The distinct roles of zinc finger CCHC-type (ZCCHC) superfamily proteins in the regulation of RNA metabolism

ABSTRACT

The zinc finger CCHC-type (ZCCHC) superfamily proteins, characterized with the consensus sequence C-X₂-C-X₄-H-X₄-C, are accepted to have high-affinity binding to single-stranded nucleic acids, especially single-stranded RNAs. In human beings 25 ZCCHC proteins have been annotated in the HGNC database. Of interest is that among the family, most members are involved in the multiple steps of RNA metabolism. In this review, we focus on the diverged roles of human ZCCHC proteins on RNA transcription, biogenesis, splicing, as well as translation and degradation.

KEYWORDS:

• Zinc finger CCHC-type superfamily proteins
• zinc-knuckle
• RNA metabolism
• transcription
• splicing
• translation
• degradation

1. Introduction

The zinc finger CCHC-type (ZCCHC) superfamily protein contains at least one consensus ZnF-like domain with C-X2-C-X4-H-X4-C sequence (where X is any amino acid, C is cysteine and H is histidine). This motif is also called as zinc-knuckle (see Fig. 1) which are seen in all ZCCHC proteins (see Table 1) [1]. That ZCCHC domains are generally composed of two short β-strands connected by a zinc knuckle followed by a short helix. First observed in the transcription factor IIIA (TFIIIA) from Xenopus oocytes 2, ZCCHC domains have been regarded as a typical nucleic acid-recognizing motifs [3]. It is suggested that this type of zinc finger has high-affinity binding to single-stranded nucleic acids [1]. Until recently a total of 25 ZCCHC proteins have been identified in this superfamily, from ZCCHC1 to ZCCHC25. Each member plays diverse roles in both cellular physiological function and disease aetiology (see Table 2, their characteristics and cellular functions are listed). Most of them are RNA-binding proteins (RBPs), while some mediate DNA binding and protein–protein interactions through their ZCCHC domains, such as ZCCHC3, 8, 9, 10, 11, 13, 22 [4–10].

Table 1. The CCHC sequences and structure of zinc finger CCHC-type superfamily

Protein	CCHC motif/Zinc–knuckle sequences	Position	Structure
ZCCHC1	CYNCGGLDHHAKEC CHFCQSISHMVASC	137-154 159–176
ZCCHC2	CYNCGVSGHYAQDC	1131–1148
ZCCHC3	CFKCGSRTHMSGSC CFRCGEEGHLSPYC CNLCGKRGHAFAQC	335-350 352–368 372–387
ZCCHC4	CFICGELDHKRSTC	443–460
ZCCHC5	CLYCGYPGHFARDC	443–462
ZCCHC6	CSLCKREGHLKKDC CRICGKIGHFMKDC CFICGREGHIKKEC	963-980 1345–62 1451–68
ZCCHC7	CRNCDKRGHLSKNC CFLCSRRGHLLYSC CDRCHMLGHYTDAC CYHCAQKGHYGHEC	241-258 263–280 304-321 348–365
ZCCHC8	CFNCGSEEHQMKDC	227–244
ZCCHC9	CFHCRKPGHGIADC CYRCGSTEHEITKC CFVCGEMGHLSRSC CKLCGSVEHLKKDC	128-145 155–172 184-201 211–228
ZCCHC10	CQKCLEFGHWTYEC	43–60
ZCCHC11	CSICKKDGHSKNDC CRVCGKIGHYMKDC CFICGDAGHVRREC	913-930 1293–310 1357–374
ZCCHC12	CSYCGEEGHSKETC	345–362
ZCCHC13	CYNCGRSGHIAKDC CYTCGRLGHLARDC CYSCGKLGHIQKDC CYRCGEIGHVAINC	65-82 89–106 110-127 28–145
ZCCHC14	CYNCGATGHRAQDC	906–923
ZCCHC15	CYRCGKEGHWSKEC	160–177
ZCCHC16	CLYCSQSGHFTRDC	278–295
ZCCHC17	CKKCGCKGHFAKDC	131–148
ZCCHC18	CSYCGEEGHSKETC	347–364
ZCCHC19	CYECGESGHLSYAC	105-122
ZCCHC20	CYECGEKGHYAYDC	104–120
ZCCHC21	CYRCGKEGHWSKEC	160–177
ZCCHC22	CFKCGRSGHWAREC CYRCGESGHLAKDC CYNCGRGGHIAKDC CYNCGKPGHLARDC CYSCGEFGHIQKDC CYRCGETGHVAINC CYRCGESGHLAREC	4-21 52–69 72-89 96–113 117-134 135–152 156–173
ZCCHC23	CDKCGKPGNVKNDC^a	64–81
ZCCHC24	CHLCFNKGHYIKDC	132–149
ZCCHC25	CTKCGGAGHIASDC	277-296

^aZCCHC23 harbours a degenerate CCHC-type zinc finger where histidine (H) residue is substituted by asparagine (N).

Table 2. Main characterization and cellular functions of zinc finger CCHC-type superfamily

