Deregulation mechanisms and therapeutic opportunities of p53-responsive microRNAs in diffuse large B-cell lymphoma

Micro-RNAs biogenesis

Mature microRNAs are synthesized during multistage biogenesis. The canonical pathway of this process includes the stages of the microRNA gene transcription by RNA polymerase and formation of the primary microRNA transcript (pri-miR); cleavage of pri-miR in the nucleus by nuclear RNase III Drosha and its cofactor DGCR8/Pasha to a microRNA precursor (pre-miR), which is transported from the nucleus to the cell cytoplasm using the exportin protein; cleavage of pre-miR in the cytoplasm using another RNase III Dicer to a microRNA duplex consisting of mature and antisense microRNA (miR). One of the duplex strands is subsequently included in the RNA-induced silencing complex (RISC) by the Argonaute protein and serves to suppress the corresponding targets by repressing translation or by directing mRNA degradation (Gulyaeva & Kushlinskiy, 2016).

There are many reasons for the disruption of normal microRNA expression, which are based on both genetic and epigenetic changes. All of them lead to disruption of the processes of transcription and subsequent microRNA processing (Zhang, Liao & Tang, 2019).

Description of p53-responsive microRNAs

The p53-mediated cellular response to genotoxic stress is carried out due to changes in the expression spectrum of microRNAs, namely the activation of tumor suppressor microRNAs (miR-15a/16, miR-23a, miR-29, miR-107, miR-143, miR-145, miR-192, miR-194, miR-215, miR-605, let-7 and other) and suppression of oncogenic microRNAs (miR-17-92 cluster, miR-221/2222 and other), which is possible to implement both directly through transcription-dependent mechanisms and indirectly (Capaccia et al., 2022; Zhang, Liao & Tang, 2019; Jacques et al., 2020; Kaller et al., 2022).

A number of studies have shown a decrease in the level of tumor suppressor microRNAs miR-34a, miR-34b, miR-34c, miR-129, miR-203, miR-145, miR-143 in DLBCL compared with non-tumor lymphoid tissue (Hedström et al., 2013; Akao et al., 2007; lsaadi et al., 2021; Yamagishi et al., 2015; Zheng et al., 2021; Leivonen et al., 2017; Larrabeiti-Etxebarria et al., 2023). An increase in the expression of miR-145 and miR-143 in DLBCL tissue compared to healthy B-lymphocytes has been recorded only in the study by Lawrie et al. (2009).

The listed microRNAs are p53-responsive. It is important to note that some of them are involved in a feedback loop complex with p53, which contributes to enhanced, better controlled and fine-tuning of the cellular response to genotoxic stress (Cabrita et al., 2016). For example, miR34a may create a positive feedback loop with p53 by inhibiting the deacetylase SIRT1, which lead to p53 acetylation and activation (Gong et al., 2023; Ong & Ramasamy, 2018), whereas miR-129 and miR-145 directly target the anti-apoptotic factors MDM2 and MDM4, and, thus, counteract the MDM2 and MDM4-mediated suppression of p53 signaling (Yao et al., 2021; Zeinali et al., 2019).

According to miRTarBase, which contains data on microRNA-target interactions validated experimentally by reporter assay, western blot, microarray and next-generation sequencing experiments, p53-responsive microRNAs, such as microRNAs miR-34 family, miR-143/145 cluster, miR-129 and miR-203 have many common targets, namely pro-oncogenic transcription factors, positive regulators of the cell cycle at the G1/S transition checkpoint, anti-apoptotic factors, neoangiogenesis factors, participants in oncogenic signaling pathways, as well as DNA methyltransferases, which are significant in lymphomagenesis and oncogenesis in general (Table 1) (Larrabeiti-Etxebarria et al., 2019).

The MIR-34A gene is mapped on 1p36.22. The MIR-34B/C gene is mapped on 11q23.1 and is responsible for the sequence of the bicistronic transcript, from which mature microRNAs miR-34b and miR-34c are subsequently formed. MIR-34A and MIR-34B/C are located in the host genes of long non-coding RNAs, such as EF570048 and BC021736, respectively (Asmar et al., 2014).

