A long-awaited taxogenomic investigation of the family Halomonadaceae

The family Halomonadaceae is the largest family composed of halophilic bacteria, with more than 160 species with validly published names as of July 2023. Several classifications to circumscribe this family are available in major resources, such as those provided by the List of Prokaryotic names with Standing in Nomenclature (LPSN), NCBI Taxonomy, Genome Taxonomy Database (GTDB), and Bergey’s Manual of Systematics of Archaea and Bacteria (BMSAB), with some degree of disagreement between them. Moreover, regardless of the classification adopted, the genus Halomonas is not phylogenetically consistent, likely because it has been used as a catch-all for newly described species within the family Halomonadaceae that could not be clearly accommodated in other Halomonadaceae genera. In the past decade, some taxonomic rearrangements have been conducted on the Halomonadaceae based on ribosomal and alternative single-copy housekeeping gene sequence analysis. High-throughput technologies have enabled access to the genome sequences of many type strains belonging to the family Halomonadaceae; however, genome-based studies specifically addressing its taxonomic status have not been performed to date. In this study, we accomplished the genome sequencing of 17 missing type strains of Halomonadaceae species that, together with other publicly available genome sequences, allowed us to re-evaluate the genetic relationship, phylogeny, and taxonomy of the species and genera within this family. The approach followed included the estimate of the Overall Genome Relatedness Indexes (OGRIs) such as the average amino acid identity (AAI), phylogenomic reconstructions using amino acid substitution matrices customized for the family Halomonadaceae, and the analysis of clade-specific signature genes. Based on our results, we conclude that the genus Halovibrio is obviously out of place within the family Halomonadaceae, and, on the other hand, we propose a division of the genus Halomonas into seven separate genera and the transfer of seven species from Halomonas to the genus Modicisalibacter, together with the emendation of the latter. Additionally, data from this study demonstrate the existence of various synonym species names in this family.


