Open AccessArticle Evidence for Genotype-Associated Differences in Disease Severity and Limitations of Serotype-Based Classification in Glaesserella parasuis Revealed by Whole-Genome Sequencing in Japan by Manao Ozawa Manao Ozawa SciProfiles Scilit Preprints.org Google Scholar *, Motoshi Kawano Motoshi Kawano SciProfiles Scilit Preprints.org Google Scholar and Shoko Iwamoto Shoko Iwamoto SciProfiles Scilit Preprints.org Google Scholar National Veterinary Assay Laboratory, Ministry of Agriculture, Forestry and Fisheries, 2-1-22, Kannondai, Tsukuba 305-8535, Ibaraki, Japan * Author to whom correspondence should be addressed. Pathogens 2026, 15(6), 619; https://doi.org/10.3390/pathogens15060619 (registering DOI) Submission received: 18 May 2026 / Revised: 1 June 2026 / Accepted: 8 June 2026 / Published: 9 June 2026 Abstract We conducted whole-genome sequencing to investigate serotypes, virulence-associated genes, antimicrobial resistance determinants, and genetic relationships among Glaesserella parasuis isolates from diseased pigs in Japan, focusing on underrecognized aspects of disease epidemiology and control. Although Glässer’s disease is well recognized in swine production, its epidemiology remains incompletely understood, particularly regarding the relationship between serotype, genotype, and pathogenicity. Serotypes 5 or 12 (5/12) (28.9%) were predominant, followed by serotype 7 (10.8%). Phylogenetic analysis based on core-genome single nucleotide polymorphisms and cluster analysis classified the isolates into three genetic groups, with no clear association between serotype and genetic grouping. One genetic group tended to exhibit a lower proportion of severe clinical cases compared with the others, with a statistically significant difference observed in one comparison but not in the other. These findings provide evidence suggesting genotype-associated differences in disease severity, indicating that pathogenic potential may be more closely linked to genetic background than to serotype. These findings suggest a potential limitation of serotype-based vaccine strategies. Although 86.7% of isolates lacked antimicrobial resistance genes, resistance determinants were identified on contigs predicted to be of plasmid origin. These results indicate that antimicrobial resistance, while not widespread, may be underestimated and could disseminate. Overall, our findings highlight underexplored aspects of Glässer’s disease relevant to improving control and prevention. Keywords: antimicrobial resistance; Glaesserella; whole genome sequencing; vaccine 1. Introduction Fifteen G. parasuis serotypes have been reported previously. Serotypes 4, 5, and 13 are prevalent globally [ 3, 4, 5, 6]. Several untypable (UT) isolates have also been identified. Among G. parasuis isolates from sick pigs in Japan between 1987 and 1988, serotype 5 (14.2%) was the most prevalent, followed by serotype 4 (9.2%) [ 7]. It has been reported that serotypes 1, 5, 10, 12, 13, and 14 are highly virulent; serotypes 2, 4, 8, and 15 are moderately virulent; and serotypes 3, 6, 7, 9, and 11 are nonvirulent [ 8]. However, the relationship between serotype and pathogenicity remains inconsistent and poorly understood [ 9]. Antimicrobials are commonly used to treat Glässer’s disease, and treatment typically includes agents such as penicillins, macrolides, tetracyclines, phenicols, and fluoroquinolones [ 18]. However, the inappropriate use of antimicrobials leads to the selection of resistant bacteria on farms, as observed in G. parasuis isolates globally [ 19, 20, 21, 22]. Nevertheless, antimicrobial resistance in G. parasuis has been relatively understudied in certain regions, including Japan. Although serotypes and pathogenicity of G. parasuis isolates in Japan have been reported previously [ 7, 23], comprehensive genomic analyses of field isolates remain limited. In particular, the relationships among genotype, pathogenicity, and antimicrobial resistance have not been systematically investigated using WGS. Therefore, this study aimed to elucidate the genetic characteristics of G. parasuis isolates from Japan, including virulence-associated genes and antimicrobial resistance determinants, using WGS, and to provide insights into underrecognized aspects of disease epidemiology and control. 2. Materials and Methods 2.1. Bacterial Isolates We collected G. parasuis isolated and identified from pig-derived specimens that were sent to the Livestock Hygiene Service Centers nationwide for disease diagnosis. Eighty-three G. parasuis isolates were obtained from diseased pigs in 2013 (35 isolates), 2014 (17 isolates), 2015 (20 isolates), and 2022 (11 isolates) from 19 prefectures in Japan ( Table S1). The 19 prefectures used for sample collection are located across the Hokkaido/Tohoku (HT), Kanto/Koshinetsu (KK), Tokai/Hokuriku (TH), Chugoku/Shikoku (CS), and Kyushu (KY) regions. All isolates were recovered from the lungs, heart, or brain of pigs with clinical signs of Glässer’s disease (death, respiratory symptoms, arthritis, neurological symptoms, and stunted growth). Each isolate was obtained from a single animal. Bacterial species were identified biochemically using the NN-20 Rapid ID test (SHIMADZU, Kyoto, Japan). G. parasuis isolates were classified according to the clinical information recorded during routine diagnostic investigations conducted by veterinarians at the Livestock Hygiene Service Centers. Classification was based on the predominant clinical outcome reported for each affected pig. Isolates were assigned to the death group ( n = 34) when the animal died or was euthanized because of severe disease. The systemic group ( n = 36) included pigs showing systemic manifestations compatible with Glässer’s disease, such as polyserositis, polyarthritis, meningitis, neurological signs, or severe generalized illness. The respiratory group ( n = 13) included pigs presenting predominantly respiratory signs without reported systemic involvement. Clinical information was obtained from diagnostic records and interviews with farmers conducted by attending veterinarians. The isolation and identification of G. parasuis isolates were performed by the veterinarians of the Livestock Hygiene Service Centers in the prefectures. 