Chromosome-level genome of the poultry shaft louse Menopon gallinae provides insight into the host-switching and adaptive evolution of parasitic lice

Abstract Background Lice (Psocodea: Phthiraptera) are one important group of parasites that infects birds and mammals. It is believed that the ancestor of parasitic lice originated on the ancient avian host, and ancient mammals acquired these parasites via host-switching from birds. Here we present the first chromosome-level genome of Menopon gallinae in Amblycera (earliest diverging lineage of parasitic lice). We explore the transition of louse host-switching from birds to mammals at the genomic level by identifying numerous idiosyncratic genomic variations. Results The assembled genome is 155 Mb in length, with a contig N50 of 27.42 Mb. Hi-C scaffolding assigned 97% of the bases to 5 chromosomes. The genome of M. gallinae retains a basal insect repertoire of 11,950 protein-coding genes. By comparing the genomes of lice to those of multiple representative insects in other orders, we discovered that gene families of digestion, detoxification, and immunity-related are generally conserved between bird lice and mammal lice, while mammal lice have undergone a significant reduction in genes related to chemosensory systems and temperature. This suggests that mammal lice have lost some of these genes through the adaption to environment and temperatures after host-switching. Furthermore, 7 genes related to hematophagy were positively selected in mammal lice, suggesting their involvement in the hematophagous behavior. Conclusions Our high-quality genome of M. gallinae provides a valuable resource for comparative genomic research in Phthiraptera and facilitates further studies on adaptive evolution of host-switching within parasitic lice.


Introduction
Lice (Insecta: Phthiraptera) are parasites that infest birds and mammals with more than 4,500 species of chewing lice (Amblycera, Ischnocera, Trichodectera and Rhynchophthirina) and 500 species of blood-feeding sucking lice (Anoplura) [1,2].Chewing lice feed on the feathers, sebaceous secretions, and skin of their avian and mammalian hosts [1], while sucking lice which parasitize only mammals have piercing-sucking mouthparts and feed exclusively on blood [3].These parasites entirely rely on the body of the host and they affix their eggs to hairs or feathers of the host [1,3].As an obligate parasite of domestic chickens (Gallus gallus), the poultry shaft louse Menopon gallinae is a main vector for chicken diseases.These lice live on the skin, penetrate within the skin, or even burrow into the air sacs or under the feathers of chickens.Infestation by these lice can lead to annoyance, decreased weight gain, reduced egg production, egg abandonment in brooding hens, and chick mortality [4].Additionally, they can cause high morbidity, which adversely affects the economic production of poultry [5].
To date, only two louse genomes have been published: the human body louse Pediculus humanus (Anoplura) and the pigeon wing louse Columbicola columbae (Ischnocera).
Compared with other insects, lice have a reduced number of protein-coding genes (PCGs), including fewer opsin genes, odorant receptors, and detoxification pathways [6,7].Our understanding of the genomic signatures of parasitism in Phthiraptera is limited largely to these two species.Whole-genome data of the chewing lice suborder Amblycera, which represents the earliest diverging group of lice, is still lacking.Recent studies suggested that parasitic lice have an avian ancestral host and the ancestor of Afrotheria mammals acquired these parasites via host-switching [8,9].After this host-switching from bird to mammals, parasitic lice have colonized other lineages of mammals through host-switching and codiversified with their host.Parasitic lice have specific morphological and behavioral adaptations for attachment and avoiding host defenses [1,2].In contrast to bird lice, mammal lice are morphologically adapted to live on their mammal hosts with tibial tarsal claws to attach to host hairs and highly derived mouthparts for feeding directly from host blood vessels [10].
During host-switching of lice, their genomes accumulate mutations, some of which may be directly linked to functional adaptations.Identifying such genomic feature and linking them to phenotypic differences is critical for deciphering the genomic drivers of species adaptability.Expansion or contraction of key gene families may facilitate the emergence of novel functions, leading to successful host-switching of lice.Therefore, it is necessary to reveal significant variations in the genome during the host-switching of parasitic lice from birds to mammals.
In this study, we presented a high-quality chromosome-level genome of M. gallinae (representing the earliest diverged lice Amblycera) using a combination of Illumina short-read sequencing, PacBio high-fidelity (HiFi) long-read sequencing, and Hi-C technology.
Combining the genome of human body louse P. humanus, the latest diverged lice (Anoplura), together with other various representative insect species, we performed comparative genomic analyses to evaluate the evolution of genes putatively involved in host-switching of parasitic lice from birds to mammals.These data would supply a useful genetic resource for future research of parasitic lice.

