Profiling chromatin accessibility responses to seasonal heat stress in Angora rabbit testes

Abstract

Background

Seasonal heat stress negatively impacts spermatogenesis, yet the underlying molecular mechanisms remain poorly understood. To investigate the epigenetic responses to heat stress in Angora rabbit spermatogenesis, we performed an Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) on testicular tissues collected during spring and summer.

Results

High-quality sequencing data were obtained, yielding 196 to 310 million clean reads per sample with mapping rates exceeding 97%. Chromatin accessibility signals were predominantly enriched near transcription start sites, and principal component analysis (PCA) clearly segregated the spring and summer samples, revealing distinct epigenetic landscapes across seasons. We identified 416 differentially accessible chromatin regions (DARs), comprising 346 up-regulated and 70 down-regulated peaks in the summer group compared to the spring group. Motif analysis of these regions revealed enrichment of binding sites for transcription factors (TFs) such as E-box, MITF, and Usf2. Genes associated with these DARs were significantly enriched in reproduction-related biological processes and critical signaling pathways, including p53, TGF-beta, MAPK, Wnt, and mTOR. Integrating the ATAC-seq data with RNA-seq profiles identified 88 overlapping differentially expressed genes, among which BAX, CDK4, UBB, and TNP2 exhibited consistent or complementary alterations in both chromatin accessibility and gene expression. These key genes are intrinsically involved in spermatogenesis, stress response, and apoptosis, and are potentially regulated by the transcription factors identified in this study.

Conclusions

Our findings provide novel insights into the epigenetic regulation of testicular function under heat stress, highlighting essential candidate genes and pathways for further functional studies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12864-026-13133-8.

Background

Seasonal heat stress during summer is becoming an increasingly severe challenge due to global warming, negatively affecting animal health, welfare and production. Exposure in high ambient temperatures during warmer months can severely impair the animal reproductive system and disrupt the testicular microenvironment [1]. Although thermal tolerance varies among species, heat stress consistently diminishes fertility [2]. As homoeothermic animals, Angora rabbits are particularly sensitive to temperature fluctuations due to their dense coats and lack of functional sweat glands, making heat dissipation difficult under high-temperature conditions [3, 4]. Consequently, heat stress can decrease rabbit production performance and increase mortality by disrupting endocrine regulation and suppressing both immune and reproductive functions [5, 6]. This heat-induced decline in summer fertility has resulted in substantial economic losses for the rabbit industry [5, 7]. Understanding the molecular mechanisms driving heat stress-induced testicular dysfunction is therefore essential for developing effective strategies to mitigate reproductive damage and enhance breeding efficiency during high-temperature seasons.

The optimal growth environment for rabbits typically requires a temperature of 15℃–20℃ and a relative humidity of 60%–65% [3]. Elevated temperatures are known to reduce rabbit reproduction performance [8]. In mammals, sperm is generated within the testes, where germ cells–particularly spermatocytes and early round spermatids–exhibit high mitotic activity and are extremely sensitive to elevated temperatures [9]. Once testicular heat stress occurs, the risk of infertility increases significantly [10]. Heat stress can elevate the production of reactive oxygen species, induce germ cell DNA damage, and disrupt the spermatogenic process, ultimately leading to a decline in semen quality and failed spermatogenesis [10–12]. The impacts of natural summer heat stress in temperate climates are well-documented; for instance, bull semen quality is affected by season, with spring samples generally exhibiting superior quality compared to those collected in other seasons [13]. Similarly, the sperm density, vitality, and kinematic parameters of Suffolk sheep are reduced, while sperm acrosome malfunctions and DNA damage increase during the transition from spring to summer [14].

Testicular development is a highly complex process precisely regulated in time and space by multiple TFs and epigenetic regulators [15]. For example, SRY acts as the primary testis-determining factor during the morphogenetic process of vertebrate gonadogenesis [16]. Furthermore, sperm production requires the coordinated regulation of gene expression by various members of the SRY-related box (SOX) transcription factor family [17–20], while GATA factors play important roles in regulating genes essential for male testis differentiation and function [21]. At the epigenetic level, histone H2B.W1 has been suggested to alter the chromatin structure of spermatogonia [22]. Additionally, the dynamic expression and distribution of histone acetyltransferases (HATs), histone deacetylases (HDACs), and the lactate-regulating enzyme hexokinase 2 (HK2) underscore importance of histone lactylation in modifying chromatin accessibility across different testicular developmental stages [23].