MicroRNAs of the miR-34 family are characterized by a high degree of homology in the seed sequence, which is necessary for binding to mRNA targets. Using bioinformatics and experimental studies, microRNAs of the miR-34 family have been shown to have a large number of common targets (Zhang, Liao & Tang, 2019).

Human microRNA miR-129 is encoded by MIR-129-1 and MIR-129-2, which located intergenic on 7q32.1 and 11p11.2, respectively (Yu et al., 2013). The targets of miR-129 are mRNAs of several oncogenes, the most studied of which are CDK6, PDGFRA, FOXP1, HMGB1 and the stem cell transcription factors SOX4, EIF2C3 and CAMTA1 (Yu et al., 2013; Wang, Luo & Guo, 2014; Voropaeva et al., 2022a).

The MIR-143 and MIR-145 genes are located 1.7 kb apart on 5q32. The host-gene of this microRNA cluster is the CARMN non-coding RNA gene. It is assumed that miR-143 and miR-145 are formed from a common precursor and their functions are co-operated (Pidíkova, Reis & Herichova, 2020). It has been shown that the levels of miR-143 and miR-145 increase in response to stress through the PI3K/Akt and p53-mediated pathways. The targets of miR-143 are such well-known oncogenes as SOX2, KLF4, DNMT3A, MAPK1/7, MAP3K, BCL2, MDM2 and PDGFRA. miR-145 is involved in the post-transcriptional regulation of SOX2/9/11, STAT1, CDKN1A, KLF4/5, MAP3K3, MAP2K4, MAP3K11, MYC, CDK4/6, VEGFA, MDM2, etc. (Voropaeva et al., 2023d; Zhang et al., 2022).

The MIR-203 gene is located in the intergenic region of the chromosomal locus 14q32.33. Studies show that the microRNA miR-203 encoded by this gene, besides targeting CDK6, SOX2/4, VEGFA, MYD88, MDM4, BCL2L2, is also involved in post-transcriptional control of PI3K/AKT, SRC, RAS/MAPK and JAK/STAT3 signaling pathways participants. Its predicted targets are proto-oncogenes such as DNMT3B and CREB1 (Voropaeva et al., 2022a).

Impaired p53 function as a cause of deregulation of p53-responsive microRNAs

Mutations of the TP53 gene are the commonest molecular genetic changes in malignant neoplasms including DLBCL. A recent review has addressed the importance of TP53 gene mutations in DLBCL that is relevant to poor prognosis. The better understanding of abnormalities of p53 is meant for the basis of development of better therapeutic strategy for DLBCL (Wen et al., 2024). Most of them affect the DNA-binding domain (Fig. 1A) and can be divided into three types based on their effect: loss-of-function, gain-of-function and dominant-negative. It is critically important because by its function the p53 protein is a transcription factor that is involved in the expression of numerous protein and microRNA-coding genes (Wen et al., 2024). Over the years, several published research strongly established the gain of function mutant p53 promoting aberrant expression of several miRNAs leading to cells reprogramming and pluripotency, provoke chemoresistance and cell survival to drive different cancer phenotypes (Ghatak et al., 2021; Madrigal et al., 2021). On the other hand, the loss-of-function mutations lead to disruption of the ability of p53 to contact regulatory elements of canonical target microRNA genes (Strano et al., 2007). A predominant repression effect on miRNA expression was found for p53 mutants (Madrigal et al., 2021; Zhang et al., 2016).

Also, both wild type and mutant p53 may affect microRNA processing. Wild-type p53 has been shown to regulate microRNA maturation by facilitating Drosha-mediated processing of pri-miRNA into pre-miRNA (Liu et al., 2017) whereas mutant p53 may break biogenesis and decrease production of mature microRNAs not only by inhibiting the interaction between Drosha and pri-microRNA (Yao et al., 2021; Asmar et al., 2014; Speciale et al., 2020; Navarro & Lieberman, 2015), but also by inhibiting p63 and reducing Dicer expression (Liu et al., 2017).

