Extracellular vesicle long non-coding RNA-mediated crosstalk in the tumor microenvironment: Tiny molecules, huge roles.

Emerging evidence has shown that dynamic crosstalk among cells in the tumor microenvironment modulates the progression and chemotherapeutic responses of cancer. Extracellular vesicles comprise a crucial form of intracellular communication through horizontal transfer of bioactive molecules, including long non-coding RNA (lncRNA), to neighboring cells. Three main types of extracellular vesicles are exosomes, microvesicles and apoptotic bodies, exhibiting a wide range of sizes and different biogenesis. Over the last decade, dysregulation of extracellular vesicle lncRNA has been revealed to remodel the tumor microenvironment and induce aggressive phenotypes of tumor cells, thereby facilitating tumor growth and development. This review will focus on extracellular vesicle lncRNA-mediated crosstalk between tumor cells and recipient cells, including tumor cells as well as stromal cells in the tumor microenvironment, and overview the mechanisms by which lncRNA are selectively sorted into extracellular vesicles, which may pave the way for their clinical application in cancer diagnosis and treatment.

adipose‐derived mesenchymal stem cells

cancer‐associated fibroblasts

colorectal cancer

density gradient centrifugation

extracellular matrix

epithelial ovarian cancer

Erb‐B2 receptor tyrosine kinase 2

esophageal squamous cell carcinoma

extracellular vesicles

forkhead box protein O1

hepatocellular cancer

human lymphatic endothelial cells

heterogeneous nuclear ribonucleoprotein A2B1

human umbilical vein endothelial cells

locked nucleic acid

long non‐coding RNAs

microRNAs

multiple myeloma

mesenchymal stem cells

normal fibroblast

natural killer

non‐small cell lung cancer

nanoparticle tracking analysis

oral squamous cell carcinoma

pancreatic ductal adenocarcinoma

RNA‐binding proteins

small extracellular vesicles

short hairpin RNA

small interfering RNA

tumor‐associated macrophages

transmission electron microscopy

tumor microenvironment

western blot

X‐ray repair cross complementing 4

INTRODUCTION

The tumor microenvironment (TME), composed of cancer cells, stromal cells and the extracellular matrix (ECM), creates a niche for their residence and interactions. The representative stromal cells include endothelial cells, mesenchymal stem cells, cancer‐associated fibroblasts (CAF), adipocytes and infiltrating immune cells. , , It is well accepted that the reciprocal communication among cells in the TME plays a significant role in the ECM remodeling, angiogenesis, drug resistance, energy metabolism reprogramming and anti–tumor immune responses.

Tumor cells can exchange information with recipient cells through cell‐to‐cell contact, secretion of soluble factors, as well as release of extracellular vesicles (EV). EV, heterogeneous membrane‐enclosed phospholipid vesicles, are implicated in cancer initiation, angiogenesis, tumor immunity and drug resistance. They are usually subdivided into three main types based on their size and biogenesis: exosomes (40‐100 nm), microvesicles (50‐1000 nm) and apoptotic bodies (800‐5000 nm). , Among these, exosomes, particles that are derived from endosomal origin, have drawn increasing attention in the field of cancer research. According to their endosomal origin, knockdown or overexpression experiments of ESCRT‐pathway molecules like Rab27a, TSG101 and Hrs are necessary for determining exosomes. Particles only detecting surface markers or particle size are not defined as exosomes. Hence, we use the term small EV (sEV) instead of exosomes in the references which do not perform studies for determining EV as of endosomal origin.

Extracellular vesicles have emerged as extracellular messengers to regulate signaling pathways and gene expression by transferring diverse cargoes, including long non–coding RNA (lncRNA). LncRNA are defined as RNA transcripts longer than 200 nucleotides with a lack of protein‐coding capacity, which modulate the occurrence and development of cancer. Recently, EV‐enriched lncRNA have been shown to shape the local cellular microenvironment and mediate phenotypic alterations of cancer cells. In this review, we aim to summarize the EV lncRNA‐mediated crosstalk between tumor cells and the recipient cells in the TME. The article further discusses the underlying mechanisms of cancer cells selectively sorting lncRNA into EV, and highlights the promising clinical applications of EV lncRNA in cancer diagnosis and treatment.

