Molecular mechanisms underlying hematophagia revealed by comparative analyses of leech genomes

Abstract Background Leeches have been used in traditional Chinese medicine since prehistoric times to treat a spectrum of ailments, but very little is known about their physiological, genetic, and evolutionary characteristics. Findings We sequenced and assembled chromosome-level genomes of 3 leech species (bloodsucking Hirudo nipponia and Hirudinaria manillensis and nonbloodsucking Whitmania pigra). The dynamic population histories and genome-wide expression patterns of the 2 bloodsucking leech species were found to be similar. A combined analysis of the genomic and transcriptional data revealed that the bloodsucking leeches have a presumably enhanced auditory sense for prey location in relatively deep fresh water. The copy number of genes related to anticoagulation, analgesia, and anti-inflammation increased in the bloodsucking leeches, and their gene expressions responded dynamically to the bloodsucking process. Furthermore, the expanded FBN1 gene family may help in rapid body swelling of leeches after bloodsucking, and the expanded GLB3 gene family may be associated with long-term storage of prey blood in a leech's body. Conclusions The high-quality reference genomes and comprehensive datasets obtained in this study may facilitate innovations in the artificial culture and strain optimization of leeches.

Leeches have been used in traditional Chinese medicine since pre-historic times to treat a spectrum of ailments but very little is known about their physiological, genetic, and evolutionary standpoint. Here we sequenced and assembled chromosome-level genome assemblies of three leech species (bloodsucking Hirudo nipponia and Hirudinaria manillensis , and non-bloodsucking Whitmania pigra ) and both bloodsucking leeches have similar dynamic population history and genome-wide expression patterns compared to non-bloodsucking leech. Combined analysis of the genomic and transcriptional data revealed that bloodsucking leeches presumably enhanced auditory other than visual sense for prey location in relatively deep fresh water. As expected, the copy number of genes related to anticoagulation, analgesia, and anti-inflammation obviously increased in the bloodsucking leeches, and their dynamic gene expressions respond to the bloodsucking process. We also found that the expanded FBN1 gene family might help for leech body rapid swelling after bloodsucking, and the expanded GLB3 gene family was potentially used to store prey blood for a long time in the leech's body.

Introduction
Leeches are obligate blood-feeding arthropods distributed from tropical to subarctic regions around the globe. Hematophagous species, such as bats, ticks, and mosquitos are the most versatile vectors capable of transmitting a wide range of pathogens to humans, livestock, and wildlife, including protozoa, bacteria, nematodes, fungi, and viruses [1], whereas leeches have been found to transmit a few infectious diseases. Furthermore, leeches have been used in traditional Chinese medicine since prehistoric times to treat a spectrum of ailments. Most obviously, leeches secrete the most potent natural thrombin inhibitor hirudin [2] and exhibited a variety of fascinating behavioral and physiological characteristics that are of interest to evolutionary, biochemical, and pharmaceutical studies. Leeches have also developed persistent adaptive strategies and characteristics to perceive their environment during long-term evolution. Leeches continuously receive sensory information from their surroundings either by mechanical or visual sensation to locate and target their prey. Additionally, the sanguivorous behaviour of leeches is capable to reduce natural host reflexes (blood coagulation, pain and inflammation) during bloodsucking [3]. Meanwhile, the prey localization, sanguivorous behaviour and medicinal value of leeches, therefore, necessitate to explore the fundamental knowledge of leech genomes and genetic diversity which would undoubtedly open new avenues for research on leech biology, host interactions, and control strategies at the molecular level which have not yet been elucidated. Heretofore, leech research was primarily focused on strain optimization, artificial culturing, and the identification and development of therapeutic strategies however, well annotated genome or genetic data is still unavailable. Currently, a few leech species genomic data including one nonblood sucking leech (Helobdella robusta) and low coverage genome sequence data of two lineages Amynthas cortices have been published [3][4][5][6], but none of them reach to chromosome level. Here we provided three high-quality leech genomes and abundant transcriptomes which illustrated the gene expression dynamics of bloodsucking leeches including anticoagulation, analgesic, and anti-inflammation that would facilitate the understanding at the genetic level and could be crucial for drug candidate prospecting.