2.2. Serotyping The contig sequences of the isolates obtained using WGS were serotyped in silico using UniPro UGENE v45.1, an online PCR data analysis tool [ 24]. Previously reported primers were used [ 25]. 2.3. Genome Sequencing and Analysis Genomic DNA was prepared for WGS using a DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA, USA). The 260/280 ratio of the extracted DNA was calculated using a Nanodrop instrument (Thermo Fisher Scientific, Waltham, MA, USA), and DNA samples with a ratio of 1.8 or higher were used. DNA samples were subjected to tagmentation and PCR amplification using a Nextera XT DNA Sample Preparation Kit (Illumina, Inc., San Diego, CA, USA). For library preparation, the DNA samples were diluted to 0.2 ng/μL. WGS was performed using a MiSeq instrument (Illumina, Inc., San Diego, CA, USA) in paired-end mode (2 × 300). De novo assembly was performed using Shovill software (version 1.0.4; https://github.com/tseemann/shovill, accessed on 21 August 2024). Contigs shorter than 300 bp were excluded from the Shovill assembly. To identify antimicrobial resistance genes, contig sequences were analyzed using ResFinder 4.0 ( https://genepi.food.dtu.dk/resfinder, accessed on 21 August 2024). The identity and coverage thresholds used for ResFinder were 90% and 60%, respectively. bcrA, bcrB, bcrC, bacA (bacitracin resistance), and bcr-1 (bicyclomycin resistance) were detected using the Comprehensive Antibiotic Resistance Database ( https://card.mcmaster.ca, accessed on 21 August 2024) within ABRicate (version 1.0.1; https://github.com/tseemann/abricate, accessed on 2 September 2024). The identity and coverage thresholds used in ABRicate were 75% and 0%, respectively. Virulence-factor-encoding genes were identified using ABRicate software. A database including bioB (biotin synthase), F357_gp34 (Mu-like prophage protein gp36), hicA and hicB (corresponding to the toxin and antitoxin components, HicA and HicB, respectively), K756_02745 (bacteriophage protein/plasmid stabilization system protein), K756_06920 (CMP-N-acetylneuraminate-beta-galactosamide-alpha-2,3-sialyltransferase/lipopolysaccharide biosynthesis protein), and capD (polysaccharide biosynthesis protein) (GenBank accession No. KC795345.1) was constructed using previously reported sequences. These genes have been reported to be associated with G. parasuis isolated from pigs affected by Glässer’s disease [ 14, 26]. Additionally, the Virulence Factor Database ( http://www.mgc.ac.cn/VFs/, accessed on 2 September 2024) was used within ABRicate to detect genes encoding virulence factors [ 27]. The contigs (chromosomal or plasmid-derived) were classified using PlasFlow (version 1.1.0) [ 28]. The ratio of the core genome to the pan-genome was calculated using Prokka and Roary with default settings [ 29, 30]. Species identity was further confirmed by average nucleotide identity (ANI) analysis using FastANI v1.34 ( https://github.com/ParBLiSS/FastANI, accessed on 30 May 2024) against the genome of the Glaesserella parasuis type strain ATCC 19417 T (CCUG 3712; GCA_002015085.1). 2.4. Single-Nucleotide Polymorphism (SNP) Phylogenetic Analysis Repeat and prophage regions of the reference genome (strain YHP1818, accession no. CP071487.1) were identified using NUCmer (MUMmer v3.23) [ 31] and PhiSpy v4.2.21 [ 32], respectively. Simulated short reads of contig sequences were generated using the Ngsngs program version 0.9.2 [ 33]. Simulated reads were aligned to the reference sequence using bwa-mem v0.7.17 [ 34] with default parameters. Mapping coverage and statistics were calculated using BEDTools v2.31.1 [ 35] and SAMtools v1.20 [ 36], respectively. Duplicate reads were marked using the MarkDuplicates tool from GAKT v4.5.0.0 [ 37], followed by variant calling with HaplotypeCaller. SNPs were extracted using GATK SelectVariants and filtered with VariantFiltration based on QD, QUAL, SOR, FS, MQ, MQRankSum, ReadPosRankSum, heterozygosity, and depth. Depth filtering applied a threshold of DP < max(mean depth/10, 5) to avoid excessive filtering in low-coverage regions. Base quality score recalibration was performed using GATK BaseRecalibrator and ApplyBQSR, followed by final variant calling and filtering. Recombination regions were predicted using consensus sequences constructed with BCFtools v1.21 [ 36] and BEDTools, followed by Gubbins v3.4 [ 38]. SNP datasets were merged using VCFtools v0.1.16 [ 39], and SNPs in non-core, repeat, recombination, and predicted prophage regions were excluded. Remaining SNPs were concatenated to construct a pseudo-sequence for core-genome phylogenetic analysis using IQ-TREE v2.4.0 [ 40] with parameters ‘-m MFP+ASC -b 100’. The phylogenetic tree was visualized using iTOL (version 6) [ 41]. 2.5. Population Structure Analysis Population structure was inferred using the Bayesian hierarchical clustering algorithm implemented in the R package fastbaps (version 1.0.7) ( https://github.com/gtonkinhill/fastbaps, accessed on 28 January 2026). A core-genome alignment generated from single nucleotide polymorphisms (SNPs) created in 2.4 was used as input for the analysis. Clustering was performed in R (version 4.3.1) using fastbaps with default parameters [ 42]. The fastbaps function was applied to identify genetic population structure without prior assumptions regarding the number of clusters. Hierarchical Bayesian clustering was subsequently performed to resolve subpopulation structure at multiple levels. The resulting clusters were mapped to the phylogenetic tree for visualization and further interpretation of the population structure. 