Transcriptome Analysis of Right Ventricular Outow Tract Myocardium in Repaired TOF Patients with Pulmonary Regurgitation Revealed the Genetic Features of the Right Ventricle Dysfunction

Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart defect. Surgical repair can prolong the survival time of TOF patients, but the right ventricle (RV) function of repaired TOF (rTOF) patients will gradually deteriorate because of the residual pulmonary regurgitation (PR). Although pulmonary valve replacement (PVR) can delay the progression of RV dysfunction, it cannot avoid the ultimate RV failure. We conducted a comparative transcriptome analysis of RV outow track (RVOT) myocardium specimens collected from 6 rTOF patients with PR to dene the genetic features of RV dysfunction in this set of patients. We found signicant differential expression of a total of 30 genes among which 15 genes were up-regulated and the other 15 genes were down-regulated. There differentially expressed genes were signicantly enriched for the biological pathways ‘calcium signaling’, ‘cAMP signaling’, ‘Adrenergic signaling in cardiomyocytes’ and ‘Dopaminergic synapse’. This study presents the rst global comparative transcriptome proling of RV dysfunction using RNA-seq and reveals the different gene expression pattern in rTOF patients with PR. These ndings have potential translational value because they identify new candidate prognostic markers and targets for the treatment of RV dysfunction.


Background
Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart defect. From the genetic perspective, the etiology of TOF is multifactorial. Despite successful surgical repair, many patients have chronic volume loading caused by pulmonary regurgitation (PR) that leading to a 40% probability of right ventricular (RV) failure by the 3 rd decade of life [1,2] . In addition, some patients developed RV failure before 30 years of age.
Heart failure in repaired TOF (rTOF) is recognized as a signi cant complication. Its occurrence is strongly associated with adverse outcome. Although pulmonary valve replacement (PVR) can delay and improve RV failure, the optimal timing of PVR is still unknown because of the unnatural history of rTOF [3] . Therefore, it is necessary to better understand the molecular and cellular mechanisms which lead to progressive RV enlargement and dysfunction in rTOF patients.
Messenger RNA (mRNA) are coding RNA is a single-stranded nucleotides that carries genetic information to guide protein synthesis which are transcripted from one strand of DNA. Messenger RNA accounts for only 2%-5% of the total RNA of human cells, but it widest in variety and most active in metabolism. Based on its rapid expression, simple sequence composition, low immunogenicity and exible modi cation, great progress has been made in mRNA therapy which has been used in the treatment of rare genetic metabolic diseases [4][5][6] .
It was reported that interfering with the mRNA of PCSK9 can reduce the production of PCSK9 in the liver and the level of low-density lipoprotein cholesterol, so as to reduce the risk of coronary heart disease [7] . The research on mRNA therapy for heart disease is in progress [8] , so mining the mRNA pro le of RV in rTOF patients is helpful to nd potential therapeutic targets and provide necessary theoretical basis of mRNA therapy for RV dysfunction in rTOFpatients.
We hypothesized that mRNA pro les in RV cardiomyocyte would provide insight into the mechanism of disease progression which could then aid in risk strati cation and early prediction of RV failure in rTOF patients. For this purpose, we used RNA-seq to analyze the data pertaining to expression levels and performed high-resolution transcriptome analyses of the myocardial tissue of RV. The RNA-seq library was sequenced using the DNBSEQ platform, which uses the DNA nanoball technology.

Study population
Six rTOF patients (4 males, 2 females) undergoing open heart PVR surgery with CPB at the Department of Cardiovascular Surgery of the Sichuan University West China Hospital were enrolled in the study from December 2019 to January 2020. The rTOF patients were divided into two groups, according to the time interval from initial surgical TOF repair to PVR, PVR earlier (PVRe, n=3) and PVR later (PVRl, n=3). PVRe indicated the time interval less than 15 years, PVRl meant the time interval more than 15 years.
The study and protocol were reviewed and approved by Bioethics Committee of Sichuan University West China Hospital. The study was carried out in accordance with the approved guidelines. Written informed consents have been given from all patients and their legal guardian. All procedures in this study were performed in compliance with Helsinki Declaration and national laws.
Tissue sample collection RV myocardial tissue samples were obtained from rTOF patients who underwent PVR. The samples from RV were the redundant myocardial at RVOT. Experimental subject enrollment criteria included: rTOF patients receive open heart PVR procedure for the rst time. RV myocardial tissues were harvested from six patients with PR after TOF repair. Immediately following surgical resection the RV myocardial was immersed in liquid nitrogen and then stored at -80°C for RNA extraction.