HGNC ID (gene) Approved symbol	Approved name	Previous symbols Synonyms	Chromosome	Main functions	Refs
HGNC:15,986 LIN28A (ZCCHC1)	Lin-28 homolog A	LIN28, LIN-28, FLJ12457, ZCCHC1, CSDD1	1p36.11	miRNA biogenesis Cancers Stem cell pluripotency and differentiation Reprogramming Normal development Glucose metabolism	[154,184,185]
HGNC:22,916 ZCCHC2	Zinc finger CCHC-type containing 2	C18orf49, FLJ20281, KIAA1744, FLJ20222	18q21.33	Suppress retinoblastoma tumorigenesis Type I and IV hypersensitive reaction Haplotype Sharing	[11,186,187]
HGNC:16,230 ZCCHC3	Zinc finger CCHC-type containing 3	C20orf99, dJ1103G7.7	20p13	Transcription Viral-triggered innate immune response	[8,54,56]
HGNC:22,917 ZCCHC4	Zinc finger CCHC-type containing 4	HSPC052, FLJ23024, ZGRF4	4p15.2	Translation Tumour biology	[130,188,189]
HGNC:22,997 RTL3 (ZCCHC5)	Retrotransposon Gag like 3	ZCCHC5, FLJ38865, Mar3, Mart3, ZHC5, SIRH9	Xq21.1	Lung squamous cell carcinoma Establishment of muscle stem cells	[27,190]
HGNC:25,817 TUT7 (ZCCHC6)	Terminal uridylyl transferase 7	ZCCHC6, KIAA1711, FLJ13409, PAPD6, TENT3B	9q21.33	mRNA/miRNA degradation Reverse transcription Innate immune responses Osteoarthritis pathogenesis Embryo development Spermatogenesis	[143,147,165,168,191^]
HGNC:26,209 ZCCHC7	Zinc finger CCHC-type containing 7	FLJ22611, AIR1	9p13.2	RNA degradation Acute lymphoblastic leukaemia	[13,177]
HGNC:25,265 ZCCHC8	Zinc finger CCHC-type containing 8	DKFZp434E2220	12q24.31	RNA degradation Blastocyst Formation Familial pulmonary fibrosis	[15,181,192]
HGNC:25,424 ZCCHC9	Zinc finger CCHC-type containing 9	DKFZp761J139, PPP1R41	5q14.1	Transcription Non-small cell lung cancer HPV/cervical cancer	[6,26,28,193]
HGNC:25,954 ZCCHC10	Zinc finger CCHC-type containing 10	FLJ20094	5q31.1	Transcription DNA methylation Apoptosis Tumorigenesis Inflammation pathway	[9,10]
HGNC:28,981 TUT4 (ZCCHC11)	Terminal uridylyl transferase 4	ZCCHC11, KIAA0191, PAPD3, TENT3A	1p32.3	mRNA/miRNA degradation Tumour growth and metastasis Embryo development Spermatogenesis	[143,152,194]
HGNC:27,273 ZCCHC12	Zinc finger CCHC-type containing 12	FLJ16123, SIZN, SIZN1, PNMA7A	Xq24	Transcription papillary thyroid cancer X-linked mental retardation	[17,28,35,195]
HGNC:31,749 ZCCHC13	Zinc finger CCHC-type containing 13	4930513O09RIK, Cnbp2, ZNF9L	Xq13.2	DNA methylation Liver cancer Non-obstructive azoospermia	[10,25]
HGNC:24,134 ZCCHC14	Zinc finger CCHC–type containing 14	BDG29, MGC14139	16q24.2	Chronic hepatitis B virus infection Intracerebral haemorrhage Mental impairment Congenital malformation	[196–198]
HGNC:28,842 RBM4B ZCCHC15	RNAbinding motif protein 4B	RBM30, MGC10871, ZCCHC15, RBM4L, ZCRB3B, ZCCHC21B	11q13	/	/
HGNC:25,214 RTL4 (ZCCHC16)	Retrotransposon Gag like 4	ZCCHC16, Mart4, Mar4, FLJ46608, SIRH11	Xq23	Cognition function	[199]
HGNC:30,246 ZCCHC17	Zinc finger CCHC-type containing 17	PS1D, HSPC251, pNO40	1p35.2	Splicing Alzheimer’s disease Protein–protein interaction	[18,65,66]
HGNC:32,459 ZCCHC18	Zinc finger CCHC-type containing 18	SIZN2, PNMA7B	Xq22.2	Murine cytomegalovirus-associated disease in central nervous system	[200]
HGNC:29,620 ZCRB1 (ZCCHC19)	Zinc finger CCHC-type and RNA-binding motif containing 1	MADP-1, MADP1, RBM36, ZCCHC19, SNRNP31	12q12	Viral replication and transcription	[201]
HGNC:10,789 SRSF7 (ZCCHC20)	Serine- and arginine-rich splicing factor 7	SFRS7, 9G8, ZCRB2, HSSG1, AAG3, RBM37, ZCCHC20	2p22.1	Constitutive and alternative splicing mRNA export Apoptosis Tumorigenesis Alternative polyadenylation Nonsense mediated RNA decay Translation	[21,69,75,80]
HGNC:9901 RBM4 (ZCCHC21)	RNA binding motif protein 4	LARK, RBM4A, ZCRB3A, ZCCHC21	11q13.2	Alternative splicing Translation RNA silencing Supraspliceosome assembly	[64,87]
HGNC:13,164 CNBP (ZCCHC22)	CCHC-type zinc finger nucleic acid binding protein	DM2, ZNF9, RNF163, ZCCHC22, CNBP1	3q21.3	Transcription Translation RNA stability Myotonic dystrophy type 2 Innate immune response	[19,46,48,50,202]
HGNC:32,963 ZCCHC23	Zinc finger CCHC-type containing 23	FLJ45231	4q25	/	/
HGNC:26,911 ZCCHC24	Zinc finger CCHC-type containing 24	C10orf56, FLJ90798, Z3CXXC8	10q22.3	Apoptosis	[203]
HGNC:12,950 SF1 (ZCCHC25)	Splicing factor 1	ZNF162, ZFM1, ZCCHC25	11q13.1	Alternative splicing Tumorigenesis Spliceosome assembly	[105,115]

Figure 1. The CCHC/zinc-knuckle motif in the CCHC-type zinc finger proteins. C and H refers to cysteine and histidine, respectively. X is any amino acid, whereas there is still regularity to conform to. The conservatively substituted residues have been mentioned in the figure. That conserved array has high affinity to Zn+ and participates directly in nucleic acid recognition and binding, especially the single-stranded nucleic acids

Several members of the superfamily are correlated with diverse diseases. For examples, ZCCHC2 regulates retinoblastoma tumorigenesis by inhibiting the activity of c-Myc [11]; ZCCHC4 associates with the aetiology of attention deficit hyperactivity disorder (ADHD) [12]; ZCCHC7 disrupted by sequences 3ʹ of FRG1B and LOC1499503 is potential genetic targets in acute lymphoblastic leukaemia [13]; ZCCHC8 is identified as a fusion partner of ROS1 in congenital glioblastoma multiforme [14] and a novel cause of human short telomere syndromes–familial pulmonary fibrosis [15]; ZCCHC12 is referred to as a novel oncogene of papillary thyroid carcinoma [16,17]; ZCCHC17 regulates synaptic gene expression in Alzheimer’s disease [18]; abnormal expression of ZCCHC22 is a potential onset in myotonic dystrophy type 2 (DM2) [19]. ZCCHC1, ZCCHC5, ZCCHC9, ZCCHC13, ZCCHC20 and ZCCHC21 are related to various solid tumours, such as breast, lung, liver, colorectal and renal cancers [20–27].

More importantly, increasing literatures have implicated the ZCCHC superfamily in the regulation of RNA metabolism. RNA metabolism includes four distinct steps. First, selected segments of the genome are transcribed into primary RNAs by an RNA polymerase. Second, these primary transcripts are processed to be functionally mature RNAs. Splicing, the most striking event in RNA processing, occurs to remove introns and join exons together in eukaryotic cells. Other processing events include protection of both ends of the transcript. Third, these specific RNAs may encode proteins (coding RNAs) or play structural or catalytic roles (non-coding RNAs). Finally, the RNAs are degraded after finishing their functions. All these steps are essential to sustain cellular functions. Despite the growing published studies on the ZCCHC superfamily, many key questions remain to be answered. Based on the existing progresses, this review summarizes and discusses the emerging concept of ZCCHC superfamily as vital molecules in the regulation of RNA metabolism including transcription, splicing, translation and degradation.

2. Transcription

RNA transcription is a process in which genetic information is transformed from DNA to RNA, including the synthesis of messenger RNA and non-coding RNA (transfer RNA, ribosomal RNA, etc.). This process follows the base pairing principle and consists of three stages: initiation, chain elongation, and termination. By the action of an RNA polymerase, segments of DNA are transcribed into RNA under the strict regulation. Transcriptional co-activation/co-suppression occurs via multi-protein complexes that dock on transcription factors and recruit chromatin-modifying enzymes. Several ZCCHC proteins may function as a transcriptional activator or suppressor and then influence the progress of RNA transcription. Some are well illustrated, whereas others remain unfathomable so far.

2.1 ZCCHC12/Sizn1

ZCCHC12/Sizn1, identified as a novel modulator that functions as a transcriptional co-activator of bone morphogenic protein (BMP) signalling, has an impact on the progress of transcription initiation [28]. In the nucleus, the Smad complex can directly bind to DNA and then recruit either transcriptional co-activators or co-repressors to enhance or repress the transcription, respectively, [29]. CBP, a chromatin-modifying enzyme for acetylation, interacts with components of the basal transcription machinery and various different DNA-binding factors so as to activate the transcription [30,31]. Although ZCCHC12 does not directly interact DNA or RNA in vitro, ZCCHC12 come up close to DNA through its interaction with Smad1 and CBP in vivo, thus positively regulating Smad1-mediated transcription [28]. The N-terminus of ZCCHC12 interacts with CBP, while its C-terminus associates with Smad1 [32]. Unlike other ubiquitously expressed transcriptional co-regulators, ZCCHC12 shows restricted spatial and temporal expression patterns [28]. ZCCHC12 localizes to promyelocytic leukaemia protein nuclear bodies (PML-NBs) [32], which are large ring-shaped protein complexes involved in diverse nuclear functions including transcription [33,34], through two SIMs (SUMO interaction motif) and the MA homologous domain. The necessity of SIMs in ZCCHC12 for full co-activation of BMP signalling pathway is verified by the SBEx4-luciferase reporter assay [32]. ZCCHC12 also activates the transcriptional activities mediated by three important transcription factors, c-Jun, AP-1 and CREB, for which CBP is a co–activator [35–37].