2.6. Statistical Analysis Statistical analyses were performed using R software (version 4.3.1). Analysis of the association between the isolates and the clinical signs exhibited by the pigs from which they were isolated was evaluated using Fisher’s exact test. For post hoc pairwise comparisons, Fisher’s exact test was applied with Bonferroni correction for multiple testing. All tests were two-sided, with a significance level set at 0.05. 3. Results 3.1. Serotypes and Virulence Genes Prior to downstream analyses, species identity was confirmed by average nucleotide identity (ANI) analysis against the genome of the G. parasuis type strain ATCC 19417 T (CCUG 3712). All isolates exhibited ANI values ranging from 96.2% to 98.9%, exceeding the accepted species threshold of 95–96%, thereby confirming their classification as G. parasuis. There were 27 isolates from Hokkaido/Tohoku, 32 from Kanto/Koshinetsu, 9 from Tokai/Hokuriku, 0 from Kinki, 2 from Chugoku/Shikoku, and 13 from Kyushu (Figure S1). Of the 83 isolates, 24 (28.9%) and 7 (10.8%) belonged to serotypes 5/12 (5 or 12) and 7, respectively. Other serotypes included 4 (9.6%), 1 (8.4%), and 13 (8.4%) ( Table 1). Three isolates were identified and products were detected for both serotypes 1 and 2 using in silico PCR (serotype 1/2). UT isolates accounted for 25.3% of all isolates ( Table 1). The virulence genes and their respective prevalence among the 83 isolates were as follows: gmhA (83 isolates, 100%), kdsA (81 isolates, 97.6%), F33357_gp34 (71 isolates, 85.5%), bioB (62 isolates, 74.7%), K756_02745 (59 isolates, 71.1%), capD (55 isolates, 66.3%), group 1 vtaA (39 isolates, 47.0%), hicA and hicB (27 isolates, 32.5%), K756_06920 (25 isolates, 30.1%) group 2 vtaA (17 isolates, 20.5%), and group 3 vtaA (9 isolates, 10.8%) ( Table 2). 3.2. Phylogenetic Analysis Table S2 presents the assembly statistics for all 83 G. parasuis isolates analyzed in this study. The contig numbers ranged from 83 to 203, with N50 values between 31,421 and 97,951 bp, and mean coverage values generally exceeding 50× across isolates. The size of the pan-genome was 5821 genes in the 83 isolates used in this analysis. Of these, 1345 genes were found within all isolates. The percentage of core genes was 23.1% (1345 genes/5821 genes). A phylogenetic tree of the 83 G. parasuis isolates derived from Japan was generated based on SNP sites, showing two well-supported clades ( Figure 1). Clade 1 included isolates from all regions, whereas clade 2 included isolates from all regions except the Chugoku/Shikoku region. Genetically related isolates were identified in the same region of each clade. Based on population structure analysis, all isolates were classified into one of three groups: 1, 2, and 3 ( Figure 1). Group 1 belonged to clade 1 and did not have bioB, hicA, hicB, or K756_06920. The rates of the presence of group 1 vtaA genes were 50.0%, 49.0%, and 33.3% in groups 1, 2, and 3, respectively, and no significant differences were observed among the groups. Groups 1, 2, and 3 were dominated by serotype 7 isolates (31.8%, 7/22), serotype 5/12 isolates (44.9%, 22/49), and serotype UT isolates (75.0%, 9/12), respectively. Groups 1 and 2 contained a variety of serotypes. Group 3 consisted mostly of UT isolates, but also included isolates of serotype 1 and 5/12. Thus, there was no clear relationship between the groups and serotypes. The relationship between the groups and clinical signs of the pigs from which each isolate was obtained was analyzed ( Table 3). The rates of death, systemic symptoms, and respiratory symptoms were 36.4%, 45.5%, and 18.2% in group 1; 46.9%, 49.0%, and 4.1% in group 2; and 25.0%, 16.7%, and 58.3% in group 3, respectively. The combined rate of death and systemic symptoms was significantly lower in group 3 (41.7%) than in group 2 (95.9%). The relationship between serotype and clinical symptoms was such that the combined rates of death and systemic symptoms ranged from 66.7% to 100%; however, there was no significant difference in the rates between serotypes ( Table S3). 3.3. Antimicrobial Resistance Genes No resistance genes were detected in 72 of the 83 isolates (86.7%). The tet(B) and cat genes were dominant (6.0%, 5/83), followed by blaROB-1 (4.8%, 4/83; Figure 1). Only one isolate (1.2%, 1/83) harbored the macrolide resistance gene, erm(B). A combination of aminoglycoside-, β-lactam-, tetracycline-, and sulfonamide-resistance genes was observed in one isolate. The classification of contigs with resistance genes revealed six plasmid-derived contigs ( Table S4), which harbored resistance genes for aminoglycosides, β-lactams, macrolides, phenicols, or sulfonamides. aph(3′)-Ia, aph(3″)-Ib, aph(6)-Id, and sul2 were seen on the same plasmid-derived contig. 4. Discussion G. parasuis is a major health threat to pig production and requires antimicrobial agent administration to control. Controlling G. parasuis involves multifaceted measures including vaccination, antimicrobials, and strict husbandry hygiene management. Information on the serotypes, virulence profiles, and predicted antimicrobial resistance patterns of G. parasuis obtained using WGS is useful for vaccine prevention and antimicrobial therapy. Furthermore, molecular epidemiological analysis using WGS helps estimate the transmission routes of G. parasuis in pig farms [ 15, 16]. However, despite its importance, key aspects of Glässer’s disease, including the relationship between genotype and disease severity, limitations of serotype-based control, and the dynamics of antimicrobial resistance, remain insufficiently addressed. In this study, serotypes 5/12 were the most common among G. parasuis isolates ( Table 1). Serotype distribution