Samples collection
For genome sequencing of the poultry shaft louse Menopon gallinae, approximately 1600 individuals were collected from natural populations infesting chickens (Gallus gallus) in Chongqing, China.

DNA extraction, RNA extraction, library construction and sequencing
Genomic DNA used for the SMRTbell library preparation was extracted from about 1,000 adults with the Blood & Cell Culture DNA Midi Kit (Qiagen, Hilden, Germany).After assessing the quality of the isolated DNA, a ∼20-kb library was constructed using the SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences, California, USA).HiFi long clean reads produced by circular consensus sequencing (CCS) on the PacBio Sequel II platform were used for contig-level genome assembly.
For genome survey and assembly polishing, genomic DNA was extracted from 50 adults and an Illumina sequencing library was constructed according to the manufacturer's instructions (Illumina, California, USA).The library was then sequenced on the Illumina NovaSeq 6000 platform in paired-end 150-bp mode to generate approximately 50 Gb data.
For genome annotation, total RNA was extracted from 50 adults using the Tiangen RNA extraction kit (Beijing, China).After reverse transcription of mRNA into cDNA, another Illumina RNA-seq library was constructed and sequenced with the same parameters, generating approximately 6 Gb data.In addition, a PacBio Iso-Seq library was constructed using the SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences, California, USA) from 50 adults and sequenced on the PacBio Sequel II platform, generating approximately 60 Gb data.
To construct a chromosomal-level assembly of the genome, we constructed the Hi-C library.In brief, 600 adults of M. gallinae were immersed in 2% formaldehyde for cross-linking of cellular protein.The purified nuclei were digested with 100 units of DpnII enzyme.Then Hi-C samples were extracted by biotin labelling, flat end ligation, DNA purification and random shearing of DNA into 300-600 bp fragments.Finally, the Hi-C libraries were quantified and sequenced using the Illumina NovaSeq platform with paired-end 150-bp reads.

Genome assembly and evaluation
The genome of M. gallinae was sequenced using the Illumina HiSeq 2000 and PacBio Sequel II platform.The Hi-C sequencing technology was used to assist the assembly of the genome.The genome size, heterozygosity, and duplication of the genome were estimated by the K-mer method.Specifically, 17-base oligonucleotide K-mers were counted using JELLYFISH version 2.1.3[11].The genome features were then evaluated using GenomeScope version 2.0 [12].The PacBio CCS reads were processed using WTDBG2 version 2.5 [13] to generate a draft assembly.The purge_dups version 1.2.6 [14] was used to remove potential haplotypic duplications and contig overlaps.Clean reads sequenced from the Hi-C library were aligned to the contig-level genome with an end-to-end algorithm implemented in BWA-MEM version 0.7.17 [15].Juicer version 1.6 [16] and 3D-DNA version 180419 [17] were used to assemble the scaffolds into a chromosome-level genome.The chromosome-level genome was reviewed using Juicebox version 1.11.08 (https://github.com/aidenlab/Juicebox).The completeness of the genome was assessed using BUSCO version 3.0.2[18].
RepeatModeler version 2.0.4 [21] was used to construct a de novo repeat library.LTR FINDER version 1.0726 [22] and LTR retriever version 2.9.028 [23] were used to identify LTR retrotransposons.Tandem Repeats Finder (TRF) version 4.09.1 [24] was used to annotate tandem repeats.
Additionally, we utilized the bam2hints program in AUGUSTUS version 3.2.3[36] to transfer the sorted and mapped bam file of RNA-seq data into a hints file.These trained gene sets and hint files were then combined as inputs for AUGUSTUS version 3.2.3[36] to predict coding genes from the assembled genome.Finally, the high-confidence gene set was generated by merging ab initio, RNA-seq-based, and homology-based genes using MAKER version 2.31.10 [37].