Chromatin architecture dynamically modulates gene expression by controlling the accessibility of transcription factors to regulatory elements [24, 25]. Open chromatin regions permit transcription factors and other regulatory complexes to bind to cis-regulatory elements, initiating gene transcription, whereas closed chromatin hinders this interaction, leading to gene silencing [26, 27]. The ATAC-seq is a robust tool for analyzing chromatin dynamics; it accurately maps open chromatin regions, predicts transcription factor binding sites, and elucidates the relationship between chromatin structure and gene expression [28–30]. This technology enables genome-wide profiling of chromatin accessibility with high sensitivity and efficiency [31, 32].

In our previous study, we analyzed the testicular transcriptomes of Angora rabbits across different seasons using RNA-seq, identifying numerous genes related to heat stress and reproduction [33]. While ATAC-seq has been widely applied in studies investigating reproductive development and environmental stress responses, the impact of seasonal heat stress on chromatin accessibility in Angora rabbit testicular tissues remains largely unexplored. Here, we performed ATAC-seq on testicular tissues collected from Angora rabbits during spring (no heat stress) and summer (intense heat stress) to systematically characterize seasonal variations in the epigenetic landscape. By identifying differentially accessible chromatin regions, determining their associated transcription factor binding motifs, and integrating these data with our previously published transcriptomic profiles [33], we aimed to construct gene regulatory networks responsive to heat stress. This study seeks to elucidate the chromatin-level mechanisms underlying heat stress-induced impairment of testicular development, identify key regulatory factors, and provide a comprehensive molecular basis for understanding heat stress-related reproductive dysfunction in rabbits.

Materials and methods

Experimental animals and sample collection

Eight healthy, 1-year-old male Angora rabbits with normal reproductive function were selected from breeding group (more than 500 Angora rabbits) from the Animal Husbandry Institute of the Anhui Academy of Agriculture Sciences (Fig. 1). The rabbits were housed individually in cages, provided with free access to water, and fed a fixed amount of diet at scheduled times. Continuous monitoring of ambient temperature and relative humidity was conducted during spring and summer. The temperature-humidity index (THI), calculated according to our previous study which revealed no heat stress in spring but severe heat stress in summer in the rabbit house [33]. The equation is as follows:

Fig. 1.

Photo of Angora rabbits

The db °C means dry bulb temperature, RH means relative humidity percentage / 100. The obtained THI values were categorized as follows: THI < 27.8, no heat stress; 27.8 ≤ THI < 28.9, moderate heat stress; 28.9 ≤ THI < 30.0, severe heat stress; THI ≥ 30.0, very severe heat stress.

Testis samples were obtained from Angora rabbits sacrificed by injecting 20–30 ml of air into the ear vein during both seasons and immediately snap-frozen in liquid nitrogen for subsequent ATAC-seq analysis.

ATAC-seq library construction and sequencing

Four biological replicates of Angora rabbit testis samples from both spring and summer were utilized for ATAC-seq. Approximately 30–50 mg frozen testis tissue per sample was ground into powder in a pre-cooled lysis buffer. Nuclei were subsequently isolated by centrifugation at 550 rcf for 10 min at 4 °C. The resulting nuclei suspensions were resuspended in the Tn5 transposase reaction mixture, and the transposition reaction was incubated at 37 °C for 1 h. Following transposition, the DNA libraries were amplified using PCR, and purified with the Qiagen MinElute PCR Purification Kit (Qiagen, Hilden, Germany). Library quality was assessed using an Agilent 2100 Bioanalyzer to confirm successful construction. Finally, the libraries were sequenced on the Illumina novaseq 6000 platform by Wuhan Yingzi Gene Technology Co., Ltd. (Wuhan, China).