Recent studies indicate the important impact of ZCCHC12 on the oncogenesis of papillary thyroid cancer [17] and specific gene expression of basal forebrain cholinergic neuron [28]. ZCCHC12 is also a potential molecular marker of Cajal–Retzius cells, which is worth investigating in the development and pathogenesis of mental retardation [38], and a potential endometrial receptivity-related marker as well [39].

2.2 ZCCHC22/CNBP

ZCCHC22/CNBP, containing seven Zn knuckles and an Arg-Gly-rich box, regulates gene expressions majorly at the transcriptional levels. It acts as either a positive or negative transcription factor. Binding to the CT-rich element, ZCCHC22 enhances the transcription of a proto-oncogene, c-Myc [40]. Further studies show that ZCCHC22 binds the G-rich sequence in the nuclease hypersensitivity element (NHE) III of c-Myc promoter to promote the formation of parallel G-quadruplexes (G4), which are folded from guanine-rich DNA strands to function as cis-regulatory elements surrounding transcription start sites [41,42]. In contrast, a different conclusion is drawn that in vitro ZCCHC22 unfolds the G4 in the NHE III of c-MYC promoter to stimulate c-Myc transcription through relieving the suppression caused by G4 formation [43,44]. More interestingly, by stabilizing G4 structure, ZCCHC22 may inhibit heterogeneous ribonucleoprotein K (hnRNPK) gene transcription [45]. Again, on the contrary, by unfolding parallel G4 structure, it enhances c-KIT, KRAS, PDGF, RET and VEGF transcriptions while down-regulates the transcription of NOGGIN/noggin3 [43]. Additionally, ZCCHC22 has been identified as a transcription regulator of β-myosin heavy chain (β-MHC), components of the Wnt signalling pathway [46,47], and inflammatory genes such as IL-6 and IL-12 [48,49].

Moreover, at the post-transcriptional level ZCCHC22 cooperates with a long non-coding RNA, LncRNA-assisted stabilization of transcripts (LAST), to stabilize the Cyclin D1 (CCND1) messenger RNA via binding to 5ʹ untranslated region of CCND1 mRNA, and then exerts a protective effect against the nuclease degradation [50].

2.3 ZCCHC-containing factors have an impact on rRNA biogenesis

Ribosomal RNA (rRNA) processing starts in the nucleolus. Combined with other components, rRNAs form a nucleoprotein called a ribosome serves as the place for protein synthesis. ZCCHC17, also named pNO40, acts as a negative regulator of rDNA transcription. When ZCCHC17 interacts with the upstream binding factor (UBF), whose binding to the rDNA promoter is reduced [51]. Human ZCCHC4 has been reported as a novel m⁶A methyltransferase that methylates human 28S rRNA [52]. Moreover, a knockdown of ZCCHC9 that results in an overexpression of 21S pre-rRNA [53].

2.4 Other ZCCHC-containing proteins that are also involved in RNA transcription

ZCCHC3 is considered as a positive regulator of RIG-I and MDA5 transcription [54]. Overexpression of ZCCHC3 enhances the transcription efficiency of the poly(I:C)-induced downstream antiviral genes [55]. By contrast, knockdown of ZCCHC3 in 293-TLR3 cells and HT1080 cells downregulates the transcriptions of IFNB1, ISG56 and CXCL10 genes [56].

ZCCHC9 is proved to inhibit the transcription of NF-kappaB and SRE, in which manner MAPK signalling cascade is modulated [6]. Overexpression of ZCCHC9 has a positive correlation with transcriptions of p‐JNK, Cyclin D1, and MMP7 and then facilitates the proliferation and invasion in human non‐small cell lung cancer [26]. Moreover, ZCCHC10 cooperates with pared-like homeodomain 1 (PITX1) to repress the transcription of the human telomerase reverse transcriptase (hTERT) [10]. Through changing telomere length, hTERT serves as a contributor to cancerous unlimited replication; thus, ZCCHC10 may have a function of tumour suppressor. Meanwhile, miR-410-3p promotes the NF-κB pathway activation by targeting ZCCHC10 [57].

3. Splicing

Splicing, as a step of mRNA maturation, is a part of post-transcriptional regulation. Constitutive splicing refers to the process of removing introns and retaining exons, whereas alternative splicing (AS) is responsible for exon selection and isoform regulation [58,59]. Disregulation of splicing including exon skipping, intron retention, or alternative exon usage, etc., may produce deleterious transcripts and lead to RNA-related diseases. Apart from the functional classification, splicing can also be classified according to numbers of pre-mRNA targets. The traditional cis-splicing takes place in single-molecule mRNAs, whereas the less common trans-splicing procedure represents exon recombinations of multiple independent pre-mRNAs [60]. No matter which form it happens to be, a spliceosome is assembled most of the time to identify splicing sites, which is generally believed to be constituted by five major small nuclear ribonucleoprotein particles (snRNPs) and 100 non-snRNP protein splicing factors (SFs) [61,62]. The whole process requires trans-acting proteins to conduct positive or negative regulation of splicing by binding to cis-regulatory elements that can be divided into four categories: exonic splicing enhancers (ESEs), exonic splicing silencers (ESSs), intronic splicing enhancers (ISEs) and intronic splicing silencers (ISSs) [58,63]. Those factors and their associated proteins, for example, SR proteins (serine/arginine-rich proteins), can be stored and assembled in subnuclear compartments called nuclear speckles, during which they undergo post-translational modifications, enabling their recruitment to trancriptional sites [64,65].

Several members of ZCCHC superfamily have different roles in splicing as illustrated in Figure 2. Both ZCCHC17 and ZCCHC20 act like a shuttle cargo that either recruits splicing factors to splice sites or accompanies splicing-completed mRNAs export to cytoplasm. ZCCHC20, ZCCHC21 and ZCCHC25 are distinct alternative splicing (AS) regulators, participating in various physiological and pathological processes by altering functional protein isoform profiles. ZCCHC25 is also indispensable in spliceosome assembly and 3ʹ splice site recognition.

Figure 2. Four members of the ZCCHC family are associated with pre-mRNA splicing. Constitutive splicing is the process of removing intron, while alternative splicing takes part in the selection of exons. Together with other mRNAs, trans-splicing is performed. A spliceosome is assembled during the process. Splicing-associated proteins localize in some special nuclear compartments named nuclear speckles, meanwhile can be released into the nucleoplasm once activated. ZCCHC17/pNO40 is a ribosome protein located within the nucleolus, but it can shuttle to nuclear speckles and bind to splicing factor SRSF2 indirectly affecting the splicing process. ZCCHC21/RBM4 is a critical RNA binding protein (RBP) which decides the inclusion or omission of exons by interacting with the cis-regulatory elements. Apart from its role in alternative splicing, evidence shows that it can also regulate trans-splicing. ZCCHC20/SRSF7 is also RBP, and it acts to regulate constitutive and alternative splicing of various proteins on binding the exon/intron splicing enhancer (ESE/ISE) region of pre-mRNA. SRSF7 also mediates mature mRNA nuclear export by recruiting NXF1. ZCCHC25/SF1, however, participates in the pre-spliceosome assembly

3.1 ZCCHC17/pNO40

ZCCHC17, also named as pNO40, is initially known as an SR-like protein pinin (pnn)-interacting protein. As a core component of 60S ribosomal subunit, it often resides in the nucleoli to negatively regulate rRNA biogenesis [51]. It is highly conserved between human and mouse, and has its highest expressions in brain, heart and testis [66,67]. ZCCHC17 has an S1 RNA binding region on its N-terminus, a CCHC-type zinc finger motif in the middle, and clusters of basic amino acid representing potential nucleolar targeting signal on its C-terminus [66]. Moreover, the N-terminus also mediates protein–protein interaction, as ZCCHC17 is proved to interact with SR-family splicing factors including Pnn and recruit them from nuclear speckles to nucleoli, contributing to rRNA splicing regulation [66]. Aberrant expression of ZCCHC17 also influences overall mRNA quality control. For example, pNO40 overexpression reduces pre-mRNA splicing, ploy(A)+ RNA nuclear export, and RNAP II activity [65].