Identification of orthologous genes and inference of phylogenetic relationships
To identify the gene families, we utilized the protein sequences of ten insects and selected two dipteran insects as an outgroup.OrthoFinder version 2.5.4 [38] was used to find orthologues and orthogroups.The phylogenetic tree was inferred using single copy orthologues in each species.Sequence alignment was performed using MAFFT version 7.520 [39], and the resulting alignment was trimmed with the option "automated1" using trimAl version 1.4.rev15[40].We estimated the phylogenetic tree using the concatenated sequences of aligned proteins in RAxML version 8.2.12 [41] with the option "-m PROTGAMMAJTTnb 100".
Based on the results of OrthoFinder, gene family clusters were divided into five categories, (1) single-copy genes in all species, (2) multiple-copy genes in at least one species, (3) species-specific genes (genes absent in other N-1 species), and (4) other genes.

Gene family expansion, contraction and annotation
The divergence time was estimated using r8s version 1.81 [42]  To analyze gene family, we downloaded the proteins of corresponding gene families of P.
humanus, D. melanogaster, and Acyrthosiphon pisum from the NCBI database.The hidden Markov models (HMMs) were obtained from the Pfam database.The HMMs and proteins were fed as input for HMMER version 3.0 [45] and BLAST version 2.12.0 [46] to search for homology genes.The annotated genes were further manually filtered based on gene length and the presence of conserved domains to finally obtain candidate genes for each gene family.

Positive selection and dN/dS ratios analysis
We used MAFFT version 7.520 [39] to align the protein sequences, and subsequently converted the multiple protein sequence alignment and corresponding coding sequences (CDS) into a codon alignment using Pal2Nal version 14 [49].To calculate dN/dS ratios across pairwise alignments of each gene pair between bird lice M. gallinae and mammal lice P. humanus, we employed the Yn00 algorithm in PAML version 4.9 [50].To identify potential positively selected genes (PSGs) in M. gallinae and P.humanus, we utilized the branch-site model of CodeML in PAML version 4.9 [50] with single-copy orthologs of ten insect species.
Specifically, we set M. gallinae/P.humanus as the foreground branch, and the remaining species as background branches.We chose p < 0.05 as the significance threshold after FDR correction to identify a particular orthogroup as positively selected.

Genome sequencing and assembly
We assembled a high-quality chromosome-level genome of Menopon gallinae by using a combination of PacBio long reads (28.51 Gb, 184-fold), Illumina short reads (50.79 Gb, 328.32-fold) and Hi-C reads (20.86 Gb, 134-fold).The genome size was estimated to be 145 Mb with a heterozygosity rate of 0.363% by calculating the frequency with 17 k-mer analysis (Fig. 1a).At the contig level, we generated a final genome assembly of 155 Mb, consisting of 100 contigs with an N50 of 27.42 Mb (Table 1).This final genome size of M. gallinae (155 Mb) is comparable to our preliminary estimation (145 Mb).However, it is larger than the previously published genome size of mammal lice P. humanus (108 Mb) [7] and smaller than that of the booklice Liposcelis brunnea [51] genome (174 Mb).Our M. gallinae genome also possessed a remarkably higher contig N50 (27.42 Mb) and GC content (41%) compared to the P. humanus genome (contig N50 = 34 Kb; GC content = 28%).Five complete chromosomes were obtained in our assembly (N50 = 27.95Mb), consisting of 97% of the whole genome.
Chromosome lengths ranges from 25.47 Mb to 37.36 Mb (Figs. 1b and 1c).Moreover, BUSCO evaluation showed high completeness and accuracy of the genome assembly, with 97.8% of genes being successfully identified, including 97.4% single-copy genes and 0.4% duplicated genes.These results justify a high-quality M. gallinae genome that could be used in the downstream analyses.