Genome assemblies
We sequenced with the Nanopore platform and performed genome assembly for three leech species, including Hirudo nipponia, Hirudinaria manillensis, and Whitmania pigra (Fig. 1A), which are ubiquitously used in the Chinese pharmacopeia.  Fig. 1). The assemblies presented larger scaffold N50 sizes and lower scaffold numbers, indicating higher continuity than previously reported genomes (Supplementary Table 1).
Furthermore, we aligned the short reads and transcriptome assemblies to the genome to assess the completeness of our genome assemblies and found that more than 98% of the reads and 95% of transcriptome data were mapped to the assemblies (Supplementary Table 2 Table 5 and Supplementary Fig. 3).

Population history of leeches
We used the pairwise sequentially Markovian coalescent (PSMC) analysis to infer changes in the effective population size (Ne) of the ancestral populations of leeches. The population of the non-bloodsucking leech W. pigra underwent two expansions while the populations of the two bloodsucking leeches including H. nipponia and H. manillensis experienced only one expansion (Fig. 1C). The different fluctuations of Ne might hint at different enviromental adaptations of the two type of leeches. Interestingly, the Ne of the bloodsucking leeches begins to decrease at the onset of Pleistocene (~2 Mya), which is characterized by repeated cycles of glaciations. The glaciations likely have reduced the contact between leeches and animal hosts, resulting in a decline in the size of bloodsucking leech populations.  Table 9). Besides, thousands of ncRNA genes and secreted genes were also identified in each of the three leech genomes (Supplementary Tables 10-13).

Gene annotation and gene family construction
Meanwhile, we constructed the gene family and performed a phylogenetic analysis for 14 species. W. pigra and H. nipponia shared a common ancestor about 50 million years ago, whereas H. manillensis diverged at an earlier date ( Fig. 2A). It implied that the blood-sucking behavior might have existed in the ancestors of leeches, but this behavior was lost in the lineage of W. pigra. Consistent with evolutionary relationships, the four leech species were found to share most of their gene families (Fig. 2B). Additionally, a total of 1,289, 925 and 719 expanded, and 927, 2,164 and 1,312 contracted gene families have also been identified for each of H. nipponia, H. manillensis and W. pigra genomes. The GO analysis depicted that the expanded gene families including the ATP-binding cassette transporter complex, transcription factor IIA complex and calcium ion binding functions were significantly enriched in both bloodsucking leech species (Fig. 2C).

Transcriptome dynamics
To explore the gene expression patterns of the three leeches, we sequenced and analyzed the transcriptome of 32 samples (three replicates for each sample) including different developmental stages, tissues, and a series of bloodsucking behavior at 5 different time points (Supplementary Fig. 5-9). We used DEseq2 to identify differentially expressed genes (DEGs) between bloodsucking leeches and nonbloodsucking leech, and found that the majority of DEGs of two bloodsucking leeches shared similar expression patterns (Fig. 3B). We further investigated the transcriptomic dynamics during the process of bloodsucking in H. manillensis. Mostly the DEGs respond quickly after bloodsucking and continuously changed in the following 60 minutes (Fig. 3C, F). After 24 hours, the expression patterns of these DEGs had virtually restored to that of the pre-bloodsucking group (Fig. 3C). Furthermore, the GO and KEGG pathway analyses identified that most of the DEGs were significantly enriched to calcium, indicating that calcium-related regulation might play an important part in bloodsucking behavior of leeches (Fig. 3D, E).

Genetic basis of the prey location for leeches
Leeches are efficient predators because they can utilize their mechanical and auditory systems to acquire information and locate their prey. The genetic diversity of specific phenotypic traits could impersonate a better prototype of biologically substantial diverse genes especially having a crucial role in the development of the auditory and visual systems in mammals and in leeches these were identified by homolog-based functional annotation (Fig. 4). In leeches, among hearing-related genes (Fig. 4A), SIX1 plays a crucial role for audio sensation and we found that there is a single copy of SIX1 gene in the non-bloodsucking leech, while two or four copies of SIX1 gene are present in the bloodsucking leech (Supplementary Table 14) and the expression pattern of SIX1 gene was generally higher in bloodsucking than non-bloodsucking leech ( Supplementary Fig. 10) which clearly indicated that bloodsucking leech might possess better auditory perception.
Similarly, various mechanisms go on in the eye for the detection of visual signals, with the coordinated involvement of genes and their related proteins or enzymes, and intriguingly, the gene PDE6D, encoding the delta subunit of rod-specific photoreceptor phosphodiesterase, existed in non-bloodsucking leech while lost in the bloodsucking leech (Supplementary Table 14). Thus, we speculated that bloodsucking leech potentially strengthened audition and weakened vision to hide in the relatively deep fresh water.