varies depending on the country and region. Serotypes 4, 5, and 13 are common in China, Brazil, and Germany [ 21, 22, 43], whereas serotypes 7, 13, and 2 are found in North America [ 15]. Serotypes 4, 5/12, and 1 were previously reported in Europe [ 44]. Recently, serotype 7 isolates have been frequently detected in clinical cases in North America, Europe, and China [ 13, 15, 44]. Japan imports breeding pigs from North America and Europe [ 45]. Although the genotype of serotype 7 isolates in this study was not compared with the genotypes of isolates originating from North America, the possibility that serotype 7 isolates entered Japan through pig imports cannot be ruled out. In the present study, we combined core-genome SNP analysis with Bayesian analysis of population structure (BAPS) to more precisely understand the population structure of G. parasuis [ 46]. While core-genome SNP-based phylogenetic analysis can capture vertical evolutionary history with high resolution, the extremely high recombination rate exhibited by G. parasuis [ 14] makes it difficult to accurately distinguish between actual phylogenetic groups and genetic population structures. Therefore, combining core-genome SNP analysis with BAPS, which statistically absorbs the effects of recombination and estimates genetically consistent clusters, may overcome the limitations of core-genome SNP analysis. In particular, for bacteria with high recombination rates, such as G. parasuis, phylogenetic proximity does not necessarily imply the same evolutionary group. Howell et al. [ 14] demonstrated that BAPS analysis divided a population into five clades, which corresponded to the two major clades in the phylogenetic tree. Gong et al. [ 13] also classified 764 strains into 10 BAPS groups that showed high concordance with clades 1/2 based on core-genome SNPs. In this study, clade 1 corresponded to BAPS group 1 and clade 2 corresponded to other BAPS groups. The concordance between the phylogenetic and BAPS clustering results supports the robustness of the inferred population structure despite the high recombination rate of G. parasuis. In G. parasuis, the relationship between the presence of virulence factors and the manifestation of virulence remains unclear [ 14, 47]. In this study, all isolates were classified into three groups based on phylogenetic and cluster analyses. Furthermore, a significant association between genetic grouping and clinical signs was observed in one comparison. Specifically, significantly fewer deaths and systemic symptoms were observed in group 3 than in group 2. These findings suggest that disease severity may be associated with genetic background., although statistical significance was observed in only one comparison, suggesting that the relationship between genotype and clinical outcome is complex and may be influenced by additional factors. Group 1 vtaA genes have been reported to be associated with the virulence of G. parasuis [ 48, 49]. In this study, the prevalence of the group 1 vtaA genes was 33.3%, 45%, and 50% in BAPS groups 1, 2, and 3, respectively. Although the prevalence was lower in group 1 than in the other groups, this difference was not statistically significant. These results suggest that virulence in G. parasuis cannot be explained by vtaA alone and may instead be determined by a combination of multiple genetic factors, highlighting an underexplored aspect of genotype–phenotype relationships in this pathogen. Several limitations should be considered when interpreting the observed association between genetic background and clinical severity. First, clinical classification was based on information collected during routine diagnostic investigations and was not derived from a standardized clinical scoring system. Therefore, some degree of subjectivity and inter-observer variability may have been present. Second, host-related factors that could influence disease outcome, including genetic background, breed susceptibility, management conditions, and immune status, were not available for analysis. Although information regarding vaccination history was available for some animals, these data were incomplete and not sufficiently standardized across all cases to permit meaningful statistical evaluation and were therefore not incorporated into the statistical analyses. Third, information regarding concurrent infections with immunomodulatory pathogens, such as porcine reproductive and respiratory syndrome virus (PRRSV) or porcine circovirus type 2 (PCV2), was unavailable. Such co-infections may influence disease severity independently of bacterial genotype. Consequently, the associations observed in this study should be interpreted as evidence of a potential relationship between genetic background and clinical outcome rather than definitive evidence of a causal relationship between genotype and disease severity. In Japan, inactivated bacterins containing serotypes 2 and 5 are commercially available; however, their protective capability against all isolates within these serotypes remains uncertain, and cross-protection against heterologous serotypes appears limited [ 50]. These findings further emphasize the limitations of current serotype-based vaccination strategies and the need for careful consideration when applying these vaccines in the field. In addition to vaccination, strict hygiene and management practices are essential for effective prevention of Glässer’s disease. Overall, 86.5% of the isolates did not harbor any resistance genes. However, reports from the United States and China have shown that almost all isolates possess ksgA, bcr, bacA, sul2, and aph(3″)-Ib, which confer resistance to kasugamycin, bicyclomycin, bacitracin, sulfonamide, and streptomycin, respectively [ 15, 16]. However, Gong et al. [ 13] reported that the detection rates of sul2 and aph(3″)-Ib