G C A T
T A C G
G C A T
genes
Article
Identification of Circular RNAs of Testis and Caput Epididymis
and Prediction of Their Potential Functional Roles in Donkeys
Yan Sun 1,† , Yonghui Wang 2,† , Yuhua Li 2 , Faheem Akhtar 2 , Changfa Wang 2, *
Citation: Sun, Y.; Wang, Y.; Li, Y.;
Akhtar, F.; Wang, C.; Zhang, Q.
Identification of Circular RNAs of
Testis and Caput Epididymis and
Prediction of Their Potential
Functional Roles in Donkeys. Genes
2023, 14, 66. https://doi.org/
10.3390/genes14010066
Academic Editors: Xiang Li and
Wuzi Dong
Received: 16 November 2022
Revised: 19 December 2022
Accepted: 22 December 2022
Published: 25 December 2022
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Genes 2023, 14, 66. https://doi.org/10.3390/genes14010066
and Qin Zhang 1, *
1 Shandong Provincial Key Laboratory of Animal Biotechnology and Disease Control and Prevention,
College of Animal Science and Technology, Shandong Agricultural University, Taian 271018, China
2 Liaocheng Research Institute of Donkey High-Efficiency Breeding and Ecological Feeding,
Liaocheng University, Liaocheng 252059, China
* Correspondence: wangcf1967@163.com (C.W.); qzhang@cau.edu.cn (Q.Z.)
† These authors contributed equally to this work.
Abstract: Circular RNAs (circRNAs) are a class of noncoding RNAs with a covalently closed loop.
Studies have demonstrated that circRNA can function as microRNA (miRNA) sponges or competing
endogenous RNAs. Although circRNA has been explored in some species and tissues, the genetic
basis of testis development and spermatogenesis in donkeys remain unknown. We performed
RNA-seq to detect circRNA expression profiles of adult donkey testes. Length distribution and
other characteristics were shown a total of 1971 circRNAs were differentially expressed and 12,648
and 6261 circRNAs were detected from the testis and caput epididymis, respectively. Among these
circRNAs, 1472 circRNAs were downregulated and 499 circRNAs were upregulated in the testis.
Moreover, KEGG pathway analyses and Gene Ontology were performed for host genes of circRNAs.
A total of 39 upregulated circRNA host genes were annotated in spermatogenesis terms, including
PIWIL2, CATSPERD, CATSPERB, SPATA6, and SYCP1. Other host genes were annotated in the focal
adhesion, Rap1 signaling pathway. Downregulated expressed circRNA host genes participated in
the TGF-β signaling pathway, GnRH signaling pathway, estrogen signaling pathway, and calcium
signaling pathway. Our discoveries provide a solid foundation for identifying and characterizing
critical circRNAs involved in testis development or spermatogenesis.
Keywords: circRNAs; testis; caput epididymis; donkey
1. Introduction
Donkeys (Equus asinus) belong to the equine family. Thousands of years ago, donkeys
were used for transporting goods and facilitated communications in distant areas. The don-
key industry has produced more economic benefits by exploiting donkey skin, milk, and
meat. It is essential for carrying out the breeding of donkeys because of plummeting stocks.
Jackass fertility plays a decisive role in donkey reproduction. Besides the requirements of
morphological assessment, testicular development and sperm quality are the most critical
indicators. In the early stages, we investigated the mRNAs and microRNAs (miRNAs)
expression of testis and cauda epididymis in jackass. Noncoding RNAs (ncRNAs), such as
miRNAs, Piwi-interacting RNAs (piRNAs), and long noncoding RNAs (lncRNAs), are vital
regulators for gene expression. The essential genes, regulated events, and critical signaling
pathways related to reproductions were summarized [1]. In recent years, circular RNAs
(circRNAs) have ubiquitously been in covalently closed-loop form in mammalian cells as
an important class of post-transcriptional regulators [2]. CircRNAs presented significant
tissue and developmental stage specificity; their primary role is modulating microRNAs
as sponges to protect their mRNA targets from degradation [3]. Therefore, circRNAs are
known as part of the competitive endogenous RNA network (ceRNET). Besides tethering
activity, circRNAs also regulate the transcription of their host genes [4] and the splicing of
their cognate mRNA through R-loop formation [5]. Recent studies suggested that circRNAs
https://www.mdpi.com/journal/genes
Genes 2023, 14, 66 2 of 10

act as microRNA sponges and play an important role in regulating gene expression through
a circRNA-microRNA-gene pathway [2,6].
CircRNAs involved in reproductive development have been reported. The well-known
circRNA related to reproductive processes was from the sex-determining region Y (SRY)
gene and performed specific functions in adult mouse testis [7]. The number of circRNAs
in the testis rank second to the amount found in rats’ brains [8,9]. The expression profiles
of circRNAs of human testis tissue were delineated using high-throughput sequencing,
and it was discovered that circRNA-forming genes are mostly related to spermatogenesis
and fertilization [10]. Liang et al. constructed a circRNA database of pigs [11]; in the testis
tissue, circRNAs showed the most incredible abundance and the most significant number of
specific circRNAs. Gao et al. detected circRNAs expression in neonatal and adult Chinese
Qinchuan cattle testes and performed functional analysis for the host genes of circRNAs [12].
The comparative analysis of circRNAs from three livestock species (pig, sheep, and cattle)
suggested that the number of exonic circRNAs in the testis seems associated with the
sexual maturity of the animal [13]. However, no further studies discussed the regulatory
function of these circRNAs in mammalian testis development and spermatogenesis. Our
understanding of circRNAs, their production, and their function remains limited.
We assembled one male Dezhou donkey genome, the first chromosome-level donkey
reference genome [14]. This reference and annotations version is more helpful for analyzing
the role of regulators. Therefore, in this study, to further explore the complicated regulatory
network of circRNAs, microRNA, and protein-coding genes in donkey testis and caput
epididymis, high-throughput sequencing and bioinformatics analyses were performed. The
general molecular biological characteristics of circRNAs were also determined. A circRNA-
miRNA-mRNA network related to reproduction was subsequently constructed to explore
the interaction among different molecules.