Genetic basis of the sanguivorous behavior of bloodsucking leeches
To avoid being detected by hosts during the bloodsucking process, leeches executed three key operations including inhibition of blood coagulation, suppression of inflammation and alleviating pain. We identified the related genes for each process ( Fig. 5) and anticipatedly found that the total copy number of these genes in bloodsucking leech was higher than non-bloodsucking leech (Supplementary Table  14).
We found that two hyaluronidase (LHYAL) copies reached their highest expressions in only 5 minutes during the bloodsucking process ( Supplementary Fig. 11). Leeches inhibited hemagglutination mainly via three ways: supressing the thrombin cascade (HIRM1, ANTA and PROS1), inhibiting platelet aggregation (DECO, MMP13, ADAMTS18, APY and HIRM1) and dilating vessels (NEP1 and HIRM1) (Fig. 5A). In this study, HIRM1 mainly express in the oral suckers of the three leech species and the expression of one HIRM1 copy was comparatively increased at the time point of 30 minutes ( Supplementary Fig 11), which was consistent with the time of physiological coagulation. As expected, the gene expressions of five antistasin (ANTA) copies and two PROS1 copies retained relatively higher during the process of bloodsucking (5-60 minutes) ( Supplementary Fig. 11). Besides, all gene copies of ADAMTS18 and NEP1 acquired their peak expression levels in 10 minutes, suggesting that they potentially played a pivotal role in the process of inhibiting hemagglutination.
Additionally, during the bloodsucking process leeches also produced agrin (AGRN), cystatin (CYT), neprilysin-1 (NEP1) and membrane metalloendopeptidase like 1 (MMEL1) (Fig. 5B), which may reduce pain sensitivity and help them to avoid being recognized by the host. All copies of AGRN and MMEL1 as well as CYT generally kept high expressions in the bloodsucking process ( Supplementary Fig. 11). Further analysis showed that the leeches also expressed various anti-inflammatory genes, including NEP1, MMEL1, CYT, eglin (ICIC), cystatin (CYT), LeukoCYTe elastase inhibitor (SERPINB1), Toll-like receptor 4 (TLR4) and lipoprotein receptor-related protein 1 (LRP1) (Fig. 5C). Surprisingly, the expression levels of most of TLR4 and LRP1 copies reduced quickly at the beginning of the bloodsucking process and increased after bloodsucking (24 hours) process ( Supplementary Fig. 11). It indicated that leeches always kept anti-inflammatory proteins on hand to swiftly release into the body of prey.
Moreover, we noticed that GLB3 and FBN1 were the most expanded genes associated with sanguivorous behaviour in the bloodsucking leech. Seven or eight copies of the GLB3 were tandemly arranged in the two bloodsucking leeches, while only one copy in W. pigra and no copy in H. robusta (Supplementary Table 13 and Supplementary  Fig. 12). Moreover, the expression levels of the GLB3 family increased in the bloodsucking process. Noticeably, three GLB3 copies displayed significant expression level changes after the bloodsucking process ( Supplementary Fig. 12). Presumably that could explain why leeches can store prey blood in their body for months. Twelve copies of FBN1 were detected in the two bloodsucking species, and only four or zero copies were found in non-bloodsucking leeches (Supplementary Table 15). In general, The FBN1 family continually increased their expressions during the bloodsucking process ( Supplementary Fig. 12), indicating that FBN1 might be associated with the adaptability of leech body swelling after bloodsucking.

