The potential of using blood circular RNA as liquid biopsy ...R EVIEW The potential of using blood circular RNA as liquid biopsy biomarker for human diseases Guoxia Wen1, Tong Zhou2&,

REVIEW

The potential of using blood circular RNAas liquid biopsy biomarker for human diseases

Guoxia Wen1 , Tong Zhou2& , Wanjun Gu1&

1 State Key Laboratory of Bioelectronics, School of Biological Sciences and Medical Engineering, Southeast University,Nanjing 210096, China

2 Department of Physiology and Cell Biology, Reno School of Medicine, University of Nevada, Reno, NV 89557, USA& Correspondence: [email protected] (T. Zhou), [email protected] (W. Gu)Received September 21, 2020 Accepted October 9, 2020

ABSTRACT

Circular RNA (circRNA) is a novel class of single-stranded RNAs with a closed loop structure. Themajority of circRNAs are formed by a back-splicingprocess in pre-mRNA splicing. Their expression isdynamically regulated and shows spatiotemporal pat-terns among cell types, tissues and developmentalstages. CircRNAs have important biological functions inmany physiological processes, and their aberrantexpression is implicated in many human diseases. Dueto their high stability, circRNAs are becoming promisingbiomarkers in many human diseases, such as cardio-vascular diseases, autoimmune diseases and humancancers. In this review, we focus on the translationalpotential of using human blood circRNAs as liquidbiopsy biomarkers for human diseases. We highlighttheir abundant expression, essential biological func-tions and significant correlations to human diseases invarious components of peripheral blood, includingwhole blood, blood cells and extracellular vesicles. Inaddition, we summarize the current knowledge of bloodcircRNA biomarkers for disease diagnosis or prognosis.

KEYWORDS peripheral blood circular RNA, liquidbiopsy, human diseases, translational biomarkers

INTRODUCTION

Liquid biopsy is a biopsy that uses body liquids as thesample source to diagnose, predict the outcome of or mon-itor the development of human diseases (Rubis et al., 2019;Luo et al., 2020a). Compared to traditional tissue biopsy,liquid biopsy has the advantages of being noninvasive,performed in real-time and accurate (Rubis et al., 2019; Luo

et al., 2020a). It has been proven to be applicable to themanagement of many human diseases, including cancers(Heitzer et al., 2019; Mattox et al., 2019; Rubis et al., 2019),prenatal genetic disorders (Zhang et al., 2019a), heart dis-eases (Zemmour et al., 2018), schizophrenia (Chen et al.,2020b), transplant rejection (Bloom et al., 2017) and infec-tious diseases (Burnham et al., 2018; Hong et al., 2018;Blauwkamp et al., 2019; Han et al., 2020a). To date, mostliquid biopsy studies have focused on its clinical applicationin human cancers (reviewed in Siravegna et al., 2017;Heitzer et al., 2019; Mattox et al., 2019; Rubis et al., 2019).For example, a cell-free DNA (cfDNA)-based liquid biopsytest that determines the mutational status of the epidermalgrowth factor receptor (EGFR) gene was used to guide theresponse of EGFR tyrosine kinase inhibitors in non-small celllung cancer (NSCLC) patients, which was approved by theFDA in clinical practice (Kwapisz, 2017). Another FDA-ap-proved liquid biopsy test, Epi proColon, assessed themethylation status of the Septin9 gene in whole blood, whichwas used to screen colorectal cancer patients from healthycontrols (Lamb and Dhillon, 2017). In addition to cfDNA(Wan et al., 2017; Cescon et al., 2020), several other ana-lytes within circulating body fluids were investigated as liquidbiopsy biomarkers, such as circulating tumor cells (CTCs)(Yu et al., 2013), extracellular vesicles (EVs) (Torrano et al.,2016), cell-free RNA (cfRNA) (Zaporozhchenko et al., 2018),circulating proteins (Surinova et al., 2015), circulatingmetabolites (Crutchfield et al., 2016) and platelets (Joosseand Pantel, 2015; Best et al., 2017). Among them, RNA-based liquid biopsy biomarkers have gained much moreattention in recent years since they have dynamic expres-sions and are closely related to different disease conditions(Zaporozhchenko et al., 2018; Sole et al., 2019). Circulatingnoncoding RNAs, especially microRNAs (miRNAs), haveshown promising potential as stable blood-based biomarkers

© The Author(s) 2020

Protein Cellhttps://doi.org/10.1007/s13238-020-00799-3 Protein&Cell

Protein

&Cell

http://orcid.org/0000-0001-6544-8242http://orcid.org/0000-0003-2361-1931http://orcid.org/0000-0003-4501-0539http://crossmark.crossref.org/dialog/?doi=10.1007/s13238-020-00799-3&domain=pdf

in liquid biopsy (Mitchell et al., 2008; Anfossi et al., 2018;Pardini et al., 2019).

Circular RNAs (circRNAs) are a group of endogenousnoncoding RNA molecules (Chen, 2016, 2020; Li et al.,2018d). They were first found in plant viroids (Sanger et al.,1976) and eukaryotic cells (Hsu and Coca-Prados, 1979) in1970s, and were recently observed to be functional (Hansenet al., 2013; Memczak et al., 2013). They are joined head totail to generate a covalently closed loop structure throughback-splicing (Vicens and Westhof, 2014; Li et al., 2018d).They lack a 5-prime cap and 3-prime poly-A tail, which isquite different from canonical linear RNAs (Vicens andWesthof, 2014; Li et al., 2018d). CircRNAs have beenidentified in almost all organisms across the eukaryotic treeof life (Wang et al., 2014). Some circRNAs are the pre-dominant transcript isoform of their host genes expressed inspecific tissues or cell types (Salzman et al., 2012, 2013;Rybak-Wolf et al., 2015). Furthermore, they are expressed ina tissue- (Zhou et al., 2018a), cell-type- (Salzman et al.,2013) and developmental stage-specific manner (Zhouet al., 2018a). Accumulating studies have revealed manyregulatory roles and versatile cellular functions that cir-cRNAs can perform (Li et al., 2018d; Chen, 2020). They canact as miRNA decoys (Hansen et al., 2013), RNA bindingprotein (RBP) sponges (Ashwal-Fluss et al., 2014; Huanget al., 2020a) and protein scaffolds (Li et al., 2015c, 2019a).Moreover, a small percentage of circRNAs can be translatedinto proteins (Legnini et al., 2017; Pamudurti et al., 2017;Heesch et al., 2019). Aberrant expression of circRNAs hasbeen related to many human diseases, including cancers(Shang et al., 2019; Vo et al., 2019), neurodegenerativediseases, cardiovascular diseases (Aufiero et al., 2019) andimmune diseases (Chen et al., 2019c; Zhou et al., 2019b).Due to their high stability (Enuka et al., 2015), abundantexpression (Jeck et al., 2013) and high specificity (Zhouet al., 2018a), circRNAs are becoming promising biomarkersfor human diseases (Zhang et al., 2018j).

Due to the importance of liquid biopsy biomarkers inprecision medicine (Vargas and Harris, 2016) and thesuperior characteristics of circRNAs as disease biomarkers(Zhang et al., 2018j), recent studies have put great effortsinto researching the use of circRNAs as liquid biopsybiomarkers for human diseases. Here, we review currentprogress that has been made on this topic. We focus oncircRNAs in the peripheral blood, although circRNAs areabundantly expressed in other body fluids, such as saliva(Bahn et al., 2014; Ghods, 2018) and urine (Kölling et al.,2019; Lam and Lo, 2019; Vo et al., 2019). In this review, wefirst briefly introduce the biogenesis, function, and expres-sion patterns of circRNAs and their close correlations tohuman diseases. Then, we emphasize the functional rolesthey may play in different blood components, including bloodcells, serum, plasma, platelets and EVs in the blood. Wesummarize peripheral blood circRNA biomarkers that havebeen constructed for the management of human diseases.

Finally, we discuss future opportunities and challenges oftranslating blood circRNAs into clinical practice.

THE BIOGENESIS OF CIRCRNAS

CircRNAs are derived from pre-mRNAs and formed by back-splicing, in which a downstream 5-prime splice site (ss) iscovalently joined with an upstream 3-prime ss (Fig. 1) (Liet al., 2018d). There are three major types of circRNAs,exonic circRNAs (ecircRNAs), exon-intron circRNAs (EI-circRNAs) and circular intronic RNAs (ciRNAs) (Fig. 1).EcircRNAs only contain exons, and most ecircRNAs arelocated in the cytoplasm. Two models have been proposedto explain the production of ecircRNAs (Fig. 1A) (Chen,2016; Wu et al., 2017a; Kristensen et al., 2019; Xiao et al.,2020). The first model suggests that when the pre-mRNA ispartially folded during transcription, canonical splicing willcause an “exon jump” and generate a linear RNA containingskipped exons (Starke et al., 2014; Kelly et al., 2015). Asubsequent back-splicing event will turn the ‘lariat interme-diate’ into a closed circular transcript (Barrett et al., 2015; Liet al., 2018d). The second model proposes that when the5-prime ss is pulled closer to the 3-prime ss by base pairingof flanking intronic complementary sequences (Ivanov et al.,2014; Vicens and Westhof, 2014; Zhang et al., 2014; Wilusz,2015) or specific bindings of intronic sequences to RBPs(Conn et al., 2015), a back-splicing event may occur, and acircRNA may be generated (Jeck et al., 2013). Thesemodels are supported by accumulating evidence showingthat cis-acting elements (Jeck et al., 2013; Zhang et al.,2014; Yoshimoto et al., 2019) and trans-acting factors(Ashwal-Fluss et al., 2014; Conn et al., 2015; Agirre et al.,2019) play crucial regulatory roles in circRNA formation.EIcircRNAs are generated when an intron or several intronsare alternatively retained in splicing (Li et al., 2015c).Therefore, EIcircRNAs are mainly considered intermediatesof ecircRNAs (Li et al., 2015c). CiRNAs, which only containintrons, are produced through escape from debranching ofintron lariats (Zhang et al., 2013). This process depends on aGU-rich motif near the 5-prime ss and a C-rich motif close tothe branch-point site (Zhang et al., 2013). Unlike ecircRNAs,EIcircRNAs and ciRNAs are mainly located in the nucleus.

