Blood Res (2024) 59:33
Published online October 17, 2024
https://doi.org/10.1007/s44313-024-00037-3
© The Korean Society of Hematology
Correspondence to : Yuansheng Lin
linys202012@163.com
Hongmei Jing
hongmeijing@bjmu.edu.cn
Weilong Zhang
zhangwl2012@126.com
Full list of author information is available at the end of the article
†Xue He, Changjian Yan, Yaru Yang and Weijia Wang contributed equally to this work.
© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Background SM-like (LSM) genes a family of RNA-binding proteins, are involved in mRNA regulation and can function as oncogenes by altering mRNA stability. However, their roles in B-cell progression and tumorigenesis remain poorly understood.
Methods We analyzed gene expression profiles and overall survival data of 123 patients with mantle cell lymphoma (MCL). The LSM index was developed to assess its potential as a prognostic marker of MCL survival.
Results Five of the eight LSM genes were identified as potential prognostic markers for survival in MCL, with particular emphasis on the LSM.index. The expression levels of these LSM genes demonstrated their potential utility as classifiers of MCL. The LSM.index-high group exhibited both poorer survival rates and lower RNA levels than did the overall transcript profile. Notably, LSM1 and LSM8 were overexpressed in the LSM.index-high group, with LSM1 showing 2.5-fold increase (p < 0.001) and LSM8 depicting 1.8-fold increase (p < 0.01) than those in the LSM.index-low group. Furthermore, elevated LSM gene expression was associated with increased cell division and RNA splicing pathway activity.
Conclusions The LSM.index demonstrates potential as a prognostic marker for survival in patients with MCL. Elevated expression of LSM genes, particularly LSM1 and LSM8, may be linked to poor survival outcomes through their involvement in cell division and RNA splicing pathways. These findings suggest that LSM genes may contribute to the aggressive behavior of MCL and represent potential targets for therapeutic interventions.
Keywords LSM genes, Mantle cell lymphoma, RNA degradation, LSM1, LSM8
RNA degradation is a conserved and ubiquitous process in all cells critical for the proper regulation of genetic information [1]. In eukaryotic cells, mRNA degradation occurs primarily via two pathways: the 5′ to 3′ pathway and the 3′ to 5′ pathway [1]. The cytoplasmic LSM1-7 complex (comprising LSM1–LSM7) and the nuclear LSM2–8 complex (comprising LSM2–LSM8) are distinct, conserved heptamer complexes that facilitate the 5’ to 3’ RNA degradation pathway [2–7]. The subunits LSM1 and LSM8 are critical for distinguishing the two complexes: the LSM1-7 complex (cytoplasmic, involved in mRNA degradation but not for RNA splicing) and the LSM2-8 complex (nuclear, involved in pre-mRNA degradation and necessary for RNA splicing) [2–7]. The mRNA decay mediated by the LSM1–7 complex in the cytoplasm and the pre-mRNA decay mediated by the LSM2–8 complex in the nucleus share several similarities. In both complexes, deadenylation initiates mRNA decay [8, 9], followed by decapping and degradation via the 5′ to 3′ exonucleolytic pathway [10, 11]. In the LSM1–7 complexmediated pathway, this complex promotes decapping by interacting with decapping activators such as Dhh1 and Pat1 [12, 13]. The mRNA decay rate is slower in LSM1–7 mutants [14].
Mantle cell lymphoma (MCL) is a subtype of non-Hodgkin B cell lymphoma that primarily affects older individuals [15, 16]. Although chemotherapy and stem cell transplantation have improved the prognosis, MCL still has a short median survival of approximately 5 to 7 years due to its aggressive behavior and rapid progression [17]. Understanding the molecular mechanisms driving MCL’s aggressive nature and identifying new treatment strategies are crucial. LSM1 overexpression has been observed in both primary and metastatic tumors [18, 19]. In human breast cancer, the chromosome region 8p11-12 containing LSM1 gene is frequently amplified [20, 21]. However, little is known about the prognostic significance and biological implications of LSM family genes in MCL. In this study, we demonstrated the prognostic relevance and biological implications of LSM genes in MCL.
We obtained 123 MCL gene expression arrays (Affymetrix Human Genome U133 Plus 2.0 Array) from the NCBI Gene Expression Omnibus (GEO) database (GSE93291) [22]. Additionally, 64 MCL gene expression arrays (Affymetrix Human Genome U133 Plus 2.0 Array) were obtained from the NCBI GEO database (GSE21452) [23]. Dataset GSE21452 represents the first phase of dataset GSE93291 [23]. These samples were derived from diagnostic lymph node tissues in formalin-fixed paraffin-embedded biopsies with a tumor content of ≥ 60%. All patients underwent treatment, including R-CHOP therapy, followingdiagnostic biopsy. Notably, dataset GSE21452 represents the first phase of GSE93291 [24]. This study was conducted in accordance with the principles of the Declaration of Helsinki.
Probeset measures for all arrays (123 MCL samples) were calculated using the Robust Multiarray Averaging algorithm. The relative RNA expression values for each probe were log-transformed (log2). Data comparing the LSM.index-high group to the LSM.index-low group were analyzed using an unpaired t-test and presented as mean ± SEM. Only genes with a fold change (log2) > 1 or < -1 and P-value < 0.05 were considered differentially expressed genes.
A comprehensive LSM.index was defined to predict survival in patients with MCL. The LSM.index was calculated using Eq. (1).
Where LSM.indexj indicates the index of LSM genes of jth sample used in survival prediction.