Genome annotation
To predict bona fide protein-coding genes in the M. gallinae genome, we employed three different approaches: de novo prediction, homologous gene prediction, and RNA-seq-based prediction (see Materials and Methods for details).In total, we predicted 11,950 PCGs that were supported by three approaches.The number of PCGs in M. gallinae was comparable to that observed in mammal lice P. humanus (10,773 PCGs) [7], but lower than the number reported in the genome of L. brunnea (15,543 PCGs) [51], a booklice which is the outgroup of the two parasitic lice M. gallinae and P. humanus.This result indicated that the coding genes in two parasitic lice might have experienced an ancestral loss due to their limited habitats and simple dietary regimes compared to the wide-spread booklice L. brunnea.
Annotation of the 11,950 PCGs in M. gallinae revealed an average of 8.26 exons and 7.26 introns per gene.The average length of the mRNA transcripts was 2,763.21bp, while the average length of the coding sequence (CDS) was 1,815.98 bp (Table S1).Functional annotation revealed that 10,664 (89.24%), 9,178 (76.80%), and 8,999 (75.31%) genes matched with proteins recorded in databases NR, SwissProt, and Pfam, respectively.
Repeat sequences (transposable elements, TEs) only made up 4.1% of the M. gallinae genome.This fraction of repeats is considerably lower than those in P. humanus (7.3%) and L. brunnea (15.9%).For specific types of TEs, 0.01% of the M. gallinae genome is short interspersed nuclear elements (SINEs), 0.27% is long interspersed nuclear elements (LINEs), 0.58% is long terminal repeats (LTRs), 0.56% is DNA transposon, and the other 2.30% is tandem repeats (TRs) (Table S3).TRs are much more abundant than any other types of TEs in M. gallinae.

Ortholog identification and phylogenetic inference
To get an accurate landscape of the evolutionary gains and losses of functional genes in two parasitic lice and understand how these evolutionary dynamics is related to the phenotypic innovations of the species, we looked for orthologous genes in M. gallinae, P.
humanus, and other eight representative insect species.We identified 16,518 gene family clusters using OrthoFinder (Materials and Methods).The gene family clusters were divided into four categories, single-copy genes, multiple-copy genes, species-specific genes (unique genes), and other genes.A phylogenetic tree generated using single-copy orthologous genes showed that all species of Paraneoptera (Psocodea, Hemiptera, and Thysanoptera) formed a clade.A total of 137,947 genes belonging to 16,518 unique gene families (orthogroups) were identified.Particularly, there were 115 orthogroups specific to bird lice, while only 23 orthogroups were specific to mammal lice (Fig. 2; Table S4).