THE FUNCTION OF CIRCRNAS

With the rapid application of high-throughput RNA-se-quencing (RNA-seq) technology (Stark et al., 2019) andrelated bioinformatics tools that identify circRNAs from RNA-seq datasets (Jeck and Sharpless, 2014; Szabo and Salz-man, 2016; Li et al., 2017a; Gao and Zhao, 2018), more thanone million circRNAs have been annotated in model organ-isms (Glažar et al., 2014; Vo et al., 2019; Cai et al., 2020; Wuet al., 2020a). However, the functional characterization ofcircRNAs is still at its early stage (Chen, 2016, 2020; Li et al.,2018d). Their circular conformation and the large sequenceoverlap with their linear mRNA counterparts have posed


Protein

&Cell

REVIEW Guoxia Wen et al.

Figure

1.Thebiogen

esis

andfunctionofcircR

NAs.(A)Circ

RNAsare

form

edbyback

-splicingofpre-m

RNAsbydifferentmech

anisms,

includ

ing:laria

tdriv

encirculariz

atio

n,

intronpairingdriv

encirculatio

n,RBPdriv

encirculariz

atio

n,a

nddebranc

hingescapeofintronlaria

ts.(B)Circ

RNAsca

nperform

diversebiologicalfunctions.

Firs

t,EIcirc

RNAsca

n

interact

with

RNAPolIIa

ndU1sn

RNPto

regulate

genetranscrip

tionin

thenuc

leus(i).S

eco

nd,e

circRNAsca

naccumulate

inthecy

toplasm

and

act

asmiRNAdeco

ys(ii),protein

regulators

(iii),

andtranslatio

ntemplates(iv

).Third

,circ

RNAsca

nbese

cretedinto

EVsbymanyce

lltypesandtran

sportedto

recipientc

ells

byEVs.

EVcircRNAsca

nalsoact

as

importantgeneregulators.

Translational potential of blood circular RNAs REVIEW


Protein

&Cell

Table

1.Cell-freecircRNA

biomarkersin

circulatingperipheralblood.

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

Breas

tca

nce

rHsa

_circ_000

1785

Plasm

aUp

57breast

canc

er/17HC

Ass

ociatedwith

histological

grade,TNM

stageand

distantmetastasis;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.784

Microarray

RT-qPCR

Yin

etal.

(2018)

Bladder

cance

rHsa

_circ_000

0285

Serum

Down

97Bladder

canc

er/97HC

Ass

ociatedwith

tumorsize

,differentia

tion,LNM,distant

metastasis,

TNM

stage

and

cisp

latin

resp

onse

.

NA

RT-qPCR

Chie

tal.

(2019)

CA

Hsa

_circ_000

3204

Plasm

aderived

EV

Up

35CA/32HC

Associatedwith

proliferatio

n,

migratio

nandtube

form

atio

nofend

othelial

cells.

0.770

RT-qPCR

Zhangetal.

(2020b

)

CAD

Hsa

_circ_000

5540

Plasm

aderived

exo

some

Up

108CAD/89HC

Ass

ociatedwith

the

Framingham

HeartStudy

riskfactors.

0.853

RNA-seq

RT-qPCR

Wuetal.

(2020b

)

CHB

Circ

MTO1

Serum

Down

360CHB/360

HC

Ass

ociatedwith

liverfib

rosis

progress

ionandprognosis.

0.914

RT-qPCR

Wangetal.

(2019c

)

CHD

Hsa

_circ_004

183

Hsa

_circ_07926

5Hsa

_circ_10503

9

Plasm

aDown

40CHD/40HC

NA

0.965

Microarray

RT-qPCR

Wuetal.

(2019a

)

CLL

Circ

-RPL15

Plasm

aUp

150CLL/65HC

Associatedwith

progression

andoutcom

e.

0.840

Microarray

RT-qPCR

Wuetal.

(2020c

)

CRC

Hsa

_circ_000

6990

Plasm

aUp

60CRC/43HC

Associatedwith

TNM

stage.

0.724

RT-qPCR

Lie

tal.

(2019c

)

CRC

Circ

-CCDC66

Circ

-ABCC1

Circ

-STIL

Plasm

aDown

45CRC/61HC

Circ

-ABCC1wasassociated

with

tumorgrowth

and

progress

ion;significa

ntexp

ress

iondifference

of

circ-C

CDC66betweenpre-

trea

tmentand

post-

trea

tment.

0.780

RT-qPCR

Lin

etal.

(2019)

CRC

Hsa

_circ_000

4771

Serum

derived

exo

some

Up

135CRC/45HC

Ass

ociatedwith

TNM

stage

anddistantmetastasis;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.880

RT-qPCR

Panetal.

(2019)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

CRC

Hsa

_circ_000

0826

Serum

Up

100CRC/100

HC

Ass

ociatedwith

liver

metastasis.

0.778

RT-qPCR

Shie

tal.

(2020)

CRC

Hsa

_circ_010

1802

Serum

derived

exo

some

Up

221CRC/221

HC

NA

0.854

RNA-seq

RT-qPCR

Xie

etal.

(2020c

)

CRC

Hsa

_circ_008

2182

Hsa

_circ_000

0370

Hsa

_circ_00354

45

Plasm

aUp

Up

Down

156CRC/66HC

Thefirst

twocircRNAswere

associatedwith

LNM

and

hadsignificantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment;thethird

was

associatedwith

TNM

stage.

0.835

Microarray

RT-qPCR

Yeeta

l.(201

9)

CRC

Hsa

_circ_000

7534

Plasm

aUp

112CRC/46HC

Associatedwith

progression

ofclinicalc

lass

ifica

tion,

metastatic

pheno

type,and

differentia

tion.

0.780

RT-qPCR

Zhangetal.

(2018f)

EC

Hsa

_circ_010

9046

Hsa

_circ_000

2577

Serum

derived

EV

Up

10EC/10HC

NA

NA

RNA-seq

RT-qPCR

Xuetal.

(2018a

)

EOC

Circ

BNC2

Plasm

aDown

83EOC/83

ben

ignova

rian

cyst/83HC

Ass

ociatedwith

histological

grade,se

rioussu

btyp

e,

LNM

anddistant

metastasis.

0.923

RT-qPCR

Huetal.

(2019b

)

ESCC

Hsa

_circ_000

1946

Hsa

_circ_00436

03

Plasm

aDown

50ESCC/50HC

Ass

ociatedwith

recu

rrence

,ove

rallsu

rvival

and

disea

se-freesu

rvival.

0.894

0.836

Microarray

RT-qPCR

Fanetal.

(2019)

ESCC

Circ

GSK3ß

Plasm

aUp

86ESCC/11

ben

ignlesion

/43HC

Ass

ociatedwith

recu

rrence

,metastasis,

clinicals

tage

andoutcom

e;significa

nt

exp

ress

iondifference

betweenpre-treatm

entand

post-treatm

ent.

0.793

Microarray

ddPCR

RT-qPCR

Huetal.

(2019a

)

ESCC

Circ

-TTC17

Plasm

aUp

30ESCC/25HC

Ass

ociatedwith

TNM

stage,

LNM

andsu

rvival

time.

0.820

RT-qPCR

Wangetal.

(2019b

)

ESCC

Circ

-SLC7A5

Plasm

aUp

87ESCC/53HC

Associatedwith

TNM

stage

andsu

rvivaltim

e.

0.772

Microarray

RT-qPCR

Wangetal.

(2020c

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

GBM

Circ

FOXO3

Hsa

_circ_002

9426

Circ

-SHPRH

Plasm

aDown

100GBM/100

HC

NA

0.906-

0.980

RT-qPCR

Chenetal.

(2020a

)

GC

Hsa

_circ_000

0190

Plasm

aDown

104GC/104HC

Associatedwith

blood

carcinoembryo

nic

antig

en

leve

l.

0.600

RT-qPCR

Chenetal.

(2017c

)

GC

Hsa

_circ_002

1087

Hsa

_circ_000

5051

Plasm

aDown

70GC/70HC

Ass

ociatedwith

tumorsize

andTNM

stage;significa

nt

exp

ress

iondifference

of

hsa

_circ

_002

1087between

pre-treatm

entandpost-

trea

tment.

0.773

RT-qPCR

Hanetal.

(2020c

)

GC

Hsa

_circ_000

0745

Plasm

aDown

60GC/60HC

Associatedwith

TNM

stage.

0.683

RNA-seq

RT-qPCR

Huangetal.

(2017a

)

GC

Hsa

_circ_000

01649

Serum

Down

20GC

(pre-

ope

ratio

nvs.

pos

t-op

eration)

Ass

ociatedwith

pathological

differentia

tion;significa

nt

exp

ress

iondifference

betweenpre-treatm

entand

post-treatm

ent.

0.834

RT-qPCR

Lie

tal.

(2017d

)

GC

Hsa

_circ_000

1017

Hsa

_circ_006

1276

Plasm

aDown

121GC/121HC

Ass

ociatedwith

distal

metastasis,

ove

rallsu

rvival,

prognosisand

outcome.

0.912

Microarray

RT-qPCR

RT- ddPCR

Lie

tal.

(2018c

)

GC

Hsa

_circ_000

0467

Plasm

aUp

40GC/20HC

Associatedwith

TNM

stage;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.790

RT-qPCR

Luetal.

(2019a

)

GC

Hsa

_circ_000

6848

Plasm

aDown

30GC/30HC

Associatedwith

differentia

tionandtumor

size

;significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.733

RT-qPCR

Luetal.

(2019b

)

GC

Hsa

_circ_001

0882

Plasm

aUp

66GC/66HC

Associatedwith

tumorsize

andhistologicalg

rade,

ove

rallsu

rviveand

prognosis.

NA

RT-qPCR

Pengetal.

(2020)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

GC

Circ

PSMC3

Plasm

aDown

106GC/21HC

Ass

ociatedwith

TNM

stage,

LNM

andove

rallsu

rvival.

0.933

Microarray

RT-qPCR

Rongetal.

(2019)

GC

Hsa

_circ_006

5149

Plasm

aderived

exo

some

Down

39GC/41HC

NA

0.640

RT-qPCR

Shaoetal.

(2020)

GC

Circ

KIAA1244

Plasm

aDown

62GC/25HC

Ass

ociatedwith

TNM

stage,

LNM

andove

rallsu

rvival.

0.748

Microarray

RT-qPCR

Tangetal.