Discussion
Precise non-redundant reference genomes with vindicated annotations are critical for functional as well as evolutionary analyses and indeed, it was remained a challenge to produce a highly accurate chromosome-level assembly, particularly for leeches' chromosomes. Although leeches have been used to treat diverse ailments since ancient times, most of our information about them is based on psychometrics. A comprehensive catalog of their genome and gene expression pattern is fundamental to understanding the genetic basis of the behavior and will be crucial for drug candidate prospecting. Though the genome of medicinal leech has been sequenced in several other studies, the assembling results rest on fragmental level [3][4][5][6]. But, in this study, we developed three chromosome-level genome assemblies of the H. nipponia, H. manillensis and W. pigra by integrating short-read sequencing, Nanopore sequencing and Hi-C technology.
Leeches are efficient predators owning to their specialized predation adaptation, with acute senses such as hearing vision and chemosensation. Leeches trace and locate their prey via mechanical and visual cues from water waves on the basis of S cells [8].
Ethological experiments have shown that leeches can quickly identify and locate the source of sound by analyzing the distribution of water waves [8]. Among hearingrelated genes, SIX1 mediated the relative numbers of sensory hair cells and statoacoustic ganglion neurons [9]. Overexpression of the SIX1 gene could result in more hair cells [9]. In our study higher expression of the SIX1 gene in bloodsucking leech than non-bloodsucking leech clearly indicated that bloodsucking leech might possess better auditory perception. Furthermore, genes that encode opsin had very early origins and were recruited repeatedly during eye evolution [10]. The opsin family can be divided into seven subfamilies, and rhodopsin and Gq-coupled opsin/melanopsin are the most abundant proteins in rod cells [11]. Phototransduction is initiated when rhodopsin absorbs photons and triggers the exchange of GDP for GTP on the G-protein, which leads to an increase in cGMP hydrolysis by the phosphodiesterase (PDE) complex [12] and surprisingly, the gene PDE6D, which encodes the delta subunit of rod-specific photoreceptor phosphodiesterase, was present in non-bloodsucking leeches but not in bloodsucking leeches. We demonstrated that bloodsucking leech possibly prefers enhancing audition to hide in the relatively deep fresh water for prey.
In many species of invertebrates [13] or vertebrates [14], the choice of feed with low risk seems to be evaluated as a cost-benefit analysis influenced by hunger cues which face immediate risks including nociception that led to be identified by the host. To prevent detection by hosts throughout the bloodsucking process, leeches performed three crucial operations: inhibition of blood coagulation, suppression of inflammation, and pain relief. Hyaluronidase (LHYAL) boosts the diffusion and penetration of bioactive substances into tissues, and it can be used to ameliorate various complications associated with hyaluronic acid [15]. Hirudin (HIRM1) not only prevents fibrinogen clotting but also hinders other thrombin-catalyzed haemostatic reactions and activation of thrombin-induced platelets [16]. Additionally, hirudin can dissolve clots that have already formed by promoting the release of T-PA [17], so it may help in the clearance of thrombus. Apart from hirudin, antistasin (ANTA) can inhibit the function of coagulation factor Xa [18], and protein S (PROS1) blocks anticoagulant protease coenzyme C and factor VIII [19]. Moreover, throughout the bloodsucking process, leeches could also activate the anti-inflammatory proteins that might lower pain sensitivity and avoid being detected by the host. Similarly, for bloodsucking leeches the genes related to sanguivorous behaviour such as GLB3 and FBN1 are crucial such as GLB3 which is associated with oxygen binding and carrier, haeme and iron ion binding, is involved in the formation of the haemoglobin complex [20] whereas FBN1 that is a major structural component of microfibrils, and was found to be the largest influential factor for a height-associated variation in a human population [21]. In this study, we found that the copy number of genes related to sanguivorous behaviours in bloodsucking leech is higher than in non-bloodsucking leech. Furthermore, the dynamic gene expression of these genes responded to the bloodsucking process.
Overall, we have provided three leech genomes with optimal assemblies and raised some profoundly interesting questions on the environmental perception and sanguivorous behaviours of leeches. The chromosome-level reference genomes and underlying genetic mechanisms possibly provide insights into the genetic basis of leeches to the bloodsucking lifestyle. The comprehensive genomic and transcriptomic datasets may serve as a powerful platform to facilitate innovations in the artificial culture and strain optimization of leeches, identification of novel bioactive compounds, and candidate drug prospecting.