MIR-34A, MIR-34B/C, MIR-129-1, MIR-129-2, MIR-143 and MIR-145 genes, which contain the sequence of the corresponding microRNAs, are p53-responsive. It has been proven that the promoters of all of these genes except MIR-203 contain p53-responsive elements triggering transcription by binding of 53 protein (Asmar et al., 2014; Poli, Seclí & Avalle, 2020; Rihani et al., 2015; Madrigal et al., 2021). The participation of p53 in the regulation of expression at the post-transcriptional level has been described for microRNAs miR-203 and the miR-143/145 cluster (Larrabeiti-Etxebarria et al., 2023; Yao et al., 2021; Asmar et al., 2014; Davis-Dusenbery & Hata, 2010; Bruijn et al., 2023).

The decrease in the expression of microRNAs miR-34a, miR-34b, miR-34c, miR-129, miR-203, miR-145, miR-143 in tumors with mutant p53 status has been shown (Navarro & Lieberman, 2015; Poli, Seclí & Avalle, 2020; Fang et al., 2023). Unfortunately, there is no large-scale current study of the expression repertoire of p53-responsive microRNAs depending on the mutational status of TP53 in tumor samples from patients with DLBCL.

However, mutations in the TP53 gene are verified in 20% of cases of DLBCL at the time of diagnosis, with an even higher frequency in relapsed lymphoma (Li et al., 2023).

Considering the mentioned above, impairment of p53 function as a result of mutations could potentially be one of the significant reasons for the decrease in the level of p53-responsive microRNAs of the miR-34, miR-203, miR-129 family and the miR-143/145 cluster described in DLBCL. An additional reason to think so is the fact that the most common hot-spots of mutations in the TP53 gene in DLBCL are codons 175 and 248. The mutations in codons 135 and 273 also commonly occur in this disease (Fig. 1A) (Bykov et al., 2018; Gurtner et al., 2016; Muller et al., 2014). In studies on various malignancies, missense mutations in these codons have been correlated with lower activity in processing pri-miRNA to pre-miRNA. In particular, p53 is associated with Drosha/DGCR8 through interaction with p68 and p72. Mutant p.C135Y, p.R175H, p.R273H and p.R248Q p53 disrupt the assembly of the complex with Drosha by sequestering p68, while mutant p.R175H and p.R273H variants of p53 inhibit Drosha activity by direct binding proteins p72 and p68 (Fig. 1B) (Gurtner et al., 2016). The observations of impaired processing of pre-miR to microRNA duplex under the influence of Dicer in the cytoplasm due to inhibition of p63 activity under the influence of p.R175H and p.R273H mutant p53 protein have been described (Muller et al., 2014).

In addition to mutations, of course, there are other mechanisms for dysfunction of the TP53 gene. However, they are less common in DLBCL. There are a small number of published studies which purpose was to carry out assessment of other types of aberrations that potentially lead to impaired microRNA expression in DLBCL. For example, it is known that loss of 17p (locus of the TP53 gene location) occurs in 16% (according to standard karyotyping) and TP53 methylation—in 5–6% of cases at the time of diagnosis of DLBCL. Often these aberrations are combined in lymphoma tissue (Voropaeva et al., 2019; Fiskvik et al., 2013; Voropaeva et al., 2023a).

The relationship between p53-responsive microRNA expression and host-gene expression

The relationship between p53-responsive microRNA expression and host-gene expression seems to be a less studied aspect. Transcription of microRNA genes is carried out by RNA polymerase (Pol II or Pol III). Intergenic miRNAs are transcribed independently using their own promoters, whereas intragenic miRNAs ones may be co-transcribed with the host gene. The dependence of miRNA expression on host gene expression and transcription factors has been demonstrated in several tumors. However, experimental evidence of such a correlation is still insufficient (Kaller et al., 2022).