(2018)

GC

Hsa

_circ_000

0419

Plasm

aDown

44GC/43HC

Associatedwith

tumorstage,

lymph

atic

and

distal

metastasis,

venou

sand

perin

euralinva

sion.

0.840

RT-qPCR

Taoetal.

(2020)

GC

Hsa

_circ_000

0936

Serum

derived

exo

some

Up

32GC/20HC

Ass

ociatedwith

TNM

stage

andprognosis;

significa

nt

exp

ress

iondifference

before

andafte

rgastrectomy.

NA

RT-qPCR

Xie

etal.

(2020a

)

GC

Hsa

_circ_000

0181

Plasm

aDown

102GC/105HC

Associatedwith

differentia

tionand

carcinoembryo

nic

antig

en.

0.756

RT-qPCR

Zhaoetal.

(2018b

)

Glioma

Circ

NF1X

Serum

derived

exo

some

Up

69glioma/10

HC

Ass

ociatedwith

temozo

lomideresp

ons

eandprognosis.

0.885

RT-qPCR

Dingetal.

(2020)

HCC

Hsa

_circ_005

1443

Plasm

aderived

exo

some

Down

60HCC/60HC

Ass

ociatedwith

progress

ion.

0.809

Microarray

RT-qPCR

Chenetal.

(2020c

)

HCC

Circ

-ZEB1.33

Serum

Up

64HCC/30HC

Ass

ociatedwith

TNM

stage

andprognosis.

NA

RT-qPCR

Gongetal.

(2018)

HCC

Hsa

_circ_100

338

Serum

derived

exo

some

Up

39HCC

(pre-

ope

ratio

nvs.

pos

t-op

eration)

Associatedwith

metastasis,

proliferatio

n,angiogen

esis

andprognosis.

NA

RT-qPCR

Huangetal.

(2020b

)

HCC

Circ

SMARCA5

Plasm

aDown

135HCC/143

cirrho

sis/117

hep

atitis/103

HC

NA

0.938

RT-qPCR

Lie

tal.

(2019e

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

HCC

Hsa

_circ_000

3998

Plasm

aUp

100HCC/50

hep

atitis/50

HC

Significa

ntexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.892

RT-qPCR

Qiaoetal.

(2019)

HCC

Hsa

_circ_000

4001

Hsa

_circ_00041

23

Hsa

_circ_007

5792

Serum

Up

21HCC/32HC

Ass

ociatedwith

TNM

stage

andtumorsize

.0.890

RT-qPCR

Sunetal.

(2020b

)

HCC

Circ

_FOXP1

Serum

Up

30HCC/16HC

Ass

ociatedwith

tumorsize

,microva

scularinva

sionand

adva

nce

dTNM

stageand

survivaltim

e.

0.932

RT-qPCR

Wangetal.

(2020d

)

HCC

Hsa

_circ_104

075

Serum

Up

101HCC/60

HC/23

hep

atitis

B/26

hep

atitis

C/23

cirrhosis/20

LC/19GC/30

colonca

nce

r/21breast

canc

er

Ass

ociatedwith

TNM

stage;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.973

RT-qPCR

Zhangetal.

(2018g

)

HCC

Hsa

_circ_000

1445

Plasm

aDown

104HCC/57

cirrhosis/44

hep

atitis

B/52

HC

Ass

ociatedwith

serum

alpha-

fetoprotein

leve

l.0.862

RT-qPCR

Zhangetal.

(2018h

)

HCM

Circ

DNAJC

6Circ

TMEM56

Circ

MBOAT2

Serum

Down

64HCM/53HC

Ass

ociatedwith

left

ventricularoutflow

tract

gradientandthickn

essof

interventricularse

ptum

inpatie

nts

with

obs

tructive

HCM.

0.819

0.756

0.738

RT-qPCR

Sonnen

schein

etal.(2019)

Hyp

ertension

Hsa

_circ_000

5870

Plasm

aDown

54 Hyp

ertens

ion/

54HC

NA

NA

Microarray

RT-qPCR

Wuetal.

(2017b

)

Heartfailure

Hsa

_circ_006

2960

Plasm

aUp

30Heartfailure/

30HC

Ass

ociatedwith

B-typ

enatriuretic

peptid

ese

rum

leve

ls.

0.838

Microarray

RT-qPCR

Sunetal.

(2020c

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

IPAH

Hsa

_circ_006

8481

Serum

Up

82IPAH/82HC

Associatedwith

heart

function,6-m

inwalk

distance

,se

rum

N-term

inal

pro-B-typ

enatriu

retic

peptid

e,se

rum

H2S,ris

kstratifica

tion,rig

htheart

failure,anddeath.

0.895

RT-qPCR

Zhangetal.

(2019b

)

KD

Circ

ANRIL

Hsa

_circ_012

3996

Serum

Down

Up

56KD/56HC

Ass

ociatedwith

multip

leclinicallaboratory

factors;

significantexp

ress

ion

difference

ofcircANRIL

betweenpre-treatm

entand

post-treatm

ent.

0.624

0.747

RT-qPCR

Wuetal.

(2019b

)

LAC

Hsa

_circ_005

6616

Plasm

aderived

exo

some

Up

42LAC

with

LNM/48LAC

with

outLNM

Ass

ociatedwith

theleve

lof

CXCR4protein,Tstage,

Mstage,

andTNM

grade.

0.812

RT-qPCR

Heetal.

(2020)

LAC

Hsa

_circ_001

3958

Plasm

aUp

30LAC/30HC

Associatedwith

TNM

stage

andLNM.

0.794

Microarray

RT-qPCR

Zhuetal.

(2017)

LC

Hsa

_circ_000

0190

Plasm

aUp

231LC/41HC

Associatedwith

tumorsize

,histologicaltyp

e,stage,

distantmetastasis,

extrathoracicmetastasis,

survival,prognos

is,PD-L1

leve

landtherapyresp

onse

.

0.950

RNA-seq

RT-qPCR

RT- ddPCR

Luoetal.

(2020c

)

LN

Hsa

_circ_002

453

Plasm

aUp

59SLE(30with

LN

and

29

with

outLN)/26

RA/32HC

Ass

ociatedwith

seve

rityof

renalinvo

lvementand

24-hourproteinuria

.

0.906

Microarray

RT-qPCR

Ouya

ngetal.

(2018)

LUAD

Hsa

_circ_000

5962

Hsa

_circ_008

6414

Plasm

aUp

Down

153LU

AD/54

HC

Hsa

_circ_008

6414was

associatedwith

EGFR

mutatio

ns;

significa

nt

exp

ress

iondifference

of

hsa

_circ

_000

5962between

pre-treatm

entandpost-

trea

tment.

0.810

RT-qPCR

Liu

etal.

(2019b

)

LUAD

Hsa

_circ_002

178

Serum

derived

exo

some

Up

120LUAD/30

HC

Ass

ociatedwith

programmed

celldeath

protein

1(PD1)

exp

ress

ion.

0.997

RT-qPCR

Wangetal.

(2020b

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

MCL

Circ

CDYL

Plasm

aUp

18MCL/17HC

Associatedwith

cell

proliferatio

n.

0.856

RT-qPCR

Meie

tal.

(2019)

MDD

Circ

DYM

Plasm

aDown

60MDD/32HC

Ass

ociatedwith

thesc

oresof

the24-item

Hamilton

Depress

ionRatin

gSca

le,

retardatio

nsu

bscale

and

trea

tmentresp

onse

.

0.643

RT-qPCR

Songetal.

(2020)

NPC

Hsa

_circ_000

0285

Serum

Up

150NPC/100

HC

Ass

ociatedwith

tumorsize

,differentia

tion,LNM,distant

metastasis,

TNM

stage

,su

rvivalrate

and

radiotherapyresp

onse

.

NA

RT-qPCR

Shuaie

tal.

(2018)

NPC

Hsa

_circ_006

6755

Plasm

aUp

86NPC/86HC

Associatedwith

clinical

stage.

0.904

RT-qPCR

Wangetal.

(2020a

)

NSCLC

Circ

FARSA

Plasm

aUp

50NSCLC

/50

HC

NA

0.710

RNA-seq

RT-qPCR

Hangetal.

(2018)

NSCLC

Hsa

_circ_010

9320

Plasm

aUp

90NSCLC

Associatedwith

progression-

free

survivala

ndgefi

tinib

resp

onse

.

0.805

Microarray

RT-qPCR

Liu

etal.

(2019c

)

NSCLC

Hsa

_circ_000

2130

Serum

derived

exo

some

Up

28drug-

resistan

ceNSCLC

/32

drug-sens

itive

NSCLC

Ass

ociatedwith

osimertinib

resp

onse

.0.792

RT-qPCR

Maetal.

(2020)

Osteo

sarcoma

Hsa

_circ_000

0885

Serum

Up

55 osteo

sarcoma/

27benign

bon

etumor/25

HC

Ass

ociatedwith

clinical

prognosis;

significa

nt

exp

ress

iondifference

betweenpre-treatm

entand

post-treatm

ent.

0.783

RNA-seq

RT-qPCR

Zhuetal.

(2019a

)

PBC

Hsa

_circ_402

458

Plasm

aUp

35PBC/36HC

NA

0.710

Microarray

RT-qPCR

Zhengetal.

(2016)

PC

Circ

-LDLRAD3

Plasm

aUp

31PC/31HC

Ass

ociatedwith

CA19-9,N

classifica

tion,ve

nous

inva

sion,lymphatic

inva

sion,stage,

metastasis.

0.670

RT-qPCR

Yangetal.

(2017a

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

PDAC

Hsa

_circ_003

6627

Plasm

aderived

exo

some

Up

93PDAC/20HC

Ass

ociatedwith

duode

nal

inva

sion,va

scularinva

sion,

Tfactor,TNM

stage

and

survivalrate.

NA

Microarray

RT-qPCR

Lie

tal.

(2018e

)

PDAC

Circ

-IARS

Plasm

aderived

exo

some

Up

20PDAC

with

metastasis/20

PDAC

with

out

metastasis

Ass

ociatedwith

tumorve

ssel

inva

sion,liver

metastasis,

TNM

stage,andprognosis.

NA

Microarray

RT-qPCR

Lie

tal.

(2018a

)

PoAF

Hsa

_circ_025

016

Plasm

aUp

769underw

ent

off-pump

coronary

artery

byp

ass

grafting/15HC

Ass

ociatedwith

fastingblood

gluco

se.