Introduction
First described in 1988 to accommodate the genera Halomonas and Deleya based on 16S rRNA oligonucleotide cataloging, the family Halomonadaceae (Franzmann et al., 1988) is the largest bacterial family harboring halophilic microorganisms.Since then, 13 new genera with validly published names and one non-validly published name have been added to the family, as stated in the List of Prokaryotic names with Standing in Nomenclature (LPSN) (Parte et al., 2020).As of July 2023, it comprises 168 species with validly published names (according to the LPSN) that share the following features: Gram-stainnegative, non-endospore-forming rods or coccoid cells, chemoorganotrophic, aerobic or facultatively anaerobic, catalase-positive and oxidase-variable, and halophilic or halotolerant bacteria (Ventosa et al., 2021b).In accordance with the LPSN, this family belongs to the order Oceanospirillales (Garrity et al., 2005b), class Gammaproteobacteria (Garrity et al., 2005a), and other taxonomic sources indicate that it is most closely related to three genera belonging to the family Oceanospirillaceae (Garrity et al., 2005c): Oceanospirillum, Marinospirillum, or Pseudospirillum.The most widely used prokaryotic classification resources have different taxonomic opinions on this family.According to Bergey's Manual of Systematics of Archaea and Bacteria (BMSAB), the most recent description of the family Halomonadaceae comprises 12 genera with validly published names: Halomonas (type genus), Aidingimonas, Carnimonas, Chromohalobacter, Cobetia, Halotalea, Kushneria, Larsenimonas, Modicisalibacter, Pistricoccus, Salinicola, and Zymobacter (Ventosa et al., 2021b), as well as one genus with an effectively published name, "Phytohalomonas" (Liu et al., 2020).As of July 2023, LPSN and NCBI Taxonomy also assign these 13 genera to the Halomonadaceae and 2 additional genera: Halovibrio and Terasakiispira in LPSN (Parte et al., 2020) and Halovibrio and Salicola in NCBI Taxonomy (Schoch et al., 2020), while Genome Taxonomy Database (GTDB) (Parks et al., 2018) release 08-RS214 comprises the 13 BMSAB genera together with the genus Terasakiispira, the 3 Oceanospirillaceae genera listed above, and an unnamed genus (TA22) based on environmental sequence data.GTDB also proposes the division of the genus Halomonas into five distinct genera (denoted with alphabetical suffixes) for a total of 22 genera in this family.Hence, currently, there is some disagreement on the circumscription of this family, which warrants further investigation.
For the present study, we followed the taxonomy proposed by BMSAB, but irrespective of the classification adopted, most genera of the family Halomonadaceae are well-defined and conserved, with the notable exception of the type genus of the family, Halomonas.Halomonas is the 15th largest genus across the prokaryotes in terms of the number of species with validly published names (121 as of July 2023) according to the LPSN. 1 The species of this genus seldom form a monophyletic group and are instead scattered amongst the other genera of the family Halomonadaceae (de la Haba et al., 2010Haba et al., , 2012Haba et al., , 2014;;Ventosa et al., 2021a,b).The polyphyly of Halomonas probably stems from the use of this genus as a default parent taxon for newly described members of the family that did not fall in any of the welldefined genera.Members of this genus exhibit a high degree of heterogeneity, as can be evidenced by their broad DNA G + C range (51.4-74.3 mol%), the widest across the family (Ventosa et al., 2021a).
1 https://lpsn.dsmz.de/text/largest-generaThese observations point to the necessity of dividing the genus Halomonas and reclassifying some of its species.
After its description, three studies have proposed amendments to the family Halomonadaceae, accompanied by the inclusion of new genera into this family (Dobson and Franzmann, 1996;Ben Ali Gam et al., 2007;Ntougias et al., 2007).Additionally, the type genus of the family, Halomonas, has been amended on one occasion to accommodate nine new species in the genus (Dobson and Franzmann, 1996).Initially, a number of characteristic signatures for the 16S rRNA gene sequence were defined for members of this family (Dobson et al., 1993).Nevertheless, as new members have been described, it has been necessary to readapt the 16S rRNA gene sequence signatures (Ben Ali Gam et al., 2007;de la Haba et al., 2014).As whole-genome sequencing has become easier and cheaper, the analysis of clade-specific signature genes has been proposed as a tool to supplement the use of 16S rRNA gene sequence signatures (Zheng et al., 2020).
Several taxonomic studies have been carried out on the Halomonadaceae, some of them based on the 16S rRNA gene sequence analysis (Dobson et al., 1993;Mellado et al., 1995;Dobson and Franzmann, 1996), and the most recent focused on alternative phylogenetic markers, such as the 23S rRNA gene (Arahal et al., 2002;de la Haba et al., 2010) and multilocus sequence analysis based on single-copy housekeeping genes (de la Haba et al., 2012).Those approaches allowed, at that time, to establish the phylogenetic relationships among the members of this family and to propose taxonomic rearrangements accordingly.More than 10 years later, the number of species within the Halomonadaceae has doubled, and many of the type strains have a publicly available genome sequence.To the best of our knowledge, no studies specifically addressing the taxonomic status of the family Halomonadaceae in the genomic era have been conducted.
Therefore, in this study, we re-evaluate the genetic relationship, phylogeny, and taxonomy of the species and genera within the present family Halomonadaceae, as delineated in BMSAB, with the primary focus on its type genus.For this, we considered various overall genome relatedness indexes (OGRIs) based on average nucleotide identity (ANI), digital DNA-DNA hybridization (dDDH), average amino acid identity (AAI), core-protein average amino acid identity (cAAI), and the core-genome and GTDB phylogenies together with a signature gene analysis.To resolve the polyphyly within the genus Halomonas, we propose a division of the genus into seven different genera and the reclassification of several Halomonas species into the genus Modicisalibacter.Furthermore, the present study demonstrated the need to exclude the genus Halovibrio from the family Halomonadaceae and the existence of nine synonym species across this family.
Genome Portal, 5 and Global Catalogue of Type Strain (Shi et al., 2021).A total of 147 genome sequences of distinct type strains of species of the family Halomonadaceae (as per the BMSAB resource) and the genera Halovibrio and Terasakiispira were available in those public databases at the beginning of this study.Since some databases searched contained redundant data, the final set of genomes used in this research was only retrieved from the NCBI Assembly and NCBI SRA databases.One out of those 147 available genomes was discarded due to a failed quality check according to GTDB (Parks et al., 2018), while another marked as contaminated and excluded from RefSeq was finally retained after the verification of its contamination and completeness in GTDB, giving a total number of 146 genome sequences (Supplementary Table 1).All except one of the downloaded genomic data were assembled either at contig, scaffold, or chromosome level.The remaining one was only available in the NCBI SRA database, and thus, only raw reads could be retrieved, which were processed as indicated below.
To make our study as comprehensive as possible, we sequenced the genomes of an additional 17 type strains of species of Halomonadaceae that were not publicly available when the analyses were designed and performed (Supplementary Table 1).Seven out of those type strains were sequenced and assembled by the Global Catalogue of Microorganisms (Shi et al., 2021) as part of the type strain sequencing project (Wu and Ma, 2019).The genomic DNA of the other 10 type strains, retrieved from culture collections, was extracted and purified as described elsewhere (Galisteo et al., 2022) and further sequenced by Novogene Europe (Cambridge, United Kingdom) on the Illumina NovaSeq PE150 platform after library preparation of genomic material using the Novogene NGS DNA Library Prep Set (Cat.No. PT004).Raw reads were quality trimmed and filtered with BBTools v.38.44 (Bushnell, 2020) and further assembled with SPAdes v3.15.2 (option "-isolate") (Prjibelski et al., 2020).
For all the retrieved or obtained sequences, assembly quality, and basic statistics were estimated with QUAST v.2.3 (Gurevich et al., 2013).Genome sequence completeness and contamination were verified by using CheckM v.1.1.5(Parks et al., 2015).Prodigal v.2.6.3 (Hyatt et al., 2010) was employed to extract the translated coding sequences from the assembled genomes, and Prokka v.1.14.6 (Seemann, 2014) was employed for the automated annotation.When more than one genome sequence was available for the same type strain, we selected the one with a better assessment of contig number, completeness, contamination, 16S rRNA gene sequence presence and length, and the presence of 20 essential amino acids coded by distinct tRNA-coding genes, resulting in a final set of 161 genomes of type strains of the family Halomonadaceae.