Hj indicates the product of the expression of harmful genes with a P-value < 0.05 in the jth sample. Three of the eight LSM genes (LSM1, LSM2, and LSM4) with a hazard ratio of > 1 were included.
Fj indicates the product of favorable gene expression in the jth sample. One of the eight LSM genes (LSM8) with a hazard ratio of < 1 was included.
Pathway enrichment analysis for differentially expressed genes between the LSM.index-high and LSM.index-low groups in MCL was conducted using the DAVID tool with default parameters [25]. The enriched GO pathway terms shown in the main figure were manually curated by selecting nonredundant GO terms from the biological process category.
R software v3.1.3, with the ggplot2 and survminer packages, was used for statistical analysis. Data were expressed as mean ± SEM in bar plots. A P-value < 0.05 was considered statistically significant.
To investigate the relationship between the LSM genes (LSM1, LSM2, LSM3, LSM4, LSM5, LSM6, LSM7, and LSM8) and survival in MCL, we analyzed the expression profiles of 123 MCL samples from the GSE21452 dataset. The expression levels of five out of the eight LSM genes were significantly associated with MCL survival (p < 0.05, log-rank test). The eight LSM genes were classified based on their hazard ratio values. Two of the LSM genes (LSM3, LSM8) had a hazard ratio of less than 1 and were defined as “favorable genes,” which are considered beneficial for MCL survival. Among these, LSM8, with a hazard ratio of 0.45 (95%CI, 0.24–0.87), was the most significant among the “favorable genes.” The other six LSM genes (LSM1, LSM2, LSM4, LSM5, LSM6, and LSM7) had a hazard ratio greater than 1 and were classified as “harmful genes,” which are considered detrimental to MCL survival (Fig. 1). LSM4, with a hazard ratio of 3.09 (95%CI, 1.35–4.48), was the most significant among the “harmful genes.” Furthermore, Kaplan–Meier survival curves for the four LSM genes were compared for 123 patients with MCL using the log-rank test (Fig. 2; LSM8, P = 1.7E-02; LSM2, P = 4.6E-03; LSM1, P = 4.5E-03; LSM4, P = 1.6E-05, log-rank test). Both the “favorable” and “harmful” LSM genes showed significant associations with MCL survival predictions.
We also compared 123 MCL samples from the GSE93291 dataset with 12 reactive lymph node samples from the GSE78513 dataset. We found that in the reactive lymph node group, the expression levels of LSM4, LSM8, and LSM12 were higher, whereas in the MCL group, LSM2, LSM5, LSM6, LSM7, and LSM14B were highly expressed (all p < 0.05; Supplementary Fig. 1A). No significant differences were observed in the expression of the remaining genes. Although LSM1 was identified as a key differentiator between prognostic groups (P = 4.5E-03, Fig. 2), there was no significant difference in its expression levels between MCL and reactive lymph nodes (p < 0.05, Supplementary Fig. 1A), contrary to LSM8. This finding suggests that although LSM1 may play a role in the biological differences between prognostic groups, it may not be a specific biomarker for distinguishing between MCL and reactive lymph nodes.
A correlation plot of the expression levels of the eight LSM genes in the MCL is shown in Fig. 3. Considerably, some LSM genes exhibited positive correlations, including LSM5 and LSM6 (correlation coefficient, cor = 0.48), LSM2 and LSM4 (cor = 0.47), LSM3 and LSM8 (cor = 0.46), and LSM1 and LSM6 (cor = 0.38). Conversely, some LSM genes displayed negative correlations, such as LSM3 and LSM4 (cor = -0.4) and LSM2 and LSM3 (cor = -0.37). Additionally, some LSM genes, such as LSM4 and LSM7, showed no correlation (cor = 0.01). The “favorable genes” LSM3 and LSM8, which are associated with better survival outcomes in MCL, exhibited a positive correlation (cor = 0.46). In contrast, the “harmful gene” LSM4, which is associated with poorer survival outcomes, was negatively correlated (cor = -0.4) with the “favorable genes” LSM3.
We performed unsupervised clustering of the expression levels of the eight LSM genes in 123 patients with MCL using cosine correlation similarity (Fig. 4A). Considerably, the eight LSM genes were grouped into two clusters, one containing LSM3 and the other containing the remaining seven LSM genes. Notably, all “harmful genes” were grouped together, suggesting that the expression patterns of LSM genes are linked to MCL survival outcomes and exhibit distinct expression characteristics. Furthermore, we identified that MCL could be categorized into two groups based on fuzzy clustering analysis of the expression of the eight LSM genes in 123 MCL samples (Fig. 4B; R, ggplot2). To quantify the imbalance between the expression levels of “harmful” and “favorable” genes, we calculated a ratio termed the LSM.index (refer to methods for the definition). The LSM.index was strongly associated with MCL survival (Fig. 4C; P = 3.29E-06, log-rank test). The hazard ratio for the LSM.index was 1.63 (95%CI, 1.33–2.00). A high LSM. index was associated with poor survival in patients with MCL, whereas a low LSM.index was associated with better survival outcomes.
The LSM.index-high and LSM.index-low groups represent two distinct classes of MCL. To explore the differences between these groups, we compared their gene expression profiles (Fig. 5A). We identified a total of 19 upregulated and 190 downregulated genes in the LSM.index-high group compared to those in the LSM. index-low group (Fig. 5B, p < 0.05). The LSM.index-high group exhibited a higher number of downregulated genes than upregulated genes, suggesting a different RNA metabolism process compared with the LSM.index-low group. The cumulative distribution of DEG RNA expression levels for differentially expressed genes between the LSM.index-high and LSM.index-low groups also showed that the LSM.index-high group had lower overall RNA levels (Fig. 5C, P = 2.57E-07). This finding was further validated using another dataset (GSE21452, 64 samples, P = 0.0048).