DNA isolation, Nanopore library preparation and sequencing
Three leech species, namely, Hirudo nipponia, Hirudinaria manillensis and Whitmania pigra, were obtained from the bank of Changjiang river, and their intestinal tracts were removed and washed with saline solution. The genomic DNA was collected using the DNeasy Blood & Tissue Kit (Qiagen, Wroclaw, Poland). The DNA quality was assessed, a long-read library was constructed (insert size, 20 kb), and Nanopore platform was used to perform long-read sequencing. Hi-C was performed using the following protocol: The leech tissues were fixed in 1% formaldehyde solution. Nuclear chromatin was obtained from the fixed tissue and digested using HindIII (New England Biolabs, NEB, USA). The overhangs were blunted with bio-14-dCTP (Invitrogen, California, USA) and Klenow enzyme (NEB). After dilution and re-ligation using T4 DNA ligase (NEB), the genomic DNA was extracted and sheared to 350-500 bp with a Bioruptor (Diagenode, Belgium). Then, the biotin-labeled DNA fragments were enriched with streptavidin beads (Invitrogen).

Genome size estimation
The genome size was estimated using 17-mer analysis. First, short reads were mapped on the genomes of bacteria and leeches by using minimap2 (v2.17-r941) [22]. The reads that aligned best on the bacterial genomes were filtered. Fastp (v0.20.0) [23] was used to filter the low-quality reads. Jellyfish (v2.3.0) [24] was used to divide the short reads into 17-mers and calculate 17-mer frequency. The 17-mer distributions of the three leeches generated using GenomeScope [25] followed Poisson distribution. The genome sizes were estimated by dividing the total number of 17-mers by the peak of the distribution and found to be 206 Mb, 155 Mb and 172 Mb for H. nipponia, H. manillensis and W. pigra, respectively.

Genome assembly and assessment
Nanopore long reads (~43 Gb for H. nipponia, ~48 Gb for H. manillensis and ~47 Gb for W. pigra) were used to establish de novo genome assemblies by using Flye (v2.6) [26]. Three rounds of correction were conducted using Racon (v1.4.7) [27] with the default parameters based on alignments of long reads by using minimap2 (v2.17-r941) [22]. The resulting assemblies were further polished using two rounds of Pilon (v1.23) [28]. Contigs that covered more than 50% of the bacterial genome sequences were filtered. Finally, LACHESIS [29] was used to hierarchically cluster the contigs and obtained the pseudo-chromosome assemblies. The completeness and accuracy of the final assemblies were estimated using both BUSCO (v5. 4

Repeat annotation
Both de novo and homology approaches were used to identify repetitive sequences in the leech genomes. RepeatModeler (v1.0.11) (http://www.repeatmasker.org/RepeatModeler/) was used to construct the de novo libraries. Then, RepeatMasker (http://www.repeatmasker.org/) was run for the three leech genomes by using the de novo libraries and a known repeat library (Repbase-20181026). A total of 25 -33% repeat content was obtained by combining the annotation results of the two approaches.

Gene and functional annotation
Three gene prediction methods based on de novo prediction, homologous genes, and transcriptomes were used to annotate protein-coding genes in the three leech genomes. Two de novo programs, Augustus (v3.0.3) [31] and SNAP (v2006-07-28) [32], were used to predict genes in the repeat-masked genome sequences. Transcriptome assemblies processed with PASA (r20140417) [33] were used to train gene model parameters for the two de novo programs. For homology-based prediction, protein sequences from C. teleta, H. robusta and E. andrei were aligned over the leech genomes by using tblastn (e-value < 10-5). GenblastA [34] was used to cluster adjacent high-scoring pairs from the same protein alignments, and GeneWise (version 2.4.1) [35] was used to identify accurate gene structures. After quality control and filtering, reads from all RNA libraries were mapped to the leech genomes by using hisat2 (v2.1.0) [36], and StringTie (v2.0.6) [37] was subsequently used to predict the gene models. All predicted genes from the three approaches were combined with EVM (r2012-06-25) [38] to generate high-confidence gene sets.
To obtain gene function annotations, SwissProt and TrEMBL [39] protein databases were searched using blastp (e-value<1e-05). The best blastp hits were used to assign homology-based gene functions. KOBAS (v3.0.3) [40] was used to search the KEGG [41] database for KO assignments. The functional classification of GO categories and InterPro entries was performed using InterProScan (version 5.39-77.0) [42].

Prediction of secreted proteins
For the secreted protein analysis, three methods, SignalP5.0 [46], Phobius [47] and SPOCTOPUS [48], were used. SignalP5.0 focuses on the prediction of signal peptides (SPs), and the other two algorithms can predict both transmembrane regions and SPs. The protein with at least one SP predicted using at least two out of the three methods was identified as a secreted protein.