3.2 ZCCHC20/SRSF7

ZCCHC20, also named as SRSF7 or 9G8, is a member of SR family serine–arginine-rich splicing factors (SRSF). It is highly conserved across species and expressed most abundantly in bone marrow and lymph node [67]. Sharing characteristics with other members of SRSF family, ZCCHC20 possesses an N-terminal RNA recognition motif (RRM), a C-terminal repeated serine-arginine dipeptide domain (RS), and a CCHC-type zinc-finger motif in between [68]. By binding to the pre-mRNA exon or intron splicing enhancers (ESE, ISE) through RRM, ZCCHC20 promotes the inclusion or exclusion of exons, respectively, highlighting its regulative role in both constitutive and alternative splicing [69]. For examples, by binding to splicing enhancers, ZCCHC20 indeed regulates constitutive and alternative splicing of various transcripts including CD44, BRCA1 and Tau [70,71]. CCHC-type zinc-finger motif of ZCCHC20 increases the binding specificity[70], while its RS domain mediates protein–protein and/or RNA–protein interaction [72]. It undergoes phosphorylation–dephosphorylation cycle, which is crucial for the dynamic splicing process including the spliceosome assembly, splicing catalysis and termination [72].

In some circumstances, ZCCHC20 acts competitively with other SRSFs in AS regulation of CD44 [73]. ZCCHC20 regulates alternative splicing of Fas/CD95 to balance the ratio of pro/anti-apoptotic isoform, and knockdown of ZCCHC20 inhibits proliferation and induces apoptosis in multiple cancer cell lines [74]. Interestingly, its zinc-containing motif is antagonized by iron, so as to be regulated by local iron homoeostasis [74], demonstrating an unique functional regulation pathway of CCHC-type zinc-finger superfamily. In addition to its role in AS-dependent exon selection, ZCCHC20 also antagonizes other SRSFs in splicing-independent alternative polyadenylation (APA) [74]. Depletion of ZCCHC20 results in more inclusion of exons and longer 3ʹUTR mRNA isoforms [75].

ZCCHC20 also participates in other downstream events including mRNA export, nonsense-mediated RNA decay and translation [75,76]. Splicing-completed mRNA bound by ZCCHC20 can recruit an adaptor of nuclear transporter 1 (NXF1) facilitating mRNA export from nucleus to cytoplasm [75]. Interestingly, the hypophosphorylation state of ZCCHC20 RS domain is requisite for its binding to NXF1, suggesting that ZCCHC20 carries primarily mature, splicing-completed mRNAs to the cytoplasm [77]. Therefore, ZCCHC20 indeed serves as a shuttle cargo that links mRNA splicing and APA to mRNA export. Moreover, ZCCHC20 also binds intron-derived non-coding RNAs (ncRNAs) and helps them export to cytoplasm [69]. However it turns out that ZCCHC20 binds to a much lower extent to intergenic long non-coding RNA (lncRNA) exons than protein-coding exons [78].

ZCCHC20 has established its role in RNA metabolism and is crucial for common physiological processes. It is actually responsible for yielding juvenile-specific splicing isoforms in multiple organs examined in C57BL/6 mice, which is indispensable in deciding proper anabolic state for juveniles [79]. Furthermore, dysregulation of ZCCHC20 leads to carcinogenesis. ZCCHC20 is overexpressed in lung and colon cancer tissues [80]. Single-nucleotide-polymorphisms (SNP) rs10197412 of ZCCHC20 is associated with an increased risk for lung cancer in a Chinese population [81].

3.3 ZCCHC21/RBM4

ZCCHC21, also named as RNA-binding motif protein 4 (RBM4) or Lark, is initially reported as a protein involved in circadian rhythm control of Drosophila, but has also been brought in sight in mammalian cells. It has a ubiquitous expression, with particularly high levels in testis, ovaries, heart and pancreas [82]. ZCCHC21 structurally contains two RNA recognition motifs (RRMs) and a CCHC-type zinc finger in its N-terminus [83], as well as an alanine repeat-containing C-terminal domain (CAD) in its C-terminus [84]. Its C-terminus and RRMs are both responsible for binding a set of pre-mRNA cis-regulatory elements, necessary for conducting splicing regulation [85]. The C-terminus also mediates protein–protein interactions, as well as nuclear speckle localization [84,86]. RRM2 is crucial for the nucleolar localization [83]. The cellular function of zinc-finger motif has not yet been discovered, but it may act in concert with RRM domains and contribute to RNA binding specificity [83]. ZCCHC21 mainly resides in nucleoli and nuclear speckles, where it colocalizes with SR-proteins and participates in pre-mRNA alternative splicing regulation [83]. It can also translocate to cytoplasm with a perinuclear distribution, acting a role in mRNA translation regulation [86].

ZCCHC21 is initially recognized as a regulator of AS. By binding with ISE downstream of Tau exon10, ZCCHC21 inhibits AS process and promotes the inclusion of Tau exon10, similarly but in opposite to ZCCHC20 [87,88]. Although its role is predominantly negative, ZCCHC21 also positively regulates AS by binding different cis-regulatory elements, as is the case for alpha-tropomyosin (TPM1) splicing regulation [89]. ZCCHC21 antagonizes the suppressive splicing regulation of PTB protein by competitive binding to intronic CU-rich elements, stimulating exon skipping [90]. To summarize, ZCCHC21 probably exhibits a binding site-dependent regulatory pattern: a binding to both exonic and intronic elements promotes splicing and leads to exon exclusion; on the contrary, a binding only to downstream intronic elements inhibits splicing and facilitates exon inclusion [85]. Besides its bidirectional role in cis-splicing, ZCCHC21 also undergoes trans-splicing regulation. Trans-splicing between pre-mRNAs of ZCCHC21 and another splicing regulator CoAA (RBM14) yields a zinc finger-containing coactivator CoAZ and a non-coding splicing variant ncCoAZ during stem/progenitor cell neural differentiation, which influences Tau exon10 selection, presumably important for neural cell differentiation [91]. In this feedback model, the trans-splicing product CoAZ antagonizes ZCCHC21 through the shared zinc-finger motif.

Splicing process regulated by ZCCHC21 seemingly requires the cooperation of other factors in order to expect a precise modulation. LncRNA TPM1-AS, which is naturally anti-sense transcribed from TPM1 intronic region, interacts and therefore hinders ZCCHC21 from binding to TPM1 pre-mRNA and inhibits its alternative splicing function, resulting in an increase of exon 2a inclusion[92]. WT1 (+KTS), an isoform of tumour suppressor protein WT1, directly binds ZCCHC21 and abrogates its effect on splicing site selection[93]. Moreover, ZCCHC21 is a part of supraspliceosome assembly, regulating alternative splicing sites selection of minigenes SMN2 and SRp20 [86,93].

ZCCHC21 is involved in common physiological process particularly cell differentiation and organogenesis by balancing of protein isoforms. Pyruvate kinase M2 (PKM2) to PKM1 isoform shift occurs at the time of neuron induction when ZCCHC21 generates more PKM1 through AS regulation, promoting neural differentiation [94]. ZCCHC21 also plays a role in neuronal migration via regulating the expression of Dab1 splicing isoforms[95]. By modulating IR-B, Pref-1 and PPAR-r isoform shift, ZCCHC21 manages the reprogramming splicing cascade during differentiation of brown adipocytes[96]. ZCCHC21 alters isoform balance of Pax4 and Isl1 to activate the transformation to insulin-producing cells in β-islets[97]. In myogenesis, ZCCHC21 downregulates and antagonizes PTB/mPTB, acting together to generate a set of muscle-specific protein isoforms through AS [90]. Moreover, ZCCHC21 modulates splicing profiles change upon enteric infection in Drosophila, suggesting its potential role in host against infection [98].