2. Materials and Methods

2.1. Sample Preparation, Ethics Statement
The Institutional Animal Care and Use Ethics Committee approved the animal ex-
periments and study protocol of Shandong Agricultural University (SDAUA-2018-018).
Normal and intact donkey testis tissues were collected from male Dezhou donkeys. Three
adult donkeys were derived from Yucheng Huimin Agricultural Science and Technology
Co. Ltd (Yucheng, Shandong, China). The testis and caput epididymidis samples were
sliced and immediately preserved in liquid nitrogen until use. Finally, tissue samples were
sent to the Beijing Omics Biotechnology Co. LTD for library construction and sequencing.

2.2. RNA Isolation, Library Construction, and Illumina Sequencing

The total RNA from six samples (the testis and caput epididymidis, each with 3 sam-
ples) was extracted using Trizol reagent (Invitrogen, Carlsbad, CA, USA). Then, ribosomal
RNA (rRNA) was removed from the DNA-free RNA using the Ribo-Zero™ Kit (Epicen-
tre, Madison, WI, USA) according to the manufacturer’s instructions. RNase R removed
linear RNA molecules. RNA integrity and quality were evaluated using the Agilent 2100
Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and the NanoDrop ND-2000
Spectrophotometer (NanoDrop, Wilmington, DE, USA).
Total RNA was randomly fragmented into small pieces and purified into fragments
with 200–600 bp lengths. Cleaved RNA fragments were subsequently reverse-transcribed
to create the final library, whose sequencing data will eventually identify mRNAs and
circRNAs. For the identification of microRNAs, the library construction was performed
using a small RNA Sample Prep Kit (TruSeq, Illumina). After detecting the insert size of
the library, sequencing was completed using the Illumina NovaSeq 6000 platform.
The sequencing data in Fastq format were uploaded to the Sequence Read Archive
under Accession Number PRJNA869196.
Genes 2023, 14, 66 3 of 10

2.3. Read Quality Control and Mapping

High-quality clean data were generated from raw data by removing adapter sequences,
duplicated sequences, and sequences that showed more than 10% unidentified nucleotides
and more than 50% low quality (Quality score ≤ 10). Read quality was assessed through
the FastQC program (v0.11.9). BWA software (v0.7.17) [15] was used to map the clean data
to the Dezhou donkey reference genome, a chromosome-level reference genome [14].

2.4. Identification of circRNAs and microRNAs

After the clean reads were aligned to the reference genome, CIRI software [16] was ap-
plied to identify backsplice junction reads using the BWA-MEM algorithm [17] for circRNA
prediction. The types of circRNA are classified according to their origin. The identified
circRNAs were subjected to statistical analysis of chromosome distribution and length
distribution. The expression of circRNAs was quantified by mapped backsplicing junc-
tion reads per million mapped reads (TPMs). The final criterion was the circRNAs were
expressed at least in two biological replicates.
The analysis steps of sRNA sequencing data for microRNA identification were de-
scribed in a previous report [1]. MicroRNAs were identified by using unique reads blasting
to the microRNA database (Release 22.1) and predicting the microRNAs by mirdeep2
software (v0.1.3) according to microRNA homology and high conservation [18].

2.5. Differentially Expressed RNA Identification and Pathway Analysis

Differentially expressed microRNA (DEM) and circRNA (DEC) identification between
testis and caput epididymis was performed using the DESeq2, with fold change ≥1.5 and
p < 0.05 as evaluation criteria. The differences in gene expression levels and statistical signif-
icance can be quickly detected through a volcano plot. RNAhybrid [19] and miRanda [20]
software were used to predict the target genes of DEMs. We conducted functional enrich-
ment analysis of host genes to study the main functions of circRNAs. To obtain a functional
annotation, Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes
(KEGG) were performed for host genes of circRNAs. Finally, a corrected p-value < 0.05 was
the threshold for statistically significant correlation.

2.6. DEC and Target microRNA Interaction

Integrating ncRNA and mRNA expression may identify the functional links between
ncRNAs and their target genes. Any RNA transcripts (circRNAs, lncRNAs, and mRNAs)
with miRNA binding sites can compete for and mutually affect the same miRNA; these
species are competing for endogenous RNAs [21]. The putative target microRNAs of DECs
were performed using the miRanda (v 3.3a) software and TARGETSCAN (V7.2).