Expansion, contraction and positively selected of gene families
We used CAFÉ version 4.2.1 [43] to study the expansions and contractions of gene families during the evolution of parasitic lice.Compared to the common ancestor of the parasitic lice, we found 511 expanded and 1,866 contracted gene families in bird lice and 381 expanded and 1,728 contracted gene families in mammal lice.Similarly, 922 expanded and 902 contracted gene families were founded in booklice compared to the ancestral node of booklice and parasitic lice (Fig. 2).Enrichment analysis of GO and KEGG revealed that expanded gene families in bird lice are enriched in obsolete drug binding, drug metabolismother enzymes and Hippo signaling pathway (Fig. 3a; Tables S5 and S6), while contracted gene families are enriched in chemosensory behavior, cytochrome P450, digestive system and immune system (Fig. 3b; Tables S7 and S8).Correspondingly, for mammal lice, while the expanded gene families are enriched in negative regulation of neurogenesis, positive regulation of nitrogen compound metabolic process (Fig. 3c; Table S9), those contracted gene families are enriched in G protein-coupled receptor activity, obsolete drug binding and signaling receptor activity (Fig. 3d; Table S10).The fact that the contracted gene families of chemosensory-related pathways in two parasitic lice implies that these genomic features may be relevant to their parasite behavior after split from their common ancestor.Accordingly, we re-estimated the expanded and contracted gene families in the ancestor of two parasitic lice and found that the contracted genes are enriched in response to xenobiotic stimulus, response to insecticide, defense response, cytochrome P450 and digestive system (Table S11, Table S12).Next, our analysis on booklice revealed that the expanded gene families are enriched in response to xenobiotic stimulus, lectins and drug metabolism-cytochrome P450 (Table S13, Table S14).
To understand whether the species-specific genes among the three lice contribute to their distinct behavior, we looked for the gene families present in one lice species but absent in all other nine insect species we used.We uncovered 115, 23, and 198 species-specific gene families in bird lice, mammal lice, and booklice, respectively (Table S4).Gene enrichment analysis showed that bird lice-specific gene families are enriched in pathways of response to temperature stimulus, response to heat, and response to xenobiotic stimulus (Fig. 3e; Tables S15 and S16).Mammal lice-specific gene families are enriched in innate immune response and epidermal growth factor receptor signaling pathway (Fig. 3f; Table S17).Booklicespecific gene families are enriched in cellular response to oxygen-containing compound, digestive system, and cytochrome P450 (Table S18 and S19).This result agrees with the a previous finding that P450 genes were expanded in booklice [51].
Positive selection is an important source of evolutionary innovation and one of the major forces driving species divergence.We identify the positively selected genes in single-copy genes from these species for branch-site model analysis by Maximum Likelihood (PAML) (Materials and Methods).We found that 135 and 201 genes were positively selected in bird and mammal lice respectively (likelihood ratio test, P<0.05).Seven hematophagy-related genes in mammal lice were positively selected, such as heme-related genes (coproporphyrinogen III oxidase), iron-related genes (nuclear hormone receptor, NADH ubiquinone 75kD subunit, peroxinectin like, mitoferrin, transferrin2) and salivary IP5P (Table S20), which may be due to the unique hematophagous behavior of mammal lice.Based on the above results of gene expansion/contraction and positively selected genes, we will focus on hematophagy, digestion, chemosensory, temperature, immune and detoxification gene families in the following analyses.

Hematophagy-related genes
We compared hematophagy-related genes to observe the signature change of feeding habit during host-switching.We found that proteins involved in iron/heme binding, transport and metabolism, oxidative stress, urea cycle enzymes, and other hematophagy-related genes are generally conserved in two lice genomes (Fig. 4; Table S21).Among the hematophagyrelated genes in mammal lice, five were absent in bird lice: heme-related genes (ferrochelatase), iron-related genes (neverland, ND-pdsw), oxidative stress-related genes (phosphoserine phosphatase), and urea cycle enzymes (arginase).Ferrochelatase is a crucial enzyme involved in heme synthesis in insects [52].ND-pdsw plays a key role in oxidative phosphorylation and is overrepresented in blood feeding insects [53].

Major digestive enzyme analysis
As important as hematophagy-related genes, digestive enzyme are also essential for diet of lice.The digestive enzyme of parasitic arthropods are mainly involved in the digestion of host blood and other ingested proteins present in the skin [54][55][56].Feather feeding in bird lice and hematophagy in mammal lice may result in different adaptations and alterations in the composition of digestive enzymes in two lice.We found bird lice (168 genes) have a similar amount of digestive enzyme to mammal lice (152 genes) (Fig. S1; Table S21).Compared to other insects or even the booklice (236 genes), digestive enzymes of two parasitic lice were substantially contracted (Fig. S1).