were 12.7% and 9.82%, respectively. Mugabi et al. [ 15] used SRST2 [ 52] to detect resistance genes. In contrast, Gong et al. [ 13] used ResFinder. In the present study, we used ResFinder, and the detection rate of resistance genes was similar to that reported by Gong et al. Differences in resistance gene detection rates are thought to be largely influenced by differences in analysis tools, sequencing data quality, and bioinformatics analysis conditions. In fact, it has been pointed out that in antimicrobial resistance analysis using next-generation sequencing, data quality, the database used, and the choice of analysis tools have a significant impact on the reproducibility and accuracy of results [ 53]. However, previous studies and this study have not confirmed the phenotypes of isolates harboring resistance genes. Therefore, future studies should investigate the extent to which resistance genotypes and phenotypes match. Plasmids play an important role in the horizontal transfer of resistance genes. In this study, six contigs presumably derived from plasmids were detected, containing resistance genes, including those for β-lactams and aminoglycosides. Small plasmids of 5–7 kbp have previously been reported in G. parasuis [ 16, 55], and some of the contigs in this study were of a similar size. Although we did not obtain the complete sequences of the plasmids in this study, they may be similar to previously reported small plasmids. To prevent the spread of such plasmids in G. parasuis, careful use of antimicrobials is required when treating Glässer’s disease. In conclusion, G. parasuis isolates were divided into three groups based on phylogenetic and clustering analysis. Although we did not find a clear correlation between serotype and genotype, we observed a potential association between clinical symptoms and the genotype of some isolates. Our findings suggest genotype-associated differences in clinical severity and reinforce the limitations of serotype-based approaches to disease control. These findings suggest that vaccine efficacy cannot be reliably predicted based on serotype alone, highlighting an important limitation of current control strategies. In addition to vaccination, strict hygiene and management practices are essential for the effective prevention of Glässer’s disease. Although antimicrobial resistance genes may spread via plasmids or ICEs, only a few isolates possess resistance genes, indicating that antimicrobial resistance is not widespread among G. parasuis isolates in Japan. However, antimicrobial agents should be used carefully after confirming the susceptibility of the causative bacteria to prevent further spread of antimicrobial resistance. Overall, this study highlights that key aspects of Glässer’s disease—including genotype–pathogenicity relationships, vaccine limitations, and antimicrobial resistance dynamics—remain insufficiently understood and warrant further investigation. Supplementary Materials The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/pathogens15060619/s1, Figure S1: Geographical distribution of isolates included in this study across Japan; Table S1: G. parasuis isolates used in this study ( n = 83); Table S2: Assembly statistics of G. parasuis ( n = 83); Table S3: Relationship between serotypes and clinical signs; Table S4: Classification of contigs and resistance genes harbored in those contigs. Author Contributions Conceptualization, M.O.; methodology, M.O. and M.K.; formal analysis, M.O.; investigation, M.O.; data curation, M.O.; writing—original draft preparation, M.O.; writing—review and editing, S.I.; visualization, M.O.; supervision, S.I. All authors have read and agreed to the published version of the manuscript. Funding This research received no external funding. Institutional Review Board Statement This study did not involve any experimental procedures on animals. All Glaesserella parasuis isolates analyzed in this work were obtained from routine diagnostic investigations conducted by veterinarians at Livestock Hygiene Service Centers across Japan. The samples were originally collected for the purpose of disease diagnosis in clinical cases, and no additional sampling, handling, or intervention involving animals was performed specifically for this research. The present study was conducted solely on bacterial isolates and associated diagnostic data provided after routine veterinary procedures. Therefore, according to institutional and national guidelines for animal research in Japan, ethical approval was not required for this study. According to institutional and national guidelines for animal research in Japan, ethical approval from an Institutional Animal Care and Use Committee was not required for this study. Informed Consent Statement No identifiable information regarding the animal owners was collected or used in this study. Therefore, informed consent from animal owners was not required in accordance with applicable guidelines. Data Availability Statement All raw whole-genome sequencing reads generated in this study have been deposited in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive under BioProject accession number PRJDB19113. Acknowledgments We are grateful to the staff of the Livestock Hygiene Service Centers across Japan for providing us with G. parasuis isolates. We thank Tsuyoshi Sekizuka of SEKIZUKA Industry Co., Ltd. for assistance in developing the core-genome phylogenetic analysis pipelines. Conflicts of Interest The authors declare no conflicts of interest. References Oliveira, S.; Pijoan, C. Haemophilus parasuis: New trends on diagnosis, epidemiology and control. Vet. Microbiol. 