2.7. Quantitative Real-Time PCR Validation

To confirm individual circRNA, after incubation with RNase R (Epicentre), first-strand
cDNA was synthesized with a random primer transcribed using the FastKing RT Kit (with
gDNase) (Tiangen, China) according to the manufacturer’s protocol. qPCR was performed
using the TB Green PCR Kit (TaKaRa, Dalian, China) on the LightCycler/LightCycler 480
System (Roche Diagnostics), according to the manufacturer’s instructions. The primers of
circRNAs were outward-facing, which primed divergently and amplified only circularly,
but not linear RNA [22]. Each 20 µL real-time RT-PCR reaction included 10 µL TB Green
Premix Ex Taq II (Tli RNaseH Plus, 2×), 2 µL cDNAs (100 ng) and 0.8 µL primers. PCR
conditions consisted of one cycle at 95 ◦ C for 30 s, followed by 40 cycles at 95 ◦ C for 5 s
and 60 ◦ C for 20 s, with fluorescence acquisition at 95 ◦ C for 5 s, 60 ◦ C for 1 min, 95 ◦ C for
15 s, and 50 ◦ C for 30 s. Relative expression level was determined using the 2−∆∆ct method
with β-actin as the control. Finally, the relative expression results were compared with the
RNA-seq data.
 and amplified only circularly, but not linear RNA [22]. Each 20 µ L real-time RT-PCR re-
action included 10 µ L TB Green Premix Ex Taq Ⅱ (Tli RNaseH Plus, 2×), 2 µ L cDNAs
(100 ng) and 0.8 µ L primers. PCR conditions consisted of one cycle at 95 °C for 30 s, fol-
lowed by 40 cycles at 95 °C for 5 s and 60 °C for 20 s, with fluorescence acquisition at 95
°C for 5 s, 60 °C for 1 min, 95 °C for 15 s, and 50 °C for 30 s. Relative expression level was
Genes 2023, 14, 66 determined using the 2−ΔΔct method with β-actin as the control. Finally, the4 relative
of 10 ex-
pression results were compared with the RNA-seq data.

3. Results
3. Results
3.1. General Characteristics of circRNA in Donkey Testis and Caput Epididymis
3.1. General Characteristics of circRNA in Donkey Testis and Caput Epididymis
The average clean data of transcriptome sequencing was 10,276,318,400 bp, and the
The average clean data
high-quality rateof(Q20)
transcriptome
was not lowersequencing
than 97%was 10,276,318,400
in each bp, and the Table
sample (Supplementary
high-quality rate (Q20) was not lower than 97% in each sample (Supplementary
S1). We identified one circRNA containing at least two backspliced junction Table S1). Based
reads.
We identifiedon one circRNA containing at least two backspliced junction reads. Based
their location on the genome, all circRNAs were widely scattered on all chromosomes on
their locationand
on the genome, all circRNAs were widely scattered on all chromosomes
mainly distributed on chromosomes 2 and 3 (Figure 1A). The host genes of these
and mainly distributed
circRNAs were on chromosomes
derived from 2exonic
and 3regions
(Figure(16232,
1A). The host intronic
82.3%), genes ofregions
these (2114,
circRNAs were 10.7%),
derivedandfrom
intergenic
exonicregions
regions (1345, 7.0%)
(16232, (Supplementary
82.3%), Table S2).
intronic regions For10.7%),
(2114, exonic circR-
and intergenicNAs,
regions61.69% circRNAs
(1345, were composedTable
7.0%) (Supplementary of noS2).
moreForthan 10 circRNAs,
exonic exons. The 61.69%
lengths were
widely distributed, and the most exonic circRNAs (62.4%) were
circRNAs were composed of no more than 10 exons. The lengths were widely distributed, no more than 20 kb nu-
cleotides (nt) (Figure 1B). A total of 12,648 and 6261 circRNAs
and the most exonic circRNAs (62.4%) were no more than 20 kb nucleotides (nt) (Figure 1B).were identified from the
testis and caput epididymis, respectively. We found that 3928
A total of 12,648 and 6261 circRNAs were identified from the testis and caput epididymis,circRNAs were shared by
testis and caput epididymis, and circRNA abundance was quantified
respectively. We found that 3928 circRNAs were shared by testis and caput epididymis, and as TPM (Supple-
mentary Table S3). According to the expression profiles of the circRNAs, the correlation
circRNA abundance was quantified as TPM (Supplementary Table S3). According to the
coefficients among three testis samples were >0.97, and Pearson’s correlation coefficients
expression profiles of the circRNAs, the correlation coefficients among three testis samples
of three caput epididymides were >0.88, indicating high relativity between samples
were >0.97, and Pearson’s correlation coefficients of three caput epididymides were >0.88,
(Supplementary Figure S1).
indicating high relativity between samples (Supplementary Figure S1).

Figure 1. General characteristics of circRNAs in donkey testis and caput epididymis: (A) chromoso-
mal distribution of circRNAs; (B) length distribution of circRNAs.
Figure 1. General characteristics of circRNAs in donkey testis and caput epididymis: (A) chromo-
somal
3.2. Feature and distribution
Function of circRNAs;
Analysis (B) length
of circRNAs distribution
and Their of circRNAs.
Host Gene
All circRNAs were mapped to 5102 corresponding host genes in donkey testis and
3.2. Feature and Function Analysis of circRNAs and Their Host Gene
caput epididymis. Of these genes, 40.5% produced only one circRNA, and 59.5% generated
more than two circRNAs (Supplementary Table S2). The circRNAs expressed in the testis
were derived from 3904 host genes, and circRNAs expressed in caput epididymis were
derived from 2802 host genes. We conducted KEGG annotation for the coexpressed and
specifically expressed host genes in testis and caput epididymis, a total of 202 pathway
terms, showing a high level of significance (corrected p-value < 0.05) including focal
adhesion (Ko04510), oocyte meiosis and maturation (Ko04114, Ko04914), GnRH signaling
pathway (Ko04912), and estrogen and calcium signaling pathway (Ko04915 and Ko04020)
(Supplementary Table S4).