We also examined the differences in known prognostic molecular markers for MCL between the high and low LSM index groups, along with the differential expression of genes within the LSM gene family and their correlation with survival rates. Among the known MCL molecular markers, MKI67 expression was significantly lower in the low LSM.index group (p = 0.0004, Supplementary Table 1, Supplementary Fig. 1B). Within the LSM gene family, LSM4, LSM1, LSM6, and LSM2 had lower expression in the low LSM index group, whereas LSM8, LSM3, LSM14A, and LSM12 showed higher expression (all p < 0.05). No statistically significant differences were observed for LSM5, LSM7, LSM14B, or LSM10 levels between the two groups. The low LSM index group was associated with a longer survival time and lower mortality rate (all p < 0.05).
We further investigated differences in MCL35 classification genes between the high- and low-LSM index groups. In the low LSM index group, the expression levels of GLIPR1, CHD4, GSK3B, IK, and ATL1 were higher (all p < 0.05; Supplementary Fig. 2A). In contrast, the LSM index-high group exhibited elevated expression levels of UBXN4, CDKN3, FOXM1, H2AFX, FAM83D, MKI67, CDC20, NCAPG, CCNB2, ESPL1, KIF2C, and ZWINT (all p < 0.05). These differentially expressed genes (GLIPR1, CHD4, GSK3B, IK, ATL1, UBXN4, CDKN3, FOXM1, H2AFX, FAM83D, MKI67, CDC20, NCAPG, CCNB2, ESPL1, KIF2C, and ZWINT) may represent downstream targets and potential therapeutic targets; however, further investigation and validation are required.
Univariate and multivariate Cox regression analyses were conducted on the LSM index, additional LSM gene family members (LSM10, LSM11, LSM12, LSM14A, and LSM14B), SOX11, and TP53. Variables with p-values less than 0.15 were included in the multivariate Cox regression analysis. Our findings indicated that the LSM index can serve as an independent prognostic factor (p < 0.05; Supplementary Table 2).
We focused on the differential gene expression pathways between the LSM.index-high and LSM.index-low groups in MCL. The most significantly enriched pathway among all differentially expressed genes was the positive regulation of the B cell activation pathway, followed by the cell division and RNA splicing pathways (Fig. 6A). Among the differentially expressed genes, ATAD3B, TRR , and WASL were identified as being either upregulated or downregulated in the cell division pathway (Fig. 6B). Suppression of LSM1 expression causes cell cycle arrest in the G2-M phase 18. Therefore, ATAD3B, TRR , and WASL may be the target genes regulated by LSM that mediate cell cycle arrest. The differential expression patterns of these genes were also observed in another dataset (GSE21452, 64 samples). These findings suggest that LSM genes may regulate the cell division pathway, potentially contributing to poor survival outcomes in patients with MCL.
LSM1, also known as “CaSm” (“Cancer-associated Smlike”), is overexpressed in various cancer, including esophageal, lung, bladder, and prostate cancer. It functions as an oncogene by altering mRNA stability [18, 19]. However, the prognostic significance and biological implications of LSM family genes in MCL remain unclear. In this study, we analyzed the expression of LSM genes in MCL and found that their dysregulated expression predicts poorer survival outcomes and affects RNA expression across the entire transcriptome.
MCL is one of the rarest types of non-Hodgkin lymphoma and a subtype of B-cell lymphoma that is challenging to treat and is rarely considered curable. The median survival time for patients with MCL ranges from approximately 3 to 6 years. Therefore, identifying new biomarkers to predict survival outcomes in MCL is crucial [26]. The MCL International Prognostic Index (MIPI) is currently the most commonly used prognostic model [27]. It was derived from a cohort of 455 patients with advanced-stage MCL treated in a series of clinical trials in Germany and Europe. The MIPI incorporates the Eastern Cooperative Oncology Group (ECOG) performance status, age, leukocyte count, and lactic dehydrogenase levels and stratifying patients with MLC into three risk groups: low-risk, intermediate-risk, and high-risk [28]. Although various prognostic indicators for MCL have been considered, there is still no universal consensus on forecasting outcomes. KI67 and P53 abnormalities are critical prognostic factors in MCL. High KI67 expression is associated with poor clinical outcomes, and P53 abnormalities often indicate resistance to standard therapies. Prognostic indices such as the MIPI and MIPI-C, which include KI67, are valuable for patient stratification. The MCL35 classification based on RNA expression analysis utilizes genomic and transcriptomic profiling to identify molecular subsets of MCL, thereby providing enhanced prognostic precision and guiding personalized therapeutic strategies [29]. However, these models lack molecular biological predictive features such as gene expression factors. Our analysis revealed that the expression levels of five LSM genes were significantly associated with survival in patients with MCL. Additionally, we developed a comprehensive LSM.index to predict poor survival outcomes in these patients. The LSM.index demonstrated superior predictive power for survival compared to individual LSM proteins (P = 3.29E-06).
Notably, more than half of the LSM genes (five out of eight) were associated with survival outcomes in patients with MCL (p < 0.05, log-rank test). This suggests that LSM genes are associated with survival in patients with MCL. Moreover, the LSM.index serves as a better predictor of survival, indicating that the imbalanced expression of LSM genes may correlate with shorter survival times in these patients. Additionally, we provide three key pieces of evidence to support a strong connection between LSM genes and MCL prognosis. First, LSM gene expression serves as an effective classifier for MCL. Second, based on the LSM.index, we categorized the MCL samples into two groups: LSM.index-high and LSM.index-low. The LSM.index-high group exhibited lower RNA expression levels throughout the transcriptome. Third, in the LSM. index-high group, we discovered that the differentially expressed genes were related to the positive regulation of cell division and RNA splicing pathways, which may contribute to poorer survival outcomes in MCL patients.