0.802

Microarray

RT-qPCR

Zhangetal.

(2018c

)

SAI

Circ

FUNDC1

Plasm

aUp

26AIS

with

SAI/

42AIS

with

out

SAI

Ass

ociatedwith

neutrophils

counts,

white

bloodce

lland

neutrophilratio

s.

0.661

RT-qPCR

Zuoetal.

(2020)

SCC

Circ

FoxO

3a

Serum

Down

103SCC/30HC

Associatedwith

stromal

inva

sion,LNM

and

prognosis.

NA

RT-qPCR

Tangetal.

(2020)

SCLC

Circ

FLI1

Serum

derived

exo

some

Up

61SCLC

/55HC

Ass

ociatedwith

tumor

survivala

ndch

emotherapy

resp

onse

.

NA

RT-qPCR

Lie

tal.

(2018b

)

SLE

Hsa

_circ_407

176

Hsa

_circ_001

308

Plasm

aDown

126SLE/102

HC

NA

0.599

0.662

Microarray

RT-qPCR

Zhangetal.

(2018e

)

SOC

Circ

SETDBI

Serum

Up

60SOC/60HC

Ass

ociatedwith

clinical

stage,

LNM,ch

emotherapy

resp

onse

andprogress

ion-

free

survival.

0.830

RT-qPCR

Wangetal.

(2019d

)

TB

Hsa

_circ_000

1204

Hsa

_circ_00017

47

Plasm

aDown

195TB/50

pne

umonia/50

LC/50COPD/

170

HC

Ass

ociatedwith

the

radiologicals

eve

rityscores;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.928

RT-qPCR

Huangetal.

(2018b

)

TB

Hsa

_circ_000

1953

Hsa

_circ_000

9024

Plasm

aUp

123TB/103HC

Ass

ociatedwith

TBse

verity;

significantexp

ress

ion

difference

between

pre-

trea

tmentand

post-

trea

tment.

0.915

Microarray

RT-qPCR

Huangetal.

(2018a

)



Protein

&Cell

Table

1co

ntinued

Disease

CircRNA

biomarker

Source

Expression

chan

ge

Cohort

size

Clin

icalsignificance

AUC

Method

Reference

TB

Hsa

_circ_051

239

Hsa

_circ_02996

5Hsa

_circ_404

022

Serum

Up

131TB/50

pne

umonia/53

HC

Hsa

_circRNA_05123

9was

associatedwith

TBdrug

resp

onse

.

0.992

Microarray

RT-qPCR

Liu

etal.

(2020)

TB

Hsa

_circ_103

571

Plasm

aDown

35TB/32HC

NA

0.838

Microarray

RT-qPCR

Yietal.(2018)

UCB

Circ

PRMT5

Serum

derived

exo

some

Up

71UCB/50HC

Ass

ociatedwith

LNM

and

tumorprogress

ion.

NA

RT-qPCR

Chenetal.

(2018b

)

Abbreviation:AIS:a

cute

isch

emicstroke

;CA:c

erebrala

therosclerosis;

CAD:c

oronaryartery

disea

se;C

HB:c

hronichep

atitisB;C

HD:c

onge

nita

lheartdisea

ses;

CLL:

chroniclymphocytic

leuke

mia;COPD:ch

ronicobstructivepulm

onary

disea

se;C

RC:co

lorectalcance

r;ddPCR:dropletdigita

lPCR;E

C:e

ndometrialc

ance

r;EOC:epith

elialo

varia

nca

nce

r;ESSC:eso

phagea

l

squamou

sce

llca

nce

r;EV:extrace

llularve

sicle;GBM:glioblastoma;GC:gastric

cance

r;HC:health

yco

ntrol;HCC:hep

atoce

llularca

rcinoma;HCM:hyp

ertrophic

cardiomyo

pathy;

IPAH:

idiopathic

pulm

onaryarteria

lhyp

ertension

;KD:Kawasa

kidisea

se;LAC:lungadenoc

arcinoma;LC:lungca

nce

r;LN:lupusnephritis;LNM:lymphnodemetastasis;

LUAD:lungadeno-

carcinoma;MCL:

man

tlece

lllymphoma;MDD:majordep

ress

ivedisorder;NA:notapplicab

le;RT-qPCR:reve

rsetranscrip

tionandquantitativePCR;NPC:naso

pharyng

ealca

rcinoma;

NSCLC:non-smallce

lllungca

nce

r;PBC:prim

ary

biliary

cholangitis;PC:pan

creatic

cance

r;PDAC:pancreatic

ductal

adenoc

arcinom

a;PoAF:postop

erativeatrialfib

rillatio

n;SAI:stroke

ass

ociatedinfection;SCC:sq

uam

ousce

rvicalc

ance

r;SCLC:sm

allce

lllungca

nce

r;SLE:sy

stemic

lupuserythematosu

s;SOC:se

rousova

rianca

nce

r;TB:tuberculosis;

TNM:tumornode

metastasis;

UCB:urothelialc

arcinomaofthebladde

r..



Protein

&Cell

substantial challenges in investigating the functions of cir-cRNAs (Li et al., 2018d). Based on our current knowledge,circRNAs have diverse functions as miRNA decoys, protein

regulators and translation templates (Fig. 1B, see (Chen,2016, 2020; Li et al., 2018d) for excellent reviews).

Figure 2. Peripheral blood circRNAs are implicated in human diseases and can be used as potential disease biomarkers in

liquid biopsy. (A) Peripheral blood circRNAs are abundantly expressed and can be reliably detected in cell-free circulating blood

components (such as exosomes, EVs, plasma, and serum) and blood cells (including PBMCs, macrophages, RBCs, and platelets).

(B) Some exosomal circRNAs, such as circ-DB (i), circSHKBP1 (ii), and circUHRF1 (iii), are important regulators in oncogenic

pathways, while some circRNAs in exosomes, such as circHIPK3 (iv), play key roles in the release of inflammatory cytokines.

(C) Peripheral blood circRNAs, both cell-free circRNAs and intracellular circRNAs in blood cells, have potential clinical applications as

liquid biopsy biomarkers in many human diseases, such as the diagnosis, prognosis and treatment guidance of many human

diseases, including autoimmune diseases, cancers, cardiovascular diseases, immuno-deficiency diseases, infectious diseases, and

neurodegenerative diseases.



Protein

&Cell

MiRNA decoys

The most well-known function of circRNAs is that they canact as a sponge to inhibit the function of miRNAs and indi-rectly regulate the expression of miRNA target genes at theposttranscriptional level (Fig. 1B) (Hansen et al., 2013;Memczak et al., 2013; Weng et al., 2017). For example,Cdr1as can compete with mRNAs by binding miR-7 throughmiRNA response elements (MREs) (Hansen et al., 2013). Afollow-up in vivo experiment that knocked down the Cdr1aslocus in the mouse genome confirmed the importance of theCdr1as and miR-7 interaction in normal brain function (Pi-wecka et al., 2017). Although it may not be a common

phenomenon for most circRNAs (Guo et al., 2014), severalabundant circRNAs also function as miRNA sponges,including circASAP1 (Hu et al., 2019c), circBIRC6 (Yu et al.,2017), circHIPK2 (Huang et al., 2017b), circHIPK3 (Zhenget al., 2016) and circSry (Hansen et al., 2013).

Protein regulators

In addition to miRNAs, circRNAs can interact with proteins orprotein complexes to regulate protein expression and func-tion (Fig. 1B) (Huang et al., 2020a). First, circRNAs can bindto some RBPs and act as RBP sponges. CircMbl, a circRNAoriginating from the second exon of the splicing factor

Figure 3. CircRNAs are actively involved in host immune responses to exogenous pathogens. (A) CircRasGEF1B, a circRNA

induced by LPS, can protect cells from bacterial infection by regulating the expression of ICAM-1 mRNA in the TLR4/LPS pathway.

(B) Exogenous circRNAs released by viruses can be recognized by RIG-I, thus activating the host innate immunity to viruses (i).

Moreover, NF90/NF110 not only promotes circRNA production in the nucleus but also interacts with mature host circRNAs to form

circRNP complexes in the cytoplasm. Upon viral infection, NF90/NF110 can be released from circRNP complexes and bind viral

mRNAs to inhibit viral replication (ii). In addition, circRNAs can form RNA duplexes and act as inhibitors of PKR under normal

conditions. When a virus invades the cells of its host, RNase L is activated to efficiently degrade circRNAs, and PKR is released and

activated to initiate the early cellular innate immune response (iii).



Protein

&Cell

Table

2.CircRNA

biomarkers

inbloodcells

orwhole

blood.

Disease

CircRNA

biomarker

Source

Expression

change

Cohort

size

Clin

ical

significan

ce

AUC

Method

Reference

AIS

Circ

-DLGAP4

PBMC

Down

170AIS

/170

HC

Associatedwith

Health

Stroke

Sca

lesc

ore

andleve

lsof

C-reactiveprotein,TNF-α,IL-

6,IL-8,IL-22.

0.816

RT-qPCR

Zhuetal.

(2019b

)

ALS

Hsa

_circ_00239

19

Hsa

_circ_00634

11Hsa

_circ_0088

036

Leuko

cyte

Down

Up

Up

60ALS

/15HC

Hsa

_circ

_0063

411

was

ass

ociatedwith

thedisease

duratio

nandsu

rvival

time.

0.950

Microarray

RT-qPCR

Dolinar

etal.

(2019)

AML

Hsa

_circ_00042

77

Mononu

clear

cell

Down

115AML/12HC

Associatedwith

progress

ive

stage.

0.957

Microarray

RT-qPCR

Lie

tal.

(2017c

)

CAD

Circ

ZNF609

Leuko

cyte

Down

330CAD/209

HC

Associatedwith

infla

mmatory

proce

sses.

0.761

RT-qPCR

Liangetal.

(2020)

CAD

Hsa

_circ_00018

79

Hsa

_circ_0004

104

PBM

Up

436CAD/297

HC

Hsa

_circ

_0001

879was

ass

ociatedwith

bod

ymas

sindex(BMI),systolic

blood

press

ure,diastolic

blood

press

ure

andGensini

score;

hsa

_circ_0004

104was

ass

ociatedwith

high-density

lipoprotein

cholesterol.

0.703

0.700

Microarray

RT-qPCR

Wangetal.

