and their associated transcription factors. One notable motif is "taipale_tf_pairs__MYBL1_MAX_NCACGTGNNYAACSGNN_CAP_rep," representing binding sites for transcription factors MYBL1 and MAX. MYBL1, a member of the MYB family, regulates cell cycle and proliferation⁵⁰, while MAX, as a MYC partner, influences cell growth and apoptosis⁵¹. These transcription factors play crucial roles in cell proliferation, differentiation, and apoptosis, which are central to cancer development.

The ceRNA regulatory mechanism plays a pivotal role in various biological processes and is critical in cancer treatment. To delve deeper into this mechanism, a ceRNA regulatory network was constructed, revealing 108 mRNA-miRNA relationship pairs through reverse miRNA prediction. Key miRNAs identified include hsa-miR-17-5p, hsa-miR-193b-3p, hsa-miR-106b-5p, hsa-miR-16-5p, and hsa-miR-34a-5p. miR-17-5p is linked to the progression and metastasis of colorectal cancer and acute myeloid leukemia⁵². Elevated plasma levels of miR-17-5p predict treatment response in CRC, making it a valuable non-invasive biomarker for liver metastasis and treatment efficacy⁵³. miR-34a-5p interacts with breast cancer markers CD44 and CD24 and influences drug response by regulating the cholesterol pathway⁵⁴. miR-106b-5p functions as a tumor suppressor, regulating critical processes such as cell cycle, proliferation, apoptosis, differentiation, invasion, angiogenesis, drug resistance, and metastasis in various cancers, including cervical, gastric, bladder, and colorectal cancers. Additionally, miR-106b-5p serves as a diagnostic and prognostic biomarker for cervical cancer⁵⁵. miR-16-5p, produced by the MIR16-1 gene, was first identified in chronic lymphocytic leukemia due to its deletion and downregulation⁵⁶. This miRNA is downregulated in various cancer cell lines and clinical samples of multiple cancers, including neuroblastoma, osteosarcoma, hepatocellular carcinoma, cervical cancer, breast cancer, brain tumors, gastrointestinal cancers, lung cancer, and bladder cancer, playing significant roles in their pathogenesis⁵⁷.

In addition to miRNAs, 10 lncRNAs were identified that can regulate core genes by competitively binding to miRNAs. Among these, HAGLR (also known as HOXD-AS1 and Mdgt) plays crucial roles in intestinal development and various cancers, promoting cancer cell proliferation and invasion⁵⁸. FGD5-AS1, a newly discovered lncRNA, is abnormally highly expressed in multiple cancer tissues and is closely associated with lymph node metastasis, tumor invasion, survival time, and recurrence rates in cancers such as bladder and gastric cancer. It is involved in multiple signaling pathways, including TGF-β, ERK/AKT, Wnt/β-catenin, and PD1/PD-L1⁵⁹. Notably, the Wnt/β-catenin signaling pathway mediates the proliferation, migration, drug resistance, and development of osteolytic lesions in MM cells⁶⁰. These findings underscore the complex roles of miRNAs, lncRNAs, and transcription factors in cancer biology and suggest potential links between them and MM and MGUS, offering new insights for diagnosis, prognosis, and treatment.

Several limitations should be noted in this study. Firstly, only four hub genes were identified. Secondly, the detailed molecular mechanisms by which hub genes, miRNAs, and transcription factors affect these diseases remain unclear. In summary, this study explores the common pathogenesis of MGUS and MM, elucidates the roles of these four core genes in the malignant progression of these diseases, and identifies them as potential therapeutic targets for patients with MGUS and MM.

In conclusion, this systematic bioinformatics analysis highlights the critical role of four core genes (DAP3, HIST1H1C, MRPL4, and UBE2S) in the pathogenesis and progression of MM. These genes regulate key biological processes such as cell apoptosis, mitochondrial function, chromatin structure, and protein synthesis. Their involvement is closely linked to the malignant transformation from MGUS to MM, as well as the progression and prognosis of MM. By exploring the roles of these genes in transcriptional regulatory networks and miRNA/lncRNA regulatory mechanisms, the study provides valuable insights into MM pathogenesis, identifies potential therapeutic targets, and supports the advancement of precision medicine.