Dysregulation of alternative splicing results in protein isoform imbalance, and is relevant to carcinogenesis signatures including immortality, migration, angiogenesis, etc[89].. ZCCHC21 is recognized as a tumour suppressor in various malignancies, including lung, breast, ovarian, pancreatic, liver, prostate and gastric cancer [89,99,100]. Notably, the reduced expression of ZCCHC21 is associated with poor overall survival in lung, breast, and ovarian cancer in a Chinese population [24]. ZCCHC21 can modulate isoform proportions of Bcl-x through alternative splicing towards a pro-apoptotic profile, suppressing proliferation and migration of various cancer cell lines [89]. In breast cancer cells, the phosphorylation at Ser309 in the RDRSP domain of ZCCHC21 mediated by an SR protein kinase 1 (SRPK1) induces an cytoplasmic accumulation of ZCCHC21 [85], which reduces pro-apoptotic protein isoforms MCL-1s and IR-B [85,101]. ZCCHC21 also mediates inflammatory response via splicing regulation of TNIP1 in HeLa cells [102]. In colorectal cancer (CRC) cells, ZCCHC21 participates in reprogramming splicing profiles of FGFR2 and PKM2 genes, as well as mediates alternative splicing cascades of a set of genes including Nova1, SRSF6, VEGF165, modulating cell migration, metabolic and angiogenetic signature [65,103]. In human CRC tissues, ZCCHC21 can also antagonize another splicing factor, SRSF3, which binds to MAP4K4 exon16 CU-elements yielding a MAP4K4 variant that promote migration and invasion of CRC cells [23,100]. Similarly, ZCCHC21 and SRSF1 antagonistically regulate splicing profiles of HIF-1α, influencing oncogenic signatures of lung cancer cells [104].

3.4 ZCCHC25/SF1

ZCCHC25, also named as splicing factor 1 (SF1), participates in early steps of pre-spliceosome assembly. It is highly conserved and has a particular high expression in bone marrow [67]. It contains an N-terminal U2AF ligand motif (ULM), a central hnRNPK homology motif which specifically recognizes branch point sequence (BPS), as well as a CCHC-type zinc finger, and a C-terminal proline-rich domain [105]. The spliceosome assembly follows a series of ordered events, during which ZCCHC25 cooperates with U2 ribonucleoprotein auxiliary factor 65-kDa subunit (U2AF[64]) in recognizing 3ʹ splice-sites of pre-mRNAs and facilitating the ongoing of complex E to A evolution. As ZCCHC25 being a part of early complex E formation, U2AF[64] is recruited by ZCCHC25 to splice sites [106]. With its hnRNPK homology motif, ZCCHC25 can recognize BPS in the first place; however, this interaction is strengthened upon the presence of U2AF [64,107]. Additionally, ZCCHC25 associates with U2AF[64] in nuclear speckles where they accumulate when not engaged in splicing [108]. As for U2AF[64], it bears the ability to recognize pre-mRNAs polypyrimidine (py) tract and 3ʹ splice site. Once the ZCCHC25/U2AF[64]/pre-mRNA complex is established, it undergoes a conformational change with the presence of ATP [109], which joins U2 snRNP and the branch point together [105,110], making the completion of complex A. Notably, ZCCHC25 is regulated by post-translational modification that is crucial for the machinery. Phosphorylation of Ser80/82 within a highly conserved SPSP motif of ZCCHC25 enhances the interaction of ZCCHC25/U2AF[64] and pre-mRNAs [111], whereas its phosphorylation by cGMP-dependent protein kinase-I (PKG-I) inhibits its binding to U2AF [64,111,112]. Computational methods predict a potential phosphortylation site of ZCCHC25 nuclear localization signal (NLS), may turn out to be another regulatory pathway [113]. Apart from spliceosome assembly and 3ʹ splice site recognition, ZCCHC25 is expected to participate in alternative splice site selection, especially by contacting branch-site like sequences. ZCCHC25 silencing results in disturbance of several endogenous AS variants, including changes of exon inclusion/skipping ratio [114]. ZCCHC25 is regulated by the β-catenin/TCF4 complex, and conversely it interacts the complex to trans-activate gene expression and splicing activities[115].

Evidence is accumulating that ZCCHC25 is a potential regulator of reproduction and germ cell tumorigenesis. The ZCCHC25 3ʹUTR variant g.423146327 T > C polymorphism increases litter size in small Tail Han sheep, suggesting its role in mammalian reproduction [116]. Cross-breeding of Sf1± mice with 129-Ter and M19 mice exhibits decreased incidence of testicular germ cell tumours (TGCTs), indicating its oncogenic role in TGCTs [117].

4. Translation

Translation is responsible for converting genetic information coded by mRNAs into functional proteins. The biological processes of translation includes initiation, elongation, termination and ribosome recycling. The first step, initiation, is the most regulated, with ZCCHC21 and ZCCHC22 involved (see Figure 3 for details). However, the translation initiation procedure differs a lot among species. In eukaryotes, the most predominant way is the canonical cap-dependent translation initiation, which involves m⁷G cap recognition at the 5ʹ end of mRNAs and ribosomal scanning [118,119]. The m⁷G cap is added at the 5ʹ end of mRNAs during transcription and is recognized by eukaryotic initiation factor 4E (eIF4E), which together with helicase eIF4A and scaffold protein eIF4G, forms a complex eIF4F [120]. This complex then recruits the small ribosomal subunit 40S to active mRNAs, leading following events. A quite distinct mechanism of translation initiation, known as the cap-independent manner, is found in certain viral mRNAs possessing with internal ribosome entry sites (IRES) in the 5ʹ UTR [121,122]. Some eukaryotic mRNAs can also undergo cap-independent translation under certain circumstances such as stress [64,120]. N[6]-methyladenosine (m⁶A) in 5ʹ UTR of eukaryotic mRNAs allows translation initiation to bypass m⁷G cap [123]. Therefore, the concept of cap-independent translation enhancers (CITEs) is proposed to explain the cap-independent translation process in mammalians. Unlike IRESs, CITEs recognize both 5ʹ and 3ʹ UTRs and undergo ribosomal scanning as in canonical cap-dependent translation [120]. Several zinc finger CCHC-type superfamily proteins, such as ZCCHC4, ZCCHC21/RBM4 and ZCCHC22/CNBP, are found to play key roles in mRNAs translation. The details will be discussed as follows.

Figure 3. ZCCHC21 and ZCCHC22 regulate translation initiation of eukaryotic mRNAs with a 5′-end m⁷G cap and a 3′-end poly(A)tail. The cap structure is recognized and combined by a group of eukaryotic initiation factors, eIF4E, eIF4G, eIF4A and eIF4B. Then the initiation complex recruits the 43S pre-initiation complex (PIC) to bind the m⁷G cap structure to activate cap-dependent translation. ZCCHC22 positively regulates whereas ZCCHC21 plays a dual role in this manner. If the mRNAs contain IRES motifs in the 5ʹ-end, m7G cap is no longer necessary. The phosphorylated ZCCHC21 at serine 309 stimulated by cell stress can stabilize eIF4A with IRES-containing mRNAs and initiate IRES-dependent translation. ZCCHC22 also promote this way in an unknown way. In addition, ZCCHC21 associates with Ago2 in myoblast cells to suppress mRNA translation. ZCCHC22 resolves G4 structure in 5ʹ-UTR RNA to implement its translational up-regulation

4.1 ZCCHC4

The ribosome consisting of rRNAs and proteins is where the translation takes place. The post-transcriptional modifications on rRNAs may influence translation efficiency. RNA m⁶A modification is ubiquitously present in mRNAs, rRNAs and snRNAs in human. The distribution of mRNA m⁶A regulates a wide range of biological processes, such as miRNA biogenesis [96], stem cell pluripotency [124,125], RNA structural switches [126], cancer and stress responses [127,128]. However, rRNA m⁶A still remains poorly understood.