RNA-seq library preparation and sequencing
We sequenced 6 samples on DNBSEQ Platform in total and generated about 11.18 GB per sample. Sequence reads by quality lters are analyzed in two ways at the level of tran isomers: HISAT [9] and Bowtie2 [10] followed by genes expression quanti cation using a software package called RSEM12.

Enrichment analysis of DEGs
We detect DEGs with DEG-seq as requested between PRVe and PRVl. We perform hierarchical clustering for DEGs using heatmap. With the GO annotation result we perform GO functional enrichment using phyper, a function of R. With the KEGG annotation result we perform pathway functional enrichment using phyper, a function of R. The p value <0.05 means signi cant enriched.

Results
Between December 2019 and January 2021, RVOT myocardium from 6 PR patients were collected and their RNA extractedfor RNA-seq analysis. The experimental ow of RNA sequencing is shown in the Figure 1A.
Clinical characteristics of the 6 patients are described in Table 1.
In our project, we sequenced 6 samples on DNBSEQ Platform in total and generated about 11.18 GB per sample. The average clean reads ratio is 95.7%. The Q20 (error probability of 0.01) of the clean nucleotides of all samples was more than 95.88%, with an average of 96.03%. (Table 2).

Differential gene expression pro ling in PVRe and PVRl
Global gene expression levels of PVRe (n=3) and PVRl (n=3) demonstrated signi cant correlation ( Figure 1B). 203562 genes were identi ed in which 202938 of them are known genes and 624 of them are novel genes. 432117 novel transcripts were identi ed in which 12,672 of them are previously unknown splicing event for known genes, 4826 of them are novel coding transcripts without any known features, and the remaining 318593 are long noncoding RNA.

Transcriptome differences among PVRe and PVRl
By analyzing the RVOT myocardium specimen between PVRe and PVRl, 30 genes were differentially expressed and annotated, 15 DEGs (50% of all DEGs) were up-regulated and 15 DEGs (50% of all DEGs) were downregulated. Volcano plot is shown in Figure 1C. The DEGs heat map for PVRe and PVRl is shown in Figure 2A. The ten most up-and down-regulated DEGs between PVRe and PVRl are listed in Tables 3.

GO classi cations and KEGG pathway identi cation of DEGs
With DEGs, we perform GO classi cation. GO has three ontologies: molecular biological function, cellular component and biological process. For the DEGs, the samples in PVRl when compared with PVRe, generated a total of 30 unigenes that can be classi ed into 46 categories of the GO functional group. The GO classi cation results are shown as Figure 2B. Among the 46 GO classi cations, 20 biological processes were signi cantly enriched and showed in Figure 2C, the speci c GO terms and the DEGs which were involved in these biological processes are shown in Table 4. With DEGs, we perform KEGG pathway functional enrichment, the pathway functional enrichment results are shown as Figure 2D.