2004, 99, 1–12. [ Google Scholar] [ CrossRef] Duan, Y.; Hao, Y.; Feng, H.; Shu, J.; He, Y. Research progress on Haemophilus parasuis vaccines. Front. Vet. Sci. 2025, 12, 1492144. [ Google Scholar] [ CrossRef] Espíndola, J.P.; Balbinott, N.; Gressler, L.T.; Machado, G.; Klein, C.S.; Rebelatto, R.; Martín, C.B.G.; Kreutz, L.C.; Schryvers, A.B.; Frandoloso, R. Molecular serotyping of clinical strains of Haemophilus ( Glaesserella) parasuis brings new insights regarding Glässer’s disease outbreaks in Brazil. PeerJ 2019, 7, e6817. [ Google Scholar] [ CrossRef] Jia, A.; Zhou, R.; Fan, H.; Yang, K.; Zhang, J.; Xu, Y.; Wang, G.; Liao, M. Development of serotype-specific PCR assays for typing of Haemophilus parasuis isolates circulating in southern China. J. Clin. Microbiol. 2017, 55, 3249–3257. [ Google Scholar] [ CrossRef] Oliveira, S.; Blackall, P.; Pijoan, C. Characterization of the diversity of Haemophilus parasuis field isolates by use of serotyping and genotyping. Am. J. Vet. Res. 2003, 64, 435–442. [ Google Scholar] [ CrossRef] [ PubMed] Van, C.N.; Thanh, T.V.T.; Zou, G.; Jia, M.; Wang, Q.; Zhang, L.; Ding, W.; Huang, Q.; Zhou, R. Characterization of serotypes and virulence genes of Haemophilus parasuis isolates from Central Vietnam. Vet. Microbiol. 2019, 230, 117–122. [ Google Scholar] [ CrossRef] Morikoshi, T.; Kobayashi, K.; Kamino, T.; Owaki, S.; Hayashi, S.; Hirano, S. Characterization of Haemophilus parasuis Isolated in Japan. Jpn. J. Vet. Sci. 1990, 52, 667–669. [ Google Scholar] [ CrossRef] [ PubMed] Kielstein, P.; Rapp-Gabrielson, V.J. Designation of 15 serovars of Haemophilus parasuis on the basis of immunodiffusion using heat-stable antigen extracts. J. Clin. Microbiol. 1992, 30, 862–865. [ Google Scholar] [ CrossRef] Aragon, V.; Cerda-Cuellar, M.; Fraile, L.; Mombarg, M.; Nofrarias, M.; Olvera, A.; Sibila, M.; Solanes, D.; Segales, J. Correlation between clinico-pathological outcome and typing of Haemophilus parasuis field strains. Vet. Microbiol. 2010, 142, 387–393. [ Google Scholar] [ CrossRef] [ PubMed] Boerlin, P.; Poljak, Z.; Gallant, J.; Chalmers, G.; Nicholson, V.; Soltes, G.A.; MacInnes, J.I. Genetic diversity of Haemophilus parasuis from sick and healthy pigs. Vet. Microbiol. 2013, 167, 459–467. [ Google Scholar] [ CrossRef] Castilla, K.S.; de Gobbi, D.D.S.; Moreno, L.Z.; Paixão, R.; Coutinho, T.A.; dos Santos, J.L.; Moreno, A.M. Characterization of Haemophilus parasuis isolated from Brazilian swine through serotyping, AFLP and PFGE. Res. Vet. Sci. 2012, 92, 366–371. [ Google Scholar] [ CrossRef] Wang, L.; Ma, L.; Liu, Y.; Gao, P.; Li, Y.; Li, X.; Liu, Y. Multilocus sequence typing and virulence analysis of Haemophilus parasuis strains isolated in five provinces of China. Infect. Genet. Evol. 2016, 44, 228–233. [ Google Scholar] [ CrossRef] Gong, X.; Cui, Q.; Zhang, W.; Shi, Y.; Zhang, P.; Zhang, C.; Hu, G.; Sahin, O.; Wang, L.; Shen, Z.; et al. Genomic insight into the diversity of Glaesserella parasuis isolates from 19 countries. mSphere 2024, 9, e00231-24. [ Google Scholar] [ CrossRef] Howell, K.J.; Weinert, L.A.; Chaudhuri, R.R.; Luan, S.-L.; Peters, S.E.; Corander, J.; Harris, D.; Angen, Ø.; Aragon, V.; Bensaid, A. The use of genome wide association methods to investigate pathogenicity, population structure and serovar in Haemophilus parasuis. BMC Genom. 2014, 15, 1179. [ Google Scholar] [ CrossRef] Mugabi, R.; Silva, A.P.S.P.; Hu, X.; Gottschalk, M.; Aragon, V.; Macedo, N.R.; Sahin, O.; Harms, P.; Main, R.; Tucker, A.W.; et al. Molecular characterization of Glaesserella parasuis strains circulating in North American swine production systems. BMC Vet. Res. 2023, 19, 135. [ Google Scholar] [ CrossRef] [ PubMed] Wan, X.; Li, X.; Osmundson, T.; Li, C.; Yan, H. Whole-genome sequence analyses of Glaesserella parasuis isolates reveals extensive genomic variation and diverse antibiotic resistance determinants. PeerJ 2020, 8, e9293. [ Google Scholar] [ CrossRef] [ PubMed] Bak, H.; Riising, H.-J. Protection of vaccinated pigs against experimental infections with homologous and heterologous Haemophilus parasuis. Vet. Rec. 2002, 151, 502–505. [ Google Scholar] [ CrossRef] [ PubMed] Costa-Hurtado, M.; Barba-Vidal, E.; Maldonado, J.; Aragon, V. Update on Glässer’s disease: How to control the disease under restrictive use of antimicrobials. Vet. Microbiol. 2020, 242, 108595. [ Google Scholar] [ CrossRef] Dayao, D.A.E.; Kienzle, M.; Gibson, J.S.; Blackall, P.J.; Turni, C. Use of a proposed antimicrobial susceptibility testing method for Haemophilus parasuis. Vet. Microbiol. 2014, 172, 586–589. [ Google Scholar] [ CrossRef] [ PubMed] de la Fuente, A.J.M.; Tucker, A.W.; Navas, J.; Blanco, M.; Morris, S.J.; Gutiérrez-Martín, C.B. Antimicrobial susceptibility patterns of Haemophilus parasuis from pigs in the United Kingdom and Spain. Vet. Microbiol. 2007, 120, 184–191. [ Google Scholar] [ CrossRef] Silva, G.F.R.; Moreno, L.Z.; Matajira, C.E.C.; Silva, A.P.S.; Araújo, K.M.; Gomes, V.T.M.; Barbosa, M.R.F.; Sato, M.I.Z.; Moreno, A.M. Serotyping and antimicrobial susceptibility profiling of Glaesserella parasuis isolated from diseased swine in Brazil. Pathogens 2022, 11, 1443. [ Google Scholar] [ CrossRef] Zhou, X.; Xu, X.; Zhao, Y.; Chen, P.; Zhang, X.; Chen, H.; Cai, X. Distribution of antimicrobial resistance among different serovars of Haemophilus parasuis isolates. Vet. Microbiol. 2010, 141, 168–173. [ Google Scholar] [ CrossRef] Amano, H.; Shibata, M.; Kajio, N.; Morozumi, T. Pathologic observations of pigs intranasally inoculated with serovar 1, 4 and 5 of Haemophilus parasuis using immunoperoxidase method. J. Vet. Med. Sci. 1994, 56, 639–644. [ Google Scholar] [ CrossRef] Okonechnikov, K.; Golosova, O.; Fursov, M.; The UGENE Team. Unipro UGENE: A unified bioinformatics toolkit. Bioinformatics 2012, 28, 1166–1167. [ Google Scholar] [ CrossRef] Howell, K.J.; Peters, S.E.; Wang, J.; Hernandez-Garcia, J.; Weinert, L.A.; Luan, S.-L.; Chaudhuri, R.R.; Angen, Ø.; Aragon, V.; Williamson, S.M. Development of a multiplex PCR assay for rapid molecular serotyping of Haemophilus parasuis. J. Clin. Microbiol. 