The relationship between p53-dependent microRNAs expression and host gene expression also appears to be extremely complex. The genes MIR-34A, MIR-34B/C, MIR-145 and MIR-143, unlike MIR-129-1, MIR-129-2 and MIR-203, have an intragenic location, thus their expression potentially may depend on the expression host gene. However, at the present time there is no evidence of a relationship between the expression of the analyzed p53-responsive miRNA genes and their host genes in primary samples of DLBCL patients. This may be due to the following. All of the listed genes have been found to have their own promoters (Zhang et al., 2022; Cavard et al., 2023; Strmsek & Kunej, 2014; Tang et al., 2016). For example, for the genes of the miR-143/145 cluster, separate promoter regions and independent transcription factors have been identified (Zeinali et al., 2019; Pidíkova, Reis & Herichova, 2020). This is why, when CARMN is knocked down, the expression of miR-143 and miR-145 may decrease, but does not disappear completely (Zhang et al., 2022). Moreover, the transcriptional regulation of MIR-34B/C may be carried out not only from the promoter of the host gene, but also from the promoter of the oppositely oriented BTG4 gene (Cavard et al., 2023).

The significance of methylation of p53-responsive microRNA genes

The extensive analysis of genomic sequences of microRNA genes has shown that DNA methylation is one of the most common mechanisms for disrupting their transcription in malignant neoplasms (Zeinali et al., 2019). The microRNAs we describe are no exception. The promoters MIR-34A, MIR-34B/C, MIR-129-2, MIR-203, MIR-143 and MIR-145, unlike MIR-129-1 gene, also contain CpG islands, the aberrant methylation of which causes a decrease in microRNA expression in a wide range of neoplasms (Voropaeva et al., 2023d; Voropaeva et al., 2022b; Voropaeva et al., 2023c; Benati et al., 2017; Wong et al., 2013; Gao et al., 2016; Hother et al., 2012; Chim et al., 2011; Voropaeva et al., 2023b).

Methylation of microRNAs of the miR-34 family in DLBCL tumor tissue has been analyzed in several studies. According to the obtained results, the frequency of methylation in systemic DLBCL ranged from 23–28% for the MIR-34A gene and 55–78% for the MIR-34B/C gene (Asmar et al., 2014; Voropaeva et al., 2022a; Wong et al., 2013), while in DLBCL of central nervous system it was 57% and 95.2%, respectively (Munch-Petersen et al., 2016). According to literature methylation frequencies in the studied DLBCL samples were 65% for MIR-129-2 and 66% for MIR-203 (Voropaeva et al., 2022a). Moreover, comprehensive study of the gene methylation status in DLBCL samples showed that methylation of MIR-34A, MIR-34B/C, MIR-129-2 and MIR-203 in the tissue of affected lymph nodes of patients with DLBCL was a tumor-specific and combined phenomenon (Voropaeva et al., 2023a).

The works on quantitative assessment of the level of methylation of p53-responsive microRNA genes are of great interest. In particular, the data obtained by the Reduced Representation Bisulfite Sequencing method, which indicate an increase in the level of methylation of genes encoding miR-129 and miR-203 in relapsed DLBCL samples in comparison with diagnostic samples are very noteworthy and may signify an increasing role of methylation during tumor progression of lymphoma (Leivonen et al., 2017).

The above data indicate that methylation of microRNA genes in DLBCL appears to be a very promising therapeutic target. The reversibility of aberrant methylation of the promoters of the described genes was demonstrated in lymphoid tumor cell lines. Treatment of tumor cells with demethylating agents led to restoration of the level of mature microRNAs, inhibition of cell proliferation and triggering of cell death by apoptosis (Wong et al., 2013; Chim et al., 2011).