(2019a

)

CAD

Hsa

_circ_01246

44

Whole

blood

Up

179CAD/157

HC

Associatedwith

seve

rity.

0.769

Microarray

RT-qPCR

Zhaoetal.

(2017a

)

CAP

Hsa

_circ_00184

29

Hsa

_circ_00265

79

Hsa

_circ_00991

88

Hsa

_circ_00125

35

Whole

blood

Up

36CAP/36HC

NA

0.878

Microarray

RT-qPCR

Zhaoetal.

(2019)

CML

Hsa

_circ_10005

3PBMC

Up

150CML/100

HC

Associatedwith

clinical

stage,

BCR/ABLmutantstatus,

imatin

ibresp

onse

and

prognos

is.

NA

RNA-seq

RT-qPCR

Pingetal.

(2019)

CSCC

Hsa

_circ_01019

96

Hsa

_circ_0101119

Whole

blood

Up

87CSCC/55HC

NA

0.964

RT-qPCR

Wangetal.

(2017a

)

EH

Hsa

_circ_00379

11Whole

blood

Up

100EH/100HC

Associatedwith

gender,

smoking

,drin

kingand

serum

creatin

ine.

0.627

Microarray

RT-qPCR

Baoetal.

(2018)

EH

Hsa

_circ_00379

09

Whole

blood

Up

48EH/48HC

Associatedwith

serum

creatin

ineandlow‐density

lipoprotein.

0.682

RT-qPCR

Baoetal.

(2019)

EH

Hsa

_circ_00142

43

Whole

blood

Up

89EH/89HC

Associatedwith

age,high-

density

lipoprotein

leve

land

gluco

seleve

l.

0.732

RT-qPCR

Zhengetal.

(2019)



Protein

&Cell

Table

2co

ntin

ued

Disease

CircRNA

biomarker

Source

Expression

change

Cohort

size

Clin

ical

significan

ce

AUC

Method

Reference

EH

Hsa

_circ_91025

Whole

blood

Up

96EH/96HC

Associatedwith

high‐density

lipoprotein,BMI,diastolic

bloodpress

ure

andsy

stolic

bloodpress

ure.

0.620

RT-qPCR

Zhengetal.

(2020)

GC

Hsa

_circ_00018

21

Whole

blood

Down

30GC/30HC

NA

0.872

RT-qPCR

Kongetal.

(2019)

HCC

Hsa

_circ_00007

98

PBMC

Up

72HCC/30HC

Associatedwith

tumorsize

,cirrhosisandove

rallsu

rvival.

0.703

RNA-seq

RT-qPCR

Leie

tal.

(2019)

HT

Hsa

_circ_00891

72

PBMC

Up

35HT/35HC

Associatedwith

these

rum

leve

lofthethyroid

peroxidase

antibody.

0.715

RNA-seq

RT-qPCR

Xiongetal.

(2019)

IAHsa

_circ_00210

01

Whole

blood

Down

223IA/131

HC

Associatedwith

aneurysm

rupture,Hunt,Hess

leve

l,tim

ingofsu

rgery,

disease

-freesu

rvival

and

ove

rall

survival.

0.870

RT-qPCR

Tengetal.

(2017)

MI

MICRA

Whole

blood

Down

642Acu

teMI/8

6HC

Associatedwith

theris

kofleft

ventriculardysfunction.

NA

Microarray

RT-qPCR

Vauso

rtetal.

(2016)

MS

Hsa

_circ_00054

02

Hsa

_circ_0035

560

Leuco

cyte

Down

45MS/26HC

NA

0.899

0.706

Microarray

RT-qPCR

Iparraguirre

etal.

(2017)

NSCLC

Hsa

_circ_01025

33

Whole

blood

Up

41NSCLC/26

HC

Associatedwith

tumortype,

TNM

stage,

LNM

anddistant

metastasis.

0.774

Microarray

RT-qPCR

Zhouetal.

(2018b

)

OA

Hsa

_circ_00321

31

Whole

blood

Up

25OA/25HC

Associatedwith

the

pathologicalp

roce

ss.

0.846

Microarray

RT-qPCR

Wangetal.

(2019f)

KBD

/OA

Hsa

_circ_00200

14

Whole

blood

Down

25KBD/25OA

Associatedwith

early

diagnosisofOAandKBD.

0.642

Microarray

RT-qPCR

Wangetal.

(2020e

)

PE

Hsa

_circ_00049

04

Hsa

_circ_0001

855

Whole

blood

Up

35PE/35HC

Associatedwith

serum

pregnan

cy-ass

ociated

plasm

aprotein

Aleve

l.

0.611

0.621

Microarray

RT-qPCR

Jiangetal.

(2018)

PE

Hsa

_circ_10122

2Blood

corpus

cle

Up

41PE/41HC

NA

0.706

Microarray

RT-qPCR

Zhangetal.

(2016)

PMOP

Hsa

_circ_00012

75

PBMC

Up

58PMOP/41HC

Associatedwith

T-sco

re.

0.759

Microarray

RT-qPCR

Zhaoetal.

(2018a

)

RA

Hsa

_circ_00442

35

Whole

blood

Down

77RA/31SLE/

50HC

NA

0.779

RT-qPCR

Luoetal.

(2018)



Protein

&Cell

Table

2co

ntin

ued

Disease

CircRNA

biomarker

Source

Expression

change

Cohort

size

Clin

ical

significan

ce

AUC

Method

Reference

RA

Hsa

_circ_10487

1Hsa

_circ_00352

4Hsa

_circ_10187

3Hsa

_circ_10304

7

PBMC

Up

35RA/30HC

NA

0.833

0.683

0.676

0.671

Microarray

RT-qPCR

Ouya

ngetal.

(2017)

RA

Hsa

_circ_00003

96

Hsa

_circ_01304

38

PBMC

Down

32RA/20HC

NA

0.809

0.774

RNA-seq

RT-qPCR

Yangetal.

(2019)

SLE

Hsa

_circ_00004

79

PBMC

Up

97SLE

/50RA/

89HC

Associatedwith

albumin

leve

l,urin

eprotein,IgG,

leuko

cytes,

hem

oglobin

and

ESR.

0.731

RNA-seq

RT-qPCR

Guoetal.

(2019)

SLE

Hsa

_circ_00577

62

Hsa

_circ_00030

90

Whole

blood

Up

24SLE

/24HC

Hsa

_circ

_0057

762was

ass

ociatedwith

theSLEDAI-

2Ksc

ore.

0.804

0.848

Microarray

RT-qPCR

Lie

tal.

(2019b

)

SLE

Hsa

_circ_00442

35

Hsa

_circ_0068

367

PBMC

Down

79SLE

/30RA/

62HC

Hsa

_circ

_0044

235was

ass

ociatedwith

thenumbers

ofmonocytesand

autoantibodies.

0.873

0.768

Microarray

RT-qPCR

Luoetal.

(2019)

SLE

Circ

PTPN22

PBMC

Down

53SLE

/40HC

Associatedwith

SLEactivity

indexsc

ores.

0.918

RNA-seq

RT-qPCR

Miaoetal.

(2019)

SLE

Hsa

_circ_40717

6Hsa

_circ_00130

8PBMC

Down

126SLE/102

HC

NA

0.806

0.722

Microarray

RT-qPCR

Zhangetal.

(2018e

)

SLE

Hsa

_circ_00129

19

CD4+

Tce

llUp

28SLE

/18HC

Associatedwith

clinical

andlab

features

.–

Microarray

RT-qPCR

Zhangetal.

(2018a

)

Sch

izop

hrenia

Hsa

_circ_10459

7PBMC

Down

102 Sch

izop

hrenia/

103HC

Significa

ntexp

ression

difference

betweenpre-

treatm

entandpost-

treatm

ent.

0.885

Microarray

RT-qPCR

Yaoetal.

(2019)

T2DM

Circ

ANKRD36

Leuco

cyte

Up

43T2DM/45HC

Associatedwith

gluco

se,

glyco

sylatedhemoglobin

and

interle

ukin.

NA

RNA-seq

RT-qPCR

Fangetal.

(2018)

T2DM

Hsa

_circ_00546

33

Whole

blood

Up

90T2DM/83

Pre-diabetes/

86HC

NA

0.751

Microarray

RT-qPCR

Zhaoetal.

(2017b

)

TB

Hsa

_circ_10301

7Hsa

_circ_05991

4Hsa

_circ_1011

28

PBMC

Up

31TB/30HC

Hsa

_circ

_1011

28was

ass

ociatedwith

theleve

lof

let‐7a.

0.870

0.821

0.817

Microarray

RT-qPCR

Fuetal.

(2019)



Protein

&Cell

Table

2co

ntin

ued

Disease

CircRNA

biomarker

Source

Expression

change

Cohort

size

Clin

ical

significan

ce

AUC

Method

Reference

TB

Hsa

_circ_00434

97

Hsa

_circ_00012

04

Monocyte

Up

101TB/88HC

Significa

ntexp

ression

difference

betweenpre-

treatm

entandpost-

treatm

ent.

0.860

0.848

Microarray

RT-qPCR

Huang

etal.,

(2017c

)

TB

Hsa

_circ_00193

7PBMC

Up

178TB/40PE/

40COPD/40

LC/133HC

Associatedwith

radiological

scores;

significa

nte

xpress

ion

difference

betweenpre-

treatm

entandpost-

treatm

ent.

0.873

Microarray

RT-qPCR

Huang

etal.,

(2018c

)

TB

Hsa

_circ_00013

80

PBMC

Down

32TB/31HC

NA

0.950

RT-qPCR

Luoetal.

(2020b

)

TB

Hsa

_circ_00004

14

Hsa

_circ_00006

81

Hsa

_circ_00021

13

Hsa

_circ_00023

62

Hsa

_circ_00029

08

Hsa

_circ_00087

97

Hsa

_circ_00631

79

PBMC

Up

12TB/13HC

NA

0.946

RNA-seq

Microarray

RT-qPCR

Qianetal.