3.3. Feature and Function Analysis of microRNAs in Donkey Testis and Caput Epididymis
For small RNA sequencing, the average number of unique reads was 13,825,528 bp
after conducting secondary quality control in each sample (Supplementary Table S5). Finally,
we identified 647 microRNAs; among them, 578 microRNAs were expressed in testis, and
511 microRNAs were expressed in caput epididymidis. There were 135 microRNAs specifi-
cally expressed in testis, and 68 microRNAs specifically expressed in caput epididymidis
(Supplementary Table S6).
We identified 232 DEMs between testis and caput epididymidis, including 137 upregu-
lated microRNAs and 95 downregulated microRNAs (Supplementary Table S7). The DEM
 nally, we identified 647 microRNAs; among them, 578 microRNAs were expressed in
testis, and 511 microRNAs were expressed in caput epididymidis. There were 135 mi
croRNAs specifically expressed in testis, and 68 microRNAs specifically expressed in
caput epididymidis (Supplementary Table S6).
We identified 232 DEMs between testis and caput epididymidis, including 137 up
Genes 2023, 14, 66 5 of 10
regulated microRNAs and 95 downregulated microRNAs (Supplementary Table S7)
The DEM result was also presented using a volcano plot (Figure 2A). A total of 21,892
target genes of DEMs were predicted; the result is shown in Supplementary Table S8.
result was also presented using a volcano plot (Figure 2A). A total of 21,892 target genes of
DEMs were predicted; the result is shown in Supplementary Table S8.

Figure 2. Differential microRNA (A) and circRNA (B) expression between testis and caput epididymis.
Volcano plots showing −log10 (p-value) versus log2 (fold change) in RPM. Red dots denote significantly
Figure 2.and
upregulated microRNAs Differential
circRNAs, microRNA
whereas(A) and dots
green circRNA (B) expression
denote between
significantly testis and caput epidi
downregulated
dymis. Volcano plots showing −log10 (p-value) versus log2 (fold change) in RPM. Red dots denot
microRNAs and circRNAs.
significantly upregulated microRNAs and circRNAs, whereas green dots denote significantly
downregulated microRNAs and circRNAs.
3.4. Analysis of Differentially Expressed circRNAs between Testis and Caput Epididymis
According to3.4.the circRNA
Analysis expressionExpressed
of Differentially profiles,circRNAs
1971 circRNAs were and
between Testis differentially ex-
Caput Epididymis
pressed (Supplementary Table S9). Compared with caput epididymis, 1472 DECs were
downregulated, and 499 DECs were upregulated in the testis (Figure 2B). To some extent,
the function of circRNAs is reflected through their host genes. We conducted functional
pathway and Gene Ontology analysis on the host genes. A total of 307 GO terms signifi-
cantly enriched on host genes of upregulated expressed circRNAs are shown in Supple-
mentary Table S10. The prominent GO terms include germ cell development (GO:0007281),
male germ cell nucleus (GO:0001673), and sperm-egg recognition (GO:0035036). The result
showed that 39 host genes were annotated in spermatogenesis terms (GO:0007283), includ-
ing essential genes associated with reproduction of piwi-like RNA-mediated gene silencing
2 (PIWIL2), cation channel sperm-associated auxiliary subunit delta (CATSPERD), cation
channel sperm-associated auxiliary subunit β (CATSPERB), spermatogenesis-associated 6
(SPATA6), and sperm flagellar 2 (SPEF2). In total, 298 circRNAs were from these host genes.
Moreover, boule homolog, RNA-binding protein (BOLL), and synaptonemal complex pro-
tein 1 (SYCP1) produced 22 circRNAs. Pathway analysis can provide more information
about the biological functions of genes. We enriched 77 significantly clustering functional
pathways (Supplementary Table S10), including oocyte maturation and meiosis (ko04914
and ko04114), focal adhesion (ko04510), and Rap1 signaling pathway (ko04015).
For host genes of downregulated expressed circRNAs, we predicted 72 significantly
enriched signal transduction pathways (Supplementary Table S11), including the TGF-
β signaling pathway (ko04350), GnRH signaling pathway (ko04912) and estrogen sig-
naling pathway (ko04915), calcium signaling pathway (ko04020), and oocyte meiosis
(ko04114). A total of 22 host genes were annotated in the above pathways, including inosi-
tol 1,4,5-trisphosphate receptor type 2 (ITPR2), ATPase plasma membrane Ca 2+ transporting
(ATP2B1), calcium/calmodulin-dependent protein kinase II, delta (CAMK2D), and epidermal
growth factor receptor (EGFR). The GO annotation indicated that the host genes were involved
in many biological processes (Supplementary Table S11), such as protein binding, signal
Genes 2023, 14, 66 6 of 10

transduction, and metabolic process, in which genes ITPR2 and ATP2B1 were annotated in
calcium ion transmembrane transporter activity (GO:0015085).