ATAC-seq data analysis

For all sequenced samples, raw reads were stored in the FASTQ format. To ensure data quality, adapters, low-quality bases with Phred quality score < 20, and reads containing more than 10% unknown nucleotides were trimmed and filtered using cutadapt (v2.5) [34]. The resulting high-quality clean reads were aligned to the Oryctolagus cuniculus reference genome (GCF_009806435.1) using Bowtie2 (v2.3.4.1). Peak calling was performed with MACS2 (v2.1.1) [35] to identify open chromatin regions for each sample, utilizing a strict significance threshold of P value < 10^{− 5}. Peak annotation across different genomic features was executed using the ChIPseeker (v1.32.1) R package [36]. Peak Transcripts Per Million quantification and motif enrichment analysis within the peak regions were conducted using Homer (v4.11.1) [37]. The DESeq2 (v1.36.0) R package [38] was employed to identify differentially accessible peaks, defining |log2 Fold change| > 1 and P value < 0.05 as the strict thresholds for significance between the two seasonal groups. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analyses were performed using the clusterProfiler (v4.4.4) R package [39].

Integration of ATAC-seq and RNA-seq

To investigate the correlation between chromatin accessibility and gene transcription levels, we integrated our ATAC-seq data with previously generated RNA-seq data (Accession number: PRJNA1220349) [33]. The expression profiles of differentially expressed genes identified in the RNA-seq dataset were cross-referenced with the genes assigned to the differential peaks discovered via ATAC-seq. Genomic sequencing tracks were visualized using the Integrated Genomics Viewer (IGV v2.3.61) [40]. The overlapping genes—those exhibiting both differential expression in the RNA-seq data and association with differential chromatin peaks in the ATAC-seq data—were subsequently subjected to GO and KEGG pathway enrichment analyses. The co-expression network of genes and TFs was constructed using Cytoscape (v 3.10.1) [41].

Results

ATAC-seq quality control of Angora rabbit testicular tissues

To explore the epigenetic molecular mechanisms underlying heat stress response, ATAC-seq was performed to detect genome-wide chromatin accessibility differences in testicular tissues of Angora rabbits collected during spring and summer. Following the filtration of low-quality raw reads, 196 to 310 million clean reads were obtained per sample (Table 1). Furthermore, over 97% of these clean reads were successfully mapped to the rabbit reference genome for downstream analysis. Library quality was rigorously evaluated based on insert fragment lengths and peak signal distributions. The results demonstrated that the majority of insert fragments fell within the 0–200 bp and 200–400 bp ranges, indicative of abundant nucleosome-free regions and mono-nucleosome fragments, respectively (Fig. 2A, B). Chromatin accessibility signals were predominantly highly enriched within 3 kb of transcription start sites (TSSs), highlighting the significance of these open chromatin regions in active transcriptional regulation (Fig. 2C). PCA based on ATAC-seq peak read signals revealed that the spring testicular samples (T1–T4) clustered tightly together and were clearly distinct from the summer samples (T5–T8) (Fig. 2D). Collectively, these results confirm that the generated sequencing data are of high quality and reliability.

Table 1.

Summary of ATAC-seq libraries for each testis sample

Sample	Raw Reads	Clean Reads	Q30 (%)	Mapping Rate (%)	Properly Mapped (%)
T1	310,116,770	310,112,986	86.56	97.57	95.69
T2	244,769,664	244,767,486	91.81	97.21	96.29
T3	232,967,968	232,965,260	91.23	97.08	96.08
T4	228,471,014	228,467,824	90.84	97.77	96.86
T5	278,974,574	278,971,142	90.75	97.67	96.72
T6	196,478,694	196,476,288	91.12	98.01	97.06
T7	294,918,288	294,915,776	91.32	97.62	96.76
T8	242,479,280	242,476,398	90.81	97.54	96.68

Fig. 2.

Quality control of ATAC-Seq. A, B Distribution of insert size in testis tissues from spring and summer. T1–4 represent testes in spring, T5–8 represent testes in summer; C ATAC-seq signals around the transcript start site (TSS). The x-axis represents the 3 kb range upstream and downstream of TSS, and each row on the y-axis represents degree of read enrichment; D The principal component analysis (PCA) of testis samples from spring and summer. CT represents testis samples from spring, including T1, T2, T3, and T4; XT represents testis samples from summer, including T5, T6, T7, and T8

Identification of global chromatin accessibility

The chromosomal distribution of all identified peaks indicated that the overall peak signal intensity in the spring samples was substantially higher than that in the summer samples (Fig. 3A). A total of 9,466 peaks were specifically identified in the spring samples, while 1,271 peaks were specific to the summer samples, with 1,956 peaks shared between the two seasons (Fig. 3B). To annotate all the peaks, corresponding region of the genome were separated into promoters, exons, introns, intergenic regions, 5ʹ untranslated regions (UTRs), and 3ʹ UTRs. We found that the vast majority of peaks were localized within intergenic regions, introns, and promoters (Fig. 3C). Additionally, distance analysis relative to TSSs revealed that most peaks were concentrated within 10–100 kb of the TSSs across the genome (Fig. 3D).

