Long noncoding RNAs in intestinal homeostasis, regeneration, and cancer

1 INTRODUCTION

The intestinal epithelial layer forms a crypts–villus structure to maximize the absorption of water and nutrients as well as form a protective barrier against harsh luminal environment, pathogenic microorganisms, and ingested toxins. To maintain the barrier function, the intestinal epithelium is renewed every 3–5 days by a small number of stem cells (SCs) at the bottom of the crypt (Gehart & Clevers, 2019).

Intestinal epithelial injury can happen in various pathological settings, such as sepsis, burn injury, radiation injury, and inflammatory bowel diseases (IBD). These injuries may cause exacerbated inflammation in the intestinal mucosa. Regenerative response is critical for wound healing of damaged intestinal tissue. During regeneration, replenishment of epithelial cell loss is ensured by the proliferation of crypt SCs and their progenitors in response to inflammation (Varga & Greten, 2017). Intestinal regeneration/wound healing response involves a delicately tuned, self-limiting molecular event orchestrated by the transient activation of specific signaling pathways. While precise regulation of these pathways is indispensable for the repair of damaged tissue, aberrant activation of wound healing processes may promote cancer, including intestinal cancer (MacCarthy-Morrogh & Martin, 2020). In humans, colorectal cancer (CRC) represents the main type of intestinal cancer, which is one of the leading human malignancies and cause of cancer-related mortality (Haggar & Boushey, 2009). Several signaling pathways and molecular mechanisms that are essential for intestinal homeostasis, regeneration, and repair have been found to be dysregulated both in chronic inflammation and intestinal cancer. For example, activation of Wnt signaling is required for homeostasis and injury-induced regenerative response, whereas deregulated Wnt signaling is a hallmark of intestinal cancer (Pinto et al., 2003; Schepers & Clevers, 2012). Thus, understanding the precise mechanism of intestinal homeostasis and regeneration may provide insights into delineating the mechanisms involved in intestinal tumorigenesis.

With the recent advancement and implementation of high-throughput sequencing technologies for genomics and transcriptomics, it is now apparent that <2% of the genome translates proteins, whereas about 75% is actively transcribed into noncoding RNAs, which include small interfering RNA (siRNA), microRNA (miRNA), piwi-interacting RNA (piRNA), and long noncoding RNA (lncRNA) (Djebali et al., 2012; Pandya et al., 2020). Among these noncoding RNAs, lncRNAs portray a distinct transcript family that is more than 200 nucleotides and do not have protein-coding capacity (Derrien et al., 2012). LncRNAs are produced from promoter regions, antisense sequences, enhancer sequences, 3ʹ and 5ʹ untranslated regions (UTRs), exons, introns, intergenic and intragenic regions of the genome (Pandya et al., 2020). LncRNAs are expressed at low levels but exhibit high tissue and cell specificity (Derrien et al., 2012; Han et al., 2015). LncRNAs are localized in both nucleus and cytoplasm, with a modest nuclear propensity (Cabili et al., 2015; Ulitsky & Bartel, 2013). The unique localization of lncRNAs determines their functions by allowing them to bind and regulate different molecules. For example, nuclear lncRNAs interact with genomic DNA, transcription factors, and nuclear proteins and are involved in transcriptional and epigenetic regulation (Schmitt & Chang, 2017) whereas, cytoplasmic lncRNAs are involved in the transcriptional and posttranscriptional regulation of key components of epigenetic and signaling pathways (Schmitt & Chang, 2016). As lncRNAs can regulate various cellular processes, it is not surprising that lncRNAs have emerged as important regulators of various biological processes related to tissue homeostasis and pathological conditions, including cancer (Schmitz et al., 2016). Several lncRNAs have been identified in cancer with anti- or protumorigenic potential (Arun et al., 2018); thus, a comprehensive understanding of the function of lncRNAs in normal tissue homeostasis as well as during tumorigenesis will help to exploit them in diagnosis and as therapeutic targets. In this review, we present some of the recently identified lncRNAs and their functions in the regulation of intestinal homeostasis, regeneration, and intestinal cancer initiation and progression. We also summarize lncRNAs as potential diagnostic and therapeutic markers in intestinal pathologies such as IBDs and intestinal cancer.