3.5. The Target microRNAs of Differentially Expressed circRNAs in Donkey Testis and
Caput Epididymis
The resulting circRNA-microRNA association network provided nodes and connec-
tions between circRNAs and their target microRNAs. Finally, 229 target microRNAs of
DECs were predicted (Supplementary Table S12). According to our data, one circRNA may
show tethering activity toward multiple microRNAs as targets. For example, 22 target microR-
NAs were predicted for 7 upregulated expressed circRNAs from the PIWIL2 host gene. On
the other hand, more than two circRNAs show tethering activity toward the same miRNA as
targets. eca-miR-205 was the target microRNA of multiple DE_circRNAs from host genes
EGFR, CATSPERD, ATP2B1, PIWIL2, and the microRNA Eca-miR-324-5P was the target
of circRNAs from genes CATSPERB, SPEF2, CAMK2D, and PIWIL2. Eca-miR-100 was the
target of circRNA from genes CATSPERD, ATP2B1, and CATSPERB.

3.6. Experimental Validation of Predicted circRNAs

To validate whether the differentially expressed circRNAs found by sequencing are
authentic, we selected exonic circRNAs that may have a significant differential expression
Genes 2021, 12, x FOR PEER REVIEW
for qRT-PCR verification. From these predicted circRNAs, six circRNAs of different abun- 7 of 10
dance and lengths were selected. The six host genes are PIWIL2, CATSPERD, SPATA6,
ITPR2, CAMK2D, EGFR. Primers spanning the backsplicing junction used in quantitative
real-time PCR
shows that are shownpatterns
expression in Supplementary Table S13.
of these selected Figurewere
circRNAs 3 clearly shows that
concordant withexpres-
RNA-
sion
seq patterns of these
results. These selected
results circRNAs were
demonstrated that concordant with RNA-seq
the ribo-depleted results. These
total RNA-seq results
results demonstrated that the ribo-depleted total RNA-seq results were reliable.
were reliable.

Figure3.3.Validation
Figure Validationof of DE-circRNAs
DE-circRNAs by by qRT-PCR.
qRT-PCR. Histograms
Histograms of relative
of the the relative expression
expression levels.levels.
The
The y-axis shows the fold change between two groups (log(−ΔΔCt,2) for qRT-PCR,
y-axis shows the fold change between two groups (log(−∆∆Ct,2) for qRT-PCR, log2 (fold change)log2 (fold
for
change) for sequencing). In total, the qRT-PCR results were consistent with the RNA-seq results.
sequencing). In total, the qRT-PCR results were consistent with the RNA-seq results.

4.4.Discussion
Discussion
In
Inthis
thisstudy,
study,weweperformed
performed the first
the demonstration
first demonstration of circNRA
of circNRAexpression in donkey
expression in don-
testis and caput
key testis epididymis.
and caput UsingUsing
epididymis. RNA-seq data, data,
RNA-seq 19,72119,721
circRNAs were were
circRNAs identified, and
identified,
we noted that almost all are exonic circRNAs. This characteristic was similar
and we noted that almost all are exonic circRNAs. This characteristic was similar to pre-to previous
observations on cattle,
vious observations on pig, and
cattle, sheep
pig, andtesticular tissues tissues
sheep testicular [13]. [13].
Functional
Functional annotation
annotation ofof host
host genes
genes found
found that
thatmultiple
multiplegenes
geneswere
wereannotated
annotatedinin
spermatogenesis
spermatogenesisterms.
terms. Two
TwocircRNAs
circRNAshighly
highlyexpressed
expressedin inthe
thetestis
testiswere
werederived
derivedfromfrom
the PIWIL2 gene. PIWIL2 belongs to the piwi family gene, a novel class of evolutionarily
conserved genes [23]. The PIWIL2 homolog Miwi has been identified in the mouse ge-
nome. Miwi is essential for male fertility and initiating spermiogenesis, a process that
transforms round spermatids into mature sperm [24]. Four upregulated circRNAs from
Genes 2023, 14, 66 7 of 10