Fig. 3.

Identification and analysis of peaks. A Distribution of all peaks in Chromosomes in testis tissues from spring and summer; B Venn diagram illustrating the overlapping peaks between the testis tissues from spring and summer; C Distribution ratio of peaks on the promoter, 5ʹUTR, 3ʹUTR, exon, intron, and intergenic regions; D Common peaks relative TSS distance distribution ratio map in the testis tissues from spring and summer. CT and XT represent the testis tissues from spring and summer, respectively

Analysis of differentially accessible regions (DARs) and motif enrichment

To identify the specific open chromatin regions responsive to summer heat stress during testicular development, a differential peak analysis was conducted. A total of 416 differentially accessible peaks were identified, comprising 346 up-regulated and 70 down-regulated peaks in the summer group compared to the spring group (Fig. 4A). Hierarchical clustering of differential peaks demonstrated high intra-group consistency and distinct inter-group segregation between the two seasons (Fig. 4B). Motif enrichment analysis of the DARs using Homer software successfully identified binding motifs for 47 known TFs (P < 0.05) (Supplementary Table 1). Notably, the top 10 TFs annotated to these differentially accessible chromatin regions included BIM3, E-box, Cbf1, MITF, Usf2, BIM1, Ronin, bHLHE40, GFY-Staf, and NRF (Fig. 4C).

Fig. 4.

Analysis of differential peaks and prediction of transcription factors. A Scatter plot of differential peaks ( |log2 Fold change| > 1; P value < 0.05 ). The yellow dot represents the up-regulated difference peak and the blue dot represents the down-regulated difference peak; B Cluster analysis of differential peaks; C Motif analysis of differential peaks

Functional enrichment of genes associated with DARs

We identified a total of 368 genes related to the differential peaks, including 301 genes linked to up-regulated peaks and 67 genes linked to down-regulated peaks. To elucidate their potential biological functions, GO and KEGG pathway enrichment analyses were performed (P < 0.05). GO enrichment analysis revealed that these genes were significantly enriched in biological processes essential for testicular function, such as reproduction, germ cell development, cell death, cell cycle regulation, spermatid differentiation and development, and the positive regulation of meiotic chromosome separation (Fig. 5A). Concurrently, KEGG analysis identified 27 significantly enriched pathways, prominently featuring the p53, TGF-beta, MAPK, Wnt, and mTOR signaling pathways (Fig. 5B). These findings strongly suggest that summer heat stress profoundly impacts the chromatin accessibility of key gene networks governing testicular development.

Fig. 5.

Enrichment analyses of genes related to differential peaks (P < 0.05). A GO enrichment analysis; B KEGG enrichment analysis

Integrative analysis of ATAC-seq and RNA-seq data

To further investigate the regulatory relationship between open chromatin regions and actual gene expression, we integrated our ATAC-seq data with parallel RNA-seq datasets. This combined analysis identified 88 overlapping differentially expressed genes (DEGs), comprising 62 up-regulated and 26 down-regulated genes (Fig. 6A). Correlation analysis between chromatin accessibility and transcriptional levels revealed a weak positive correlation (r²= 0.15; Fig. 6B). We combined with RNA-seq data [33] and specifically focused on the ATAC-seq and RNA-seq signals of several critical genes involved in testicular development (BAX, CDK4, UBB, and TNP2). The results demonstrated that both the transcript levels and chromatin accessibility of BAX and CDK4 were elevated in summer compared to spring, while those of TNP2 were reduced in summer. Interestingly, UBB exhibited a discordant trend, showing increased transcript levels despite a decrease in chromatin accessibility (Fig. 6C, D).

Fig. 6.