ZCCHC4 is a newly discovered RNA m⁶A methyltransferase, which primarily catalyzes human 28S rRNA methylation at the position A⁴²²⁰ both in vitro and in vivo [52,129]. ZCCHC4 resides in the nucleolus and interacts with nucleolar proteins such as ASC-1 complex as well as m⁶A-reader YTHDC2 [130]. The functional domains of ZCCHC4 are shown in Fig. 4. Structural analysis implicates that ZCCHC4-mediated rRNA methylation is critical for ribosome structure maintenance and subunit assembly[52]. Through its 28S rRNA m⁶A methylation activity as well as interaction with a subset of mRNAs, ZCCHC4 influences global translation and cell proliferation. Loss of 28S rRNA m⁶A modification by knockout of ZCCHC4 reduces the strict control of translation, thus increasing mRNAs translation [52]. ZCCHC4 is overexpressed in human hepatocellular carcinoma (HCC) tissues, while knockout of which inhibits the growth of HCC tumours in vivo [52]. However, the detailed implication of ZCCHC4 in tumour biology remains to be further explored.

Figure 4. ZCCHC4 as RNA m⁶A methyltransferase methylates human 28S rRNA. By recognizing a stem-loop structure inherent in the human 28S rRNA, ZCCHC4 is activated to form a multi-domain RNA-binding surface. The integrated RNA-binding platform of ZCCHC4 is composed of a methyltransferase (MTase) domain, an N-terminal GRF-type and C2H2 zinc finger domains, and a C- terminal CCHC domain

4.2 ZCCHC21/RBM4

ZCCHC21, also known as RNA-binding motif protein 4 (RBM4), is a multifunctional nucleocytoplasmic shuttling protein which not only participates in alternative splicing of pre-mRNA, but also has an impact on RNA translation [86]. RBM4 is involved in the cell stress-induced signalling on translation initiation control. Arsenite exposure induces phosphorylation of RBM4 at serine 309 and drives its cytoplasmic accumulation to stress granule and nucleus via the MKK3/6-p38 signalling pathway, which subsequently suppresses cap-dependent translation in a cis-element-dependent manner [64]. Accordingly, under arsenite exposure, RBM4 activates internal ribosome entry site (IRES)-mediated translation through enhancing the interaction between eIF4A and IRES-containing mRNAs [64]. Hypoxia is another cellular stress which sequesters eIF4E thus repressing the cap-mediated translation [131]. RBM4, along with hypoxia-inducible factor 2α (HIF-2α) and the cap-binding eIF4E2, forms an oxygen-regulated translation initiation complex that mediates selective cap-dependent protein synthesis under low oxygen tension. The complex captures the 5′ cap and recruits mRNAs to polysomes to activate translation, which ensures the protein synthesis in hypoxic cells [132]. EGFR, known as a driver of tumorigenesis in certain cancer and a biomarker of tumour drug resistance [133], is one of the representative proteins synthesized with the help of this complex under hypoxia [132]. Interestingly, EGFR is found overexpressed in tumours that harbour a wild-type EGFR gene. Since tumour microenvironment is often hypoxic, overexpression of EGFR in tumours might be explained by this protein synthesis mechanism. Moreover, upregulated expression of RBM4 in tumour tissues is correlated with a better overall survival rate and disease-free survival rate in HCC patients and gastric cancer patients [24,134]. However, whether RBM4 is a potential target for cancer treatment, more investigations on its detailed roles in tumour biology are needed.

Besides, it is also found that RBM4 can directly interact with the 3ʹ-UTR of Period1 (Per1), an important circadian clock gene, without a splicing junction, which suggests that it upregulates Per1 expression at the level of translation in a cap/poly(A)-dependent manner [135]. In myoblast cells, RBM4 co-localizes and interacts directly with Argonaute2 (AGO2) protein, a key component of miRNAs induced silencing complex (miRISC), therefore, to suppress the translation of target mRNAs. Moreover, it associates with some muscle cell-specific miRNAs thus enhancing the association between miRISCs and target mRNAs, which leads to more efficient translation suppression during muscle cell differentiation [136]. This protein translation suppression mechanism directed by miRNAs-based mechanism can be rescued while RBM4 is phosphorylated at serine 309 by p38 MAPK. The phosphorylation of RBM4 leads to its nuclear re-localization, thus disrupting the interaction between RBM4 and AGO2 in the cytosol [137].

4.3 ZCCHC22/CNBP

ZCCHC22/CNBP is involved in both transcription and translation. It participates in translational regulation and serves as a molecular chaperone through binding G-rich single-stranded nucleic acids [138]. ZCCHC22 is found to be associated with the 5ʹ-terminal oligopyrimidine tract (5ʹTOP) mRNAs which encode RPS17 (ribosomal protein), poly(A)-binding protein 1 (PABP1), and two elongation factors eEF1A and eEF2, thus regulating the rate of cap-dependent translation[139]. In Drosophila, ZCCHC22 promotes the c-myc Drosophila ortholog (dMyc) translation through an IRES-dependent mechanism [140]. Similar to RBM4 mentioned above, under cellular stress, ZCCHC22 accumulates in stress granules, which supports a potential function for ZCCHC22 in mRNAs translation suppression [141]. Based on the fact that ZCCHC22 functions as a G-quadruplex-unfolding protein in vitro, it may boost translation through resolving stable secondary structures within 5ʹ-UTR mRNAs [44].

5. Degradation

RNA degradation exists ubiquitously in all organisms, which is a complex process accomplished via different pathways in the presence of various enzymes and cofactors. It is finally achieved by two major pathways: (1) the 5ʹ-3ʹ pathway including the Lsm1-7 complex, the decapping complex and 5ʹ-monophosphate-dependent exoribonuclease; (2) the 3ʹ-5ʹ pathway including a multi-subunit exosome complex [142]. Certain members of ZCCHC superfamily play crucial roles in the regulation of RNA degradation by marking RNAs for the two separated degradation pathways, as illustrated in Fig. 5.

Figure 5. ZCCHC family play crucial roles in regulation of RNA degradation. In cytoplasm, ZCCHC11 and ZCCHC6 function redundantly as uridylyl-transferase to facilitate degradation of global mRNAs with short poly(A) tails and certain pre-miRNAs. For global mRNAs, for example, they enhance the mRNA degradation induced by miRNA (miR-1) and LINE-1 mRNA degradation in retro-transposition. For miRNA, ZCCHC11 and ZCCHC6 mark different type of miRNA for degradation. In the nucleus, ZCCHC11 participates in degradation of histone mRNA. ZCCHC8 and ZCCHC7 are a key component of the NEXT complex in nucleoplasm and the TRAMP-like complex in nucleolus, respectively. Both two complexes are related to the nuclear exosome, which is responsible for degradation of abnormal RNAs and the accurate 3′-end formation of functional RNAs

5.1 ZCCHC11/TUT4, ZCCHC6/TUT7, and ZCCHC1/Lin28A

ZCCHC11 and ZCCHC6, widespread in almost all human tissues, facilitate the global mRNA degradation as uridylyltransferases that selectively recognize and uridylate mRNA with a short poly(A) tail of fewer than ∼25 nucleotides (nt) [143]. It is worth noting that such specificity results from the intrinsic capacity of ZCCHC11 and ZCCHC6 to detect the length of the poly(A) and the disassociation of poly(A) binding protein (PABP) as the poly(A) shortened [143]. It is also indicated that oligo-uridylated mRNAs (with ≥2 uridines), rather than mono-uridylated mRNAs, are more specific products of uridylation by ZCCHC11 and ZCCHC6. Oligo-uridylated mRNAs serve as a main mark of degradation and have a stronger effect. Both 5ʹ-3ʹ and 3ʹ-5ʹ decay pathways are involved in the ZCCHC11 and ZCCHC6-regulated mRNAs degradation. The LSM1-7 complex (for 5ʹ-3ʹ pathway) and DIS3L2 (for 3ʹ-5ʹ pathway) are the downstream factors to recognize the oligo-U marks.