Genes associated with chemosensory
Genes involved in chemosensory systems, especially the OR, GR and IR subfamilies, play a critical role in feeding, mating, and predator avoidance of insects [57,58].We collected six subfamilies of chemosensory-related genes and the two parasitic lice have remarkably reduced chemosensory-related genes.Mammal lice have fewer chemosensory-related genes (totally 63 genes) than bird lice (106 genes) (Fig. 5a; Table S21).Bird lice have 21 GR, 16 OR and 36 IR genes, while mammal lice have only four GR, nine OR and 27 IR genes (Figs. 5a, S2a and S2b).Notably, GR genes showed the most reduction in mammal lice.GR genes primarily mediate gustation, specifically detecting sweet and bitter tastants, as well as to sense carbon dioxide (CO2) [59][60][61][62].Previous study has reported that some GR genes (such as sugar receptor genes and CO2 receptor genes) were absent in mammal lice [7].Here we confirmed that the bird lice genome encodes two sugar receptor and two CO2 receptor.We also found a fructose receptor gene in the bird lice genome which is absent in mammal lice (Fig. S2a).The remaining members of GR subfamily all belong to GR28 genes: bird lice have 16 GR28 genes while mammal lice only have four GR28 genes.GR28 are temperature sensors that can help identify hosts dependent on warmth, the strategy of which has been widely used by parasites such as tsetse flies and mosquitoes [63].

Temperature-related genes
As molecular chaperones, heat shock proteins (Hsps) play important roles in helping insects cope with various ambient stresses, such as extreme temperatures, oxidation, heavy metals, and other abiotic factors [64,65].We have collected five Hsp subfamilies and found fewer Hsp genes in mammal lice (61 genes) than in bird lice (83 genes).The main difference is that mammal lice have ten Hsp60 and nine Hsp70 genes, while bird lice have 20 Hsp60 and 19 Hsp70 genes (Figs. 5a and 6a; Table S21).Hsp40 and Hsp90 also show higher copy numbers in bird lice than in mammal lice while only sHSP is more abundant in mammal lice (seven genes) than in bird lice (three genes) (Fig. 5a; Table S21).

Immunity-related genes
In defense against pathogens, insects rely mainly on their innate immune system [66,67].
Ninety-two immunity-related genes were identified from the mammal lice genome while 90 were identified from the bird lice genome (Fig. 5b; Table S21).All components of the Toll, JAK/STAT, and JNK pathways existed in both lice, however, several gene families involved in the humoral immune system were considerably diminished or missing in two lice genomes.
In case of pathogen recognition-related genes, bird lice have three PGRP while mammal lice have only one, and the GNPB protein was absent in two lice.Several components of the Imd pathway (Imd and its adaptor protein FADD) are absent in bird lice, as reported in mammal lice [68].A similar result was observed in the pea aphid A. pisum and kissing bug Rhodnius prolixus, where a more extensive loss of the Imd pathway genes purportedly allowed the development of its obligate endosymbiont [69].Furthermore, hemocytin gene was found in mammal lice genome which is absent in bird lice.

Detoxification gene family analysis
Detoxification genes are involved in the metabolic detoxification of xenobiotics, such as plant allelochemicals and synthetic insecticides [70].The number of detoxification genes was similar in two lice as five detoxification families (P450s, GSTs, ESTs, UGTs, and ABC) of bird lice included 43, 20, 13, 9, and 45 genes, whereas mammal lice had 37, 18, 11, 4, and 38 genes for each family (Fig. S3a; Table S22).When mapping 43 P450 genes to the chromosomes of bird lice, one gene cluster with ten CYP3 genes was found on chromosome 4 (Figs.6b and S3b), consistent with the expansion of ten CYP3 genes (Table S7).This expansion of CYP3 has been associated with pesticide resistance and xenobiotic metabolism, as studied in several dipteran and lepidopteran insects [71,72].For ABC transporter, only ABCH subfamilies is slightly more abundant in two lice (6 genes) than in D. melanogaster (3 genes) (Figs.6c and S3a; Table S22), while ABCG5 gene was positively selected in mammal lice (Table S20).

dN/dS ratios analysis
We calculated the dN/dS ratios (between mammal and bird lice) of hematophagy, chemosensory, detoxification, temperature, digestion, and immunity-related genes with those of other gene families.Higher dN/dS represents faster evolution rate.We found that chemosensory genes exhibit slightly higher dN/dS ratios than other gene categories (p<0.001,t-test) (Fig. 7a), suggesting the rapid evolution of chemosensory genes in lice.Among chemosensory genes, GR genes have higher dN/dS ratios than other subfamilies (Fig. 7b), suggesting that even among the fast-evolving chemosensory genes, the GR subfamily is under relaxed selective constraints.The dN/dS ratios of hematophagy-related genes are significantly lower than those of chemosensory genes and other genes (p<0.001 in both comparisons, t-test) but show no significant difference with the remaining groups (Fig. 7a).Among these hematophagy-related genes, iron-related genes have the lowest dN/dS ratios than other genes (Fig. 7b), indicating that conservation is more evident across hematophagy-related genes.