Integration analysis of ATAC-seq and RNA-seq data. A Overlapping analysis of differentially expressed genes identified in RNA-seq and genes related to differential peaks from ATAC-seq; B Correlation of genes related to significantly differential peaks (ATAC-Seq) and gene expression (RNA-seq); C Cluster analysis of BAX, CDK4, UBB, and TNP2 expression levels; D Visualization of BAX, CDK4, UBB, and TNP2 from ATAC-seq data in the spring (blue) and summer (red) groups; E, F GO and KEGG enrichment analyses of overlapping genes, respectively (P < 0.05). CT and XT represent the testis tissues from spring and summer

Functional annotation of these overlapping genes revealed significant enrichment in terms related to response to stress, sexual reproduction, cell cycle, cell death, spermatogenesis, male gamete generation, spermatid differentiation, male gonad development, flagellated sperm motility, positive regulation of apoptotic signaling pathway, and positive regulation of apoptotic DNA fragmentation and p53 signaling pathway (Fig. 6E, F). Specifically, BAX, TNP2, and UBB are involved in regulating male gamete generation and spermatid differentiation, TNP2 also are involved in regulating sperm motility and flagellated sperm motility, BAX and UBB regulate cell death, BAX, CDK4, and UBB are key responders to stress. Furthermore, BAX and CDK4 serve as critical regulators within the p53 signaling pathway. By mapping the motifs of the differential peaks to the TFs of these specific differentially expressed genes, we constructed a comprehensive regulatory network highlighting the close interactions between these four key genes (BAX, CDK4, UBB, and TNP2) and their upstream TFs (Fig. 7). In these interactions, transcription factors USF1, USF2, MITF, KLF1, KLF3, KLF5, and KLF14 regulate BAX gene. UBB gene is only regulated by Sp2. Multiple transcription factors govern CDK4 expression. Among them, transcription factors bHLHE41, SPCH, KLF14, CLOCK, PIF4, bHLH34, and bHLH40 are also involved in the transcriptional regulation of the TNP2 gene. The results underscore their pivotal roles in reproductive regulation during summer heat stress.

Fig. 7.

Interaction network between genes and transcription factors. The green ovals represent transcription factors, red diamonds represent genes

Discussion

Heat stress is widely recognized as a detrimental factor affecting livestock reproductive performance. Consistent with this, our previous study demonstrated that natural summer heat stress not only decreased semen volume and sperm density but also induced severe morphological damage in the testes, alongside widespread transcriptomic alterations in Angora rabbits [33]. In the current study, we extended these findings by performing ATAC-seq to investigate genome-wide changes in chromatin accessibility within the testicular tissues across seasons, aiming to elucidate the epigenetic mechanisms driving heat stress-induced reproductive dysfunction.

Our ATAC-seq profiling revealed that summer heat stress led to an overall decrease in the number of accessible peaks, with signals predominantly enriched near TSS regions, reinforcing the established principle that open chromatin is functionally coupled to transcriptional activity [42]. Differential analysis identified 416 DARs, heavily skewed toward up-regulation in the summer group (346 up-regulated vs. 70 down-regulated). This disproportionate increase in specific accessible regions may reflect an adaptive, stress-induced epigenetic activation of regulatory elements governing cellular defense mechanisms or apoptotic pathways. Transcription factors binding to these accessible DNA regions are primary drivers of gene expression related to testicular development and spermatogenesis [43, 44]. Motif enrichment within the DARs highlighted several critical TFs (e.g., E-box, MITF, Usf2, NRF, Cbf1, and BIM) known to regulate apoptosis, spermatogenesis, and cellular stress response. For instance, MITF and Usf2 have previously been implicated in mediating thermotolerance and controlling Sertoli cell proliferation and differentiation [45–47]. E-box elements are fundamental regulators of circadian clock-controlled genes [48], suggesting that thermal stress might perturb diurnal rhythms essential for optimal testicular function. Furthermore, Cbf1 and BIM are established regulators of stress resistance [49, 50]. Collectively, these results indicate that a complex network of TFs is recruited to modulate testicular development and survival under summer heat stress.