EXTRACELLULAR VESICLE LONG NON–CODING RNA MEDIATE CROSSTALK BETWEEN TUMOR CELLS

As a key mediator of cell‐to‐cell communication, tumor‐derived EV could package and transfer lncRNA to target cells, including neighboring tumor cells and stromal cells, thereby modulating their phenotypes and remodeling the TME. , EV lncRNA mediating the progression and chemoresistance of tumor cells in the microenvironment are included in Table 1.

Table 1

EV lncRNA mediate the progression and chemoresistance of tumor cells in the TME

System	Tumor type	EV lncRNA	EV type	EV identification	Target	Function	Reference
Digestive System	PDAC	lncRNA‐Sox2ot	sEV	TEM, WB	miR‐200	Promote progression and metastasis	12
	Gastric cancer	ZFAS1	sEV	TEM,NTA,WB	/	Promote cell proliferation and migration	10
		HOTTIP	sEV	TEM,NTA,WB	miR‐218	Confer cisplatin resistance	13
	ESCC	PART1	sEV	TEM, WB	miR‐129/Bcl‐2 pathway	Confer gefitinib resistance	14
	CRC	UCA1	sEV	TEM, WB	/	Confer cetuximab resistance	15
	HCC	TUC339	Large EV	TEM, DGC	/	Promote cell growth and inhibit cell adhesion	16
		lincRNA‐ROR	Large EV	TEM,NTA	p53 signaling	Confer sorafenib, camptothecin, or doxorubicin sensitivity	17
Gynecological system	Breast cancer	AFAP1‐AS1	sEV	TEM,NTA,WB	ERBB2	Confer trastuzumab resistance	18
		AGAP2‐AS1	sEV	TEM, WB	/	Confer trastuzumab resistance	19
		SNHG14	sEV	TEM,NTA,WB	Bcl‐2/Bax	Confer trastuzumab resistance	20
		UCA1	sEV	NTA,WB	/	Confer tamoxifen resistance	23
		H19	sEV	TEM,WB	/	Confer doxorubicin resistance	24
Respiratory system	NSCLC	H19	sEV	TEM,NTA,WB	/	Confer gefitinib resistance	26
		RP11‐838N2.4	sEV	TEM,WB	FOXO1	Confer erlotinib resistance	27
Urogenital system	Bladder cancer	UCA1	sEV	TEM,NTA,WB	/	Promote cell proliferation, invasion and migration	29
	Renal cancer	lncARSR	sEV	TEM,NTA,WB	miR‐34/miR‐449	Confer sunitinib resistance	30
	Prostate cancer	PCSEAT	sEV	TEM,NTA,WB	miR‐143‐3p/miR‐24‐2‐5p	Promote cell proliferation and invasion	31
Neural system	Pituitary adenoma	H19	sEV	TEM,NTA,WB	4E‐BP1	Inhibit cell growth	32
	Glioblastoma	SBF2‐AS1	sEV	TEM,NTA,WB	miR‐151a‐3p/XRCC4	Confer temozolomide resistance	33

/, not disclosed; CRC, colorectal cancer; DGC, density gradient centrifugation; ERBB2,Erb‐B2 receptor tyrosine kinase 2; ESCC, esophageal squamous cell carcinoma; EV, extracellular vesicles; FOXO1, forkhead box protein O1; HCC, hepatocellular cancer; lncRNAs, long non–coding RNA; NSCLC, non–small cell lung cancer; NTA, nanoparticle tracking analysis; PDAC, pancreatic ductal adenocarcinoma; sEV, small extracellular vesicles; TEM, transmission electron microscopy; TME, tumor microenvironment; WB, western blot; XRCC4, X‐ray repair cross complementing 4.