the PIWIL2 gene. PIWIL2 belongs to the piwi family gene, a novel class of evolutionarily
conserved genes [23]. The PIWIL2 homolog Miwi has been identified in the mouse genome.
Miwi is essential for male fertility and initiating spermiogenesis, a process that transforms
round spermatids into mature sperm [24]. Four upregulated circRNAs from CATSPERB
and CATSPERD genes. These two genes belonged to CatSper complex parts, were restricted
to the testis, and were predicted to be active in the cilium. CatSperbeta (CATSPERB) is
the first identified auxiliary protein to the CatSper channel [25]. Mice lacking the sperm
tail-specific CATSPERD are infertile, and their spermatozoa lack both Ca(2+) current and
hyperactivated motility [26]. Spermatogenesis-associated (SPATA) genes are a large family
of genes that play a vital role in testis development and spermatogenesis. Six circRNAs
highly expressed in the testis were from SPATA genes, including SPATA1, SPATA6, SPATA7,
and SPATA22. In mice, inactivation of SPATA6 leads to acephalic spermatozoa and male
sterility [27].
For the host genes of circRNAs highly expressed in caput epididymis, we found a
particular TGF-β signaling pathway in KEGG pathway annotations. In total, 14 circRNAs
from eight host genes were involved in this pathway, including transforming growth factor,
β receptor II (TFGBR2), latent transforming growth factor β binding protein 1 (LTBP1),
activin A receptor type 2A (ACVR2A), and bone morphogenetic protein receptor (BMPR)
family genes (BMPR1A, BMPR1B, and BMPR2). These genes’ progressive expression at
key stages influences testis development and controls germline differentiation [28]. Our
results enriched many significant signaling pathways related to reproductive regulation
and neuroendocrine activity, including the GnRH signaling, estrogen, and calcium signal-
ing pathways. Inositol 1,4,5-trisphosphate receptors (ITPR) are a family of intracellular
Ca2+ release channels on the ER membrane. Calcium signaling plays essential roles in
mammalian fertilization, such as the calcium wave in the egg and the acrosome reaction
(AR) in sperm. The existence of ITPR types 1 and 3 in human sperm and their changes
during the reaction, but ITPR type 2 was undetectable [29]. Our sequencing data only
detected the host gene ITPR type 2 (ITPR2); this difference needs to be explored further.
ITPR1 and ITPR2 play an essential and redundant role in maintaining the integrity of the
fetal–maternal connection and embryonic viability [30]. In donkey’s caput epididymidis,
the high expression of circRNAs may influence reproductive endocrine activity.
circRNAs were identified as efficient microRNA sponges [6]. Interestingly, the inter-
actions of circRNAs and microRNAs show a peculiar behavior: multiple DE-circRNAs
act on the same unique microRNA. This may be a redundant mechanism to regulate the
expression of that microRNA. In our results, eca-miR-100 and eca-miR-205 were DEMs,
which interacted with DECs. Previous studies have reported that miR-205 and miR-100
were explicitly expressed in human testis [31]. Among the target microRNAs of DECs, mir-
532 and mir-34 families were also shown to be involved in cattle spermatogenesis [32,33].
MiR-122 is enriched in mice’s late-stage male germ cells, reducing mRNA expression of the
post-transcriptionally regulated germ cell transition protein 2 (Tnp2) to affect male fertil-
ity [34,35]. Mir-122 was highly expressed in donkey testis, and there were many interacted
circRNAs, including those derived from the SPATA7 gene but no circRNAs derived from
the Tnp2 gene. It demonstrated that microRNA could act on multiple target genes to per-
form a function. Another microRNA, miR-184, was highly expressed in donkey testis and
interacted with DECs from the PIWIL2 gene. Wu et al. reported that miR-184 expression
levels increased with the postnatal development of the mouse testis, and its expression was
restricted to testicular germ cells [34,35]. In the future, more molecular experiments will be
conducted to verify the regulation of circRNAs on downstream miRNA targets.

5. Conclusions
In summary, we detected abundant circRNAs in donkey testes and epididymis and
revealed the length distribution, genomic features, and other characteristics of circRNAs.
Among these circRNAs from testis, 39 upregulated circRNA host genes were annotated in
spermatogenesis terms, including PIWIL2, CATSPERD, CATSPERB, SPATA6, and SYCP1.
Genes 2023, 14, 66 8 of 10

Other host genes were annotated in the focal adhesion, Rap1 signaling pathway. Down-
regulated expressed circRNA host genes participated in the TGF-β signaling pathway,
GnRH signaling pathway, estrogen signaling pathway, and calcium signaling pathway.
These findings have laid the foundation for further study of noncoding RNA-related
donkey reproduction.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/genes14010066/s1, Figure S1: Pearson Correlation of all Samples.
Table S1: The general information of clean data from transcriptome sequencing.; Table S2: all
information of identified circRNAs and their host genes.; Table S3: The expression of circRNAs
identified from testis and caput epididymis.; Table S4: The KEGG pathways for the the co-expressed
and specifically expressed host genes in testis and caput epididymis.; Table S5: The data quality
of small RNA sequencing.; Table S6: The expression of identified microRNAs in testis and caput
epididymids.; Table S7: The information of differentially expressed microRNAs between testis and
caput epididymidis.; Table S8: The predicted target genes of microRNAs.; Table S9: The differentially
expressed information of circRNAs.; Table S10: The GO annotation terms and KEGG pathway
analysis for the up_regulated expressed circRNA’ host genes.; Table S11: The KEGG pathway analysis
and GO annotation for the down_regulated expressed circRNA’ host genes.; Table S12: The predicted
target microRNAs of differentially expressed circRNAs in donkey testis and caput epididymis.; Table
S13: Primers of six validated testis derived circRNAs for general and quantitive RT-PCR.
Author Contributions: Conceptualization, Y.S. and Q.Z.; Formal analysis, Y.L.; Data curation, Y.W.;
Funding acquisition, Q.Z. and C.W.; Project administration, Q.Z. and C.W.; Supervision, C.W.;
Validation, Y.L.; Visualization, F.A.; Writing—original draft, Y.S. and Y.W.; Writing—review & editing,
Y.S. and F.A. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Shandong Province Natural Fund Project (Grant Number
ZR2020MC168), the Shandong Province Modern Agricultural Industrial Technology System Project
(Grant number SDAIT-27), the Construction of Molecular ID Card for Donkey and Camel Species,
Livestock and Poultry Seed Industry Project of Ministry of Agriculture and Rural Affair (Grant
Number 19211162), and the Key R&D Project of Shandong Province: Innovation and Demonstration
of Key Technologies for the Integrated Development of Donga Black Donkey Industry (Grant Number
2021TZXD012).
Institutional Review Board Statement: The Institutional Animal Care and Use Ethics Committee
approved the animal experiments and study protocol of Shandong Agricultural University (SDAUA-
2018-018).
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Conflicts of Interest: The authors declare that they have no competing interests.