2 LNCRNAS IN INTESTINAL HOMEOSTASIS AND REGENERATION

Intestine faces constant mechanical stress and environmental assault from ingested chemicals and microorganisms and their toxins, which take a substantial toll on the epithelia, thus imposing a stringent requirement for epithelial renewal. It is estimated that about 300 million cells die every day at the surface epithelia of the intestine (Leushacke & Barker, 2012). To compensate for the loss of these dying cells, intestinal epithelia must proliferate and generate a similar number of cells to restore homeostasis. This regeneration relies on intestinal stem cells (ISCs) at the crypt base, which are marked by a specific gene leucine-rich repeat containing G protein-coupled receptor 5 (Lgr5). Lgr5-expressing ISCs are now considered as bona fide ISCs as elegant lineage-tracing experiments have demonstrated their self-renewal and ability to generate all distinct cell types in the intestine both in vivo during homeostasis as well as in the 3D organoid system (Barker et al., 2007; Sato et al., 2009). Activation of Wnt signaling plays a crucial role in maintaining the balance between stemness and differentiation of these ISCs (Gehart & Clevers, 2019). However, the molecular mechanism that controls the stemness of Lgr5⁺ ISCs has been poorly understood until recently. A recent report suggested the role of Gata6 lncRNA (lncGata6) in the maintenance of Lgr5⁺ ISCs stemness (Zhu et al., 2018). LncGata6 is a divergent transcript expressed from the promoter of Gata6 and is highly conserved in different species. Interestingly, lncGata6 expression was specifically enriched in the Lgr5⁺ mouse and human ISCs and CRISPR/Cas9-mediated depletion of lncGata6 in intestinal organoids greatly affected the growth of organoids. Accordingly, in vivo depletion of lncGata6 impaired the stemness of ISCs and intestinal homeostasis (Zhu et al., 2018). Mechanistically, lncGata6 recruits nucleosome remodeling factor (NURF) complex to the promoter of ETS family transcription factors, enabling the expression of Ehf (ETS homologous factor). Eventually, Ehf enhances the expression of Lgr4/5 receptors, which subsequently activates Wnt signaling to induce stemness to maintain intestinal homeostasis. Additionally, lncGata6 also regulates the expression of transcription factors Myb and Sox9 (Zhu et al., 2018), which are known to regulate ISC-related genes and self-renewal (Cheasley et al., 2011; Furuyama et al., 2011). In addition to lncGata6, another lncRNA from ultraconserved regions (UCRs) was shown to regulate intestinal epithelial renewal. uc.173, a lncRNA transcribed from UCRs, regulates intestinal growth and renewal by suppressing miRNA195 expression through the decrement of primary miR-195 (Xiao et al., 2018). The molecular mechanism by which the uc.173–miRNA195 axis regulates epithelial renewal remains unclear; however, this could be through the production of cell cycle regulators (CDK4, CCND1, CDK6, and WEE1), which are known to be negatively regulated by miRNA195 (Bhattacharya et al., 2013; Xu et al., 2009).

The intestinal epithelial cells (IECs) constitute an intestinal barrier which is consists of a mucus layer, which acts as a physical barrier between the gut microbiota and epithelial layer, and interepithelial tight junctions (TJs) and adherens junctions (AJs) (Odenwald & Turner, 2017). Defects in these barriers lead to spontaneous intestinal inflammation and colitis or may increase the sensitivity to toxic agents such as DSS (Gupta et al., 2014; Laukoetter et al., 2007; Odenwald & Turner, 2017; Schmitz et al., 1999; Van der Sluis et al., 2006). Accumulating evidence suggests that lncRNAs play a crucial role in the regulation of intestinal barrier function. LncRNA SPRY4-IT1 (an intronic transcript from SPRY4 gene), a 687-base pair transcript (Khaitan et al., 2011), which is highly expressed in several human tissues including IECs (Khaitan et al., 2011; Xiao et al., 2016). SPRY4-IT1 is expressed in both cytoplasm and nuclear fraction of IECs and regulates intestinal barrier function without affecting IECs viability and proliferation. Mechanistically, SPRY4-IT1 directly binds with several TJs messenger RNA (mRNAs) to stabilize and enhance their translation, thus promoting the function of the epithelial barrier (Xiao et al., 2016). SPRY4-IT1 also interacts with RNA-binding protein (RBP) HuR, which further enhances the stability and translation of TJ mRNAs (Sharma et al., 2013; Xiao et al., 2016; Yu et al., 2011).