Digestive system

Small extracellular vesicle long non–coding RNA

Several lines of evidence support the notion that EV‐mediated delivery of lncRNA is associated with the progression of gastrointestinal cancer. sEV‐mediated overexpression of lncRNA ZFAS1 enhances the proliferation and migration of gastric cancer cells. According to Li et al, highly invasive pancreatic ductal adenocarcinoma (PDAC) cells package and transfer lncRNA‐Sox2ot to recipient PDAC cells via sEV. Upon competitively binding to the miR‐200 family, lncRNA‐Sox2ot upregulates Sox2 expression and promotes the epithelial–mesenchymal transition (EMT) and stem cell properties, aiding in the progression and metastasis of PDAC. Moreover, circulating lncRNA‐Sox2ot encapsulated in sEV is correlated with the TNM stage, lymphatic or vascular invasion, and the overall survival of PDAC patients, rendering it a useful marker for pancreatic cancer prognosis.

In contrast, EV lncRNA exert a strong influence on the chemotherapeutic responses of gastrointestinal cancer through diverse mechanisms. Transmitted from cisplatin‐resistant gastric cancer cells to sensitive cancer cells, sEV lncRNA HOTTIP sponges miR‐218 to activate HMGA1, and confers cisplatin resistance to sensitive cancer cells. Furthermore, expression of circulating lncRNA HOTTIP in sEV is significantly upregulated in cisplatin‐resistant gastric cancer patients in contrast to the cisplatin‐sensitive gastric cancer patients, indicating its potential use for the early diagnosis and treatment of gastric cancer. Like lncRNA HOTTIP, lncRNA PART1 is enriched in the gefitinib‐resistant cells and correlated with poor response to gefitinib in esophageal squamous cell carcinoma. Kang et al discovered that gefitinib‐resistant cells could deliver lncRNA PART1 to sensitive cells via sEV, which increases gefitinib‐resistant potency by regulating miR‐129/Bcl‐2 pathway. In colorectal cancer (CRC), UCA1‐containing sEV derived from cetuximab‐resistant cells could be transferred to recipient cancer cells, enhancing UCA1 expression and cetuximab resistance in vitro.

Large extracellular vesicles

Likewise, the malignant progression and chemoresistance of gastrointestinal cancers could be modulated by large EV lncRNA. lncRNA TUC339 could be exported by hepatocellular cancer (HCC) cells through large EV and taken up by adjacent counterparts, thus promoting HCC cell growth and suppressing cell adhesion to ECM. In addition to this, Takahashi et al observes that large EV‐mediated transport of lincRNA‐ROR plays a functional role in TGFβ‐dependent chemoresistance in HCC. Knockdown of lincRNA‐ROR promotes cell apoptosis in response to sorafenib, camptothecin or doxorubicin through p53 signaling, and simultaneously reduces expression of CD133 + tumor‐initiating cells, further supporting targeting lincRNA‐ROR to potentiate chemosensitivity in HCC.

Taken together, intercellular trafficking of lncRNA via EV modulates the development and chemoresistance of gastrointestinal cancer. EV lncRNA could function as not only appealing biomarkers for cancer diagnosis and prognosis but also attractive therapeutic targets to reverse chemoresistance. Clinical and preclinical trials regarding the utilization of EV lncRNA in cancer treatment have not yet been reported. It remains to be seen whether EV lncRNA‐targeted therapy is a panacea for cancer, or a poison pill with serious side effects.

Gynecological system

Dysregulation of EV lncRNA could regulate the resistance to chemotherapy in breast cancer through multiple mechanisms. In HER‐2‐positive breast cancer, sEV‐containing lncRNA AFAP1‐AS1 interacts with AU‐binding factor 1 to upregulate the translation of Erb‐B2 receptor tyrosine kinase 2 (ERBB2), resulting in dissemination of trastuzumab resistance. Recent studies have also proved that lncRNA AGAP2‐AS1 and lncRNA‑SNHG14 in sEV disseminate resistance to trastuzumab from trastuzumab‐resistant cells to sensitive cancer cells. , Dong et al performed a signal transduction reporter array to suggest that lncRNA‑SNHG14 contained in sEV mediates trastuzumab resistance by inducing Bcl‑2 expression and inhibiting Bax expression.