Pangenome and core-genome inference
The complete pangenome of the selected set of genome sequences was calculated with Anvi' o suite v.7.1 (Eren et al., 2021), using BLASTP+ (Camacho et al., 2009) as the amino acid sequence similarity search program.Specifically, the "anvi-pan-genome" script was used with default values except that the MCL inflation parameter was set to 5 (in order to increase the sensitivity of the Markov Cluster Algorithm) (van Dongen, 2008), and the minimum percent identity between two amino acid sequences was set to 40%.The pangenome database containing orthologous gene clusters was employed to extract the gene family presence/absence matrix ("anvi-computefunctional-enrichment-in-pan" script) as well as the translated singlecopy core genes present in at least 90% of the analyzed genomes, denoted as "core90" ("anvi-get-sequences-for-gene-clusters" script).

Estimation of amino acid substitution models
Model-based phylogenetic analyses of protein sequences strongly rely on amino acid substitution models, which are generally summarized in a 20-by-20 replacement matrix, designated as Q matrices.Since those matrices are computationally very expensive to estimate due to their high parametrization, they are not usually estimated during a phylogenetic analysis but selected from a pre-estimated set of Q matrices using model selection software (Minh et al., 2021).
With the aim of constructing very reliable phylogenies, in this study, we estimated empirical general time-reversible Q matrices customized for the family Halomonadaceae derived from multiple sequence alignments of selected gene sets.For that purpose, QMarker software (Minh et al., 2021), which follows a maximum-likelihood approach, was used.Two gene datasets were chosen to generate Halomonadaceae-specific Q matrices: the "core90" set and the "bac120" set (120 single-copy bacterial proteins used by GTDB taxonomy) (Parks et al., 2018).