In murine zygotic cells, ZCCHC11 and ZCCHC6 expressions are triggered by YAP1-TEAD4 transcription factor to help clear up murine maternal transcripts, a process required for mouse early embryo development beyond the four-cell stage [144]. Moreover, abnormal accumulation of maternal transcripts accompanied by decreased expression of ZCCHC11 and ZCCHC6 are observed in the development-arrested human embryos, indicating the role of ZCCHC11 and ZCCHC6 in the degradation of mammalian maternal mRNAs [145]. Similarly, ZCCHC11 and ZCCHC6 are required for 3ʹ uridylation-programmed zygotene mRNA clearance in pachytene spermatocytes that is essential for male meiotic progression [146].

Uridylation of LINE-1 mRNA 3ʹ ends by ZCCHC11 or ZCCHC6 in cooperation with the helicase/RNPase MOV10 counteracts the RNA chaperone activity of the L1-ORF1p retrotransposon protein, thus restricting the retro-transposition of LINE-1s at the mRNA level through mRNA degradation, which enhances the genome stability [147]. In addition to mRNAs with short poly(A) tails, ZCCHC11 also uridylates mRNAs that are not polyadenylated, such as histone mRNA end in a conserved stem-loop rather than a poly(A) tail [148]. Once DNA replication is inhibited, usually at the end of the S stage [149], histone mRNA will be trimmed by 3ʹhExo at the 3ʹ terminal near its stem loop and then uridylated by ZCCHC6/TUT7 [150]. The uridylation leads to degradation with both 5ʹ-3ʹ degradation pathway via decapping and 3ʹ-5ʹ degradation pathway via RNA exosome [149]. However, ZCCHC11 has not been reported to function like that.

Besides mature mRNAs, ZCCHC6 and ZCCHC11 play an important role in pre-miRNA degradation as well. In stem cells and cancer cells, ZCCHC11 and ZCCHC6 serve as suppressors of let-7 biogenesis by uridylating let-7 pre-miRNA for degradation hence augmenting mRNA of certain genes [151,152], which seems to be inconsistent with studies in global mRNA degradation. Yet, double knockdown of ZCCHC11 and ZCCHC6 does not affect the let-7 levels in the presence of ZCCHC1/Lin28A [153], instead, the overall mRNAs are upregulated [143], suggesting that ZCCHC11 and ZCCHC6 determine general mRNA levels through mRNA degradation but not let-7 miRNA biogenesis.

ZCCHC1/Lin28A is enriched in human testis and spermatids. It selectively binds to the terminal loop of let-7 pre-miRNA with a CCHC-type zinc finger RNA-binding domain, and subsequently recruit ZCCHC11 or ZCCHC6 [154], which drives the formation of a stable ternary complex between pre-let-7 and the Lin28-interacting motif (LIM) of ZCCHC11 [155,156]. As a result of recruitment, ZCCHC11 adds an oligouridine tail of 10–30 nt to the 3ʹ-terminus by recognizing a specific RNA sequences motif of GGAG in the terminal loop [151,154]. The uridylated let-7 pre-miRNA is resistant to Dicer1 and may be sequestered in the cytoplasm [157]. ZCCHC6 plays a similar role but has weaker activity [154]. A set of other pre-miRNAs with GGAG are also subject to uridylation in this ZCCHC1-dependent manner, such as pre-miR-107,-143 and −200 C [154]. In addition to the zinc finger RNA-binding domain, ZCCHC1 also harbour a cold shock domain (CSD) which recognizes a (U)GAU motif in a subclass of let-7 pre-miRNA family and enhances its binding [158].

The repression of let-7 expression is required for normal development and ZCCHC1 contributes to maintain the pluripotent state by preventing let-7–mediated differentiation of embryonic stem cells and neural stem cell commitment [154,159]. Accordingly, aberrant expression of ZCCHC1 is associated with cancer progression in multiple tissues [160–162]. We recently find that in cancer hypoxia environment, activation of ZCCHC1 is driven by SUMOylation [163]. Thereby, its inhibition of let-7 maturation is promoted, resulting in cancer cell proliferation, migration, invasion, and tumour growth in vivo [163]. Interestingly, a recent study has screened a nanobody called Nb-S2A4 for therapeutic purposes, which targets the ZCCHC1:let-7 interaction fragment of ZCCHC11/TUT4 thus blocking uridylation of let-7 [164].

In somatic cells with absence of ZCCHC1, ZCCHC6 or ZCCHC11 recognizes the 3′ trimmed group II pre-miRNAs, which the 3ʹ end of pre-miRNAs is further recessed into the stem, and generates an oligo-U tail that leads to degradation. This is deemed as a clearance mechanism of non-functional pre-miRNAs [165]. In contrast to typical group I pre-miRNAs with a 2 nt overhang at the 3ʹ end, group II pre-miRNAs acquire a shorter (1 nt) 3ʹ overhang to form a bulge next to the Drosha cleavage site [153]. As shown in Fig. 6, this mechanism expands the field of miRNAs regulation at the post-transcriptional level. Besides pre-miRNA degradation, ZCCHC6 and ZCCHC11 are also reported to uridylate cytokine-targeting mature miRNAs, thus regulating the stability of transcripts for interleukin-6 (IL-6) and IGF-1 [166,167]. For example, ZCCHC6 enhances IL-6 expression in chondrocytes by uridylating miR-26b for degradation, which then alleviates the severity of osteoarthritis in mice [168].

Figure 6. ZCCHC11 and ZCCHC6 regulate miRNA biogenesis in the presence or absence of ZCCHC1. (1) In embryonic stem cells and some cancer cells, ZCCHC1/Lin28A recruits ZCCHC6 or ZCCHC11 to oligouridylate pre-let-7 at the 3′end that inhibits its maturation. (2) In somic cells where lack of ZCCHC1, Group II pre-miRNA with a shorter 3′ overhang can be monouridylated by ZCCHC11 or ZCCHC6 that enhances miRNA biogenesis

The role of ZCCHC6 and ZCCHC11 in regulating mature miRNA dynamically has been recently reported. They play a crucial part in turnover of mature miRNAs bound by Argonaute proteins in the RNA-induced silencing complex (RISC). The exposed 3ʹ ends of miRNA, which is caused by frequent sequence modifications, are uridylated by ZCCHC6 and ZCCHC11 to facilitate DIS3L2-mediated degradation [169]. Moreover, in miRNA strand selection, uridylation of pre-miR-324 by ZCCHC11 and ZCCHC6 re-positions Dicer1 on the pre-miRNA and shifts the cleavage site. This alternative processing produces a duplex with a different terminus from which the 3ʹ strand (3p) is selected instead of the 5ʹ strand (5p), resulting in different post-transcriptional effects [170].