Apart from trastuzumab, tamoxifen and doxorubicin are widely applied as first‐line drugs for breast cancer. , Xu et al found that sEV‐mediated transfer of UCA1 from tamoxifen‐resistant LCC2 cells could significantly induce the tamoxifen resistance in tamoxifen‐sensitive MCF‐7 cells. In addition, lncRNA H19 in sEV could be delivered from doxorubicin‐resistant cells to sensitive cells, resulting in enhanced chemoresistance of doxorubicin. The involvement of EV lncRNA in chemoresistance has attracted huge attention among researchers, but their functional roles in cancer development and underlying molecular mechanisms remain to be elucidated in breast cancer.

Respiratory system

Currently, resistance to tyrosine kinase inhibitors seriously compromises the effect of chemotherapy and results in cancer‐related death in non–small cell lung cancer (NSCLC). However, the contribution of EV lncRNA to drug resistance in recipient cells is still poorly understood. Lei et al demonstrated that lncRNA H19 is responsible for the gefitinib resistance in NSCLC. In addition, sEV‐mediated transport of lncRNA H19 confers gefitinib resistance to sensitive NSCLC cells. Treatment‐sensitive NSCLC cells are endowed with resistance to erlotinib by internalizing sEV lncRNA RP11‑838N2.4 from erlotinib‐resistant cells. Acting as a transcription inhibitor, forkhead box protein O1 (FOXO1) could inactivate lncRNA RP11‑838N2.4 by binding to its promoter region in erlotinib‐resistant cell lines.

Urogenital system

Intratumoral hypoxia is one of the most important features of the solid tumor microenvironment, posing an obstacle to rapid tumor growth. Hypoxia increases the transfer of sEV lncRNA‐UCA1, promoting cell proliferation, invasion and migration in the recipient bladder cancer cells. The level of circulating lncRNA‐UCA1 packaged in sEV is remarkably higher in bladder cancer patients than normal controls, and predicts bladder cancer with a sensitivity and specificity of 80% and 83.33%, respectively. However, large‐scale clinical studies are still needed to validate its diagnostic or prognostic values. In renal cancer, Qu et al identified that sEV lncARSR facilitated AXL and c‐MET expression through sponging miR‐34/miR‐449, disseminating sunitinib resistance to the sunitinib‐sensitive cells. Locked nucleic acid‐based therapy targeting lncARSR could partially restore the sunitinib response of renal cancer. Yang et al confirmed that sEV lncRNA PCSEAT positively regulates EZH2 expression to mediate the cellular proliferation and motility by sponging miR‐143‐3p and miR‐24‐2‐5p in prostate cancer.

Neural system

Pituitary adenoma accounts for over 25% of all primary intracranial tumors. sEV‐mediated delivery of lncRNA H19 inhibits the growth of distal pituitary adenoma cells by suppressing 4E‐BP1 phosphorylation. Meanwhile, cabergoline could induce H19 expression and exert a synergic effect with sEV H19 in inhibiting pituitary tumor growth. In contrast to benign pituitary adenoma, glioblastoma in the most common primary malignant tumor in the brain. Zhang et al proposed that lncRNA SBF2‐AS1 in temozolomide‐resistant glioblastoma cells could be shuttled to peripheral sensitive cells via sEV and suppresses their temozolomide sensitivity by accelerating the DNA damage repair through the miR‐151a‐3p/X‐ray repair cross complementing 4 (XRCC4) axis. Collectively, EV lncRNA play a pivotal role in chemoresistance of neural tumors; however, their biological effects in neural tumors remain a mystery.