Phylogenomic reconstructions and overall genome relatedness indexes
The trimmed and concatenated "core90" protein alignment was used to infer the maximum-likelihood phylogeny of the family Halomonadaceae with the IQ-TREE v.2.2.0 program (Minh et al., 2020), using the custom Q_core90 reversible matrix as the substitution model combined with profile mixture models C10-C60 (Quang et al., 2008).The best-fit model parameters (mixture models, amino acid frequencies, and rate heterogeneity across sites) were determined with ModelFinder (Kalyaanamoorthy et al., 2017) according to Bayesian Information Criterion (BIC) (Schwarz, 1978).Branch support was estimated via 1,000 ultrafast bootstrap approximations (Hoang et al., 2018).Normalization of taxonomic ranks at the genus level according to relative evolutionary divergence (RED) was carried out with PhyloRank. 7 second phylogenomic tree including type and non-type strains as well as metagenome-assembled genomes (MAGs) was constructed to assess the cluster stability when new genome sequences are added to the tree and to analyze the relationships between members of the family Halomonadaceae and other closely related genera of the order Oceanospirillales (i.e., Halospina, Marinospirillum, Oceanospirillum, and Pseudospirillum).This second tree was based on the "bac120" protein set retrieved from GTDB Release 202 data files using GTDB-Tk v. 1.7.0 (Chaumeil et al., 2019).However, unlike the original GTDB-Tk pipeline that randomly selected 42 columns per marker to reduce computational requirements, we obtained the full-length multiple sequence alignments for each protein, which were later trimmed and concatenated as explained above for the "core90" dataset.Phylogenetic inference was carried out as stated for the first tree, but the Q_bac120 reversible matrix was used instead.Arrangements and visualization of both trees were accomplished by using ARB v.7.0 (Westram et al., 2011).
Commonly used OGRIs for taxonomic purposes were pairwise calculated among the analyzed genome set.ANI metrics were estimated with PyANI v.0.2.10 (Pritchard et al., 2016) using the ANIb method (i.e., BLASTN+ to align 1,020-bp fragments of the input sequences) (Goris et al., 2007).The genome-to-genome distance calculator (GGDC) webserver was used to infer dDDH relatedness according to formula 2 (Meier-Kolthoff et al., 2013).AAI values were determined using the "aai.rb"script from the Enveomics collection (Rodriguez-R and Konstantinidis, 2016).Since horizontal gene transfer (HGT) events can distort the molecular clock of bacterial evolution (Novichkov et al., 2004) and can, thus, impact the pairwise AAI values (Zheng et al., 2020), we calculated an additional genome relatedness index denoted as cAAI, similar to AAI but based on the protein sequences of core orthologous gene families, which are less impacted by HGT.For that purpose, the previously defined "core90" protein dataset (unaligned and untrimmed) was selected to calculate cAAI values with the "aai.rb"script.

Representation of clade-specific signature genes
The gene family presence/absence matrix inferred from the pangenome was used to detect "signature genes, " which are defined as gene families exclusive to specific phylogenetic clades (phylogroups), that is, gene families present in all species of a clade and absent in all other species (Zheng et al., 2020).Several phylogroups were investigated, aimed at testing different taxonomic arrangements and splitting into the family Halomonadaceae.For each phylogroup, a representative species was chosen, and the remaining species under study were assigned to the phylogroup of the representative species with whom they shared a most recent common ancestor.
The signature gene UpSet plots were generated by gathering gene families with the same presence/absence pattern across the analyzed genomes, ruling out two types of trivial patterns: gene families present in a single genome (singleton genes) and gene families detected in all genomes (core genes).The patterns were then displayed in descending order of frequency (number of gene families).R packages "tidyverse" v.1.2.1 (Wickham et al., 2019) and "tidygenomes" v.0.1.3, 8as well as other R functions and scripts, 9 were used for data processing and visualization of gene family presence/absence and signature genes.

Core-genome phylogenomics suggest several rearrangements in the family Halomonadaceae
A genome-based analysis was carried out aimed at shedding light on the relationships among species and genera of the family Halomonadaceae.The pangenome of the 161 genome sequences, which included type strains of the species belonging to the family Halomonadaceae, consisted of 608,808 genes organized into 59,992 orthologous gene clusters.Only 55 single-copy core gene clusters were detected across all genomes, so a more relaxed set of 189 orthologous genes shared by at least 90% of the genomes ("core90") encompassing a total of 23,924 residues after alignment and trimming was selected to calculate a Halomonadaceae-specific Q reversible matrix (Q_core90; Supplementary Table 2) and to infer a phylogenomic tree (Figure 1 and Supplementary Figure 1).The best-fit model selected by ModelFinder according to the BIC criterion to construct the tree was Q_core90 + C60 + F + R10.The resulting tree showed that, apart from the genus Halomonas, all the remaining genera within this family formed monophyletic groups.Noteworthy, the genera Terasakiispira and primarily Halovibrio were by far the most distantly related to the other genera within Halomonadaceae, suggesting that they might not belong to this family (see below).
Given the polyphyletic origin of the genus Halomonas and the heterogeneous characteristics of the species it harbors, we evaluated several proposals to split this genus into several genera with the following constraints: (i) the new proposed genera must be monophyletic; (ii) they must be supported by a 100% bootstrap value when possible; and (iii) they must fall within the relative evolutionary divergence (RED) interval defined for the rank of genus (Parks et al., 2018).The first attempt (proposal I) was to divide the genus Halomonas into eight phylogroups, five of them denoted according to GTDB taxonomy as  (Halomonas + Halomonas_G) supported "only" by a 97% bootstrap, which does not meet the desired constraints indicated above but is high enough to consider this option of grouping.To select the best-fitting scenario for this genus reclassification among the above proposals, we performed further comparative genomic analyses.