References
1. Tian, F.; Wang, J.; Li, Y.; Yang, C.; Zhang, R.; Wang, X.; Ju, Z.; Jiang, Q.; Huang, J.; Wang, C.; et al. Integrated analysis of mRNA
and miRNA in testis and cauda epididymidis reveals candidate molecular markers associated with reproduction in Dezhou
donkey. Livestock Sci. 2020, 234, 103885. [CrossRef]
2. Memczak, S.; Jens, M.; Elefsinioti, A.; Torti, F.; Krueger, J.; Rybak, A.; Maier, L.; Mackowiak, S.D.; Gregersen, L.H.; Munschauer, M.;
et al. Circular RNAs are a large class of animal RNAs with regulatory potency. Nature 2013, 495, 333–338. [CrossRef] [PubMed]
3. Suzuki, H.; Zuo, Y.; Wang, J.; Zhang, M.Q.; Malhotra, A.; Mayeda, A. Characterization of RNase R-digested cellular RNA source
that consists of lariat and circular RNAs from pre-mRNA splicing. Nucleic Acids Res. 2006, 34, e63. [CrossRef] [PubMed]
4. Li, Z.; Huang, C.; Bao, C.; Chen, L.; Lin, M.; Wang, X.; Zhong, G.; Yu, B.; Hu, W.; Dai, L.; et al. Exon-intron circular RNAs regulate
transcription in the nucleus. Nat. Struct. Mol. Biol. 2015, 22, 256–264. [CrossRef] [PubMed]
5. Conn, V.M.; Hugouvieux, V.; Nayak, A.; Conos, S.A.; Capovilla, G.; Cildir, G.; Jourdain, A.; Tergaonkar, V.; Schmid, M.; Zubieta,
C.; et al. A circRNA from SEPALLATA3 regulates splicing of its cognate mRNA through R-loop formation. Nat. Plants 2017,
3, 17053. [CrossRef]
6. Hansen, T.B.; Jensen, T.I.; Clausen, B.H.; Bramsen, J.B.; Finsen, B.; Damgaard, C.K.; Kjems, J. Natural RNA circles function as
efficient microRNA sponges. Nature 2013, 495, 384–388. [CrossRef]
Genes 2023, 14, 66 9 of 10