Successful sample collection and the chromosome-level genome assembly of Menopon gallinae
In this study, we present the first chromosome-level genome of Amblycera.Genome assembly is typically challenged by high heterozygosity and replication, particularly in small insects that require the extraction of DNA from multiple individuals to construct sequencing libraries [6,7].Many lice are difficult to obtain due to their low abundance on a single host (usually < 10 individuals) [73].In this study, we collected a large number of adults M.
gallinae from a chicken farm in Chongqing and used a long-read sequencing strategy (PacBio HiFi and Hi-C) to assemble its genome.This strategy has been shown to produce high integrity and continuity in genome assembly [74][75][76], making it suitable for high-quality de novo assembly of abundant small parasitic lice genomes.The genome of M. gallinae shows a low level of heterozygosity while 1,000 adult samples were used for sequencing in our research.The homogeneous environment where lice living and inbreeding in, may leading to this relatively low degree of heterozygosity.At the contig-level, the contig N50 of M. gallinae is higher in compared with the P. humanus, and the genome completeness estimated using BUSCO is also better (Table 1).These results indicate a well-assembled genome with a high degree of completeness and accuracy.

Hematophagy, digestion, detoxification and immunity-related gene families are conserved across lice genomes
In general, host-switching between birds and mammals occurred very early in the diversification of lice, and the ancestor of Afrotheria (elephants, elephant shrews and hyraxes) acquired these parasites via host-switching from an ancient avian host [8,9].After hostswitching, many lice change specific morphological characteristics and behaviors, corresponding to adaptations to different hosts.Bird lice feed on keratin tissues such as feathers typically.Most keratin possess complex protein secondary structure, making them hard to be digested.In contrast, mammal lice feed on blood, which possess relatively simple defense chemistry [77][78][79].However, although the host types and feeding habits changed during the host-switching process, our results revealed a general similarity in the number of digestive enzymes, detoxifying enzymes, and immunity-related genes in both bird and mammal lice.
For mammal lice, sucking from blood vessels could provide nutritional benefits, meanwhile also lead to potential harms caused by pro-oxidant molecules such as heme and iron.The mammal lice may have evolved adaptations to protect themselves from iron and heme-related damage, as observed in blood-feeding arthropods [80,81].Hematophagy-related genes are among the slowest-evolving gene categories in sequence divergence, suggesting that they are highly conserved (Fig. 7a).However, seven hematophagy-related genes were positively selected in mammal lice (Table S20).These genes were crucial to heme synthesis and iron transport [52,53,82].For example, coproporphyrinogen III oxidase is an enzyme crucial to the biosynthesis of heme necessary for cellular respiration and protein function [83] and mitoferrin protein transports iron into mitochondria for cellular processes like heme production and ATP synthesis [84].We speculated that these hematophagy-related genes may be associated with their adaptation to a blood-sucking lifestyle after host-switching from birds to mammals.