Thus, aberrant methylation may be one of the significant mechanisms of impaired expression of microRNAs miR-203, mir-129, miR-34a, miR-34b and miR-34c in DLBCL. At the same time, the methylation data for the miR-143/145 cluster are fundamentally different. Studies of MIR-143 and MIR-145 methylation in lymphoid tissue have shown that, on the one hand, the methylation of these genes occurs in reactive lymphadenopathy and thus is not tumor-specific feature in DLBCL. On the other hand, quantitative analysis has revealed a greater range of values of the average level of methylation of the MIR-143 gene in samples from patients with DLBCL, but in general it was significantly lower than the values in samples from patients with reactive lymphadenitis (Voropaeva et al., 2023c; Voropaeva et al., 2023b). This, along with the inconsistency of data on the direction of changes in microRNA expression in DLBCL tumor tissue, indicates the need for further research into the mechanisms of epigenetic regulation of miR-143/145 cluster (Akao et al., 2007; Lawrie et al., 2009; Voropaeva et al., 2023c; Voropaeva et al., 2023b; Roehle et al., 2008; Fischer et al., 2011; Vaisitti et al., 2018).

Copy number alteration in the loci of p53-responsive microRNA genes

According to recent data, up to half of all miRNA genes are located in genome regions that undergo amplification, deletion, or other structural rearrangements (Zeinali et al., 2019), and one of the possible reasons for the decreased expression of p53-responsive microRNAs in DLBCL may be the loss of entire chromosomes or deletion of chromosomal loci. For example, the MIR-129-1 gene is located in the fragile FRA7H site of chromosome 7 and is often lost in tumors (Gao et al., 2016), whereas recurrent chromosomal rearrangements involving the long arm of chromosome 14, where the MIR-203 gene is located, and deletion of 11p11.2, the location of the MIR-129-2 gene, are common in lymphomas including DLBCL (Aya-Bonilla et al., 2011; Aya-Bonilla et al., 2013; Ricketts, Carter & Coleman, 2003). It is shown that four protein coding genes oncosuppressors, including p53-induced protein (PIG11), also localized within 11p11.2-p12 (Ricketts, Carter & Coleman, 2003).

Interesting data from Zhu et al. (2000) showed that the deletion frequency of 11q23 (the location locus of MIR-34B/C) in a group of 17 DLBCL samples was very low (only 4.2%). At the same time, all three studied cases of transformation of indolent lymphoma into DLBCL, called the Richter syndrome, had a deletion of 11q23. This may indicate that gene loss is a late event in the evolution of lymphoma. In the work of Hezaveh et al. (2016), whole-genome-derived copy number analysis did not reveal copy number violations in the promoter regions or regions of the miR-143/145 cluster genes in any of the 19 analyzed samples. As can be seen in the available literature, only a few studies of copy number disturbances of the microRNA genes in small sets of samples are described, which increases the practical significance of the analysis of available databases accumulating information on the molecular genetic characteristics of tumors.

It should be noted that karyotyping is widely used in hematology and is the main method used to search for and characterize chromosomal abnormalities in DLBCL (Hezaveh et al., 2016; Wang & LaFramboise, 2019). In this regard, we analyzed data from the Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer (Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer, 2024), which provides data on 1,612 cases of DLBCL with cytogenetic abnormalities. We discovered that the loss of entire chromosomes or deletions of their sections in the locations of the MIR-34A, MIR-34B/C, MIR-129-1, MIR-129-2, MIR-203 genes and the miR-143/145 cluster occurred at 3.1%, 4.4%, 3.4%, 3.2%, 0.1% and 4.3%, respectively (Fig. 2A). In terms of all DLBCL cases including cases without cytogenetic aberrations these values may seem even less significant.

However, the classical cytogenetic method is known to have limitations. In particular, karyotyping does not allow detecting small chromosomal rearrangements that are outside the resolution range of the method (Shah et al., 2017). In this regard, the data presented in the cBioPortal for Cancer Genomics database are of great interest (Gao et al., 2013). The results of profiling of 396 DLBCL samples using Affymetrix SNP 6.0 microarrays were analyzed. It turned out that a decrease in genome copy number in the chr1:9,211,727-9,211,836 locus (GRCh37/hg19 by NCBI Gene), where the MIR-34A gene is located, occurred in 19% of cases. This value is lower than data by Gru et al. (2013), who described a high incidence of microdeletion of the 1p36 region in non-Hodgkin lymphomas. According to their data, nine out of 32 (28%) studied DLBCL cases had 1p36 microdeletion by fluorescence in situ hybridization (FISH) technique, not recognized by conventional cytogenetics (Gru et al., 2013). Such differences in frequency may be primarily due to different sample sizes and composition, as well as methodological approaches.