Another lncRNA H19 also regulates epithelial barrier function without affecting IECs viability. LncRNA H19 is transcribed from the H19/igf2 gene locus. Overexpression of H19 is associated with several pathological conditions and malignancies (Brannan et al., 1990; Luo et al., 2013; Shermane Lim et al., 2021). LncRNA H19 modulates intestinal barrier function by suppressing TJ and AJ proteins (ZO-1 and E-cadherin) posttranscriptionally. Two miRNAs, miR-675-5p and miR-675-3p, are encased in exon 1 of H19. Accordingly, overexpression of H19 specifically induces the expression of both miR-675-5p and miR-675-3p. Both miR-675-5p and miR-675-3p directly interact with ZO-1 and E-cadherin to decrease the stability and translation of their mRNAs and thus decrease epithelial barrier function (Zou et al., 2016). In addition to epithelial barrier function, H19 also regulates the physical barrier by exerting a suppressive effect on Paneth cells and goblet cells. This effect seems to be mediated through, at least partially, suppression of autophagy (Yu et al., 2020); however, how H19 could regulate autophagy is currently unknown. Mutations in autophagy protein ATG16L1 confer an increased risk of the development of Crohn's disease (Hampe et al., 2007) and deletion of ATG16L1 or hypomorphic ATG16L1 expression are both linked with dysfunctional Paneth and goblet cells (Lassen et al., 2014). NFkB-IKKα phosphorylates and stabilizes ATG16L1, thereby promoting epithelial regeneration after acute damage (Diamanti et al., 2017). Thus, it is plausible that H19 may act directly or indirectly via miR-675 to regulate NFkB–IKKα–ATG16L1 axis to control Paneth cell and goblet cell function.

The epithelial barrier function has a protective role in chemical and pathological stress-induced intestinal damage and regeneration as well as in pathology of IBD, such as ulcerative colitis and Crohn's disease (Gupta et al., 2014; Odenwald & Turner, 2017). During inflammation and pathological stress expression level of lncRNA H19 significantly increases (Geng et al., 2018; Yu et al., 2020; Zou et al., 2016). Upon pathological stress instigated by cecal ligation and puncture (CLP), H19 expression increases substantially, which correlates with epithelial barrier dysfunction due to markedly reduced expression of ZO-1 and E-cadherin (Yu et al., 2020). Ubiquitous RBP HuR interacts with H19 (Keniry et al., 2012) and prevents miR-675 processing from H19 and augments epithelial barrier function (Zou et al., 2016). Accordingly, in vivo HuR ablation increases miR-675 expression and obstructs epithelial barrier regeneration in an acute mesenteric ischemia/reperfusion (I/R) mouse model (Zou et al., 2016). In line with this, a recent report suggested that increased H19 levels in the intestinal tissue of ulcerative colitis patients correlate with reduced levels of vitamin D receptor (VDR) and an increased expression of miR-675 (Chen et al., 2016). miR-675 directly binds to ZO-1, E-cadherin and VDR mRNA and decreases their stability, thus affecting epithelial barrier function (Chen et al., 2016; Zou et al., 2016). On the contrary, Gend et al. (2018) reported that increased expression of H19 in the intestinal mucosa of ulcerative colitis patients and in response to inflammatory cytokine IL-22 is not associated with alterations of epithelial barrier function in vivo, but rather linked to enhanced proliferation and regeneration of intestinal epithelia. Mechanistically, the inflammatory cytokine IL-22 triggers the expression of H19 in a time- and dose-dependent manner through activation of the STAT3 pathway and protein kinase A (PKA) independently of cAMP. Inflammation-induced H19 expression markedly attenuates p53 activity, a transcription factor that mediates growth inhibition, and promotes the expression of MYCN and FOXM1, which contributes to the regeneration of intestinal epithelia (Geng et al., 2018). The discrepancies between Zou et al. (2016) and Gend et al. (2018) may be due to differences in model systems used and require further investigation.

Another lncRNA, uc.173 was recently shown to regulate intestinal barrier function. In a cell culture model, uc.173 regulates the expression of TJ protein claudin-1 by enhancing the translation of its mRNA. uc.173 binds with miR-29b and functions as a decoy RNA for miR-29b, thus reducing the accessibility of miR-29b. miR-29b interacts with the 3ʹ UTR of claudin-1 mRNA and represses claudin-1 translation, thus reducing epithelial barrier function. Accordingly, uc.173 controls intestinal barrier function by regulating claudin-1 expression during CLP-induced stress in vivo (Wang, Cui, et al., 2018). In line with this, expression of uc.173 in the intestinal tissues obtained from Crohn's disease patients displayed a substantial reduction in uc.173 compared with healthy controls, which was also correlated with reduced expression of TJ proteins and reduced proliferation (Xiao et al., 2018). Similarly, expression of lncRNA SPRY-IT1 was significantly reduced in patients of ulcerative colitis, which have increased intestinal permeability, and correlated with decreased expression of several TJ proteins. Accordingly, overexpression of SPRY-IT1 by lentiviral infection in mice prevented the reduction of TJ proteins and thus improved intestinal barrier function in mice model of CLP-induced barrier dysfunction (Xiao et al., 2016).