EXTRACELLULAR VESICLE LONG NON–CODING RNA MEDIATE CROSSTALK BETWEEN TUMOR CELLS AND STROMAL CELLS

Mesenchymal stem cells

Mesenchymal stem cells (MSC) are frequent components in the TME, contributing greatly to the growth and development of cancer. Recently, it has been reported that EV lncRNA orchestrate intercellular communication between tumor cells and MSC (Figure 1). Wang et al demonstrated that a set of lncRNA are differentially expressed in adipose‐derived mesenchymal stem cells (AD‐MSC) stimulated with lung cancer‐derived sEV, shedding new light on the involvement of sEV lncRNA in the regulation of AD‐MSC. sEV lncRNA RUNX2‐AS1 could be transmitted from multiple myeloma (MM) cells to MSC, thereby suppressing the osteogenic property of MSC by decreasing RUNX2 expression. Moreover, administration of exosome secretion inhibitor lowers lncRNA RUNX2 expression and prevents bone loss in in vivo animal models, which makes sEV lncRNA RUNX2‐AS1 a promising therapeutic target in MM patients.

FIGURE 1

Extracellular vesicle (EV) long non–coding RNA (lncRNA) mediate the crosstalk between cancer cells and stromal cells in the tumor microenvironment (TME). EV lncRNA play a key role in the interaction between cancer cells and stromal cells in the TME. EV‐mediated transfer of lncRNA from cancer cells to mesenchymal stem cells (MSC) exerts an inhibitory effect on the osteogenic property of MSC. EV lncRNA secreted from cancer cells also induce the normal fibroblast (NF)/cancer‐associated fibroblasts (CAF) transformation. Furthermore, cancer cell‐derived lncRNA are shuttled from cancer cells to endothelial cells, thereby modulating angiogenesis and inducing lymphangiogenesis. In addition, cancer cells may transport lncRNA to natural killer cells as well as tumor‐associated macrophages (TAM) via EV, resulting in enhanced cytotoxicity and immunosuppression, respectively. Reciprocally, stromal cell‐derived lncRNA are also delivered to cancer cells by EV. For example, MSC‐derived and CAF‐derived lncRNA are transferred to cancer cells by EV, leading to cancer progression and chemoresistance. Moreover, TAM transmit lncRNA to cancer cells through EV, which induces aerobic glycolysis and apoptosis resistance

Reciprocally, EV lncRNA derived from MSC could also be transferred to tumor cells to regulate tumor progression. sEV LINC00461 positively modulates BCL‐2 expression by competitively binding to miR‐15a/miR‐16, leading to enhanced cellular proliferation and reduced apoptosis of MM cells. Zhao et al indicated that lncRNA PVT1 could be encapsulated and transferred to osteosarcoma cells in bone marrow MSC‐derived sEV, which promotes tumor growth and metastasis by inhibiting the ubiquitination of ERG protein and sponging miR‐183‐5p.

Proteasome inhibitors have been regarded as a first‐line therapeutic strategy for MM patients over the past decade. However, proteasome inhibitors resistance severely impedes its therapeutic effect, and the underlying mechanism has yet to be explored. MSC could package and shuttle lncPSMA3 and lncPSMA3‐AS1 to MM cells and confer proteasome inhibitor resistance to them. Mechanistically, lncPSMA3‐AS1 and pre–PSMA3 form RNA‐RNA duplex, thus promoting lncPSMA3 expression by increasing its stability. The authors also indicate that targeting lncPSMA3‐AS1 could improve proteasome inhibitor resistance and the overall survival of MM in vivo.