Acknowledgments:

We thank Bullet Edits Limited for BulletEdits the linguistic editing and proofreading of the manuscript.

Author contributions

YJ.F and WY.G contributed equally as co-first authors to this study. YJ.F and WY.G wrote the main manuscript text. YJ.F and Y.L collect some data. Y.Cand JR.L prepared figures. YJ.F, WY.G, ZJ. Z,and Y.L review and edit the main manuscript. ZJ.Zand Y.L approve the final version of the manuscript. All authors reviewed the manuscript, contributed to the article, and approved the submitted version.

Conflict of interest

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability statement

The original contributions presented in this study are included in the article and its Supplementary Material. Further inquiries can be directed to the corresponding author.

Ethical Approval

Not applicable.

Funding:

This work was supported by the National Natural Science Foundation of China (82070148), the Science and Technology Support Plan of Changzhou Science and Technology Bureau (Social Development) (CE20235060), and the Science and Technology Bureau Foundation Application Project of Changzhou (CJ20220096). Additional support was provided by the Changzhou Clinical Medical Center and the Outstanding Talent of Changzhou "The 14th Five-Year Plan" High-Level Health Talents Training Project.

A, M. et al. Genomewide profiling of copy-number alteration in monoclonal gammopathy of undetermined significance. Eur. J. Haematol. 97, (2016).
Steiner, N. et al. Monoclonal gammopathy of renal significance (MGRS) increases the risk for progression to multiple myeloma: an observational study of 2935 MGUS patients. Oncotarget 9, 2344–2356 (2018).
Ra, K. et al. Long-Term Follow-up of Monoclonal Gammopathy of Undetermined Significance. N. Engl. J. Med. 378, (2018).
Korde, N., Kristinsson, S. Y. & Landgren, O. Monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM): novel biological insights and development of early treatment strategies. Blood 117, 5573–5581 (2011).
Pang, L. et al. Prognosis of young patients with monoclonal gammopathy of undetermined significance (MGUS). Blood Cancer J. 11, 26 (2021).
Cowan, A. J. et al. Diagnosis and Management of Multiple Myeloma: A Review. JAMA 327, 464–477 (2022).
van Nieuwenhuijzen, N., Spaan, I., Raymakers, R. & Peperzak, V. From MGUS to Multiple Myeloma, a Paradigm for Clonal Evolution of Premalignant Cells. Cancer Res. 78, 2449–2456 (2018).
Li, Y. et al. Transcription factor ZBTB42 is a novel prognostic factor associated with immune cell infiltration in glioma. Front. Pharmacol. 14, 1102277 (2023).
Z, Z. et al. A Loop-Based and AGO-Incorporated Virtual Screening Model Targeting AGO-Mediated miRNA-mRNA Interactions for Drug Discovery to Rescue Bone Phenotype in Genetically Modified Mice. Adv. Sci. Weinh. Baden-Wurtt. Ger. 7, (2020).
Edgar, R., Domrachev, M. & Lash, A. E. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 30, 207–210 (2002).
Chng, W. J. et al. Molecular dissection of hyperdiploid multiple myeloma by gene expression profiling. Cancer Res. 67, 2982–2989 (2007).
F, Z. et al. Gene-expression signature of benign monoclonal gammopathy evident in multiple myeloma is linked to good prognosis. Blood 109, (2007).
Sa, D. et al. Bone marrow microenvironments that contribute to patient outcomes in newly diagnosed multiple myeloma: A cohort study of patients in the Total Therapy clinical trials. PLoS Med. 17, (2020).
Heuck, C. J. et al. Five gene probes carry most of the discriminatory power of the 70-gene risk model in multiple myeloma. Leukemia 28, 2410–2413 (2014).
Zavidij, O. et al. Single-cell RNA sequencing reveals compromised immune microenvironment in precursor stages of multiple myeloma. Nat. Cancer 1, 493–506 (2020).
Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015).
Langfelder, P. & Horvath, S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559 (2008).
Yu, G., Wang, L.-G., Han, Y. & He, Q.-Y. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics J. Integr. Biol. 16, 284–287 (2012).
Newman, A. M. et al. Robust enumeration of cell subsets from tissue expression profiles. Nat. Methods 12, 453–457 (2015).
Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat. Methods 14, 1083–1086 (2017).
Salmena, L., Poliseno, L., Tay, Y., Kats, L. & Pandolfi, P. P. A ceRNA hypothesis: the Rosetta stone of a hidden RNA language? Cell 146, 353 (2011).
L, C. et al. Trends in the development of miRNA bioinformatics tools. Brief. Bioinform. 20, (2019).
Ab, H., D, T. & M, G. Integrated lncRNA function upon genomic and epigenomic regulation. Mol. Cell 82, (2022).
Huang, H.-Y. et al. miRTarBase update 2022: an informative resource for experimentally validated miRNA–target interactions. Nucleic Acids Res. 50, D222–D230 (2021).
Li, J.-H., Liu, S., Zhou, H., Qu, L.-H. & Yang, J.-H. starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein–RNA interaction networks from large-scale CLIP-Seq data. Nucleic Acids Res. 42, D92–D97 (2014).
Shannon, P. et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504 (2003).
Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573-3587.e29 (2021).
D, A. et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat. Immunol. 20, (2019).
Kurata, K. et al. BRD9 Is Essential for Ribosome Biogenesis and the Survival of Multiple Myeloma Cells. Blood 140, 596–597 (2022).
Xia, J. et al. Single-cell landscape and clinical outcomes of infiltrating B cells in colorectal cancer. Immunology 168, 135–151 (2023).
Uhlen, M. et al. A pathology atlas of the human cancer transcriptome. Science 357, eaan2507 (2017).
Dhodapkar, M. V. MGUS to myeloma: a mysterious gammopathy of underexplored significance. Blood 128, 2599–2606 (2016).
Rajkumar, S. V. Multiple myeloma: 2020 update on diagnosis, risk-stratification and management. Am. J. Hematol. 95, 548–567 (2020).
Landgren, O. et al. Monoclonal gammopathy of undetermined significance (MGUS) consistently precedes multiple myeloma: a prospective study. Blood 113, 5412 (2009).
Rajan, A. M. & Rajkumar, S. V. Interpretation of cytogenetic results in multiple myeloma for clinical practice. Blood Cancer J. 5, e365 (2015).
Mukherjee, D. et al. The mammalian exosome mediates the efficient degradation of mRNAs that contain AU-rich elements. EMBO J. 21, 165–174 (2002).
Kissil, J. L. et al. Isolation of DAP3, a novel mediator of interferon-gamma-induced cell death. J. Biol. Chem. 270, 27932–27936 (1995).
Sasaki, H. et al. Arg and DAP3 expression was correlated with human thymoma stage. Clin. Exp. Metastasis 21, 507–513 (2004).
Jia, Y. et al. Depletion of death-associated protein-3 induces chemoresistance in gastric cancer cells through the β-catenin/LGR5/Bcl-2 axis. J. Investig. Med. Off. Publ. Am. Fed. Clin. Res. 67, 856–861 (2019).
Song, H. et al. Death-associated protein 3 in cancer—discrepant roles of DAP3 in tumours and molecular mechanisms. Front. Oncol. 13, 1323751 (2024).
Han, J. et al. Multilayered control of splicing regulatory networks by DAP3 leads to widespread alternative splicing changes in cancer. Nat. Commun. 13, 1793 (2022).
Izzo, A., Kamieniarz, K. & Schneider, R. The histone H1 family: specific members, specific functions? Biol. Chem. 389, 333–343 (2008).
Lunning, M. A. & Green, M. R. Mutation of chromatin modifiers; an emerging hallmark of germinal center B-cell lymphomas. Blood Cancer J. 5, e361 (2015).
Wang, X. et al. A novel immune-related epigenetic signature based on the transcriptome for predicting the prognosis and therapeutic response of patients with diffuse large B-cell lymphoma. Clin. Immunol. 243, 109105 (2022).