Our study shows that LSM1 is considered to be among the “harmful genes,” with a hazard ratio greater than 1 (P = 4.5E-03); conversely, LSM8 is among the “favorable genes,” with a hazard ratio lower than 1 (P = 1.7E-02). The LSM1–7 and LSM2–8 complexes mediate cytoplasmic and nuclear mRNA decay, respectively [2–7]. Furthermore, LSM1 and LSM8 are the key subunits that differentiate the LSM1-7 complex from the LSM2-8 complex [2–7]. Our analysis suggests that LSM1–7 and LSM2–8 complexes, which mediate mRNA decay, may play distinct roles in MCL tumorigenesis. Additionally, this study revealed that although LSM1 was a significant differentiator between prognostic groups, its expression was not significantly different between MCL and reactive lymph nodes. This result suggests that LSM1 might be more involved in the cellular biological processes underlying malignant transformation in lymphoma rather than serving as a distinguishing biomarker between malignant and reactive proliferation. In contrast, LSM8 expression was significantly different between MCL and reactive lymph nodes, indicating that LSM8 may play a more direct role in these pathological conditions. Therefore, LSM1 is more likely to be a regulatory factor influencing disease progression or prognosis than a direct driver of disease onset. Future studies are required to investigate the specific functions of LSM1 further to understand its role in lymphomas better.
Cor Correlation coefficient
GEO Gene Expression Omnibus
GO Gene Ontology
MCL Mantle cell lymphoma
MIPI MCL International Prognostic Index
Contributions: (I) Conception and design: Hongmei Jing, Weilong Zhang; (II) Administrative support: Hongmei Jing; (III) Collection and assembly of data: Xue He, Changjian Yan, Weijia Wang; (IV) Data analysis and interpretation: Changjian Yan, Weijia Wang, Xiaoni Liu, Yaru Yang, Xin Huang, Wei Fu, Jing Hu, Ping Yang, Jing Wang, Mingxia Zhu, Yan Liu, Wei Zhang, Shaoxiang Li, Gehong Dong, Xiaoliang Yuan, Hongmei Jing, Weilong Zhang; (V) Manuscript writing: All authors; (VI) Final approval of manuscript: All authors.
This work was funded by the National Natural Science Foundation of China (81800195 and 81460315), Key Clinical Projects of Peking University Third Hospital (BYSYZD2019026 and BYSYZD2023014), Interdisciplinary Medicine Seed Fund of Peking University (BMU2018MB004), Beijing Natural Science Foundation (7132183 and 7182178), China Health Promotion Foundation (CHPF-zlkysx-001), Scientific Research Foundation (20141114) from Health Commission of Jiangxi Province, Science and Technology Research Foundation (GJJ14676) from Educational Commission of Jiangxi Province, China and National Clinical Key Specialty Construction Program China (2023).
No datasets were generated or analysed during the current study.
Ethics approval and consent to participate
This article contains no studies with human participants performed by any of the authors.
Consent for publication
No human participants were involved in this study.
Competing interests
The authors declare no competing interests.
Author details
1 Department of Pathology, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China. 2 Department of Hematology, Lymphoma Research Center, Peking University Third Hospital, Beijing 100191, China. 3 The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China. 4 Department of Pathology, The First Affiliated Hospital of Nanchang University, Nanchang 330006, China. 5 Department of Respiratory Medicine, The First Affiliated Hospital of Gannan Medical University, Ganzhou 341000, China. 6 Department of Respiratory Medicine, Affiliated Hospital of Jiujiang University, Jiujiang 332000, China. 7 Gannan Medical University, Ganzhou 341000, China. 8 The Second Affiliated Hospital of Fujian Medical University, Quanzhou 362000, China. 9 Department of Intensive Care Unit, Suzhou Research Center of Medical School, Suzhou Hospital, Affiliated Hospital of Medical School, Nanjing University, Suzhou 215000, China.
The online version contains supplementary material available at https://doi.org/10.1007/s44313-024-00037-3.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Blood Res 2024; 59():
Published online October 17, 2024 https://doi.org/10.1007/s44313-024-00037-3
Copyright © The Korean Society of Hematology.
Xue He1†, Changjian Yan2,8†, Yaru Yang3†, Weijia Wang4†, Xiaoni Liu5, Chaoling Wu6, Zimu Zhou7, Xin Huang2, Wei Fu2, Jing Hu2, Ping Yang2, Jing Wang2, Mingxia Zhu2, Yan Liu2, Wei Zhang1, Shaoxiang Li1, Gehong Dong1, Xiaoliang Yuan5, Yuansheng Lin9*, Hongmei Jing2* and Weilong Zhang2*
1 Department of Pathology, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China. 2 Department of Hematology, Lymphoma Research Center, Peking University Third Hospital, Beijing 100191, China. 3 The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China. 4 Department of Pathology, The First Affiliated Hospital of Nanchang University, Nanchang 330006, China. 5 Department of Respiratory Medicine, The First Affiliated Hospital of Gannan Medical University, Ganzhou 341000, China. 6 Department of Respiratory Medicine, Affiliated Hospital of Jiujiang University, Jiujiang 332000, China. 7 Gannan Medical University, Ganzhou 341000, China. 8 The Second Affiliated Hospital of Fujian Medical University, Quanzhou 362000, China. 9 Department of Intensive Care Unit, Suzhou Research Center of Medical School, Suzhou Hospital, Affiliated Hospital of Medical School, Nanjing University, Suzhou 215000, China.