Phylogenetic tree and divergence time
To perform phylogenetic analyses, peptide alignments for each single-copy family were obtained using MUSCLE [51] and concatenated to a supergene for each species. RAxML (v8.2.9) [52] with PROTGAMMAAUTO model and 100 bootstraps was used to construct the phylogenetic tree. The peptide alignments were converted to CDS sequences, which were subjected to mcmc tree in PAML (v4.9) [53] package to estimate divergence time.

Syntenic analysis
Mcscanx [54] with default parameters was used to detect syntenic genome regions among the three leeches, and jcvi was used to plot Figure 1B and show their syntenic relationships.

Transcriptome analysis
The total RNA was extracted from different leech parts at different developmental stages and before/after bloodsucking (Note that each sample included three biological replicates) by using TRIzol reagent (Invitrogen Corp., Carlsbad, CA). RNA purification was performed using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). Sequencing libraries were generated using the NEBNext Ultra RNA Library Prep Kit for Illumina (NEB, USA), according to the manufacturer's recommendations. The libraries were sequenced on an Illumina HiSeq 4000 platform, and 150 bp paired-end reads were generated. Each sample was trimmed using Trimmomatic (v.0.39) [ [36] was used to map the reads of each sample to the reference genome, and Samtools (v.1.9) [56] was used to sort and convert the SAM files to BAM. StringTie (v 2.0.6) [37] was then used to assemble and merge the transcripts of each sample. Gffcompare (v0.11.5) [57] was used to compare the merged transcripts with the reference annotation file in GTF, and StringTie (v 2.0.6) was used to estimate transcript abundances with the options '-e -B -p 20'. The abundance results were folders that ended with '.balltown', and prepDE.py was used to compare the folders. DESeq2 [58] was used for differentially expressed genes (DEGs) analysis with default parameters. To perform differential expression analysis using a genome model, the cDNA reads were mapped against the genome assembly by using HISAT2. HTSeq [59] was used to count the number of reads mapped against the annotated genes.

Competing interests
The authors declare no competing interests.

Data Availability
The genomic and transcriptomic Illumina data, Nanopore sequencing, and HiC data were uploaded at NCBI with BioProject number: PRJNA762643. The genomes and gene annotations of the three leeches were under figshare with the doi: 10.6084/m9.figshare.20400729.           Dear Editor, Our manuscript entitled "Molecular mechanisms underlying hematophagia revealed by comparative analyses of leech genomes", which we would like to be considered for publication in Gigascience.

Abbreviations
Leeches are aquatic predators that are widely distributed worldwide, and they display various fascinating behavioural and physiological characteristics that are of evolutionary, biochemical and pharmaceutical interest. Leeches have been used in the treatment of diverse ailments since ancient times in Greece, Rome, Arabia and China. Leeches secrete the most potent natural thrombin inhibitor, hirudin, and a few of bioactive proteins have been identified in its salivary gland help to eliminate microcirculation disorders, restore vascular permeability, remove hypoxia, decrease blood pressure and detoxify the organism via antioxidant pathway, which have been regarded as vital treatment of diverse human ailments.In this study, we have made exciting breakthroughs that reveal the diverse biological adaptations of leeches with respect to environmental perception and sanguivorous behaviours.
Briefly, we provided chromosome-level genome assemblies of three leech species (bloodsucking Hirudo nipponia and Hirudinaria manillensis, and non-bloodsucking Whitmania pigra) widely used as Chinese traditional medicine. we sequenced and analyzed the transcriptome of 32 samples (three replicates for each sample) including different developmental stages, tissues and a series of dynamic bloodsucking process at 5 time points. We found that the two bloodsucking leeches shared similar gene expression patterns and a common genetic basis of hematophagia. Specially, the fibrillin family (FBN1) and the globin family (GLB3) underwent significantly expansion and showed increased expression levels in bloodsucking leeches, which potentially provided adaptions for rapid swelling during the bloodsucking process and long-time storage of host blood, separately.
Our high-quality reference genomes and comprehensive catalogue may help in leech-derived candidate drug prospecting, especially for the treatment of cardiovascular and cerebrovascular diseases, thrombus, pain and other potential disorders.
We deeply appreciate your consideration of our manuscript, and expect for receiving comments from the reviewers. Please let us know if we can provide any additional information. We look forward to hearing from you soon.
Thank you and best regards.