Cancer‐associated fibroblasts

Cancer‐associated fibroblasts, one of the most prominent stromal cells in the microenvironment, are associated with tumor growth, metastasis, angiogenesis, chemoresistance, ECM remodeling, metabolic reprogramming and immune evasion. CAF generally exert inherent support to tumor cells by cell‐to‐cell contact and release of soluble factors. However, recent studies have suggested that EV lncRNA are involved in crosstalk between tumor cells and CAF (Figure 1). , Feng et al revealed that sEV lncRNA, secreted by breast cancer cells, modulated the malignant transformation of lung fibroblasts, thus constructing a pre–metastatic niche and facilitating the tumor pulmonary metastasis. Ding et al identified the participation of lnc‐CAF in the normal fibroblast (NF)/CAF transformation in oral squamous cell carcinoma (OSCC). OSCC‐derived sEV deliver lnc‐CAF to stromal fibroblasts and induce the CAF phenotype via IL‐33, thereby facilitating the growth of OSCC. Similarly, Hu et al stated that sEV‐mediated shuttling of lncRNA Gm26809 from melanoma cells transformed NF to CAF, which fuels the proliferation and migration of melanoma cells. In return, sEV lncRNA CCAL could be transferred from CAF to CRC cells and activate the Wnt/β‐catenin signaling pathway through HuR, resulting in suppression of CRC cell apoptosis and chemoresistance. sEV lncRNA H19 also stimulates the β‐catenin signaling pathway by sponging miR‐141, which induces the stemness and chemoresistance of CRC cells. In conclusion, these studies highlight the potential for preventing NF/CAF transformation and overcoming drug resistance by targeting EV lncRNA in the TME, although further studies are yet to be undertaken.

Endothelial cells

Angiogenesis is an indispensable event for the growth and metastasis of solid tumors as well as an attractive target for cancer treatment. Recent evidence has revealed that cancer cell‐secreted EV lncRNA could be shuttled to endothelial cells to regulate angiogenesis (Figure 1). LncRNA H19, shuttled from CD90 + liver cancer cells to endothelial cells, promotes its tube formation and cell‐cell adhesive properties by enhancing the expression of VEGF and ICAM‐1, respectively. In lung cancer, sEV lncMMP‐2‐2 upregulates vascular endothelial cell permeability upon TGF‐β stimulation. It has also been reported that sEV lncRNA CCAT2 and lncRNA POU3F3 derived from glioma cells enhance endothelial cell migration, proliferation, tube formation in vitro and arteriole formation in vivo. , Furthermore, sEV lncRNA CCAT2 upregulates Bcl‐2 expression and downregulates Bax and caspase‐3 expression to inhibit endothelial cell apoptosis. In epithelial ovarian cancer (EOC), Qiu et al suggested that transfer of lncMALAT1 from EOC cells to human umbilical vein endothelial cells (HUVEC) induces angiogenesis by upregulating the expression of angiogenesis‐related genes. Wu et al found that sEV isolated from tumor‐associated macrophages (TAM) in the ascites of EOC inhibited the migration of endothelial cells through the miR‐146b‐5p/TRAF6/NF‐kB/MMP2 pathway. However, EOC‐derived sEV lncRNA, ENST00000444164 and ENST00000437683, may reverse this effect of TAM on endothelial cells through activating the NF‐κB pathway.

In contrast, EV lncRNA derived from cancer cells could also exhibit anti–angiogenic properties. sEV lncRNA GAS5 suppresses HUVEC proliferation and tube formation as well as promotes their apoptosis by increasing PTEN expression and inhibiting PI3K/AKT activation, further supporting its role as a novel therapeutic target in lung cancer treatment. In addition, sEV lncRNA LNMAT2 could be transmitted from bladder cancer cells to human lymphatic endothelial cells (HLEC), and leads to lymphangiogenesis and lymphatic metastasis in bladder cancer.

Based on the above evidence, EV lncRNA participate in the angiogenesis and lymphangiogenesis of endothelial cells, which may facilitate cancer metastasis and act as a potential therapeutic target.