Further, according to data, presented in the cBioPortal for Cancer Genomics database, the reduction of genome copy number in the gene loci MIR-34B/C (chr11:111,383,663-111,384,240), MIR-129-1 (chr7:127,847,925-127,847,996), MIR-129-2 (chr11:43,602,944-43,603,0 33), MIR-203 (chr14:104,583,742-104,583,851), MIR-143 and MIR-145 genes (chr5:148,808,481-148,810,296), were observed in 5%, 13%, 16%, 14% and 12% of samples, respectively (Fig. 2B) (Gao et al., 2013; Taylor et al., 2018).

These data seem be extremely interesting. First, deletion or the loss of entire chromosome are the causes of MIR-34B/C gene loss, whereas the decrease in copy number of other described genes appears to be associated with microdeletions. Second, a decrease in copy number as a cause of decreased microRNA expression may have the greatest importance for miR-34a microRNA, and to a lesser extent for miR-129, miR-143, miR-145 and miR-203, whereas for miR-34b and miR-34c this mechanism may be insignificant. Third, as can be seen from Fig. 3B, the combined loss of two or more genes of p53-responsive microRNAs often occurs in DLBCL. The combination of deletions is statistically significant even after correction for multiple comparisons for the gene pairs indicated in Table 2.

Variants of the nucleotide sequence of p53-responsive microRNA genes

Until recently, very little attention was paid to changes in the nucleotide sequence of the genome outside the locations of protein-coding genes, because currently it is extremely difficult to assess their functional effect. The capabilities of bioinformatics for such an assessment are extremely limited. In this regard, the data on the frequency, structural and functional effect of somatic or germline variants of nucleotide sequence in miRNA genes are currently very limited (Machowska, Galka-Marciniak & Kozlowski, 2022). However, potential effects of mutations or single nucleotide polymorphisms in various parts of the miRNA gene sequence may be changes in the processes of transcription, biogenesis of miRNAs and their targets recognition. Thus, the decrease in miRNA expression may occur due to the mutations in the microRNA gene promoter or any part of the microRNA precursor sequence, including mutations in DROSHA/DICER1 cleavage sites or miRNA duplex, mutations disrupting/creating new PB motifs, etc. (Machowska, Galka-Marciniak & Kozlowski, 2022).

In the available literature, we have found only single descriptions of the spectrum of microRNA mutations that were detected at both the DNA and RNA levels in DLBCL samples. Using sequencing datasets from a large cancer genomics project, The Cancer Genome Atlas (TCGA), 36 miRNA mutations were described in 37 DLBCL samples. Most of them (92.7%) were found as substitutions, another 2.8% were nucleotide insertions. At the same time, 40.5% of the samples did not have any microRNA mutation. Among the mutated miRNA genes in DLBCL, the genes MIR-34A, MIR-34B/C, MIR-129-1, MIR-129-2, MIR-203 and the miR-143/145 cluster are not included, unlike miR-1324 and miR-142 (Urbanek-Trzeciak et al., 2020; Galka-Marciniak et al., 2021). In one of the studies in a set of 19 samples of DLBCL, the authors reported the detection of mutations in the genes of only three microRNAs, namely miR-142, miR-612 and miR-4322 (Hezaveh et al., 2016). In Kwanhian et al. (2012), no other mutations except in miR-142 were observed in the analysis of 56 lymphoma samples (Kwanhian et al., 2012).

These data suggest that nucleotide sequence variants in the p53-responsive miRNA genes we analyzed are at least a rare event in DLBCL. However, it is undoubtedly necessary to conduct studies with a large sequencing depth and a large number of DLBCL samples included as well as to develop new methods of statistical and bioinformatics processing in order to clarify the frequency and improve understanding of the effects of genetic variants in microRNA genes.