The abovementioned studies highlight the role of several lncRNAs during intestinal homeostasis, injury-induced regeneration, and pathological conditions (Figure 1). Cancer cells exhibit great similarities with regenerative responses, such as activation of SCs, enhanced proliferation as well as dysfunctional barrier function. Several signaling pathways involved in regeneration are found to be deregulated in various types of cancer, including intestinal cancer (Sundaram et al., 2018). Thus, it is likely that lncRNAs that regulate the regenerative response may also participate in intestinal tumorigenesis. In the next section, we focus on lncRNAs that are involved in intestinal tumorigenesis.

3 LNCRNAS IN INTESTINAL CANCER INITIATION AND PROGRESSION

There exists a delicate balance between regeneration following tissue injury and tumorigenesis, as similar signaling pathways have been shown to contribute to both processes. Regenerative response upon injury is a self-limiting process, while tumor initiation is categorized by the constitutive or aberrant activation of signaling pathways that regulate tissue homeostasis and injury-induced regeneration (Sundaram et al., 2018). In the intestine, Wnt signaling is indispensable to maintain homeostasis (Beumer & Clevers, 2016; Schepers & Clevers, 2012), whereas aberrant activation of Wnt signaling by mutation in APC or CTNNB (β-catenin) represents one of the first steps of intestinal tumorigenesis (Fearon & Vogelstein, 1990; Morin et al., 1997). Interestingly, activating mutations in ISCs lead to rapid development of intestinal adenomas, suggesting that ISCs are the origin of intestinal tumors (Barker et al., 2009; Sangiorgi & Capecchi, 2008; Zhu et al., 2009), where Lgr5⁺ ISCs seem to be the source of cancer stem cells (CSCs) in intestinal tumors (de Sousa e Melo et al., 2017).

Many lncRNAs have been identified that are expressed in human cancers, including intestinal cancer (Arun et al., 2018; Kim & Croce, 2018; Sharma et al., 2020). Several in vitro studies using human cancer cell lines show that lncRNAs can affect hallmarks of cancer including proliferation, apoptosis, epithelial to mesenchymal transition and metastasis (Chen & Shen, 2020; Li et al., 2017; O'Brien et al., 2020; Schmitt & Chang, 2016). In addition to these processes, lncRNAs can also regulate tumorigenesis by regulating tumor stemness (Ma et al., 2019). For example, lncGata6 is distinctly expressed in human intestinal cancer and its expression, consistent with homeostatic conditions, remnants specific to Lgr5⁺ CSCs (Zhu et al., 2018). Importantly, in vivo ablation of lncGata6⁺ tumor cells in intestinal cancer derived from patient-derived tumor organoids (PDTOs) resulted in a substantial regression of tumor mass, mimicking the effect of in vivo depletion of Lgr5⁺ CSCs (de Sousa e Melo et al., 2017; Zhu et al., 2018). Not only ablation of lncGata6⁺ tumor cells, targeting of lncGata6 transcripts either by CRISPR-Cas9 in vivo, antisense oligonucleotides (ASOs) in PDTOs also resulted in the reduction of intestinal tumor proliferation and growth. During homeostatic condition, as mentioned before, lncGata6 regulates Wnt signaling through NURF–Ehf axis mediated upregulation of Lgr4/5 receptors (Zhu et al., 2018). Secreted factor R-Spondins bind to the Lgr4/5 receptor to potentiate Wnt/β-catenin signaling (Carmon et al., 2011; Glinka et al., 2011). Given the fact that intestinal tumors harbor mutations in APC or CΤΝΝΒ, which are downstream of Lgr4/5 receptors, it is difficult to explain mechanistically the effects of targeting lncGata6. Thus, the observed effect of lncGata6 knockdown on tumor growth could be due to yet not identified targets of the NURF–Ehf axis, which should be investigated in future research.