Natural killer cells

Natural killer (NK) cells are a subset of the type I innate lymphoid cells which play a critical role in host defense against viral infection and malignant transformation. , It is well known that cell‐to‐cell contact and release of soluble factors mediate the interactions between NK and tumor cells. In the recent decade, increasing evidence has shown that EV could also mediate this bidirectional communications by transferring proteins and microRNA (miRNA). Nonetheless, there is a lack of scientific evidence supporting the involvement of EV lncRNA in this process. Fang et al found that upregulation of lncRNA GAS5 in activated NK cells suppressed the immune escape of liver cancer. Mechanistically, lncRNA GAS5 overexpression enhanced IFN‐c secretion, NK cell cytotoxicity and the percentage of CD107a + NK cells through the miR‐544/RUNX3 axis (Figure 1). It has also been reported that lncRNA GAS5‐containing sEV are produced by NSCLC cells. Therefore, it may be possible that sEV‐mediated transfer of lncRNA GAS5 is involved in the regulation of NK cell cytotoxicity, although further studies are required to validate this hypothesis.

Tumor‐associated macrophages

Macrophages that are recruited to the TME are termed TAM. TAM can promote cancer cell proliferation, induce immune evasion, enhance tumor invasion and tumor metastasis, and stimulate tumor angiogenesis. Functionally, TAM are categorized into M1 subtype with an anti–tumor role and M2 subtype with a tumor‐supportive role. TAM generally exhibit an M2‐like phenotype, which is associated with high tumor grades and poor prognosis in a variety of cancers. , It has been reported that EV lncRNA from tumor cells can regulate the activation, polarization and function of TAM (Figure 1). , sEV LINC ROR from hepatocarcinoma cells regulates the inflammation of macrophages upon LPS stimulation by inhibiting the secretion of IL‐1β. Li et al demonstrated a novel role for HCC‐derived sEV lncRNA TUC339 in macrophage activation and M1/M2 polarization. Overexpression of TUC339 reduces pro–inflammatory cytokine production, decreases co‐stimulatory molecule expression, impairs phagocytosis and drives M2 polarization in the macrophage. In CRC, Liang et al demonstrated that sEV lncRNA RPHH1 could be transferred to macrophages and mediated macrophage M2 polarization, thus promoting proliferation and metastasis of CRC cells. Compared to traditional tumor marker CEA and CA199, sEV lncRNA RPHH1 displays greater specificity and sensitivity, which may support its role as a promising diagnostic marker of CRC.

In contrast, lncRNA could act as signal transducer from TAM to tumor cells via EV to reprogram tumor metabolism. Chen et al found that TAM‐derived sEV lncRNA HISLA interfered with the interaction of PHD2 and HIF‐1α to stabilize HIF‐1α, and promoted cancer aerobic glycolysis and apoptosis resistance. Reciprocally, glycolytic tumor cells released lactate to further induce HISLA upregulation in TAM through the ERK‐ELK1 signaling pathway. Moreover, HISLA knockdown via short hairpin RNA (shRNA) inhibited glycolysis and chemoresistance of breast cancer cells in vivo, highlighting the prospect of EV lncRNA‐based targeting therapy.

ASSORTMENT AND UPTAKE OF EXTRACELLULAR VESICLE LONG NON–CODING RNA

Previous studies have suggested that noncoding RNA, mainly miRNA and lncRNA, are selectively sorted into EV; however, the underlying mechanisms remain largely unexplored. , , Based on current research, three possible mechanisms of miRNA sorting into EV have been proposed. First, RNA‐binding proteins (RBP), especially heterogeneous nuclear ribonucleoprotein A2B1 (hnRNPA2B1), could mediate the selective loading of miRNA into sEV through the recognition of specific “EXOmotifs” on miRNA. Second, the levels of endogenous miRNA targets in response to cell activation may be inviolved in the miRNA sorting to sEV. Third, the 3ʹ end posttranscriptional modifications of miRNA are key factors for miRNA sorting into sEV. Koppers‐Lalic et al confirmed that miRNA with 3ʹ end adenylation were relatively enriched in B cells, whereas those with 3ʹ end uridylation were overexpressed in sEV.