7. Capel, B.; Swain, A.; Nicolis, S.; Hacker, A.; Walter, M.; Koopman, P.; Goodfellow, P.; Lovell-Badge, R. Circular transcripts of the
testis-determining gene Sry in adult mouse testis. Cell 1993, 73, 1019–1030. [CrossRef]
8. You, X.; Vlatkovic, I.; Babic, A.; Will, T.; Epstein, I.; Tushev, G.; Akbalik, G.; Wang, M.; Glock, C.; Quedenau, C.; et al. Neural
circular RNAs are derived from synaptic genes and regulated by development and plasticity. Nat. Neurosci. 2015, 18, 603–610.
[CrossRef]
9. Gong, B.; Xu, J.; Tong, W. Landscape of circRNAs Across 11 Organs and 4 Ages in Fischer 344 Rats. Chem. Res. Toxicol. 2021, 34,
240–246. [CrossRef]
10. Dong, W.W.; Li, H.M.; Qing, X.R.; Huang, D.H.; Li, H.G. Identification and characterization of human testis derived circular
RNAs and their existence in seminal plasma. Sci. Rep. 2016, 6, 39080. [CrossRef]
11. Liang, G.; Yang, Y.; Niu, G.; Tang, Z.; Li, K. Genome-wide profiling of Sus scrofa circular RNAs across nine organs and three
developmental stages. DNA Res. 2017, 24, 523–535. [CrossRef] [PubMed]
12. Gao, Y.; Wu, M.; Fan, Y.; Li, S.; Lai, Z.; Huang, Y.; Lan, X.; Lei, C.; Chen, H.; Dang, R. Identification and characterization of circular
RNAs in Qinchuan cattle testis. R. Soc. Open Sci. 2018, 5, 180413. [CrossRef] [PubMed]
13. Robic, A.; Cerutti, C.; Kühn, C.; Faraut, T. Comparative Analysis of the Circular Transcriptome in Muscle, Liver, and Testis in
Three Livestock Species. Front. Genet. 2021, 12, 665153. [CrossRef] [PubMed]
14. Wang, C.; Li, H.; Guo, Y.; Huang, J.; Sun, Y.; Min, J.; Wang, J.; Fang, X.; Zhao, Z.; Wang, S.; et al. Donkey genomes provide new
insights into domestication and selection for coat color. Nat. Commun. 2020, 11, 6014. [CrossRef] [PubMed]
15. Li, H.; Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 2009, 25, 1754–1760.
[CrossRef] [PubMed]
16. Gao, Y.; Wang, J.; Zhao, F. CIRI: An efficient and unbiased algorithm for de novo circular RNA identification. Genome Biol. 2015,
16, 4. [CrossRef]
17. Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv 2013, arXiv:1303.3997.
18. Friedländer, M.R.; Mackowiak, S.D.; Li, N.; Chen, W.; Rajewsky, N. miRDeep2 accurately identifies known and hundreds of novel
microRNA genes in seven animal clades. Nucleic Acids Res. 2012, 40, 37–52. [CrossRef]
19. Rehmsmeier, M.; Steffen, P.; Hochsmann, M.; Giegerich, R. Fast and effective prediction of microRNA/target duplexes. RNA
2004, 10, 1507–1517. [CrossRef]
20. John, B.; Enright, A.J.; Aravin, A.; Tuschl, T.; Sander, C.; Marks, D.S. Human MicroRNA targets. PLoS Biol. 2004, 2, e363.
[CrossRef]
21. Salmena, L.; Poliseno, L.; Tay, Y.; Kats, L.; Pandolfi, P.P. A ceRNA hypothesis: The Rosetta Stone of a hidden RNA language? Cell
2011, 146, 353–358. [CrossRef] [PubMed]
22. Jeck, W.R.; Sorrentino, J.A.; Wang, K.; Slevin, M.K.; Burd, C.E.; Liu, J.; Marzluff, W.F.; Sharpless, N.E. Circular RNAs are abundant,
conserved, and associated with ALU repeats. RNA 2013, 19, 141–157. [CrossRef] [PubMed]
23. Cerutti, L.; Mian, N.; Bateman, A. Domains in gene silencing and cell differentiation proteins: The novel PAZ domain and
redefinition of the Piwi domain. Trends Biochem. Sci. 2000, 25, 481–482. [CrossRef] [PubMed]
24. Deng, W.; Lin, H. miwi, a murine homolog of piwi, encodes a cytoplasmic protein essential for spermatogenesis. Dev. Cell 2002, 2,
819–830. [CrossRef] [PubMed]
25. Liu, J.; Xia, J.; Cho, K.H.; Clapham, D.E.; Ren, D. CatSperbeta, a novel transmembrane protein in the CatSper channel complex. J.
Biol. Chem. 2007, 282, 18945–18952. [CrossRef]
26. Chung, J.J.; Navarro, B.; Krapivinsky, G.; Krapivinsky, L.; Clapham, D.E. A novel gene required for male fertility and functional
CATSPER channel formation in spermatozoa. Nat. Commun. 2011, 2, 153. [CrossRef]
27. Yuan, S.; Stratton, C.J.; Bao, J.; Zheng, H.; Bhetwal, B.P.; Yanagimachi, R.; Yan, W. Spata6 is required for normal assembly of the
sperm connecting piece and tight head-tail conjunction. Proc. Natl. Acad. Sci. USA 2015, 112, E430–E439. [CrossRef]
28. Young, J.C.; Wakitani, S.; Loveland, K.L. TGF-β superfamily signaling in testis formation and early male germline development.
Semin. Cell Dev. Biol. 2015, 45, 94–103. [CrossRef]
29. Kuroda, Y.; Kaneko, S.; Yoshimura, Y.; Nozawa, S.; Mikoshiba, K. Are there inositol 1,4,5-triphosphate (IP3) receptors in human
sperm? Life Sci. 1999, 65, 135–143. [CrossRef]
30. Yang, F.; Huang, L.; Tso, A.; Wang, H.; Cui, L.; Lin, L.; Wang, X.; Ren, M.; Fang, X.; Liu, J.; et al. Inositol 1,4,5-trisphosphate
receptors are essential for fetal-maternal connection and embryo viability. PLoS Genet. 2020, 16, e1008739. [CrossRef]
31. Liu, C.G.; Calin, G.A.; Meloon, B.; Gamliel, N.; Sevignani, C.; Ferracin, M.; Dumitru, C.D.; Shimizu, M.; Zupo, S.; Dono, M.; et al.
An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues. Proc. Natl. Acad. Sci. USA
2004, 101, 9740–9744. [CrossRef]
32. Tscherner, A.; Gilchrist, G.; Smith, N.; Blondin, P.; Gillis, D.; LaMarre, J. MicroRNA-34 family expression in bovine gametes and
preimplantation embryos. Reprod. Biol. Endocrinol. 2014, 12, 85. [CrossRef]
33. Zhang, S.; Zhang, Y.; Yang, C.; Zhang, W.; Ju, Z.; Wang, X.; Jiang, Q.; Sun, Y.; Huang, J.; Zhong, J.; et al. TNP1 Functional SNPs in
bta-miR-532 and bta-miR-204 Target Sites Are Associated with Semen Quality Traits in Chinese Holstein Bulls. Biol. Reprod. 2015,
92, 139. [CrossRef] [PubMed]
Genes 2023, 14, 66 10 of 10

34. Yu, Z.; Raabe, T.; Hecht, N.B. MicroRNA Mirn122a reduces expression of the posttranscriptionally regulated germ cell transition
protein 2 (Tnp2) messenger RNA (mRNA) by mRNA cleavage. Biol. Reprod. 2005, 73, 427–433. [CrossRef] [PubMed]
35. Wu, J.; Bao, J.; Wang, L.; Hu, Y.; Xu, C. MicroRNA-184 downregulates nuclear receptor corepressor 2 in mouse spermatogenesis.
BMC Dev. Biol. 2011, 11, 64. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.