(2018)

TB

Hsa

_circ_00058

36

PBMC

Down

49TB/45HC

NA

NA

RNA-seq

RT-qPCR

(Zhuan

getal.,

201

7)

Abbreviation:AIS:acu

teisch

emic

stroke

;ALS

:amyo

trophiclaterals

clerosis;

AML:

acu

temye

loid

leuke

mia;CAD:co

ronary

artery

disea

se;CAP:co

mmunity‐acq

uire

dpneumonia;CML:

chronic

mye

loid

leuke

mia;COPD:ch

ronic

obs

tructivepulm

onarydisea

se;CSCC:ce

rvical

squam

ousce

llca

rcinoma;EH:essentia

lhyp

ertension

;GC:gastric

cance

r;HC:hea

lthyco

ntrol;

HCC:hepatoce

llularca

rcinoma;HT:

Hash

imoto’s

thyroiditis;

IA:intracranialane

urysm

;KBD:Kash

in-Bec

kdisea

se;MI:myo

cardialinfarctio

n;MS:multiple

sclerosis;

NA:notapp

licable;

NSCLC:n

on-smallce

lllungca

nce

r;OA:osteoa

rthritis;P

BMC:perip

heralb

loodmonon

uclearce

ll;PE:pre-eclam

psia;P

MOP:postmenopau

salo

steop

orosis;

RA:rheumatoid

arthritis;SLE:

systemic

lupuserythem

atosu

s;T2DM:type2diabetesmellitus;

TB:tuberculosis..



Protein

&Cell

muscleblind (MBL), is the first circRNA that was discoveredto function as an RBP sponge (Ashwal-Fluss et al., 2014). Itwas observed that both the circular exon and the flankingintrons of circMbl contain many MBL binding sites that canbe specifically bound by MBL (Ashwal-Fluss et al., 2014).Therefore, increased expression of MBL can promote back-splicing of circMbl and lead to decreased expression ofcanonical linear MBL transcripts. At the same time, circMblcan sequester MBL in the cytoplasm and prevent it fromperforming splicing functions (Ashwal-Fluss et al., 2014).Several other circRNAs, such as circAmotl1 (Yang et al.,2017b), circANRIL (Holdt et al., 2016), circPABPN1 (Abdel-mohsen et al., 2017), circPOLR2A (Li et al., 2017e) andcircDHX34 (Li et al., 2017e), can function as protein decoysas well. Second, circRNAs can form a complex with otherRNAs and proteins to perform functions. For example, someintron-containing circRNAs, such as circEIF3J and circBPTF,can interact with U1 snRNP to form an RNA-protein com-plex, which further interacts with RNA polymerase II (Pol II)at the promoter region of their parental genes and enhancestheir transcription (Zhang et al., 2013; Li et al., 2015c).Similarly, some ciRNAs, such as ci-ankrd52, can alsoassociate with Pol II as well (Zhang et al., 2013; Li et al.,2015c). They can also regulate the expression of their par-ental gene by modulating the elongation of Pol II (Zhanget al., 2013; Li et al., 2015c). Many circRNAs, rather than aspecific circRNA, can interact with NF90/NF110 as a groupto form circRNP complexes in the cytoplasm, which mayfunction as a reservoir of NF90/NF110 (Li et al., 2017e).Upon viral invasion, NF90/NF110 could be released fromcircRNP complexes and bind to viral mRNAs to perform itsantiviral functions (Li et al., 2017e). Third, the interaction ofcircRNAs with proteins can facilitate or block the function oftheir binding proteins. For instance, increased expression ofcircFoxo3 in the cytoplasm may arrest the anti-senescentprotein ID1, transcription factor E2F1, anti-stress proteinsFAK and HIF1α and prevent the nuclear translocation ofthese transcription factors and their function, thus leading toincreased cellular senescence (Du et al., 2017). CircFoxo3can also form a tertiary circRNA-protein complex with cyclin-dependent kinase 2 (CDK2) and cyclin-dependent kinaseinhibitor 1 (p21), which facilitates the inhibition of CDK2 byp21, thus blocking cell cycle progression (Du et al., 2016a).Moreover, circFoxo3 can bind to Mdm2 and p53 in a breastcancer cell line, which leads to tumor cell apoptosis (Duet al., 2016b).

Translation

Some circRNAs with an internal ribosome entry site (IRES),such as circMbl (Pamudurti et al., 2017) and circZNF609(Legnini et al., 2017), can share the start codon of their hostgenes and be translated in a cap-independent manner. Thetranslation initiation of circRNAs may be promoted by N6-methyladenosine (m6A) modification of circRNAs (Yanget al., 2017c). In the human heart, 40 circRNAs were

observed to be associated with ribosomes and thus may betranslated (Heesch et al., 2019). Some of them, includingcircSLC8A1, circMYBPC3 and circRYR2, are heart-specificcircRNAs (Heesch et al., 2019). The translation of thesecircRNAs may be related to the physiological roles of thehuman heart. Further functional analysis observed that somecircRNA-encoded proteins can exert agonist or antagonisteffects on cancer progression. For example, PINT87aa(Zhang et al., 2018d) and AKT3-A74aa (Xia et al., 2019), twopeptides encoded by circRNAs, can interact with signalfactors and inhibit glioblastoma tumorigenicity in glioblas-toma. However, circGprc5a-secreted peptides can promotebladder oncogenesis and metastasis through its binding toGprc5a (Gu et al., 2018).

CIRCRNA EXPRESSION AND ITS IMPLICATION INDISEASES

Given the important functions that circRNAs can perform,their expression may provide significant clues in under-standing the underlying mechanisms of biological processesand disease states. Previous studies have observed thatcircRNAs are widely expressed in all tissues (Zhou et al.,2018a; Ji et al., 2019; Wu et al., 2020a) and cell types(Salzman et al., 2013) of nearly all species across theeukaryotic tree of life (Wang et al., 2014). Specifically, cir-cRNAs are enriched in brain samples (Westholm et al.,2014; Rybak-Wolf et al., 2015; Szabo et al., 2015; Venøet al., 2015) and human blood samples (Memczak et al.,2015), including peripheral blood mononuclear cells(PBMCs) (Qian et al., 2018), erythrocytes (Alhasan et al.,2016), platelets (Alhasan et al., 2016) and exosomes (Liet al., 2015b). Furthermore, circRNAs are expressed in tis-sue- (Guo et al., 2014; Szabo et al., 2015; Zhou et al.,2018a) or cell-type- (Salzman et al., 2013) specific and age-dependent manner (Rybak-Wolf et al., 2015; Szabo et al.,2015; You et al., 2015; Zhou et al., 2018a). Using the ratBodyMap dataset, we observed tissue-specific circRNAexpression in all 11 rat tissues, which may be closely relatedto the physiological functions of those tissues (Zhou et al.,2018a). The dynamic expression of circRNAs across timehas also been correlated with neuronal differentiation (Ry-bak-Wolf et al., 2015), neural development (Szabo et al.,2015; You et al., 2015), human terminal B cell differentiation(Agirre et al., 2019) and spermatogenesis (Zhou et al.,2018a).

The spatiotemporal expression of circRNAs is mediatedby the balance between circRNA generation and circRNAturnover, which can be regulated at both the transcriptionaland posttranscriptional levels (Li et al., 2018d; Chen, 2020).On the one hand, the production of circRNAs can be regu-lated by intronic complement sequences (ICSs) (Jeck et al.,2013; Liang and Wilusz, 2014; Zhang et al., 2014), cis-splicing elements (Ashwal-Fluss et al., 2014; Starke et al.,2014; Wang and Wang, 2015) and trans-acting splicing



Protein

&Cell

factors (Ashwal-Fluss et al., 2014; Conn et al., 2015; Krameret al., 2015; Khan et al., 2016; Aktaş et al., 2017; Errichelliet al., 2017; Fei et al., 2017; Li et al., 2017e). On the otherhand, the degradation of circRNAs can be affiliated bymiRNA-initiated AGO2 cleavage (Kleaveland et al., 2018)and nuclease-mediated cleavage, including RNase Lendonuclease (Liu et al., 2019a), RNase P/MRP endonu-clease complex (Park et al., 2019) and G3BP1 endonucle-ase (Fischer et al., 2020). Dysregulation of circRNAgeneration or turnover may lead to aberrant circRNAexpression in cells or tissues, which may be related to manyhuman diseases (Chen et al., 2016). Cdr1as, one of the moststudied circRNAs, harbors more than 70 binding sites formiR-7, and its overexpression strongly inhibits the activity ofthe tumor suppressor miR-7 (Hansen et al., 2013). Accu-mulating evidence has demonstrated its significant increasein colorectal cancer samples, which can inhibit miR-7 func-tion and activate EGFR and RAF1 oncogenes (Weng et al.,2017). Moreover, the dysregulation of the miR-7/Cdr1as axisis involved in some other diseases, such as Alzheimer’sdiseases (Lukiw, 2013), diabetes (Xu et al., 2015) and car-diometabolic diseases (Geng et al., 2016). Furthermore, theloss of the Cdr1as locus causes miRNA deregulation andaffects synaptic transmission in knockout mice (Piweckaet al., 2017). In addition to Cdr1as, other circRNAs are alsoimplicated in many different human diseases, includingcancers (Zhang et al., 2018j; Chen et al., 2019b; Smid et al.,2019; Su et al., 2019; Vo et al., 2019), cardiovascular dis-eases (Zhang et al., 2018j; Aufiero et al., 2019), neurode-velopment and neurodegenerative diseases (Kumar et al.,2017; Zhang et al., 2018j; Dube et al., 2019; Chen et al.,2020d; Hanan et al., 2020) and immune diseases (Chenet al., 2019c; Zhou et al., 2019b).

