Recent findings have demonstrated that lncRNAs exert a level of epigenetic control to promote and define cellular lineages, regulate gene expression, and support genomic/chromosomal architectural organization through their direct interactions across the genome [4, 32–35]. Our characterization of LINC01638, as one of the highest expressed lncRNAs in hMSCs, reveals novel insights related to osteogenesis. Multiple experimental approaches for interrogating its unique properties provide options for hMSC commitment to specific lineages controlled by LINC01638. A key finding of our studies is the essential role of LINC01638 to preserve hMSC competency to regulate distinct lineages when required for different tissues, including bone.
Examining the tissue-specific expression of LINC01638 indicated it was expressed at very low levels in most tissues (GTEx Portal). Upon initiation of hMSC commitment to the osteogenic lineage, we observed a striking down-regulation of LINC01638. We tested whether the knockdown of LINC01638 would reverse the inhibition and allow osteogenesis to proceed through stages of osteoblast differentiation. However, the knockdown of LINC01638 resulted in a proliferation decrease, cell senescence, and cell death compared to the hMSC control. We determined that the decrease in proliferation was associated with DNA damage and genomic instability mediated by γH2AX, p21, and 53BP1. Thus, we identified that the most significant role of LINC01638 is protecting a pool of proliferative hMSCs to support lineage commitment to a specific organ/cell phenotype. Our findings confirm that LINC01638 is stringently regulated to assure the ability of hMSCs to support tissue development and remodeling that is required for skeletal homeostasis.
We demonstrated that alteration of LINC01638 expression in hMSCs undergoing osteogenic commitment resulted in differential expression of genes directly upstream and downstream of the LINC01638 locus on chromosome 22 (Chr22). Genes related to cell cycle control, DNA damage and osteogenesis (GWAS, Bone Mineral Density (BMD)-associated) were identified in this cluster. The large cluster on Chr22 where this lncRNA is encoded is associated with bone-related genes, both protein and noncoding. The knockdown studies identified the up- and down-regulation of LINC01638 proximal genes, including genes critical for bone formation, such as Kremen1, that function to regulate bone formation via attenuating Wnt signaling in the developing limb to allow normal limb patterning. Kremen1 is known to be highly expressed in mature bone [36] and directly involved in maintaining bone density [36, 37]. ZNRF3 is another proximal gene that is involved in Wnt signaling through modulation of LRP4/5/6 activity [38]. This would suggest that LINC01638 regulates genes proximal to its own locus that significantly impact bone formation and cell fate. Of importance, the Chr22 LINC01638 locus was significantly enriched in other novel lncRNAs, which were of unknown function. In addition, several genes are differentially expressed and related to osteogenesis, affected by LINC01638 KD. It is interesting that several lncRNAs in the same genomic location are similarly regulated compared to LINC01638, and this warrants further investigation.
We further pursued the direct chromatin interactions with LINC01638 by ChIRP studies (ChIRP-qPCR) which confirmed chromatin binding on Chr22, identifying an enriched region of interaction. From our ChIRP PCR studies, we again identified that the KREMEN1, ZNRF3 and NOTCH genes are bound by LINC01638 suggesting a direct regulation of gene expression through genomic interactions, presumably through recruitment of transcriptional mediators, chromatin modifying complexes or stabilization of transcriptional loops or domains, as has been demonstrated for other LncRNAs [39]. Importantly, the KD of LINC01638 results in upregulation of Kremen 1 and ZNRF3 that both inhibit Wnt signaling and in turn bone formation. Taken together, given the integral role of these genes in osteogenesis and bone maintenance, it is clear that LINC01638 is an important epigenetic regulator for maintaining MSC integrity and can function when required for bone renewal.
We recognize that the mechanisms contributing to bone lineage fate are not fully understood. Additional evidence that LINC01638 directly regulates osteogenesis is by its interaction with the NOTCH2 gene located on Chr1. Notch signaling has a major role in the commitment of mesenchymal cells to the osteoblastic lineage. Notch expression in osteoblast precursors regulates femoral microarchitecture [40–42]. Our discovery provides a novel and key pathway mediated by LINC01638 in regulating bone tissue. How these important functions of LINC01638 in regulating KREMEN, ZNRF3 and NOTCH through recruitment of co-regulators during osteogenesis remain to be further studied. Nonetheless, our findings reveal a novel dimension of bone regulation by LINC01638.
LINC01638 functionally contributes to regulation of the balance between osteogenesis and adipogenesis. Studies have shown that fat-induction factors inhibit osteogenesis, and conversely, bone-inducing factors impair adipogenesis [43, 44]. The commitment of MSCs in forming fat or bone are related to pathological conditions, such as early osteoporosis in bone and/or aging-associated adipogenesis. It has been well documented that secreted factors in the bone marrow microenvironment mediate a cross-talk of lineage secreted factors between bone and fat. Numerous recent studies have demonstrated that miRNAs and lncRNAs can regulate fat and bone tissue depending on the required condition. For example, secreted frizzled-related protein sFRP-1 (osteogenic related) and Dlk1/Pref-1 (preadipocyte factor1) contribute to the regulatory effects of both adipogenesis and osteogenesis [43, 45, 46]. Our finding that LINC01638 is highly expressed in undifferentiated MSCs was somewhat surprising and could signify a context-dependent role for LINC01638 in MSC commitment to a specific lineage.
LINC01638 is expressed at low levels in normal tissues, and numerous studies have reported its expression in multiple cancers [7, 47]. Inhibition of LINC01638 reduces tumor growth in cancers, including prostate and breast, both of which metastasize to bone [47, 48]. In addition, higher expression of LINC01638 has been suggested to be a poor prognostic marker for triple-negative breast cancer [49] and induced overexpression of LINC01638 in pancreatic ductal adenocarcinoma. Overexpression promoted cell migration and invasion through epithelial-to-mesenchymal transition (EMT)-like mechanisms [50]. It should also be noted that in triple negative breast cancer (TNBC) cells, LINC01638 was found to maintain mesenchymal traits of TNBC cells, including promoting the expression of EMT signature genes and a cancer stem cell-like state [51]. In addition LINC01638 expression in papillary thyroid carcinoma cells has been demonstrated to regulate cell proliferation via interactions and modulation of the Wnt/beta-catenin pathway and activation of Axin2 [52]. These studies, along with the results presented here, would suggest that LINC01638 is a strong determinate of MSC identity, driving mesenchymal gene expression in transformed cancer cells. This LncRNA may act as a critical gene in normal MSCs regulating proliferation, protection from DNA damage and cell commitment to the osteogenic lineage. Further, our ChIRP studies reveal chromatin-associated genes that contribute to the stabilization and reorganization of chromatin interactions by LINC01638 that may be important in cancer cell survival and progression.
Taken together, it is clear from our findings that LINC01638 has unique properties to protect the self-renewing MSC population and to support growth and/or repair of bone tissue that is in a constant state of turnover. Therefore, it will be informative to obtain clinical data necessary to uncover the deeper mechanisms by which LINC01638 can mediate physiological control of bone structure and function.