To summarize, ZCCHC11 and ZCCHC6 redundantly regulate the general degradation of mRNAs with short poly(A) tails in mammalian cells [143]. Particularly, ZCCHC6 can regulate the degradation of histone mRNA. They also facilitate degradation of certain pre-miRNAs such as let–7 family after recruited by ZCCHC1 [171] and turnover of mature miRNAs bound by Argonaute proteins [166–169]. In mammalian gametes and embryos, they clear up parental mRNAs to facilitate mammalian spermatogenesis and embryonic development [144,145]. As uridylyl transferases for RNA, ZCCHC6 and ZCCHC11 are widely involved in process related to cell fate through the regulation of stability of genomes and transcripts, which makes them underlying targets for development abnormality, cancer and tissue regeneration.

5.2 ZCCHC7

ZCCHC7, together with PAP‐associated domain‐containing protein 5 (PAPD5) and RNA helicase hMTR4/DOB1/SKIV2L2, constitutes the TRAMP-like complex in mammalian cells [172]. The well-established TRAMP complex model in Saccharomyces cerevisiae suggests that TRAMP complex recognizes and polyadenylates abnormal coding and non-coding RNAs, hence working as a cofactor of nuclear exosomes to promote RNA degradation [173], but the structure and function of the mammalian TRAMP-like complex are still poorly studied. It is now clear that ZCCHC7 and PAPD5 interdependently interact with the exosome in the presence of the catalytic subunit RRP6, which takes place only in nucleoli [174], unlike their homologues in S. cerevisiae [173]. Th exosome functions include degradation of abnormal RNAs and the accurate 3ʹ‐end formation of functional RNAs, such as rRNAs, small nuclear RNAs (snRNAs), and small nucleolar RNAs (snoRNAs) [175], which may provide a hint of ZCCHC7 function in guiding RNA degradation. Indeed, it has been revealed that ZCCHC7 together with PAPD5 contribute to the 3ʹend adenylation of abortive RNA polymerase I (Pol I) transcripts as marks for degradation by the nuclear exosome [176] and interact closely with components of the small ribosomal subunit (SSU) processome which is essential in the early steps of pre‐rRNA processing [174]. Consistently, the further investigation has shown that ZCCHC7 and PAPD5 are involved in the degradation of 5ʹ-ETS by‐product RNAs excised from the pre‐rRNA [177], similar to their homologues in the yeast. In brief, ZCCHC7 is related closely to the nucleolar exosome function and acts interdependently with PAPD5 to affect pre‐rRNA 5ʹ-ETS elimination. However, whether ZCCHC7 facilitates degradation of other RNAs via exosome remains to be observed.

5.3 ZCCHC8

ZCCHC8 is first identified as a substrate of GSK-3, and Intriguingly, it is phosphorylated by GSK-3 at the spindle checkpoint [7]. Contrary to ZCCHC7, ZCCHC8 exists in the nucleoplasm rather than in the nucleoli. ZCCHC8 binds with two proteins: hMTR4, an RNA helicase, and RBM7, an RNA binding protein, to form a nuclear exosome targeting complex (NEXT complex) [174]. The NEXT complex directs a large scale of non-coding RNA for the nuclear exosome degradation, such as lncRNAs, non-coding promoter-upstream transcripts, 3ʹ-extended products of histone and small nuclear RNA transcription [174,178,179]. Taken the zinc-knuckle domain of ZCCHC8 and the similarity between the NEXT complex and the TRAMP complex into consideration, both ZCCHC8 and RBM7 might act to recognize and target RNAs for degradation. Thus, together with hMTR4, ZCCHC8/RBM7 and target RNAs into the exosome for degradation [174]. The depletion of ZCCHC8 of the NEXT complex in mouse leads to developmental defects, a shortened lifespan, and infertility, for which the underlying mechanism is proposed that the ZCCHC8-mediated degradation of transcripts of retrotransposon element LINE1 in early embryos and embryonic stem cells (ESCs) [180].

Moreover, the crystal structure of the NEXT displays a proline-rich segment of ZCCHC8 as the interaction site for the RNA-recognition motif (RRM) of RBM7 [181]. But interestingly, RBM7 is also competitively bound to a splicing factor SAP145a [182] which consists of a proline-rich segment with strong similarity to ZCCHC8 [181], thereby directing to the spliceosome other than the exosome. It still remains unclear whether there is a correlation between these traits of ZCCHC8 and its function, whether and how ZCCHC8 competes with SAP145a. In addition, ZCCHC8 contains short linear motifs that bind to the hMTR4 arch domain in a mutually exclusive manner with a ribosome processing adapter NVL, suggesting ZCCHC8 as an adapter to direct exosomes for small nuclear RNA degradation [183]. The question concerning about the exact role of ZCCHC8 in the exosome-mediated RNA degradation is still open.

In RNA degradation, similarities exist among certain ZCCHC family proteins. For instance, ZCCHC1 use its zinc finger domain to recognize the targeted let-7 pre-miRNA. ZCCHC6 and ZCCHC11 act redundantly as uridylyl-transferases to mark RNAs for degradation [143], while ZCCHC7 and ZCCHC8 are the main components of the TRAMP-like complex in nucleoli and the NEXT complex in nucleoplasm, respectively, and both are assumed to be cofactors of the nuclear exosome [173,174], which finally degrades those marked RNAs. However, the functions of ZCCHC family proteins in RNA degradation have not yet been fully understood. Whether and how their common zinc finger domain work as key motifs for their functions, such as RNA substrate selectivity and the correlations between those proteins, have to be elucidated since they seem to coordinate in the same regulating axis/pathway of RNA degradation to guide the fate of the cells.

6. Summary

The regulation of RNA metabolism is regarded to be crucial in various biological processes, such as gene expression, cell proliferation, differentiation, as well as catabolic process and cellular response to stimulus. The CCHC-type zinc fingers exist in all family members of ZCCHC proteins and some even harbours more than one consensus sequence, except for ZCCHC23, on account of its histidine (H) residue being substituted by asparagine (N). As the high affinity to single-stranded nucleic acids (not including ZCCHC18) of the distinct ZCCHC motif, the relevance of CCHC-type zinc finger proteins to RNA metabolism are doomed. It is the consensus motif that serves as the guide to target RNAs, despite the unnecessariness to play the most significant role in performing molecular functions.

ZCCHC superfamily has emerged as distinct roles in various aspects of RNA metabolism. (1) ZCCHC12 and ZCCHC22 function as the transcriptional activator thus influencing the progress of RNA transcription/biogenesis. ZCCHC3, ZCCHC9, and ZCCHC10 are also found to affect RNA transcription, but more works are needed to elucidate the mechanisms. (2) The roles of ZCCHC family in splicing are diverged. ZCCHC17, ZCCHC20 and ZCCHC25 are associated with SR protein family, while ZCCHC21 functions as a regulator of alternative splicing (AS). (3) Translation control is another important facet. ZCCHC21 and ZCCHC22 regulate the translation initiation stage, and ZCCHC4 methylates human 28S rRNA to affect global translation. (4) As for degradation, ZCCHC1 mediates the processing of miRNA let-7 together with ZCCHC6 and ZCCHC11, as uridylyl-transferases, to facilitate RNA degradation. Related closely to nuclear exosome function, ZCCHC7 and ZCCHC8 also promote RNA degradation.

The conservation of ZCCHCs in different species indicates the biologically relevant functions of the specific CCHC motif. As far as we know, it has been associated with tumorigenesis, stem cell pluripotency and differentiation, reprogramming, metabolism, inflammation and so on. Although a census of ZCCHC family as RBPs comes to the fore, research concerning their exact roles on RNA are divided, and many even achieved opposite conclusions. Besides, only a limit part of ZCCHC proteins have explicit structure, while most still lacks structural research. Many disease states are derived from long-term RNA metabolism dysregulation, thus diving deep into the regulation mechanism of RNA metabolism via ZCCHC family proteins should shed light on both the physiological and pathological functions in human beings.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by grants from National Natural Science Foundation of China (Nos 81630075, 81972585, 81902866) and China’s National Key R&D Programme (NKP) [2019YFE0110600].