Correspondence to:Yuansheng Lin
linys202012@163.com
Hongmei Jing
hongmeijing@bjmu.edu.cn
Weilong Zhang
zhangwl2012@126.com
Full list of author information is available at the end of the article
†Xue He, Changjian Yan, Yaru Yang and Weijia Wang contributed equally to this work.
© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Background SM-like (LSM) genes a family of RNA-binding proteins, are involved in mRNA regulation and can function as oncogenes by altering mRNA stability. However, their roles in B-cell progression and tumorigenesis remain poorly understood.
Methods We analyzed gene expression profiles and overall survival data of 123 patients with mantle cell lymphoma (MCL). The LSM index was developed to assess its potential as a prognostic marker of MCL survival.
Results Five of the eight LSM genes were identified as potential prognostic markers for survival in MCL, with particular emphasis on the LSM.index. The expression levels of these LSM genes demonstrated their potential utility as classifiers of MCL. The LSM.index-high group exhibited both poorer survival rates and lower RNA levels than did the overall transcript profile. Notably, LSM1 and LSM8 were overexpressed in the LSM.index-high group, with LSM1 showing 2.5-fold increase (p < 0.001) and LSM8 depicting 1.8-fold increase (p < 0.01) than those in the LSM.index-low group. Furthermore, elevated LSM gene expression was associated with increased cell division and RNA splicing pathway activity.
Conclusions The LSM.index demonstrates potential as a prognostic marker for survival in patients with MCL. Elevated expression of LSM genes, particularly LSM1 and LSM8, may be linked to poor survival outcomes through their involvement in cell division and RNA splicing pathways. These findings suggest that LSM genes may contribute to the aggressive behavior of MCL and represent potential targets for therapeutic interventions.
Keywords: LSM genes, Mantle cell lymphoma, RNA degradation, LSM1, LSM8
RNA degradation is a conserved and ubiquitous process in all cells critical for the proper regulation of genetic information [1]. In eukaryotic cells, mRNA degradation occurs primarily via two pathways: the 5′ to 3′ pathway and the 3′ to 5′ pathway [1]. The cytoplasmic LSM1-7 complex (comprising LSM1–LSM7) and the nuclear LSM2–8 complex (comprising LSM2–LSM8) are distinct, conserved heptamer complexes that facilitate the 5’ to 3’ RNA degradation pathway [2–7]. The subunits LSM1 and LSM8 are critical for distinguishing the two complexes: the LSM1-7 complex (cytoplasmic, involved in mRNA degradation but not for RNA splicing) and the LSM2-8 complex (nuclear, involved in pre-mRNA degradation and necessary for RNA splicing) [2–7]. The mRNA decay mediated by the LSM1–7 complex in the cytoplasm and the pre-mRNA decay mediated by the LSM2–8 complex in the nucleus share several similarities. In both complexes, deadenylation initiates mRNA decay [8, 9], followed by decapping and degradation via the 5′ to 3′ exonucleolytic pathway [10, 11]. In the LSM1–7 complexmediated pathway, this complex promotes decapping by interacting with decapping activators such as Dhh1 and Pat1 [12, 13]. The mRNA decay rate is slower in LSM1–7 mutants [14].
Mantle cell lymphoma (MCL) is a subtype of non-Hodgkin B cell lymphoma that primarily affects older individuals [15, 16]. Although chemotherapy and stem cell transplantation have improved the prognosis, MCL still has a short median survival of approximately 5 to 7 years due to its aggressive behavior and rapid progression [17]. Understanding the molecular mechanisms driving MCL’s aggressive nature and identifying new treatment strategies are crucial. LSM1 overexpression has been observed in both primary and metastatic tumors [18, 19]. In human breast cancer, the chromosome region 8p11-12 containing LSM1 gene is frequently amplified [20, 21]. However, little is known about the prognostic significance and biological implications of LSM family genes in MCL. In this study, we demonstrated the prognostic relevance and biological implications of LSM genes in MCL.
We obtained 123 MCL gene expression arrays (Affymetrix Human Genome U133 Plus 2.0 Array) from the NCBI Gene Expression Omnibus (GEO) database (GSE93291) [22]. Additionally, 64 MCL gene expression arrays (Affymetrix Human Genome U133 Plus 2.0 Array) were obtained from the NCBI GEO database (GSE21452) [23]. Dataset GSE21452 represents the first phase of dataset GSE93291 [23]. These samples were derived from diagnostic lymph node tissues in formalin-fixed paraffin-embedded biopsies with a tumor content of ≥ 60%. All patients underwent treatment, including R-CHOP therapy, followingdiagnostic biopsy. Notably, dataset GSE21452 represents the first phase of GSE93291 [24]. This study was conducted in accordance with the principles of the Declaration of Helsinki.
Probeset measures for all arrays (123 MCL samples) were calculated using the Robust Multiarray Averaging algorithm. The relative RNA expression values for each probe were log-transformed (log2). Data comparing the LSM.index-high group to the LSM.index-low group were analyzed using an unpaired t-test and presented as mean ± SEM. Only genes with a fold change (log2) > 1 or < -1 and P-value < 0.05 were considered differentially expressed genes.
A comprehensive LSM.index was defined to predict survival in patients with MCL. The LSM.index was calculated using Eq. (1).