GO and KEGG enrichment analyses of the genes associated with DARs further validated their biological relevance, highlighting processes such as reproduction, germ cell development, cell death, cell cycle, spermatid differentiation and development, as well as key signaling pathways including p53, TGF-beta, MAPK, Wnt and mTOR pathway. The p53 pathway is a central mediator of germ cell apoptosis under heat stress [51], and the upregulation of p53 may cause infertility [52]. TGF-beta and MAPK pathways regulate cell proliferation, differentiation, and apoptosis during spermatogenesis [53, 54]. Similarly, abnormalities in Wnt and mTOR signaling within Sertoli and germ cells are known to cause infertility [55, 56]. The enrichment of these pathways at the chromatin accessibility level suggests that broad epigenetic reprogramming is a primary driver of the transcriptional dysregulation observed in heat-stressed testes.

Integration of ATAC-seq and RNA-seq data identified 88 overlapping differentially expressed genes, and a weak positive correlation between chromatin accessibility and gene expression changes was found (r²=0.15). This modest correlation is consistent with previous studies indicating that chromatin accessibility alone is not always predictive of transcriptional output, as gene expression is also influenced by post-transcriptional regulation, transcription factor activity, and epigenetic modifications [57]. Nonetheless, the overlap includes key reproductive genes such as BAX, CDK4, UBB, and TNP2, which exhibited coordinated or opposite alterations in accessibility and expression. The pro-apoptotic gene BAX [58] demonstrated increased chromatin accessibility and transcript levels in summer, aligning with its known role in heat-induced germ cell apoptosis. Similarly, CDK4, a key cell cycle regulator [59], was upregulated, potentially reflecting aberrant cell cycle entry in response to stress. Conversely, TNP2, which is indispensable for spermatid differentiation and sperm motility [60, 61], providing a direct molecular mechanism for the impaired spermiogenesis observed during heat stress. Interestingly, polyubiquitin gene UBB, which is necessary for fertility [62], showed discordant changes with increased expression but decreased accessibility, highlighting the intricate complexity of regulatory mechanisms and the potential involvement of post-transcriptional or feedback regulation. The construction of transcription factor–target gene regulatory networks further supported the roles of these genes in heat stress response. For instance, BAX was directly linked to bHLH factors such as MITF and USF2, which are recognized modulators of apoptosis and stress adaptation [63–66]. These integrated interactions provide a framework for understanding how upstream regulators may orchestrate chromatin remodeling and gene expression changes during seasonal heat stress.

Conclusion

This study provides a comprehensive map of chromatin accessibility dynamics in rabbit testes subjected to natural seasonal heat stress. The integration of ATAC-seq and RNA-seq reveals key genes and pathways epigenetically regulated during heat stress, highlighting potential targets for mitigating heat-induced reproductive damage in male mammals. These findings contribute to a deeper understanding of the epigenetic basis of thermal stress responses and offer insights for improving animal breeding and reproductive management under changing climatic conditions. Further functional validation of candidate genes and transcription factors are required to investigate the regulation mechanisms under summer heat stress.

Abbreviations

DAR

Differentially accessible chromatin region

DEG

Differentially expressed gene

ATAC-seq

Assay for Transposase-Accessible Chromatin using sequencing

GO

Gene Ontology

HATs

Histone acetyltransferases

HDACs

Histone deacetylases

HK2

Hexokinase 2

IGV

Integrated Genomics Viewer

KEGG

Kyoto Encyclopedia of Genes and Genomes

PCA

Principal component analysis

SOX

SRY-related box

TF

Transcription factor

THI

Temperature and humidity index

TSS

Transcription start site

UTR

Untranslated region

Authors’ contributions

D.H. and H.D. conceived the study. H.Z., D.H., and Y.W. performed sample collection. H.D. and Y.W. conducted the data analysis and prepared figures and tables. H.D. wrote the manuscript. All authors reviewed the manuscript.

Funding

This work was supported by the earmarked fund for China Agriculture Research System (Grant No.: CARS-43-A-4).

Data availability

The ATAC-seq data generated by this study was submitted to the Sequence Read Archive (SRA) at the National Center for Biotechnology Information (NCBI) with accession number PRJNA1220349.

Declarations

Ethics approval and consent to participate

The present study was carried out in strict accordance with relevant guidelines and regulations by the Ministry of Agriculture of the People’s Republic of China. The verbal informed consent was obtained from the animal owners. All experimental protocols were approved by the Ethics Committee of Anhui Academy of Agricultural Sciences (protocol code AAAS2021-29).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.