c-Myc is a bona fide target gene of Wnt/β-catenin signaling that plays an essential role in intestinal tumorigenesis (Sansom et al., 2007; Wilkins & Sansom, 2008). Expression of c-Myc is deregulated in more than 50% of all cancers and is associated with poor prognosis and unfavorable survival; thus, therapeutic targeting of Myc holds great promise for the therapy of many cancers, including intestinal cancer. Attempts have been made to target Myc but have been unsuccessful due to its undruggable protein structure (Chen et al., 2018). Recently, MYU (c-Myc upregulated lncRNA) has been recognized as a bona fide target of Wnt/c-Myc axis in intestinal tumors (Kawasaki et al., 2016). Myc directly binds to the E-box-2 promoter region to induce MYU expression. Furthermore, MYU binds to RBP heterogeneous nuclear ribonucleoprotein-K (hnRNP-K) and stabilizes the expression of cyclin-dependent kinase 6 (CDK6), thus promoting cell cycle progression. Importantly, xenotransplantation of MYU-knockdown colon cancer cells displayed significant retardation of tumor growth in vivo (Kawasaki et al., 2016). Conversely, c-Myc can be regulated by lncRNA GLCC1 (glycolysis-associated lncRNA). GLCC1 directly interacts with HSP90, a protein chaperone that regulates the stability of many proteins, which further binds to c-Myc and deubiquitinating enzyme USP22 to stabilize c-Myc expression by preventing its proteasomal degradation. Accordingly, knockdown of GLCC1 substantially reduced tumor growth by inhibiting the proliferation of tumor cells in a xenograft mouse model. Importantly, expression of GLCC1 was significantly higher in intestinal tumor compared with adjacent tissue and correlated with a poor prognosis (Tang et al., 2019). Another lncRNA CCAT2 also upregulates c-Myc expression through TCF7L2 to further enhance the Wnt signaling activity in colon cancer (Ling et al., 2013). CCAT2 also induces chromosomal instability by regulating the expression of BOP1 at protein level as well as at transcriptional level through c-Myc (Chen et al., 2020). Thus, the aforementioned studies suggest that targeting of GLCC1, CCAT2, or MYU, instead of c-Myc, could be exploited as a therapeutic target in Myc-dependent intestinal tumors.

Several other lncRNAs have been identified that are implicated in the development and progression of intestinal cancer. For example, lncRNA CASC11 binds to hnRNP-K to activate canonical Wnt signaling and its targets c-Myc and cyclin-D1. Subsequently, c-Myc establishes a positive feedback loop and interacts with CASC11 promoter regions and enhances promoter histone acetylation to stimulate CASC11 expression. Notably, expression of CASC11 was upregulated in intestinal cancer tissues, which also correlated with metastasis (Zhang et al., 2016). Similarly, the expression of another lncRNA, NEAT1 is upregulated in intestinal cancers and the expression of NEAT1 correlates with advanced clinical features. NEAT1 functions as a competing endogenous RNAs (ceRNAs) to modulate the miR-34a/SIRT1 axis, leading to stimulation of canonical Wnt signaling (Luo et al., 2019). LncRNA HNF1A-AS1 also contributes to the activation of Wnt/β-catenin signaling by sponging miR-34a and subsequent suppression of miR-34a/SIRT1/p53 feedback loop (Fang et al., 2017).

The epithelial barrier also plays an important role in the progression of intestinal tumors. Both early- and late-stage intestinal tumors display reduced expression of several barrier proteins. Defects in tumor barrier function cause tumor infiltration by microbial products that elicit tumor-associated inflammation by activating the tumoral IL-17 response, which promotes tumor growth and progression (Grivennikov et al., 2012). The exact mechanism for tumor barrier dysfunction remains to be investigated; however, given the fact that several lncRNAs regulate intestinal barrier function during homeostasis and in pathological stress conditions, it is likely that lncRNAs may also play a crucial role in the regulation of the intestinal tumor barrier. In line with this idea, the expression of lncRNA H19, which plays a vital role in intestinal barrier function (Chen et al., 2016; Geng et al., 2018; Yu et al., 2020; Zou et al., 2016), was found to be upregulated in intestinal tumors (Ohtsuka et al., 2016; Wang et al., 2018); however, a causal link between increased H19 expression in tumors and tumor barrier dysfunction remains to be established. Nevertheless, H19 has been shown to affect the phosphorylation of retinoblastoma 1 (RB1) by affecting CDK4 and CCND1 expression. H19 also regulates CDK8 expression to affect β-catenin activity, thus controlling the proliferation of intestinal tumor cells by two distinct mechanisms (Ohtsuka et al., 2016).