Where LSM.indexj indicates the index of LSM genes of jth sample used in survival prediction.
Hj indicates the product of the expression of harmful genes with a P-value < 0.05 in the jth sample. Three of the eight LSM genes (LSM1, LSM2, and LSM4) with a hazard ratio of > 1 were included.
Fj indicates the product of favorable gene expression in the jth sample. One of the eight LSM genes (LSM8) with a hazard ratio of < 1 was included.
Pathway enrichment analysis for differentially expressed genes between the LSM.index-high and LSM.index-low groups in MCL was conducted using the DAVID tool with default parameters [25]. The enriched GO pathway terms shown in the main figure were manually curated by selecting nonredundant GO terms from the biological process category.
R software v3.1.3, with the ggplot2 and survminer packages, was used for statistical analysis. Data were expressed as mean ± SEM in bar plots. A P-value < 0.05 was considered statistically significant.
To investigate the relationship between the LSM genes (LSM1, LSM2, LSM3, LSM4, LSM5, LSM6, LSM7, and LSM8) and survival in MCL, we analyzed the expression profiles of 123 MCL samples from the GSE21452 dataset. The expression levels of five out of the eight LSM genes were significantly associated with MCL survival (p < 0.05, log-rank test). The eight LSM genes were classified based on their hazard ratio values. Two of the LSM genes (LSM3, LSM8) had a hazard ratio of less than 1 and were defined as “favorable genes,” which are considered beneficial for MCL survival. Among these, LSM8, with a hazard ratio of 0.45 (95%CI, 0.24–0.87), was the most significant among the “favorable genes.” The other six LSM genes (LSM1, LSM2, LSM4, LSM5, LSM6, and LSM7) had a hazard ratio greater than 1 and were classified as “harmful genes,” which are considered detrimental to MCL survival (Fig. 1). LSM4, with a hazard ratio of 3.09 (95%CI, 1.35–4.48), was the most significant among the “harmful genes.” Furthermore, Kaplan–Meier survival curves for the four LSM genes were compared for 123 patients with MCL using the log-rank test (Fig. 2; LSM8, P = 1.7E-02; LSM2, P = 4.6E-03; LSM1, P = 4.5E-03; LSM4, P = 1.6E-05, log-rank test). Both the “favorable” and “harmful” LSM genes showed significant associations with MCL survival predictions.
We also compared 123 MCL samples from the GSE93291 dataset with 12 reactive lymph node samples from the GSE78513 dataset. We found that in the reactive lymph node group, the expression levels of LSM4, LSM8, and LSM12 were higher, whereas in the MCL group, LSM2, LSM5, LSM6, LSM7, and LSM14B were highly expressed (all p < 0.05; Supplementary Fig. 1A). No significant differences were observed in the expression of the remaining genes. Although LSM1 was identified as a key differentiator between prognostic groups (P = 4.5E-03, Fig. 2), there was no significant difference in its expression levels between MCL and reactive lymph nodes (p < 0.05, Supplementary Fig. 1A), contrary to LSM8. This finding suggests that although LSM1 may play a role in the biological differences between prognostic groups, it may not be a specific biomarker for distinguishing between MCL and reactive lymph nodes.
A correlation plot of the expression levels of the eight LSM genes in the MCL is shown in Fig. 3. Considerably, some LSM genes exhibited positive correlations, including LSM5 and LSM6 (correlation coefficient, cor = 0.48), LSM2 and LSM4 (cor = 0.47), LSM3 and LSM8 (cor = 0.46), and LSM1 and LSM6 (cor = 0.38). Conversely, some LSM genes displayed negative correlations, such as LSM3 and LSM4 (cor = -0.4) and LSM2 and LSM3 (cor = -0.37). Additionally, some LSM genes, such as LSM4 and LSM7, showed no correlation (cor = 0.01). The “favorable genes” LSM3 and LSM8, which are associated with better survival outcomes in MCL, exhibited a positive correlation (cor = 0.46). In contrast, the “harmful gene” LSM4, which is associated with poorer survival outcomes, was negatively correlated (cor = -0.4) with the “favorable genes” LSM3.
We performed unsupervised clustering of the expression levels of the eight LSM genes in 123 patients with MCL using cosine correlation similarity (Fig. 4A). Considerably, the eight LSM genes were grouped into two clusters, one containing LSM3 and the other containing the remaining seven LSM genes. Notably, all “harmful genes” were grouped together, suggesting that the expression patterns of LSM genes are linked to MCL survival outcomes and exhibit distinct expression characteristics. Furthermore, we identified that MCL could be categorized into two groups based on fuzzy clustering analysis of the expression of the eight LSM genes in 123 MCL samples (Fig. 4B; R, ggplot2). To quantify the imbalance between the expression levels of “harmful” and “favorable” genes, we calculated a ratio termed the LSM.index (refer to methods for the definition). The LSM.index was strongly associated with MCL survival (Fig. 4C; P = 3.29E-06, log-rank test). The hazard ratio for the LSM.index was 1.63 (95%CI, 1.33–2.00). A high LSM. index was associated with poor survival in patients with MCL, whereas a low LSM.index was associated with better survival outcomes.