CIRCRNAS ARE PROMISING BIOMARKERS FORHUMAN DISEASES

In general, biomarkers are physiological indexes or biologi-cal molecules that can be objectively measured and canindicate either a normal or a pathogenic state (Lesko andAtkinson, 2001). Technological advances in genomics,transcriptomics, proteomics and metabolomics have led tothe discovery and validation of many biomarkers that arepushing personalized medicine forward (Chen et al., 2012;Karczewski and Snyder, 2018; Ahadi et al., 2020). A goodbiomarker with clinical significance must meet the criteria ofanalytical validity, clinical validity and clinical utility (Byronet al., 2016). As a novel type of noncoding RNA, circRNAhas several distinct advantages over canonical linear RNAsas a disease biomarker (Zhang et al., 2018j). First, circRNAis more stable than linear RNAs because it has a closed loopstructure without 5-prime and 3-prime ends (Enuka et al.,2015; Li et al., 2015b). In the MCF10A human mammaryepithelial cell line, Enuka et al. observed at least 2.5 timeslonger half-lives of circRNAs compared to linear RNAs,

including mRNAs and miRNAs (Enuka et al., 2015). Second,circRNAs are abundantly expressed in many tissue sam-ples. For example, brain samples express more than 10,000circRNAs in the rat BodyMap dataset (Zhou et al., 2018a). Insome cases, the expression values of circRNAs are muchhigher than their linear counterparts (Westholm et al., 2014;Rybak-Wolf et al., 2015; You et al., 2015; Liang et al., 2017).Third, circRNAs are expressed in a tissue-specific (Salzmanet al., 2013; Guo et al., 2014; Szabo et al., 2015; Zhou et al.,2018a) and developmental stage-specific manner (Rybak-Wolf et al., 2015; Szabo et al., 2015; You et al., 2015; Zhouet al., 2018a). Importantly, the tissue specificity of circRNA ishigher than that of mRNA of its host gene (Guo et al., 2014;Zhou et al., 2018a). These features suggest that circRNAsmay have better analytical validity (Zhang et al., 2018j),including analytical specificity, robustness, reproducibilityand repeatability (Byron et al., 2016), when used as bio-marker molecules. In a recent study, Maass et al. investi-gated circRNA expression in 20 clinically relevant tissuesamples, which underscored the feasibility of using cir-cRNAs as potential disease biomarkers (Maass et al., 2017).

To date, many circRNAs have been identified asbiomarkers for human diseases (Zhang et al., 2018j),especially human cancers (Li et al., 2015a; Meng et al.,2017; Su et al., 2019; Sheng et al., 2020). For example,Cdr1as has been revealed as a prognostic biomarker incolorectal cancer patients (Weng et al., 2017). In a trainingcohort comprising 153 primary colorectal cancer tissues and44 matched normal mucosae, significantly increased Cdr1asexpression was observed in colorectal cancer tissues, andits overexpression was associated with poor patient survival.The prognostic power of Cdr1as was further validated in anindependent validation cohort comprising 165 colorectalcancer patients (Weng et al., 2017). A four-circRNA signa-ture, consisting of hsa_circ_101308, hsa_circ_104423,hsa_circ_104916 and hsa_circ_100269, has been con-structed that can predict the early recurrence of stage IIIgastric cancer (GC) patients after surgery (Zhang et al.,2017). Moreover, Vo et al. constructed a comprehensivecatalog of circRNAs in human cancer tissues in a systematicanalysis of more than 2,000 cancer samples (Vo et al.,2019). They identified two circRNAs, circ-AURKA and circ-AMACR, as potential diagnostic biomarkers for neuroen-docrine prostate cancer (Vo et al., 2019). Although thesecircRNA biomarkers are potentially useful in cancer man-agement, most of them are derived from tissue samples (Liet al., 2015a; Meng et al., 2017; Su et al., 2019; Sheng et al.,2020). To improve biomarker accessibility, especially forbiomarkers of cancer screening and diagnosis, circRNAbiomarkers in body fluids are ideal for clinical application(Anfossi et al., 2018). Hence, the authors further investigatedand validated the reliability of detecting circRNA biomarkersin urine samples of prostate cancer patients (Vo et al., 2019),which suggests that circRNA in urine samples is a promisingstrategy for prostate cancer screening.



Protein

&Cell

BLOOD CIRCRNAS AND LIQUID BIPOSYBIOMARKERS

Liquid biopsy has been a revolutionary tool in diseasemanagement, supporting the diagnosis, prognosis andtreatment guidance of human diseases (Heitzer et al., 2019;Mattox et al., 2019; Rubis et al., 2019; Luo et al., 2020a). Incomparison to urine, saliva or cerebrospinal fluid, peripheralblood has been used as the major body fluid in liquid biopsy(Rubis et al., 2019; Luo et al., 2020a). In the circulatingblood, aberrantly expressed RNAs or fusion transcripts indifferent blood components have been associated withhuman cancers (Byron et al., 2016; Zaporozhchenko et al.,2018; Sole et al., 2019) and infectious diseases (Byron et al.,2016; Correia et al., 2017). These RNA biomarkers includecfRNAs in plasma or serum (Mitchell et al., 2008; Fehlmannet al., 2020), exosome-derived RNAs (Maas et al., 2017),EV-incorporated cfRNAs (Maas et al., 2017) and RNA tran-scripts in tumor-educated platelets (Best et al., 2017, 2018).These cell-free or EV-incorporated RNA biomarkers repre-sent the changes in expression that occur in abnormal cells,such as dysregulated genes in cancer cells (Byron et al.,2016; Rubis et al., 2019; Luo et al., 2020a). Other thancfRNAs, gene expression profiles of PBMCs or whole bloodhave been proven to be good indicators of many humandiseases (Chaussabel and Baldwin, 2014; Chaussabel,2015), since they can assess the immune status (Chauss-abel, 2015). Unlike the expression of cfRNAs in serum,plasma or EVs, RNA expression levels in PBMCs or wholeblood are measures of the host response to exogenouspathogens or autoantigens (Schnell et al., 2018; Shaked,2019). Several whole blood or PBMC gene expression sig-natures have been developed for cancer management,including the early diagnosis of colorectal cancer (Marshallet al., 2010; Ciarloni et al., 2015, 2016) and lung cancer(Showe et al., 2009; Kossenkov et al., 2018), the prognosisof adult acute myeloid leukemia (Bullinger et al., 2004; Valket al., 2004) and prostate cancer patients (Ross et al., 2012),and the monitoring of renal cell carcinoma relapse (Giraldoet al., 2017). Moreover, whole blood or PBMC RNA signa-tures have been widely investigated as liquid biopsy diag-nostic tools for infectious diseases (Ramilo and Mejias,2017), such as discriminating influenza from other respira-tory viral infections (Zaas et al., 2009), differentiating viraland bacterial infections (Tsalik et al., 2016), diagnosingseptic patients (Sweeney et al., 2018; Gunsolus et al., 2019;Mayhew et al., 2020), and discriminating active tuberculosis(TB) patients from patients with latent TB or other lung dis-eases (Bloom et al., 2013; Anderson et al., 2014; Qian et al.,2016; Zak et al., 2016; Sambarey et al., 2017; MacLeanet al., 2019; Warsinske et al., 2019; Esmail et al., 2020;Gupta et al., 2020). In addition, the transcriptome of bloodcells has been implicated in assessing the likelihood ofdeveloping obstructive coronary artery disease in symp-tomatic nondiabetic patients (Vargas et al., 2013) and

predicting antibody-mediated kidney allograft rejection inkidney transplant patients (Loon et al., 2019).

With regard to RNA molecules, both protein-codingmRNAs (Byron et al., 2016; Sole et al., 2019) and severalclasses of noncoding RNAs (Byron et al., 2016; Anfossiet al., 2018; Pardini et al., 2019; Sole et al., 2019) have beenused as blood disease biomarkers. In comparison to mRNAsor long noncoding RNAs, small noncoding RNAs, such asmiRNAs (Mitchell et al., 2008; Max et al., 2018; Fehlmannet al., 2020) and noncanonical small RNAs (Fritz et al., 2016;Pardini et al., 2019), have the advantage of high stability inthe circulating blood (Mitchell et al., 2008; Anfossi et al.,2018). This superiority is especially important when trans-lating RNA biomarkers into clinical practice (Byron et al.,2016; Anfossi et al., 2018), since fast mRNA degradation inblood sample processing may affect the performance ofmRNA biomarkers (Dvinge et al., 2014). Given the highstability of circRNAs, great efforts and substantial progresshave been made to investigate the possibility of using cir-cRNA biomarkers in liquid biopsy in recent years.

Next, we summarize circRNAs in the peripheral blood,their correlations with human diseases and their potentialapplication in liquid biopsy as disease biomarkers (Fig. 2).We classified blood circRNAs into blood cell-free circRNAsand circRNAs in blood cells since they have distinct biolog-ical meanings in the context of human diseases. Blood cell-free circRNAs, including circulating cell-free circRNAs inplasma, serum and blood EVs, are secreted from differenttissue cells into the blood. Therefore, cell-free circRNAshave corresponding tissue origins, and cell-free circRNAbiomarkers represent their clinical significance in the originaltissue (Li et al., 2019d, 2020). On the other hand, blood cellcircRNAs consist of circRNAs in various blood cells, such asmonocytes, erythrocytes, neutrophils and platelets. Cir-cRNAs in mixtures of different blood cells, such as circRNAsin lymphocytes, PBMCs and whole blood, are classified inthis group as well. The circRNA expression profiles ofperipheral blood cells or a mixture of blood cells are impor-tant indicators of the host’s immune status (Chaussabel,2015), which may undergo dynamic changes during acuteevents such as viral infections (Chen et al., 2018a; Roseet al., 2019; Zhou et al., 2019a). Hence, circRNA biomarkersin blood cells, PBMCs or whole blood represent the specificimmune response of an individual to different physiologicalaspects.

Cell-free circRNAs in peripheral blood

Mounting evidence has demonstrated abundant circRNAexpression in blood plasma (Maass et al., 2017; Yi et al.,2018; Smid et al., 2019) and serum (Gu et al., 2017; Maasset al., 2017; Sonnenschein et al., 2019; Sun et al., 2020b)(Fig. 2A). In the circulating blood, EVs, including exosomesand microvesicles, are important blood components thathave great diagnostic and prognostic value (Revenfeld et al.,2014). Li et al. investigated circRNA expression in MHCC-



Protein

&Cell

LM3 liver cancer cells and cell-derived exosomes and foundat least 2-fold circRNA enrichment in exosomes compared toproducer cells (Li et al., 2015b). They further found that theexpression level of exosomal circRNAs is largely the sameafter serum incubation at room temperature for up to 24 h (Liet al., 2015b). Using the extracellular vesicle long RNA(exLR) sequencing method, the same group explored theexpression profiles of circRNAs in 352 plasma EV samples(Li et al., 2019d). They found 137,196 circRNA candidatesthat were expressed in normal plasma EV samples, and thecircular to linear ratio was significantly higher in EVs than inPBMCs (Li et al., 2019d). A recent study also isolated pla-telet-derived EVs and demonstrated the selective release ofplatelet-specific circRNAs in exosomes and microvesicles(Preußer et al., 2018). Based on these data, a database ofblood exosomal circRNAs, exoRBase, has been developedto facilitate the development of circRNA signatures in bloodexosomes (Li et al., 2017b). In addition, two previous studieshave observed that the majority of circulating cell-free miR-NAs were associated with AGO2 protein complexes ratherthan with vesicles in blood plasma (Arroyo et al., 2011;Turchinovich et al., 2011). Like cell-free miRNAs, blood cell-free circRNAs may bind to several RNA-binding proteins orprotein complexes as well (Huang et al., 2020a). Therefore,some cell-free circRNAs may not be associated with bloodEVs. In a pilot study of circRNA expression profiles in 348primary breast cancer tissue samples, Smid et al. found thatcircCNOT2 is a prognostic biomarker of aromatase inhibitortherapy in advanced breast cancer patients (Smid et al.,2019). To investigate the possibility of using circCNOT2 as anoninvasive biomarker, they further amplified circCNOT2 inplasma samples of four breast cancer patients and foundvariable circCNOT2 expression in all plasma samples (Smidet al., 2019). All these results suggest that cell-free circRNAsin peripheral blood are stable, enriched and detectable.