Aberrant activation of Wnt signaling is a critical event for the development of intestinal adenomas (Fearon & Vogelstein, 1990; Morin et al., 1997); however, additional mutational hits (activating mutations in KRAS and inactivating mutations in TP53, SMAD4, and PTEN), which could be due to increased reactive oxygen species production by infiltrating myeloid cells in the tumor microenvironment (Canli et al., 2017), are required to gain invasive and metastatic properties (Fearon & Vogelstein, 1990; Janssen et al., 2006; Luo, Brooks, et al., 2009; Marsh et al., 2008). Epithelial–mesenchymal transition (EMT), a developmental program, plays a decisive role in tumor metastasis by enhancing cell mobility, invasion, and resistance to apoptosis (Mittal, 2018). A recent study reported that lncRNA SATB2-AS1 regulates EMT by recruiting the histone acetyltransferase p300 to SATB2 promoter to enhance the expression of SATB2, which subsequently recruits histone deacetylases (HDAC) to the Snail promoter to suppress Snail transcription and thus inhibit EMT. In line with this, expression of both SATB2-AS1 and SATB2 was downregulated in intestinal carcinoma with metastasis compared with those without metastasis and associated with shorter overall survival and poor prognosis (Wang et al., 2019). LncRNAs can also enhance metastasis of intestinal tumors by promoting protumorigenic, macrophage M2 polarization in a paracrine manner. Intestinal tumor cell-derived exosomes contain a lncRNA RPPH1 that mediates macrophage M2 polarization and thereby promotes metastasis of intestinal tumors (Liang et al., 2019).

In addition to metastasis, lncRNAs can also play a critical role in developing drug resistance during tumor progression. Adjuvant chemotherapies with 5-fluorouracil based regimens have been shown to provide overall survival benefit in subgroups of Stages II and III patients; however, so far, little is known about the predictors of these subgroups (Nelson & Benson, 2013). A recent study (Ma et al., 2016) reported that lncRNA CCAL is highly expressed in intestinal cancer and its overexpression correlates with worse survival and predicts a poor response to adjuvant chemotherapy. CCAL promotes intestinal cancer progression by inducing multidrug resistance (MDR) by modulating activator protein 2α (AP-2α), which then stimulates canonical Wnt signaling and MDR1/P-glycoprotein expression. Reduced histone H3 methylation and increased acetylation in the CCAL promoter may contribute to elevated CCAL expression in advanced intestinal cancer (Ma et al., 2016). In addition, the acquisition of de novo resistance to anti-EGFR treatment cetuximab has been attributed to several acquired mutations, especially KRAS activating mutations (Bertotti et al., 2015; Misale et al., 2014). A fraction of patients with wild-type KRAS/NRAS develops resistance to cetuximab in advanced intestinal cancer. In a recent study, Lu et al. (2017) found that the expression of a lncRNA MIR100HG and two embedded miRNA, miR-100 and miR-125b, was significantly increased in cetuximab-resistant advanced intestinal tumors that do not harbor mutations in KRAS. Furthermore, miR-100 and miR-125b downregulate the expression of several negative regulators of canonical Wnt signaling, leading to enhanced Wnt signaling and resistance to cetuximab treatment (Lu et al., 2017). Thus, inhibition of MIR100HG in cetuximab-resistant tumors with intact KRAS signaling may be of clinical relevance to overcome cetuximab resistance in advanced intestinal cancer.

In addition to activation of oncogenic pathways, sometimes cancer cells also rely on non-oncogenic pathways to sustain proliferation (J. Luo, Solimini, et al., 2009). p38 MAPK is an example of a non-oncogenic pathway in intestinal cancer (Gupta et al., 2014; Igea et al., 2015), whose activity can be controlled by various dual-specificity protein phosphatases (Owens & Keyse, 2007). A recent study showed that lncRNA FAM84B-4, which is overexpressed in intestinal cancer and correlates with worse prognosis, restrains the expression of phosphatase DUSP1 by interacting with hnRNP-K and therefore activates MAPKs including p38 MAPK (Peng et al., 2020). p38 MAPK has been reported to be activated in intestinal cancer and promotes tumorigenesis in established intestinal tumors and patient-derived xenografts (Gupta et al., 2014, 2015). Hence, lncRNA FAM84B-4 provides a missing link for the activation of p38 MAPK by downregulating the expression of DUSP1 phosphatase.