2015, 53, 3812–3821. [ Google Scholar] [ CrossRef] Zhou, H.; Yang, B.; Xu, F.; Chen, X.; Wang, J.; Blackall, P.J.; Zhang, P.; Xia, Y.; Zhang, J.; Ma, R. Identification of putative virulence-associated genes of Haemophilus parasuis through suppression subtractive hybridization. Vet. Microbiol. 2010, 144, 377–383. [ Google Scholar] [ CrossRef] Liu, B.; Zheng, D.; Zhou, S.; Chen, L.; Yang, J. VFDB 2022: A general classification scheme for bacterial virulence factors. Nucleic Acids Res. 2021, 50, D912–D917. [ Google Scholar] [ CrossRef] Krawczyk, P.S.; Lipinski, L.; Dziembowski, A. PlasFlow: Predicting plasmid sequences in metagenomic data using genome signatures. Nucleic Acids Res. 2018, 46, e35. [ Google Scholar] [ CrossRef] Page, A.J.; Cummins, C.A.; Hunt, M.; Wong, V.K.; Reuter, S.; Holden, M.T.G.; Fookes, M.; Falush, D.; Keane, J.A.; Parkhill, J. Roary: Rapid large-scale prokaryote pan genome analysis. Bioinformatics 2015, 31, 3691–3693. [ Google Scholar] [ CrossRef] Seemann, T. Prokka: Rapid prokaryotic genome annotation. Bioinformatics 2014, 30, 2068–2069. [ Google Scholar] [ CrossRef] Kurtz, S.; Phillippy, A.; Delcher, A.L.; Smoot, M.; Shumway, M.; Antonescu, C.; Salzberg, S.L. Versatile and open software for comparing large genomes. Genome Biol. 2004, 5, R12. [ Google Scholar] [ CrossRef] Akhter, S.; Aziz, R.K.; Edwards, R.A. PhiSpy: A novel algorithm for finding prophages in bacterial genomes that combines similarity- and composition-based strategies. Nucleic Acids Res. 2012, 40, e126. [ Google Scholar] [ CrossRef] Henriksen, R.A.; Zhao, L.; Korneliussen, T.S. NGSNGS: Next-generation simulator for next-generation sequencing data. Bioinformatics 2023, 39, btad041. [ Google Scholar] [ CrossRef] Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv 2013, arXiv:1303.3997. [ Google Scholar] [ CrossRef] Quinlan, A.R.; Hall, I.M. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 2010, 26, 841–842. [ Google Scholar] [ CrossRef] Danecek, P.; Bonfield, J.K.; Liddle, J.; Marshall, J.; Ohan, V.; Pollard, M.O.; Whitwham, A.; Keane, T.; McCarthy, S.A.; Davies, R.M.; et al. Twelve years of SAMtools and BCFtools. GigaScience 2021, 10, giab008. [ Google Scholar] [ CrossRef] Poplin, R.; Ruano-Rubio, V.; DePristo, M.A.; Fennell, T.J.; Carneiro, M.O.; Van der Auwera, G.A.; Kling, D.E.; Gauthier, L.D.; Levy-Moonshine, A.; Roazen, D.; et al. Scaling accurate genetic variant discovery to tens of thousands of samples. bioRxiv 2017. [ Google Scholar] [ CrossRef] Croucher, N.J.; Page, A.J.; Connor, T.R.; Delaney, A.J.; Keane, J.A.; Bentley, S.D.; Parkhill, J.; Harris, S.R. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. Nucleic Acids Res. 2014, 43, e15. [ Google Scholar] [ CrossRef] Danecek, P.; Auton, A.; Abecasis, G.; Albers, C.A.; Banks, E.; DePristo, M.A.; Handsaker, R.E.; Lunter, G.; Marth, G.T.; Sherry, S.T.; et al. The variant call format and VCFtools. Bioinformatics 2011, 27, 2156–2158. [ Google Scholar] [ CrossRef] Minh, B.Q.; Schmidt, H.A.; Chernomor, O.; Schrempf, D.; Woodhams, M.D.; von Haeseler, A.; Lanfear, R. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Mol. Biol. Evol. 2020, 37, 1530–1534. [ Google Scholar] [ CrossRef] Letunic, I.; Bork, P. Interactive Tree of Life (iTOL) v6: Recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Res. 2024, 52, W78–W82. [ Google Scholar] [ CrossRef] R Core Team. R: A Language and Environment for Statistical Computing. Available online: https://www.R-project.org/ (accessed on 14 January 2025). Wiencek, I.; Hartmann, M.; Merkel, J.; Trittmacher, S.; Kreienbrock, L.; Hennig-Pauka, I. Temporal patterns of phenotypic antimicrobial resistance and coinfecting pathogens in Glaesserella parasuis strains isolated from diseased swine in Germany from 2006 to 2021. Pathogens 2022, 11, 721. [ Google Scholar] [ CrossRef] Macedo, N.; Gottschalk, M.; Strutzberg-Minder, K.; Van, C.N.; Zhang, L.; Zou, G.; Zhou, R.; Marostica, T.; Clavijo, M.J.; Tucker, A.; et al. Molecular characterization of Glaesserella parasuis strains isolated from North America, Europe and Asia by serotyping PCR and LS-PCR. Vet. Res. 2021, 52, 68. [ Google Scholar] [ CrossRef] Furuno, M.; Uchiyama, M.; Nakahara, Y.; Uenoyama, K.; Fukuhara, H.; Morino, S.; Kijima, M. A Japanese trial to monitor methicillin-resistant Staphylococcus aureus (MRSA) in imported swine during the quarantine period. J. Glob. Antimicrob. Resist. 2018, 14, 182–184. [ Google Scholar] [ CrossRef] [ PubMed] Cheng, L.; Connor, T.R.; Sirén, J.; Aanensen, D.M.; Corander, J. Hierarchical and spatially explicit clustering of DNA sequences with BAPS software. Mol. Biol. Evol. 2013, 30, 1224–1228. [ Google Scholar] [ CrossRef] Wu, C.-F.; Hsu, C.-Y.; Chou, C.-C.; Wang, C.-M.; Huang, S.-W.; Kuo, H.-C. Serotypes, virulence factors and multilocus sequence typing of Glaesserella parasuis from diseased pigs in Taiwan. PeerJ 2023, 11, e15823. [ Google Scholar] [ CrossRef] Olvera, A.; Pina, S.; Macedo, N.; Oliveira, S.; Aragon, V.; Bensaid, A. Identification of potentially virulent strains of Haemophilus parasuis using a multiplex PCR for virulence-associated autotransporters ( vtaA). Vet. J. 2011, 191, 213–218. [ Google Scholar] [ CrossRef] [ PubMed] Pina, S.; Olvera, A.; Barcelo, A.; Bensaid, A. Trimeric autotransporters of Haemophilus parasuis: Generation of an extensive passenger domain repertoire specific for pathogenic strains. J. Bacteriol. 2009, 191, 1034–1045. [ Google Scholar] [ CrossRef] Takahashi, K.; Nagai, S.; Yagihashi, T.; Ikehata, T.; Nakano, Y.; Senna, K.; Maruyama, T.; Murofushi, J. A Cross-Protection Experiment in Pigs Vaccinated with Haemophilus parasuis Serovars 2 and 5 Bacterins, and Evaluation of a Bivalent Vaccine under Laboratory and Field Conditions. J. Vet. Med. Sci. 2001, 63, 487–491. [ Google Scholar] [ CrossRef] Alvarez-Vega, L.; De La Cruz-Monroy, K.S.M.; Garcia-Barraza, A.; Carhuaricra, D.; Llaja-Bernedo, J.L.; Huamancha-Pulido, L.L.; Medina-Gonzales, A.; Sedano-Sanchez, A.F. Genomic insights into diversity, antimicrobial resistance and virulence of Glaesserella parasuis from diseased swine in Peru. Front. Microbiol. 