Assessment on the basis of AAI and cAAI values does not provide sound clues to split the genus Halomonas
To evaluate the proposed phylogroups under the four scenarios, we plotted the all-vs-all AAI and cAAI results clustered by the phylogroup while distinguishing between intra-genus and inter-genus values (Figure 2).Theoretically, a well-delimited genus or phylogroup should not display any overlap between intra-clade and inter-clade AAI/cAAI values.As shown in Figure 2, all the currently existing genera within the family Halomonadaceae harboring more than a single species except Halomonas (sensu stricto) (i.e., Aidingimonas, Salinicola, Chromohalobacter, Kushneria, Larsenimonas, and Cobetia), as well as the new phylogroups shared by the four proposals (Halomonas_B, Halomonas_F, Halomonas_E, Halomonas_A, and the enlarged Modicisalibacter) showed minimal or no overlap between the inter-and intra-cluster AAI and cAAI values, confirming their clear separation and exclusivity as different genera within this family.
By contrast, the intra-group AAI and cAAI values largely overlapped with their inter-group counterparts for the phylogroups Halomonas, Halomonas_G, Halomonas_H, and Halomonas_I (proposal I), indicating their heterogeneity as well as the lack of clear differentiation among them.No significant overlap reduction was observed after the consecutive merger of these phylogroups (proposals II, III, and IV), although proposals II and III led to slightly decreased overlap (Supplementary Figures 2-4).Based on the above observations, we believe that AAI and cAAI data may be useful for the delineation of most of the genera of the family Halomonadaceae, but they are not reliable for fine phylogroup demarcation within the genus Halomonas.The lowest intra-clade AAI and cAAI values for the current genera and proposals I, II, and III were 66.2 and 75.9%, respectively, while the cutoffs for proposal IV were 65.5 and 73.2%, respectively.These threshold values might be useful for future genus circumscription within the family Halomonadaceae.

Signature genes support the inclusion of Halomonas_G into the genus Halomonas
Since AAI and cAAI values did not allow us to determine the best of the four proposals to split the genus Halomonas, we analyzed whether the currently existing genera and suggested phylogroups can be defined by clade-specific signature genes.For that purpose, we examined the gene family presence/absence matrix extracted from the pangenome of the family Halomonadaceae to detect gene families present in all genomes of a certain group but lacking in all other genomes.Signature genes may be the result of a shared evolutionary history within a phylogenetic clade, or they could be acquired by horizontal gene transfer reflecting common lifestyles, ecologies, and physiological properties (Zheng et al., 2020).Therefore, signature genes enable estimates of the evolutionary forces that shaped the cluster, and, thus, they might be of help to opt for one of the proposals over the others.
Except for Halomonas, all the current genera of the family Halomonadaceae according to the BMSAB classification containing more than a single species, together with Halomonas_B, Halomonas_E, Halomonas_A, the newly proposed phylogroups Halomonas_H and Halomonas_I, and the enlarged Modicisalibacter, were well supported by signature genes ranging from 5 to 548 (Figure 3 and Supplementary Figure 5).As can be expected, phylogroups/genera with a smaller number of species showed, generally, a larger number of signature genes; nevertheless, the impact of the clade size is limited, as can be evidenced from larger genera, such as Salinicola and Kushneria, displaying a relatively large number of signature genes.Considering that singleton genes (those present in a single genome) were omitted from the analysis, it was impossible to evaluate the signature genes harbored by phylogroups/genera containing only one species (i.e., Halomonas_F, "Phytohalomonas, " Carnimonas, Halotalea, and Zymobacter).Interestingly, no signature genes were identified for phylogroups Halomonas (sensu stricto) and Halomonas_G, which a priori dismisses proposal I.However, when merging these phylogroups (proposal II), a signature gene was detected (Figure 3).Although it is a single signature gene, it is conserved across a large number of genomes (40), comprising the phylogroups Halomonas and Halomonas_G and, therefore, it likely reflects a common evolutionary history of both phylogroups providing an important hint into the delineation of the genus Halomonas.Proposal III, which considers the merging of phylogroups Halomonas, Halomonas_G, and Halomonas_H, was also endorsed by a single signature gene (Supplementary Figure 5); however, proposal II is favored over proposal III due to the fact that phylogroup Halomonas_H is very well supported by 21 signature genes and, thus, it is preferable to keep it as a separate cluster.The last proposal, proposal IV, consisting of the fusion of phylogroups Halomonas, Halomonas_G, Halomonas_H, and Halomonas_I, can also be rejected because the combined clade did not show any signature genes (Supplementary Figure 5) and, in addition, phylogroup Halomonas_I was supported by five signature genes.