Numerous genome-wide association studies have identified hundreds of cancer susceptibility loci, majority of them reside in noncoding gene desert regions in human cancers including, intestinal cancer (Amundadottir et al., 2006; Zanke et al., 2007). For example, the G allele of rs6983267 was identified as a cancer susceptibility locus at 8q24.21 for intestinal cancer (Tomlinson et al., 2007). Deletion of Myc-335, a MYC enhancer that contains rs6983267, in APC ^min mice resulted in markedly reduced intestinal tumorigenesis, suggesting a functional role of rs6983267 in vivo (Sur et al., 2012). A recent study identified a lncRNA CARLo-5 in the 8q24 region. Expression of CARLo-5 significantly correlated with the risk allele rs6983267. Importantly, CARLo-5 was highly expressed in human intestinal tumors compared with adjacent tissue and downregulation of CARLo-5 greatly reduced tumor growth in xenograft models. CARLo-5 induces G1 cell cycle arrest by negative regulation of CDKN1A (cyclin-dependent kinase inhibitor 1A) (Kim et al., 2014). In addition, lncRNA OECC, which is also transcribed from 8q24, is highly expressed in intestinal cancer and metastasis. OECC functions as a ceRNA by sponging miR-143-3p to indirectly activate NFkB and p38 MAPK (Huang et al., 2018), both of which are implicated in intestinal cancer progression (Gupta et al., 2014; Xia et al., 2014). Thus, the aforementioned studies show the mechanistic link between lncRNAs in the gene deserts and cancer susceptibility variants and how they could regulate tumor progression.

In summary, we have provided several pieces of evidence that lncRNAs can play an instrumental role in the regulation of various steps involved in intestinal tumorigenesis as well as in response to therapies (Figure 1). However, given the enormous number of estimated lncRNAs and the number of cellular processes/steps they can regulate, our understanding of the role of the lncRNAs in tumorigenesis is still in its infancy. In the next section, we will focus on how we can translate our current knowledge about lncRNAs to exploit them as diagnostic and therapeutic targets.

4 LNCRNAS AS DIAGNOSTIC AND THERAPEUTIC TARGETS

As discussed above, aberrant expression of lncRNAs has been linked to various intestinal pathologies and tumorigenesis. Numerous lncRNAs are upregulated or downregulated in intestinal pathologies and cancer (Table 1), suggesting that lncRNAs may be used as biomarkers for diagnostic purposes. Preclinical studies suggest that aberrantly expressed lncRNAs can be detected in blood, saliva, or urine samples. Several groups have identified the presence of lncRNAs in body fluids from patients of IBDs and intestinal cancer (Table 1). For example, lncRNAs CCAT1 and HOTAIR both can be detected in the plasma of intestinal cancer patients and show high diagnostic performance (Zhao et al., 2015). In addition, high CCAT1 expression in intestinal tumors has been proposed to be an independent prognostic marker (McCleland et al., 2016). However, none of these biomarkers reached the clinic yet, suggesting the requirement for more rigorous data from larger cohorts from multicentric studies.

In addition to diagnostic biomarkers, aberrant expression of lncRNAs could also be exploited for therapeutic purposes. Several lncRNAs are found to be upregulated in intestinal cancer (Table 1). These upregulated lncRNAs are considered as oncogenic lncRNAs, therefore, targeting of these lncRNAs can be exploited for cancer regression. For example, lncGata6 is highly expressed in Lgr5⁺ CSCs and knockdown of lncGata6 in intestinal tumors resulted in substantial regression of tumor mass (Zhu et al., 2018). In addition, expression of specific lncRNAs has been shown to correlate with response to chemotherapy and anti-EGFR therapy in advanced intestinal cancer (Lu et al., 2017; Ma et al., 2016), which could also be used for therapeutic purposes. Several strategies such as ASOs, siRNAs, short hairpin RNAs (shRNAs), and CRISPR/Cas9 can be used for therapeutic targeting of upregulated lncRNAs (Arun et al., 2018; Toden et al., 2021; Zibbit et al., 2021). ASOs target lncRNAs for degradation via an RNase H-dependent mechanism and can efficiently deplete lncRNAs regardless of their localization within the cell (Ideue et al., 2009). Targeting of lncGata6 using ASOs in mice with established PDTO-derived xenografts resulted in substantial reduction of tumors and prolonged survival (Zhu et al., 2018). Several studies show that lncRNAs can also be targeted using siRNAs/shRNAs. The exogenous siRNA/shRNAs involve a degradation pathway that comprises a ribonuclease called dicer and a multiprotein complex RNA-induced silencing complex (RISC) in conjunction with the endonuclease AGO2 (Hannon & Rossi, 2004). The siRNA/shRNA-mediated targeting of lncRNAs resulted in inhibition of cell growth, invasion, and migration and induced cell cycle arrest (Endo et al., 2013; Ren et al., 2013). More recently, CRISPR/Cas9mediated silencing of lncRNA RAMS11, a lncRNA that is highly upregulated in metastatic CRC, resulted in reduced tumor growth and metastasis in preclinical mouse models (Silva-Fisher et al., 2020). In addition to these methods, lncRNAs can also be targeted by natural sources derived phytochemicals such as resveratrol and curcumin (Mishra et al., 2019). For example, curcumin was found to suppress the expression of lncRNAs H19 in several cancer cell lines (Novak Kujundzic et al., 2008). These phytochemicals target lncRNAs either directly or indirectly by affecting several upstream molecules (Mishra et al., 2019).