The LSM.index-high and LSM.index-low groups represent two distinct classes of MCL. To explore the differences between these groups, we compared their gene expression profiles (Fig. 5A). We identified a total of 19 upregulated and 190 downregulated genes in the LSM.index-high group compared to those in the LSM. index-low group (Fig. 5B, p < 0.05). The LSM.index-high group exhibited a higher number of downregulated genes than upregulated genes, suggesting a different RNA metabolism process compared with the LSM.index-low group. The cumulative distribution of DEG RNA expression levels for differentially expressed genes between the LSM.index-high and LSM.index-low groups also showed that the LSM.index-high group had lower overall RNA levels (Fig. 5C, P = 2.57E-07). This finding was further validated using another dataset (GSE21452, 64 samples, P = 0.0048).
We also examined the differences in known prognostic molecular markers for MCL between the high and low LSM index groups, along with the differential expression of genes within the LSM gene family and their correlation with survival rates. Among the known MCL molecular markers, MKI67 expression was significantly lower in the low LSM.index group (p = 0.0004, Supplementary Table 1, Supplementary Fig. 1B). Within the LSM gene family, LSM4, LSM1, LSM6, and LSM2 had lower expression in the low LSM index group, whereas LSM8, LSM3, LSM14A, and LSM12 showed higher expression (all p < 0.05). No statistically significant differences were observed for LSM5, LSM7, LSM14B, or LSM10 levels between the two groups. The low LSM index group was associated with a longer survival time and lower mortality rate (all p < 0.05).
We further investigated differences in MCL35 classification genes between the high- and low-LSM index groups. In the low LSM index group, the expression levels of GLIPR1, CHD4, GSK3B, IK, and ATL1 were higher (all p < 0.05; Supplementary Fig. 2A). In contrast, the LSM index-high group exhibited elevated expression levels of UBXN4, CDKN3, FOXM1, H2AFX, FAM83D, MKI67, CDC20, NCAPG, CCNB2, ESPL1, KIF2C, and ZWINT (all p < 0.05). These differentially expressed genes (GLIPR1, CHD4, GSK3B, IK, ATL1, UBXN4, CDKN3, FOXM1, H2AFX, FAM83D, MKI67, CDC20, NCAPG, CCNB2, ESPL1, KIF2C, and ZWINT) may represent downstream targets and potential therapeutic targets; however, further investigation and validation are required.
Univariate and multivariate Cox regression analyses were conducted on the LSM index, additional LSM gene family members (LSM10, LSM11, LSM12, LSM14A, and LSM14B), SOX11, and TP53. Variables with p-values less than 0.15 were included in the multivariate Cox regression analysis. Our findings indicated that the LSM index can serve as an independent prognostic factor (p < 0.05; Supplementary Table 2).
We focused on the differential gene expression pathways between the LSM.index-high and LSM.index-low groups in MCL. The most significantly enriched pathway among all differentially expressed genes was the positive regulation of the B cell activation pathway, followed by the cell division and RNA splicing pathways (Fig. 6A). Among the differentially expressed genes, ATAD3B, TRR , and WASL were identified as being either upregulated or downregulated in the cell division pathway (Fig. 6B). Suppression of LSM1 expression causes cell cycle arrest in the G2-M phase 18. Therefore, ATAD3B, TRR , and WASL may be the target genes regulated by LSM that mediate cell cycle arrest. The differential expression patterns of these genes were also observed in another dataset (GSE21452, 64 samples). These findings suggest that LSM genes may regulate the cell division pathway, potentially contributing to poor survival outcomes in patients with MCL.
LSM1, also known as “CaSm” (“Cancer-associated Smlike”), is overexpressed in various cancer, including esophageal, lung, bladder, and prostate cancer. It functions as an oncogene by altering mRNA stability [18, 19]. However, the prognostic significance and biological implications of LSM family genes in MCL remain unclear. In this study, we analyzed the expression of LSM genes in MCL and found that their dysregulated expression predicts poorer survival outcomes and affects RNA expression across the entire transcriptome.
MCL is one of the rarest types of non-Hodgkin lymphoma and a subtype of B-cell lymphoma that is challenging to treat and is rarely considered curable. The median survival time for patients with MCL ranges from approximately 3 to 6 years. Therefore, identifying new biomarkers to predict survival outcomes in MCL is crucial [26]. The MCL International Prognostic Index (MIPI) is currently the most commonly used prognostic model [27]. It was derived from a cohort of 455 patients with advanced-stage MCL treated in a series of clinical trials in Germany and Europe. The MIPI incorporates the Eastern Cooperative Oncology Group (ECOG) performance status, age, leukocyte count, and lactic dehydrogenase levels and stratifying patients with MLC into three risk groups: low-risk, intermediate-risk, and high-risk [28]. Although various prognostic indicators for MCL have been considered, there is still no universal consensus on forecasting outcomes. KI67 and P53 abnormalities are critical prognostic factors in MCL. High KI67 expression is associated with poor clinical outcomes, and P53 abnormalities often indicate resistance to standard therapies. Prognostic indices such as the MIPI and MIPI-C, which include KI67, are valuable for patient stratification. The MCL35 classification based on RNA expression analysis utilizes genomic and transcriptomic profiling to identify molecular subsets of MCL, thereby providing enhanced prognostic precision and guiding personalized therapeutic strategies [29]. However, these models lack molecular biological predictive features such as gene expression factors. Our analysis revealed that the expression levels of five LSM genes were significantly associated with survival in patients with MCL. Additionally, we developed a comprehensive LSM.index to predict poor survival outcomes in these patients. The LSM.index demonstrated superior predictive power for survival compared to individual LSM proteins (P = 3.29E-06).