Mutations in microRNA biogenesis genes

The regulation of miRNA biogenesis is complex, tightly controlled, and may be disrupted by various factors such as mutations or epigenetic modifications, and the expression level is the outcome of the transcription and subsequent biogenesis of microRNAs, in which a large number of proteins are involved (Gulyaeva & Kushlinskiy, 2016). All microRNA biogenesis genes are participants of several steps, namely nuclear step of miRNA biogenesis (SMAD2/4, FUS, SRSF3, DROSHA, DGCR8, DDX5/17 and GSK3B), export to cytoplasm (XPO5 and RAN), cytoplasmic step of miRNA biogenesis (DICER1, TARBP2, PRKRA, ADAR, KHSRP, LIN28A/B, TUT4/7, DIS3L2) and miRNA functioning (TNRC6A, AGO1/2/3/4, FMR1, MOV10, GEMIN4) (Galka-Marciniak et al., 2021; Annese et al., 2020). It has been reported that mutation frequencies in some of these genes specifically and significantly increased in certain types of cancer. However, none of the 29 microRNA biogenesis genes could be classified as over mutated genes in DLBCL (Jafari et al., 2015).

Analysis of whole exome sequencing data sets of 37 paired tumor and normal samples from patients with DLBCL, available in the TCGA repository, has shown the presence of mutations in only seven (TNRC6A, MOV10, DGCR8, DIS3L2, GEMIN4, LIN28B and RAN) out of 29 microRNA biogenesis genes. In total, 16% of samples have had mutations in microRNA biogenesis genes (Galka-Marciniak et al., 2021).

According to the analysis of high-throughput sequencing data of 1295 DLBCL samples presented in the cBioPortal for Cancer Genomics database, aberrations rarely detected in all genes except for three, namely MOV10, PRKRA and DDX5 (Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer, 2024). A total of 6.9% of samples had aberrations in microRNA biogenesis genes, while disorders that are presumably drivers of the tumor process (missense and truncating mutations, deep deletions) were identified in a small number of cases (Fig. 3). However, two samples of DLBCL are noteworthy, in which a significant combination of aberrations was identified in six functionally related genes of the nuclear (DROSHA, DGCR8 and GSK3B) and cytoplasmic (TUT4, LIN28 and ADAR) stages of microRNA biogenesis.

The function of the affected in these two samples of DLBCL genes is briefly described below. Drosha and DGCR8 are considered as major machinery components of miRNA biogenesis nuclear steps (Jafari et al., 2015). The GSK3B gene encodes the GSK3 β kinase, which has been shown recently to bind to DGCR8 and p72 in the microprocessor and thus stimulate the Drosha-cofactor and Drosha-pri-miR interaction (Fletcher et al., 2017).

The Lin28a protein, whose biochemical activity is indistinguishable from that of the homologue Lin28b, is a regulatory factor in the cytoplasmic stage of microRNA biogenesis. Lin28 binds to pre-microRNA after export from the nucleus to the cytoplasm. Subsequently, another protein (TUT4 protein) recognizes the Lin28-pre-miRNA complex and performs its uridylation, which makes the pre-miRNA resistant to Dicer processing. It has also been reported that Lin28 proteins interfere with the processing of pri-miRNA to pre-miRNA by Drosha (Heo et al., 2009). Abnormalities in the TUT4 and LIN28 genes, encoding specific suppressors of miRNA biogenesis, are involved in carcinogenesis.

The adenosine deaminase acting on RNA (ADA) enzyme converts adenosine to inosine in double-stranded RNAs, including microRNAs duplex, which leads to disruption of miRNA maturation process and activity (Tomaselli et al., 2013). ADAR knockout can cause hematological tumors in animal models (Correia De Sousa et al., 2019).

The mutations found in these genes are regarded as variants of unknown significance, but their potential effect may be a disruption of the basic biogenesis of microRNAs, and an additional study is required to prove it.