Temperature-and chemosensory-related gene families are crucial for host-switching of lice
Environmental stressors, such as high/low temperatures, can easily affect the survival, growth, and development of insects [85,86].Insects use various mechanisms to tolerate high temperatures, but these come at a cost to their energy and fitness levels.This can lead to reduced survival, fecundity, body size, and mating success [87,88].Parasitic lice can only survive for a limited time when away from host, thus they are highly sensitive to changes in body surface temperature of host [1,2].According to our analyses, mammal lice have fewer temperature-related genes compared to bird lice, especially for Hsp60 and Hsp70 genes.Since the transition of host-switching is from bird to mammal, the loss of multiple unnecessary temperature-related genes in mammal lice might reverse energy and resource for other The number of chemosensory-related genes in mammal lice have also remarkably reduced compared to bird lice.GR genes are among the fastest-evolving gene categories for both copy number variation and sequence divergence in two lice (Figs.5a and 7b).Mammal lice retained only the GR28 genes related to sensing host temperature from their avian ancestors during host-switching.Previous study has reported that sugar receptor (GR5a and GR64e) and CO2 receptor (GR21a and GR63a) were absent in mammal lice [7].The lack of sugar receptors is a common feature among various blood feeders, including kissing bug R.
Interestingly, IR25a gene, the most highly conserved olfactory receptor for CO2 attraction among insects, was positively selected in mammal lice.It is possible that mammal lice that lack CO2 receptor still respond to CO2 use the same IR25a-dependent pathway [95].Overall, after host-switching from birds to mammals, lice loss these genes of sugar, fructose, and carbon dioxide receptors.

Conclusions
In this study, we present a high-quality chromosomal-level genome assembly of Menopon gallinae with high coverage and contiguity.The M. gallinae genome provides a possibility to study the details of gene selection or loss in the process of evolution and adaptation to the host-switching of lice, including genes involved in hematophagy, digestion, chemosensory, temperature, immune, and detoxification.Our comparative analyses have revealed genetic variations of parasitic lice, which likely correlated with host-switching from bird to mammal.We observed contractions in chemosensory and temperature-related gene families and discovered seven hematophagy-related genes were positively selected in mammal lice.This study offers valuable genomic resources and insights into the genetic basis of M. gallinae, and deepens our understanding of how parasitic lice adapt to host-switching.
To confirm the findings of this study and determine the biological significance of relevant genes, broader genomic studies that include high-quality genome assemblies of more species and functional evidence based on experimental verification will be necessary.
essential biological processes to adapt to the environment.Interestingly, however, mammal lice possess a greater number of sHSP genes than bird lice.The sHSPs are the first line of cell defense, preventing irreversible denaturation of substrate proteins, especially when cells are stressed, and have critical roles in normal development in insect[89][90][91].Our results indicated a potential difference in the genetic basis of temperature-related genes in bird and mammal lice.

Figure 1
Figure 1 Genome description of Menopon gallinae.(a) GenomeScope estimation of genome size and heterogeneity using a k-mer of 17.(b) Hi-C interaction map produced by 3D-DNA.(c) Circular representation of the chromosomes.Tracks a-d represents the distribution of chromosome karyotypes, gene density, GC density, and repeat sequences density, respectively.Densities were calculated in 100 kb windows.

Figure 2
Figure 2 Phylogenetic tree with the dynamic evolution of gene families among Menopon gallinae, Pediculus humanus and other species.In the left panel, blue and red numbers on the branch shows the number of expanded and contracted gene families for each clade.Pie charts beside or on each branch of the tree show the proportion of expanded (blue) and contracted (red) gene families.The black numbers are divergence times.In the right panel, the numbers of gene families (orthogroups) were shown as barplots.Orthogroups of different categories were in different colors.

Figure 3
Figure 3 Enrichment analysis of gene families of different categories.KEGG pathway of expanded (a) and contracted (b) gene families of Menopon gallinae.GO enrichment of expanded (c) and contracted (d) gene families of Pediculus humanus.GO enrichment of specific gene families of Menopon gallinae (e) and Pediculus humanus (f).

Figure 4
Figure 4 Distribution of hematophagy-related genes in the genomes of Menopon gallinae, Pediculus humanus, and other species.The heatmap shows the numbers of hematophagyrelated genes.The numbers were transformed with log10(n+1).

Figure 2 2
Figure 2 Click here to access/download;Figure;Figure 2.pdf

Figure 3
Figure 3 Click here to access/download;Figure;Figure 3.pdf

Figure 4
Figure 4 Click here to access/download;Figure;Figure 4.pdf

Figure 5
Figure 5 Click here to access/download;Figure;Figure 5.pdf