2025, 16, 1678153. [ Google Scholar] [ CrossRef] [ PubMed] Inouye, M.; Dashnow, H.; Raven, L.-A.; Schultz, M.B.; Pope, B.J.; Tomita, T.; Zobel, J.; Holt, K.E. SRST2: Rapid genomic surveillance for public health and hospital microbiology labs. Genome Med. 2014, 6, 90. [ Google Scholar] [ CrossRef] [ PubMed] Laufer Halpin, A.; Mathers, A.J.; Walsh, T.R.; Zingg, W.; Okeke, I.N.; McDonald, L.C.; Elkins, C.A.; Harbarth, S.; Peacock, S.J.; Srinivasan, A.; et al. A framework towards implementation of sequencing for antimicrobial-resistant and other health-care-associated pathogens. Lancet Infect. Dis. 2025, 25, E235–E244. [ Google Scholar] [ CrossRef] [ PubMed] Zhao, Y.; Guo, L.; Li, J.; Huang, X.; Fang, B. Characterization of antimicrobial resistance genes in Haemophilus parasuis isolated from pigs in China. PeerJ 2018, 6, e4613. [ Google Scholar] [ CrossRef] Che, Y.; Wu, R.; Li, H.; Wang, L.; Wu, X.; Chen, Q.; Chen, R.; Zhou, L. Characterization of the plasmids harbouring the florfenicol resistance gene floR in Glaesserella parasuis and Actinobacillus indolicus. J. Glob. Antimicrob. Resist. 2023, 35, 163–171. [ Google Scholar] [ CrossRef] [ PubMed] Figure 1. Genetic relatedness, epidemiology, virulence gene profile, and resistance gene profile of Glaesserella parasuis isolates ( n = 83). Aminoglycoside-resistance genes: aph(3′)-Ia, aph(3″)-Ib, aph(6)-Id. Tetracycline-resistance genes: tet(B), tet(H). β-lactamase gene: blaROB-1. Macrolide-resistance gene: erm(B). Chloramphenicol-resistance gene: cat. Sulfonamide-resistance gene: sul2. HT: Hokkaido/Tohoku district. KK: Kanto/Koshinetsu district. TH: Tokai/Hokuriku district. CS: Chugoku/Shikoku district. KY: Kyushu district. Reference: G. parasuis strain YHP1818 (GenBank accession no. CP071487.1). Bootstrap support values are represented at the branches. The red and blue blocks indicate the presence of virulence genes and resistance genes, respectively. The reference strain was not examined for serotype or gene carriage. Figure 1. Genetic relatedness, epidemiology, virulence gene profile, and resistance gene profile of Glaesserella parasuis isolates ( n = 83). Aminoglycoside-resistance genes: aph(3′)-Ia, aph(3″)-Ib, aph(6)-Id. Tetracycline-resistance genes: tet(B), tet(H). β-lactamase gene: blaROB-1. Macrolide-resistance gene: erm(B). Chloramphenicol-resistance gene: cat. Sulfonamide-resistance gene: sul2. HT: Hokkaido/Tohoku district. KK: Kanto/Koshinetsu district. TH: Tokai/Hokuriku district. CS: Chugoku/Shikoku district. KY: Kyushu district. Reference: G. parasuis strain YHP1818 (GenBank accession no. CP071487.1). Bootstrap support values are represented at the branches. The red and blue blocks indicate the presence of virulence genes and resistance genes, respectively. The reference strain was not examined for serotype or gene carriage. Table 1. Serotypes of G. parasuis ( n = 83). Table 1. Serotypes of G. parasuis ( n = 83). Serotype Number of Isolates in Total % 2013 2014 2015 2022 1 5 2 7 8.4 1/2 2 1 3 3.6 4 4 3 1 8 9.6 5/12 10 5 8 1 24 28.9 7 5 2 2 9 10.8 10 1 1 1.2 13 1 1 2 3 7 8.4 14 1 1 1 3 3.6 UT 7 4 9 1 21 25.3 Total 35 17 20 11 83 100 Table 2. Serotypes and virulence genes of G. parasuis ( n = 83). Table 2. Serotypes and virulence genes of G. parasuis ( n = 83). Serotype Virulence Genes n gmhA kdsA capD vta_group1 vta_group2 vta_group3 F357_gp34 K756_02745 K756_06920 bioB hicA hicB 1 7 7 7 5 1 2 1 1 1 1 7 1/2 3 3 3 3 3 2 3 4 8 8 5 3 8 5 5 8 5/12 24 24 24 11 7 6 24 22 24 24 16 16 24 7 9 9 9 4 2 8 7 2 9 10 1 1 1 13 7 7 7 2 1 7 3 5 4 4 7 14 3 3 2 2 1 2 2 3 1 1 3 UT 21 20 2 10 3 2 17 15 1 19 5 5 21 Total 83 81 55 39 17 9 71 58 25 61 27 27 83 Table 3. Relationship between groups and clinical signs. Table 3. Relationship between groups and clinical signs. BAPS (Group) Clinical Signs Total Death (%) Systemic (%) Death + Systemic (%) * Respiratory (%) 1 8 (36.4) 10 (45.5) 18 (81.8) a4 (18.2) 22 2 19 (47.5) 19 (47.5) 38 (95.0) a2 (5.0) 40 3 3 (25.0) 2 (16.7) 5 (41.7) b7 (58.3) 12 * Different letters indicate significant difference ( p < 0.05). Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. © 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. Share and Cite MDPI and ACS Style Ozawa, M.; Kawano, M.; Iwamoto, S. Evidence for Genotype-Associated Differences in Disease Severity and Limitations of Serotype-Based Classification in Glaesserella parasuis Revealed by Whole-Genome Sequencing in Japan. Pathogens 2026, 15, 619. https://doi.org/10.3390/pathogens15060619 AMA Style Ozawa M, Kawano M, Iwamoto S. Evidence for Genotype-Associated Differences in Disease Severity and Limitations of Serotype-Based Classification in Glaesserella parasuis Revealed by Whole-Genome Sequencing in Japan. Pathogens. 2026; 15(6):619. https://doi.org/10.3390/pathogens15060619 Chicago/Turabian Style Ozawa, Manao, Motoshi Kawano, and Shoko Iwamoto. 2026. "Evidence for Genotype-Associated Differences in Disease Severity and Limitations of Serotype-Based Classification in Glaesserella parasuis Revealed by Whole-Genome Sequencing in Japan" Pathogens 15, no. 6: 619. https://doi.org/10.3390/pathogens15060619 APA Style Ozawa, M., Kawano, M., & Iwamoto, S. (2026). Evidence for Genotype-Associated Differences in Disease Severity and Limitations of Serotype-Based Classification in Glaesserella parasuis Revealed by Whole-Genome Sequencing in Japan. Pathogens, 15(6), 619. https://doi.org/10.3390/pathogens15060619 Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here. Article Metrics Article metric data becomes available approximately 24 hours after publication online.