Conversely, several lncRNAs are found to be downregulated in intestinal cancer (Table 1). Therefore, restoration of these tumor-suppressive lncRNAs can be another strategy to exploit lncRNAs for cancer therapy. For this purpose, exosomes could be used to deliver specific tumor-suppressive lncRNAs to the cancer cells. The main advantage of using exosomes for lncRNAs delivery is that they are easy to manipulate, easily taken up by recipient cells and most importantly lncRNAs within exosomes remain functional (Ha et al., 2016; Takahashi et al., 2014). The aforementioned studies show that lncRNAs can be exploited for therapeutic purpose.

Despite having great success in preclinical models, targeting of lncRNAs in humans may face some issues. For example, the preclinical targeting of lncRNAs has been performed using human tumor cells transplanted in mice. In these cases, only human cells will be targeted by ASOs and thus will not reflect the pleiotropic effects of ASOs in different organs, as the same lncRNA might be expressed in other tissues in humans. In addition, these xenografts lack human tumor stromal cells, which play a significant role in all the phases of tumor progression as well as therapy resistance (Balkwill et al., 2012). Therefore, before lncRNAs-based therapies become a reality, the complete expression profile and putative functions of the lncRNAs in different organs/cell types will be required to predict how well patients might tolerate ASO- or any other method-mediated lncRNAs targeting.

5 CONCLUSIONS AND FUTURE DIRECTIONS

Recent data from numerous studies suggest that lncRNAs are one of the key regulators of different cellular processes required for homeostasis, regenerative response, and tumorigenesis. In this review, we have summarized evidence that dysregulated expression of lncRNAs is associated with intestinal homeostasis, regeneration, tumorigenesis, and response to cancer therapies. Most of these studies, so far, have focused on determining the aberrant expression of lncRNAs and their autocrine effects on cellular processes. Emerging evidence suggests that lncRNAs can be detected in secreted exosomes, thereby affecting cell-to-cell communication in a paracrine manner. Therefore, future studies should identify and examine the function of exosomal lncRNAs in the regulation of regenerative response, tumorigenesis and response to different therapies. In addition, data from high-throughput sequencing suggest that mutations may occur at their functionally relevant sites and thus could alter lncRNAs function. Consequently, the relevance of such mutations in the functioning of lncRNAs should also be examined. In addition, a better understanding of the physiological roles of lncRNAs in different tissues/cell types and the molecular cause of their dysregulated expression during pathological conditions will be required to exploit them for therapeutic purposes. In this regard, 3D organoid system, which can be used to culture cells from normal tissues as well as from tumors, may play an instrumental role to delineate the physiological versus pathological roles of lncRNAs. In parallel, future studies should aim to develop methods for the accurate detection of lncRNAs in different body fluids such as blood, saliva, and urine, and so forth to develop noninvasive detection methods for diagnostic and prognostic purposes. Besides, researchers should also focus on developing/modifying methods for effective targeting of lncRNAs. Advancement of our knowledge about lncRNAs and new emerging technologies may lead to lncRNA-based therapies in the clinic in the near future.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS

Jalaj Gupta conceptualized the content and wrote the manuscript. Vipin K. Yadav, Amit Kumar, and Prem P. Tripathi helped in the discussion and editing of the manuscript, prepared figures and tables. All the authors commented on the manuscript draft.