Notably, more than half of the LSM genes (five out of eight) were associated with survival outcomes in patients with MCL (p < 0.05, log-rank test). This suggests that LSM genes are associated with survival in patients with MCL. Moreover, the LSM.index serves as a better predictor of survival, indicating that the imbalanced expression of LSM genes may correlate with shorter survival times in these patients. Additionally, we provide three key pieces of evidence to support a strong connection between LSM genes and MCL prognosis. First, LSM gene expression serves as an effective classifier for MCL. Second, based on the LSM.index, we categorized the MCL samples into two groups: LSM.index-high and LSM.index-low. The LSM.index-high group exhibited lower RNA expression levels throughout the transcriptome. Third, in the LSM. index-high group, we discovered that the differentially expressed genes were related to the positive regulation of cell division and RNA splicing pathways, which may contribute to poorer survival outcomes in MCL patients.
Our study shows that LSM1 is considered to be among the “harmful genes,” with a hazard ratio greater than 1 (P = 4.5E-03); conversely, LSM8 is among the “favorable genes,” with a hazard ratio lower than 1 (P = 1.7E-02). The LSM1–7 and LSM2–8 complexes mediate cytoplasmic and nuclear mRNA decay, respectively [2–7]. Furthermore, LSM1 and LSM8 are the key subunits that differentiate the LSM1-7 complex from the LSM2-8 complex [2–7]. Our analysis suggests that LSM1–7 and LSM2–8 complexes, which mediate mRNA decay, may play distinct roles in MCL tumorigenesis. Additionally, this study revealed that although LSM1 was a significant differentiator between prognostic groups, its expression was not significantly different between MCL and reactive lymph nodes. This result suggests that LSM1 might be more involved in the cellular biological processes underlying malignant transformation in lymphoma rather than serving as a distinguishing biomarker between malignant and reactive proliferation. In contrast, LSM8 expression was significantly different between MCL and reactive lymph nodes, indicating that LSM8 may play a more direct role in these pathological conditions. Therefore, LSM1 is more likely to be a regulatory factor influencing disease progression or prognosis than a direct driver of disease onset. Future studies are required to investigate the specific functions of LSM1 further to understand its role in lymphomas better.
Cor Correlation coefficient
GEO Gene Expression Omnibus
GO Gene Ontology
MCL Mantle cell lymphoma
MIPI MCL International Prognostic Index
Contributions: (I) Conception and design: Hongmei Jing, Weilong Zhang; (II) Administrative support: Hongmei Jing; (III) Collection and assembly of data: Xue He, Changjian Yan, Weijia Wang; (IV) Data analysis and interpretation: Changjian Yan, Weijia Wang, Xiaoni Liu, Yaru Yang, Xin Huang, Wei Fu, Jing Hu, Ping Yang, Jing Wang, Mingxia Zhu, Yan Liu, Wei Zhang, Shaoxiang Li, Gehong Dong, Xiaoliang Yuan, Hongmei Jing, Weilong Zhang; (V) Manuscript writing: All authors; (VI) Final approval of manuscript: All authors.
This work was funded by the National Natural Science Foundation of China (81800195 and 81460315), Key Clinical Projects of Peking University Third Hospital (BYSYZD2019026 and BYSYZD2023014), Interdisciplinary Medicine Seed Fund of Peking University (BMU2018MB004), Beijing Natural Science Foundation (7132183 and 7182178), China Health Promotion Foundation (CHPF-zlkysx-001), Scientific Research Foundation (20141114) from Health Commission of Jiangxi Province, Science and Technology Research Foundation (GJJ14676) from Educational Commission of Jiangxi Province, China and National Clinical Key Specialty Construction Program China (2023).
No datasets were generated or analysed during the current study.
Ethics approval and consent to participate
This article contains no studies with human participants performed by any of the authors.
Consent for publication
No human participants were involved in this study.
Competing interests
The authors declare no competing interests.
Author details
1 Department of Pathology, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China. 2 Department of Hematology, Lymphoma Research Center, Peking University Third Hospital, Beijing 100191, China. 3 The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China. 4 Department of Pathology, The First Affiliated Hospital of Nanchang University, Nanchang 330006, China. 5 Department of Respiratory Medicine, The First Affiliated Hospital of Gannan Medical University, Ganzhou 341000, China. 6 Department of Respiratory Medicine, Affiliated Hospital of Jiujiang University, Jiujiang 332000, China. 7 Gannan Medical University, Ganzhou 341000, China. 8 The Second Affiliated Hospital of Fujian Medical University, Quanzhou 362000, China. 9 Department of Intensive Care Unit, Suzhou Research Center of Medical School, Suzhou Hospital, Affiliated Hospital of Medical School, Nanjing University, Suzhou 215000, China.
The online version contains supplementary material available at https://doi.org/10.1007/s44313-024-00037-3.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Aaron Wu, Solomon A. Graf, Nicholas Burwick, Jonathan E. Grim, Zhao Ming Dong, Robert E. Richard, Thomas R. Chauncey
Blood Res 2020; 55(1): 57-61Byung Woog Kang, Sang Kyun Sohn, Joon Ho Moon, Yee Soo Chae, Jong Gwang Kim, Soo Jung Lee, Won Seog Kim, Je-Jung Lee, Se Ryeon Lee, Keon Uk Park, Ho Sup Lee, Won Sik Lee, Jong-Ho Won, Moo-Rim Park, Jae-Yong Kwak, Min Kyoung Kim, Hyo Jung Kim, Sung Yong Oh, Hye Jin Kang, and Cheolwon Suh
Blood Res 2014; 49(1): 15-21Tae-Dong Jeong, Hyun-Sook Chi, Min-Sun Kim, Seongsoo Jang, Chan-Jeoung Park, and Joo Ryung Huh
Blood Res 2016; 51(2): 127-132