Blood EVs can be secreted from cells of many differentbiological systems and can be circulated to recipient cellsthrough the bloodstream (Revenfeld et al., 2014; Lasda andParker, 2016). Therefore, blood cell-free circRNAs can per-form vital roles in many biological processes, such as cancercell proliferation, cancer metastasis, drug resistance,hemostasis and inflammation (Fig. 2B) (Wang et al., 2019e;Cui et al., 2020; Li et al., 2020; Shang et al., 2020; Xu et al.,2020b). For example, circ-deubiquitination (circ-DB) in adi-pose-secreted exosomes was found to regulate deubiquiti-nation via the suppression of miR-34a and the activation ofdeubiquitination-related USP7 in plasma samples of hepa-tocellular carcinoma (HCC) patients, which could reduceDNA damage and promote HCC cell growth (Zhang et al.,2018b). Exosomal circSHKBP1 was found to promote theprogression of GC by regulating the miR-582-3p/HUR/VEGFpathway and suppressing HSP90 degradation (Xie et al.,2020a). CircUHRF1 in plasma exosomes can inhibit thefunctions of natural killer cells by upregulating the expressionof TIM-3 via miR-449c-5p degradation and drive resistanceto anti-PD1 immunotherapy in HCC patients (Zhang et al.,

2020a). Moreover, exosomal circHIPK3 can preventischemic muscle injury by downregulating miR-421 expres-sion, increasing FOXO3a expression, inhibiting pyroptosisand releasing IL-1β and IL-18 (Yan et al., 2020).

Given the biological functions that blood cell-free cir-cRNAs can perform (Wang et al., 2019e; Cui et al., 2020; Liet al., 2020) and their implications in many human diseases,cell-free circRNAs in the peripheral blood have greatpotential as liquid biopsy biomarkers of human diseases(Fig. 2C) (Anfossi et al., 2018; Zaporozhchenko et al., 2018;Zhang et al., 2018j). To date, many blood cell-free circRNAshave been introduced for cancer management, includingearly cancer diagnosis, cancer prognosis and prediction ofcancer treatment (Preußer et al., 2018; Aufiero et al., 2019;Fraipont et al., 2019; Lu et al., 2019c; Pardini et al., 2019;Sole et al., 2019; Su et al., 2019; Wang et al., 2019e; Bel-trán-García et al., 2020; Cui et al., 2020; Li et al., 2020). In arecent study, Luo et al. measured the expression levels oftwo circRNAs in plasma samples of 231 lung cancer patientsand 41 healthy controls using reverse transcription dropletdigital PCR (RT-ddPCR) (Luo et al., 2020c). They identifiedhsa_circ_0000190 as a circRNA biomarker in human bloodplasma that can predict the survival outcomes of lung cancerpatients (Luo et al., 2020c). Furthermore, the increasedexpression of plasma hsa_circ_0000190 was also correlatedwith poor response to systemic therapy and immunotherapyin lung cancer patients (Luo et al., 2020c). Similarly, Li et al.investigated the clinical relevance of serum exosomal cir-cFLI1 in lung cancer patients in a cohort of 61 small cell lungcancer (SCLC) patients and 55 normal subjects. They foundthat serum exosomal circFLI1 levels were significantly higherin SCLC patients, especially in SCLC patients with distantmetastasis (Li et al., 2018b). Notably, they observed thatSCLC patients with lower exosomal circFLI1 expressionlevels experienced longer disease remissions, indicating itsprognostic power in SCLC. The authors also suggested thatserum exosomal circFLI1 may be used as a biomarker thatcan monitor the clinical response to chemotherapy in SCLCpatients (Li et al., 2018b). By analyzing blood plasma sam-ples of 62 GC patients and 25 healthy controls, Tang et al.proposed a novel circulating diagnostic biomarker of GC,plasma exosomal circKIAA124, that was correlated withclinical TNM stage, lymphatic metastasis and overall survivaltime of GC patients (Tang et al., 2018). In addition to theaberrant expression of blood cell-free circRNAs, the pres-ence of some fusion circRNAs (f-circRNAs) in the peripheralblood has also been used as a liquid biopsy biomarker.Guarnerio et al. found that f-circRNAs could be producedfrom cancer-associated chromosomal translocations incancer cells (Fig. 1), and f-circRNAs could promote cellulartransformation, cell viability and resistance upon therapy(Guarnerio et al., 2016). In a systematic analysis of f-cir-cRNAs in localized prostate cancer tissues, Chen et al.observed more f-circRNAs in tumors with worse prognosis(Chen et al., 2019b). Due to the lack of recurrence of thesef-circRNAs, the authors suggested that f-circRNAs are good



Protein

&Cell

biomarker candidates (Chen et al., 2019b). In NSCLCpatients, Tan et al. found that F-circEA, an f-circRNA origi-nating from the EML4-ALK fusion gene, was exclusivelyexpressed in the plasma of patients with the EML4-ALKfusion (Tan et al., 2018). Therefore, plasma F-circEA mayserve as a liquid biopsy biomarker to diagnose NSCLCpatients with EML4-ALK translocation and guide targetedtherapy for NSCLC patients in this subgroup (Tan et al.,2018). More cell-free circRNA biomarkers in the blood aresummarized in Table 1.

CircRNAs in blood cells and whole blood

CircRNA expression in blood cells and whole blood, a majorsource of liquid biopsy samples, has been extensivelyinvestigated (Fig. 2A). In a pilot study, Memczak et al.detected thousands of circRNAs in peripheral whole bloodsamples using RNA-seq (Memczak et al., 2015). They foundthat the expression levels of these blood circRNAs werecomparable to those in circRNA-rich cerebellar tissue(Memczak et al., 2015). In separate blood cell populations,circRNAs were observed to be enriched approximately100-fold in platelets and anucleate erythrocytes relative tonucleated tissues, such as lung, brain and colon (Alhasanet al., 2016). Gaffo et al. investigated circRNA expression inT cells, B cells and monocytes of healthy subjects and foundabundant circRNA expression in these mature blood cells(Gaffo et al., 2019). We also explored the expression land-scape of circRNAs in PBMCs and found that the expressionlevel of circRNAs in PBMCs, together with that in platelets,red blood cells (RBCs) and whole blood, is high enough tobe detected (Qian et al., 2018). All these results suggest thatwhole blood (Memczak et al., 2015; Qian et al., 2018),PBMCs (Qian et al., 2018), and several blood cells, includingneutrophils (Maass et al., 2017), T cells (Gaffo et al., 2019),B cells (Gaffo et al., 2019), monocytes (Gaffo et al., 2019),RBCs (Alhasan et al., 2016; Qian et al., 2018) and platelets(Maass et al., 2017; Qian et al., 2018; Gaffo et al., 2019), arereliable clinical samples for circRNA profiling in liquid biopsy.

Accumulating evidence has suggested that circRNAsplay crucial roles in the immune response and its regulation(Fig. 3) (Chen et al., 2017b, 2019c; Xu et al., 2018b, 2020a;Yang et al., 2018; Zhou et al., 2019b; Awan et al., 2020).First, circRNAs are important regulators of blood cell bio-genesis, differentiation and activation (Chen et al., 2019c;Zhou et al., 2019b; Xu et al., 2020a). In a comprehensivestudy of circRNA expression in hematopoietic progenitors,differentiated lymphoid and myeloid cells, Nicolet et al.observed a cell-type specific pattern of circRNA expressionprofiles in blood cells, and the type and number of circRNAsincreased upon hematopoietic maturation (Nicolet et al.,2018). Moreover, Holdt et al. found that circANRIL can bindto PES1 to impede the generation of pre-rRNA and ribo-somes, resulting in the biogenesis of macrophages by p53activation (Holdt et al., 2016). Second, circRNAs are activelyinvolved in antiviral immune responses (Fig. 3B) (Cadena

and Hur, 2017; Awan et al., 2020). For example, Chen et al.showed that exogenous circRNAs released by viruses canbe recognized by the pattern recognition receptor RIG-I ofhost cells, which activates host innate immunity (Chen et al.,2017a). In their subsequent work, the authors showed thatm6A RNA modification of human circRNAs inhibits innateimmunity, while unmodified circRNAs and K63-polyubiquitincan activate RIG-I and innate immune response (Chen et al.,2019a). Li et al. found that two immune factors, NF90/NF110,not only promote circRNA production in the nucleus but alsobind to mature host circRNAs to form circRNP complexes inthe cytoplasm. Upon viral infection, circRNP complexes inthe cytoplasm can release NF90/NF110, which binds viralmRNAs to inhibit viral replication (Li et al., 2017e). Moreover,Liu et al. presented that circRNAs can form RNA duplexesand act as inhibitors of innate immunity-related proteinkinase (PKR) under normal conditions (Liu et al., 2019a).Upon poly(I:C) treatment or viral infection, RNase L is acti-vated to efficiently degrade circRNAs, and PKR is thusreleased from circRNA inhibition to initiate the early cellularinnate immune response (Liu et al., 2019a). Third, circRNAsare closely associated with the antibacterial immuneresponse as well. Ng et al. characterized circRNAs inducedby lipopolysaccharide (LPS) and identified circRasGEF1B asa conserved positive regulator of the LPS r

Documents

The potential of using blood circular RNA as liquid biopsy ...R EVIEW The potential of using blood circular RNA as liquid biopsy biomarker for human diseases Guoxia Wen1, Tong Zhou2&,