Discussion
The present study describes genome-wide differential transcriptome pro ling underpinning RV dysfunction in patients of rTOF using RNA-seq. Differential gene expression pro les were observed among rTOF patients, dividing them into two groups: PVRe and PVRl. Comparative functional analysis of the DEGs revealed the presence of shared genes and pathways between the two groups. The data obtained from the study provide new insight into the molecular basis of phenotype heterogeneity seen in RV dysfunction of rTOF patients.
Comprehensive genome-wide screening of the transcriptome has helped in understanding the mechanism of disease differentiation and possible diagnostic and treatment value. Herein we compared the transcriptome pro les among the PVRe and PVRl, this helped us to elucidate the pathology of RV failure. By analyzing the expression pro les of the PVRe and PVRl, we could understand the pathology of RV failure at the genome level, and could identify DEGs which may familiarize us with the mechanism of RV failure of rTOF.
The comparative analysis of transcriptome pro les among the PVRe and PVRl indicated that the pathology of RV failure is multifactorial. Many fundamental biological processes are involved . In addition, KEGG pathway enrichment analyses revealed some speci c pathways, including dopaminergic synapse (P =0.019), adrenergic signaling in cardiomyocytes (P =0.024), cAMP signaling pathway (P value=0.048) and calcium signaling pathway (P=0.039).
With regard to the calcium signaling pathway, previous studies have reported that calcium (Ca 2+ ) is the central element of excitation-contraction coupling in cardiomyocytes, but also impacts diverse signaling cascades and in uences the regulation of gene expression, referred to as excitation-transcription coupling, Ca 2+dependent pathways are described that modulate gene expression by signal transduction to transcriptional regulators [11,12] . This ndings suggesting that Ca 2+ signaling pathway is involved in myocardial remodeling which can result in the occurrence of heart failure in rTOF patients [13] .
Cyclic adenosine monophosphate (cAMP) is a small, hydrophilic molecule, which is an important intracellular second messenger molecule regulated in many physiological processes. The production of cAMP was regulated by a variety of biological stimulus and cAMP exhibited distinct effects on cardiac function and the development of heart failure. But when the cAMP signaling pathway was inhibited the heart function was protected in rat model of heart failure [14] . In this study the cAMP signaling pathway was signi cantly enriched in both groups which suggested that the activation of cAMP pathway in rTOF patients and that the DEGs involved in cAMP signaling pathway participated in the progression of RV failure.
The adrenergic signaling was enriched in RVOT cardiomyocytes in this subset of rTOF patients. Adrenergic signaling was activated by the β 1 and β 2 adrenergic receptor (AR) which are the main β-AR subtypes expressed in the human heart [15,16] . Under normal conditions, β-ARs and their signaling pathways modulate both the rate and force of myocardial contraction and relaxation, allowing individuals to respond appropriately to physiological stress or exercise. However, sustained activation of the β-AR signaling pathways in chronic heart failure can produce harmful biological effects to heart [17] . Furthermore, there are cross-talk between βadrenergic signaling pathways and other pathways which may contribute to the progression of chronic heart failure [18] . Thus, modulating the β-AR-mediated signaling pathways have been one of the most crucial targets for heart failure therapy [15,19] .
The last but not least, the role of dopaminergic synapses in the pathogenesis of nervous diseases and mental disorder has been deeply studied, but there are few studies about that in heart failure. Dopaminergic synapses were also signi cantly enriched in RVOT cardiomyocytes by transcriptome analysis. Although there is no research elucidating the relationship between dopaminergic synapses and heart failure, our research may provide new clues for the study of the mechanism of RV dysfunction in rTOF patients.

Conclusion
This study presents the rst global comparative transcriptome pro ling of RV dysfunction in rTOF patients using RNA-seq and demonstrates the different gene expression patterns of PVRe and PVRl. The coding genes regulating calcium signaling pathway were down-regulated in PVRl, while the coding genes regulating cAMP, adrenergic and dopaminergic synapse signaling pathway were up-regulated in PVRl. This Transcriptome study shows that the coding genes regulating calcium signaling pathway may play a role in facilitating the RV dysfunction. Understanding of genetic pathways leading to RV dysfunction is important, as it is a precursor of overt clinical RV dysfunction and will facilitate the development of new therapies that halt progressive RV dysfunction. This may provide a new target for the treatment of RV dysfunction in rTOF patients. PVRl: PVR later, means that the interval from the initial TOF repair to the PVR procedure is more than 15 years. Availability of data and materials The datasets generated and analyzed during the current study are available in the SRA public repository at NCBI (https://www.ncbi.nlm.nih.gov/Traces/study/?acc=PRJNA750316) and can be accessed to through SRA Series accession number PRJNA750316.
Ethics approval and consent to participate The study and protocol were reviewed and approved by Bioethics Committee of Sichuan University West China Hospital. The study was carried out in accordance with the approved guidelines. Written informed consents have been given from all patients and their legal guardian. All procedures in this study were performed in compliance with Helsinki Declaration and national laws.
Consent for publication Not applicable.
Competing interests The authors declared that they have no competing interests. Tables   Table 1