GTDB-based phylogeny is consistent with proposal II and delineates the family Halomonadaceae
Although signature gene analyses endorsed proposal II as the most reliable one to split the genus Halomonas, this proposal involves the creation of a cluster with "only" a 97% branch bootstrap, as stated above.While this support value cannot be regarded as too low, we attempted to establish a long-term and trustworthy classification of the family for which only 100% supported phylogroups are preferred.This concern raises questions about the suitability of proposal II, making it necessary to provide some additional evidence to accept or reject it.
Accordingly, an alternative family-specific Q time-reversible matrix (Q_bac120; Supplementary Table 3) and phylogenomic tree (Figure 4) were calculated on the basis of the "bac120" marker protein dataset used to infer the bacterial GTDB taxonomy (Parks et al., 2018), comprising a total of 32,354 aligned columns after trimming.These ubiquitous single-copy proteins have been identified as being suitable for phylogenetic inference (Parks et al., 2017) and may yield better resolved trees than those obtained by using the almost entire set of core proteins ("core90" set).In fact, some of the "core90" non-curated protein-coding genes might have been laterally transferred or undergone homologous recombination (Stott and Bobay, 2020) or might lack congruent phylogenetic signals or sufficient homology to make comparisons valid and conclusive (Wu et al., 2013;Capella-Gutierrez et al., 2014).In addition, genomes from non-type strains and MAGs were also included in this complementary reconstruction to evaluate tree topology preservation.
The topology of the "bac120"-based tree (Figure 4) recovered the same phylogroups observed in the "core90"-based phylogeny (Figure 1).Although it is well-known that the selection of different phylogenetic markers can result in different topologies (Capella-Gutierrez et al., 2014;Ankenbrand and Keller, 2016;Tsai et al., 2019), in our analysis, the bias associated with marker choice did not affect the delineation of genera and phylogroups, demonstrating the stability of the obtained genome-based trees with independence of the protein set used for their inference.A good agreement between core-genome and "bac120" phylogenies has been previously demonstrated for the taxa of the Pseudomonadaceae (Lalucat et al., 2020), which is another family in the class Gammaproteobacteria.The inclusion of 82 additional GTDB representative genomes from non-type strains and MAGs of members of the family Halomonadaceae, as well as genomes from the closest genera of the order Oceanospirillales, did not alter the retrieved clades either (Figure 4).The most remarkable difference between our "core90" and "bac120" trees is the stronger support in the latter of the branch collapsing phylogroups Halomonas (sensu stricto) and Halomonas_G (corresponding to proposal II).Hence, this 100% bootstrap value recovered in the "bac120" tree after merging both phylogroups, together with the signature gene analysis, enables proposal II as the most appropriate to prune the genus Halomonas to the species comprising phylogroups Halomonas (sensu stricto) and Halomonas_G, while transferring the remaining species to six new genera and to the already described genus Modicisalibacter.
Concerning the taxonomic status of the family Halomonadaceae, the "bac120"-based phylogeny evidenced that all 12 genera currently affiliated to this family according to BMSAB, together with the non-validly published genus name "Phytohalomonas, " formed a monophyletic group of microorganisms (Figure 4), whose shared feature, which was at the same time differential to the closest related genera, was their halophily or halotolerance.The genus Halovibrio, also considered a member of the family Halomonadaceae by LPSN, was intimately related to the genus Halospina, belonging to the family Hahellaceae; 10,11 whereas the genus Terasakiispira, included into the Halomonadaceae by both LPSN and GTDB resources, was rather a taxon within the family Oceanospirillaceae given its closest relationship to the genera Marinospirillum, Oceanospirillum, and Pseudospirillum (Figure 4).GTDB taxonomy (release 08-RS214) has suggested the incorporation of these three genera into the family Halomonadaceae, even if they harbor marine (slight) halophiles in contrast to the majority of moderately halophilic bacteria characteristic of the family Halomonadaceae.Therefore, it becomes clear that the genus Halovibrio should be kept apart from the family Halomonadaceae, while further investigation including genomic, phylogenomic, and phenotypic comparisons is required to delineate the taxonomic affiliation of the genera Terasakiispira, Marinospirillum, Oceanospirillum, and Pseudospirillum at the family level.Gene family presence/absence patterns inferred from the pangenome of the family Halomonadaceae arranged according to proposal II.Each column represents a gene family pattern, where presence is indicated with a dot in the corresponding species.The absolute number of gene families that conform to each pattern is visualized in the marginal bar plot at the top.Separations between phylogroups/genera are indicated with horizontal black lines and the representative species of each phylogroup/genus is highlighted in bold.Genes that were present in all genomes of a clade and in none of the genomes outside of that clade, denoted as "signature genes," are displayed in dark blue; other genes are shown in light blue.Patterns of presence in a single species or all species are not shown.The species Halovibrio salipaludis, Halovibrio variabilis, and Terasakiispira papahanaumokuakeensis were not considered members of the Halomonadaceae and were only used as an "outgroup".