Mo, D. et al. The mitochondrial ribosomal protein mRpL4 regulates Notch signaling. EMBO Rep. 24, e55764 (2023).
Fu, Q. et al. Proteomics reveals MRPL4 as a high-risk factor and a potential diagnostic biomarker for prostate cancer. Proteomics 22, e2200081 (2022).
Zhang, M. et al. Diverse roles of UBE2S in cancer and therapy resistance: Biological functions and mechanisms. Heliyon 10, (2024).
Peng, S. et al. UBE2S as a novel ubiquitinated regulator of p16 and β-catenin to promote bone metastasis of prostate cancer. Int. J. Biol. Sci. 18, 3528 (2022).
M, L. et al. UBE2S mediates tumor progression via SOX6/β-Catenin signaling in endometrial cancer. Int. J. Biochem. Cell Biol. 109, (2019).
Anand, S. et al. From modulation of cellular plasticity to potentiation of therapeutic resistance: new and emerging roles of MYB transcription factors in human malignancies. Cancer Metastasis Rev. 43, 409–421 (2024).
Loughrey, P. B. et al. Succinate dehydrogenase and MYC-associated factor X mutations in pituitary neuroendocrine tumours. Endocr. Relat. Cancer 29, R157–R172 (2022).
Liu, M., Yu, B., Tian, Y. & Li, F. Regulatory function and mechanism research for m6A modification WTAP via SUCLG2-AS1- miR-17-5p-JAK1 axis in AML. BMC Cancer 24, 98 (2024).
Sur, D. et al. Predictive Efficacy of MiR-125b-5p, MiR-17-5p, and MiR-185-5p in Liver Metastasis and Chemotherapy Response Among Advanced Stage Colorectal Cancer Patients. Front. Oncol. 11, 651380 (2021).
Monchusi, B. & Kaur, M. miRNAs as Modulators of Cholesterol in Breast Cancer Stem Cells: An Approach to Overcome Drug Resistance in Cancer. Curr. Drug Targets 23, 656–677.
Yang, C., Dou, R., Yin, T. & Ding, J. MiRNA-106b-5p in human cancers: diverse functions and promising biomarker. Biomed. Pharmacother. Biomedecine Pharmacother. 127, 110211 (2020).
Calin, G. A. et al. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc. Natl. Acad. Sci. U. S. A. 99, 15524–15529 (2002).
Ghafouri-Fard, S. et al. A review on the role of mir-16-5p in the carcinogenesis. Cancer Cell Int. 22, 342 (2022).
Li, L. et al. HOX cluster-embedded antisense long non-coding RNAs in lung cancer. Cancer Lett. 450, 14–21 (2019).
He, N. et al. The role of long non-coding RNA FGD5-AS1 in cancer. Bioengineered 13, 11026–11041 (2022).
van Andel, H., Kocemba, K. A., Spaargaren, M. & Pals, S. T. Aberrant Wnt signaling in multiple myeloma: molecular mechanisms and targeting options. Leukemia 33, 1063–1075 (2019).

No competing interests reported.

SFigure1.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Exploring the Molecular Mechanisms and Shared Genetic Characteristics of Monoclonal Gammopathy of Undetermined Significance and Multiple Myeloma

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Data sources

Bulk Transcriptome Data

Single-Cell Transcriptome Data

Identification of DEGs among the MGUS group, the MM group, and the healthy group

WGCNA for coexpression network construction

Venn diagram plotting

Clinical significance of DEGs

Development and validation of the prognostic model

Functional enrichment analysis

Immune-related analysis

Regulatory network analysis of hub genes

Construction of ceRNA network

Single‑cell sequencing analysis

Statistical analysis

Results

Identification of Hub Genes from GSE6477 and GSE5900

ROC curves for the 12 intersection genes

Construction and verification of the prognostic model

The clinical significance of 4 hub genes

Pathway enrichment analysis of dysregulation of MGUSscore

Tumor immune microenvironment and four hub genes

Hub genes and associated transcription factors

Transcriptional regulation analysis

Overview of hub gene expression in single cells

Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1