With regards to lncRNA, Qu et al used RNA pull‐down and RNA immunoprecipitation assays to reveal that hnRNPA2B1 directs the specific loading of lncARSR into sEV in renal cancer. Subsequent studies verified that hnRNPA2B1 is responsible for the selective packaging of lncRNA AGAP2‐AS1 into sEV in breast cancer as well as that of lncRNA H19 in NSCLC. , Similarly, lncRNA LNMAT2 is exported into sEV by direct binding with hnRNPA2B1 through its specific sequence on 1930‐1960 nt in bladder cancer. This indicates that hnRNPA2B1 plays a critical role in targeting selected lncRNA into EV. Does any other RBP contribute to this process? Are other modulators involved in the process, such as endogenous RNA or posttranscriptional modifications of lncRNA? Because there are no relevant studies regarding the selective loading of lncRNA into microvesicles and apoptotic bodies, what are the differences in loading mechanisms among exosomes, microvesicles and apoptotic bodies? In vivo experiments and powerful imaging methods to track EV are urgently needed to elucidate the mechanisms of how cancer cells manipulate lncRNA into EV, which may offer new insights into EV‐based RNA therapeutics.

After release into the TME, EV appear to be taken up by recipient cells through two different endocytic pathways; namely clathrin‐dependent and clathrin‐independent pathways. The latter category includes caveolin‐dependent endocytosis, macropinocytosis, phagocytosis and lipid raft‐mediated endocytosis. Ageta et al indicated that recipient cells could internalize EV through various mechanisms in a cell‐context dependent manner. However, fewer studies have been conducted to elaborate the recipient cells’ mechanisms to take up EV, which merits further exploration in the future.

CONCLUSION

Extracellular vesicles mediate cell‐to‐cell communication by transferring their molecular cargoes to recipient cells. Growing evidence has revealed that EV‐mediated delivery of lncRNA modulates a variety of processes, such as immune response, chemosensitivity, tumor growth and development. , , As described above, tumor‐derived EV lncRNA confer aggressive and chemoresistant phenotypes to neighboring counterparts in the TME. Meanwhile, EV lncRNA mediate the interaction between tumor and stromal cells, thereby remodeling the local environment to facilitate tumor growth and progression.

EV‐containing lncRNA can reflect the biological and pathological state of noparental cells, and the existence of EV keeps them stable and resistant to endogenous RNase. These traits make EV lncRNA valuable diagnostic and prognostic biomarkers in a variety of cancers. , Further bioinformatics analyses could determine a panel of EV lncRNA, which more precisely predict cancer diagnosis and prognosis than any single EV lncRNA.

EV‐containing lncRNA could also serve as therapeutic targets for cancer treatment. It is appealing to locally inhibit EV lncRNA release or uptake to modulate the drug resistance and eventually improve the prognosis of cancer. Manipulation of EV carrying small interfering RNA (siRNA) or shRNA to target EV lncRNA may open a new avenue for cancer treatment. Kamerkar et al demonstrated that bioengineered sEV derived from normal fibroblast‐like mesenchymal cells deliver siRNA or shRNA to oncogenic KRASG12D, leading to cancer suppression and increased overall survival in PDAC. However, there are still several problems to be solved before its clinical application in cancer. For example, what are the standard procedures for EV isolation, purification, identification and lncRNA extraction? What are the exact mechanisms for cancer cells to selectively sort lncRNA into EV and how do the recipient cells take them up? What is the precise role of EV lncRNA in cancer progression and chemotherapeutic resistance? Since current studies mainly concentrate on sEV lncRNA in the TME, what are the functional differences between sEV and large EV lncRNA? Are EV lncRNA‐based therapies able to treat cancer or do they cause severe side‐effects? Therefore, applying EV lncRNA from bench to bedside is a challenging but intriguing endeavor that requires further exploration in the future.

CONFLICT OF INTEREST

The authors report no conflict of interest.

AUTHOR CONTRIBUTIONS

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.