Heterotypic synonyms revealed by ANI and dDDH values
Phylogenomic inference using both the "core90" and the "bac120" protein sets revealed the presence of some very closely related species pairs and even one set of three among the studied taxa that might be considered cases of heterotypic synonymy.To verify those hypotheses, ANI (Figure 5) and dDDH relatedness indexes were estimated between the genomes in question.It is widely accepted that cutoff values for species delineation based on these genomic indexes are 95-96% for ANI (Goris et al., 2007;Richter and Rossello-Mora, 2009;Chun and Rainey, 2014) and 70% for dDDH (Auch et al., 2010).Therefore, two or more strains can be considered to belong to the same species if both ANI values are ≥96% and dDDH values are ≥70%.According to this criterion, our results confirmed the existence of the following sets of heterotypic synonyms (ANI and dDDH values in parenthesis, respectively): Chromohalobacter israelensis -Chromohalobacter salexigens (98.1 and 83.6%); Halomonas alkaliphila -"Halomonas humidisoli" (97.7 and 80.5%); Halomonas antri -"Halomonas sulfidivorans" (97.

Taxonomic conclusions
The comparative genomic analyses (phylogenomics, OGRI, and signature genes) conducted among the type strains of the species of the family Halomonadaceae allow us to propose several taxonomic rearrangements within this family: i.The genus GTDB-derived maximum-likelihood phylogenomic tree based on the concatenation of 120 single-copy bacterial proteins ("bac120" set) showing the relationships among members of the family Halomonadaceae and related taxa.The type strain of the species Pistricoccus aurantiacus is missing because no genome sequences from this strain could be recovered.could not be sequenced when this study was accomplished, together with the new species described after the finalization of this research, should be further studied to determine their correct placement according to the taxonomy proposed here or if they may constitute new genera within the family Halomonadaceae.
Moreover, we strongly recommend the use of the clade-specific amino acid substitution matrices Q_core90 and Q_bac120, empirically calculated for the family Halomonadaceae, for future phylogenomic studies within this family instead of the pre-computed Q matrices generally selected by model selection software.
Description of Vreelandella gen.nov.
The DNA G + C content ranges between 52.1 and 63.8 mol%.

Description of
The DNA G + C content is 52.6-53.2mol%.The type strain is 86 T = ATCC 27125 T = CCUG 16063 T = CIP 103201 T = DSM 4743 T = JCM 20634 T = LMG 3445 T = NBRC 102221 T .The genome size of the type strain is 4.27 Mbp, and its DNA G + C content is 52.6 mol%.Isolated from seawater (off the coast of Oahu, Hawaii, United States).
Halomonas hydrothermalis should be regarded as a heterotypic synonym of Vreelandella venusta.The description is as given in the proposal of the basonym (Franzmann et al., 1987), with the following addition.The genome size of the type strain is 3.11 Mbp.The DNA G + C content is 60.8 mol%.
The DNA G + C content is 54.8-55.0mol%.The type strain is Eplume1 T = ATCC BAA-805 T = CCM 7107 T = CECT 5815 T = DSM 15720 T .The genome size of the type strain is 4.93 Mbp, and its DNA G + C content is 55.0 mol%.Isolated from a deep-sea hydrothermal plume (NE Pacific Ocean, Juan de Fuca Ridge).
Halomonas alkaliantarctica should be regarded as a heterotypic synonym of Vreelandella neptunia.The description is as given in the proposal of the basonym (Kaye et al., 2004), with the following addition.The genome size of the type strain is 4.48 Mbp.The DNA G + C content is 53.7 mol%.
Isolated from a deep-sea sulfide rock (NE Pacific Ocean, Juan de Fuca Ridge).
The description is as given in the proposal of the basonym (Quillaguamán et al., 2004), with the following addition.The genome size of the type strain is 4.21 Mbp.The DNA G + C content is 54.7 mol%.
Isolated from the soil around a Bolivian hypersaline lake.The type strain is LC1 T = ATCC BAA-759 T = DSM 15516 T .Type strain genome sequence accession numbers: GCA_000236035.1 and GCA_002265845.1.
The description is as given in the proposal of the basonym (Sorokin and Tindall, 2006), with the following addition.The genome size of the type strain is 3.73 Mbp.The DNA G + C content is 55.8 mol%.
Isolated from surface water from the North Arm of Great Salt Lake (United States).
The description is as given in the proposal of the basonym (Kim et al., 2007), with the following addition.The genome size of the type strain is 3.73 Mbp.The DNA G + C content is 59.8 mol%.
Isolated from saline water of the Gomseom solar saltern in Anmyeondo (Korea).
The description is as given in the proposal of the basonym (Kim et al., 2007), with the following addition.The genome size of the type strain is 3.87 Mbp.The DNA G + C content is 57.3 mol%.
Isolated from saline water of the Janggok solar saltern in Anmyeondo (Korea).

Description of Vreelandella arcis comb. nov.
Vreelandella arcis (ar' cis.L. gen.n. arcis, of a height, summit, or peak, referring to the isolation of the organism from a salt lake on the Qinghai-Tibet Plateau).
Basonym: Halomonas arcis Xu et al. 2007.The description is as given in the proposal of the basonym (Xu et al., 2007), with the following addition.The genome size of the type strain is 4.14 Mbp.The DNA G + C content is 55.9 mol%.
Isolated from the water of a salt lake located in Altun Mountain on the Qinghai-Tibet Plateau (China).
Type strain 16S rRNA gene sequence accession number: EF144147.The description is as given in the proposal of the basonym (Xu et al., 2007), with the following addition.The genome size of the type strain is 3.73 Mbp.The DNA G + C content is 58.0 mol%.
Isolated from water of a subterranean saline well at Si-Chuan Basin (China).

Description of
The DNA G + C content is 52.5-52.6 mol%.The type strain is 18bAG T = ATCC BAA-953 T = DSM 16354 T .The genome size of the type strain is 4.10 Mbp, and its DNA G + C content is 52.6 mol%.Isolated from a salt pool in Campania (Italy).

Basonym: Halomonas zhanjiangensis Chen et al., 2009.
The description is as given in the proposal of the basonym (Chen et al., 2009), with the following addition.The genome size of the type strain is 4.06 Mbp.The DNA G + C content is 54.5 mol%.
Isolated from a sea urchin, Hemicentrotus pulcherrimus, South China Sea, tidal flat of Naozhou Island near Zhanjiang (China).

Description of
Type strain 16S rRNA gene sequence accession number: AM941388.

Description of Vreelandella hamiltonii comb. nov.
Vreelandella hamiltonii (ha.mil.to'ni.i.N.L. gen.n. hamiltonii, of Hamilton, named after Dr. John R. Hamilton, a microbiologist who isolated and characterized the first strains). Basonym: The description is as given in the proposal of the basonym (Kim K. K. et al., 2010), with the following addition.Cells are 0.7-1.0 × 1.5-4.0μm in size and motile with lateral or lateral/polar flagella.Optimal growth occurs at sea-salt concentrations of 0-7.5%, w/v.Grown on cetrimide agar.Nitrate reduction to nitrite, the Voges-Proskauer test, and urease are variable.Utilization of adipate, D-galactose, malonate, L-lysine, DL-isoleucine, and L-valine as sole sources of carbon (nitrogen) and energy is variable.Acid production from L-arabinose, D-fucose, D-galactose, methyl α-D-glucoside, glycerol, D-mannitol, and melezitose is variable.Susceptibility to neomycin, penicillin G, and chloramphenicol is strain-dependent.The DNA G + C content is 60.1 mol%.
The type strain is W1025 T = DSM 21196 T = KCTC 22154 T .The genome size of the type strain is 3.93 Mbp, and its DNA G + C content is 60.1 mol%.Isolated from the blood of a dialysis machine drain at California, San Jose, Santa Clara Valley Medical Center (USA).
Halomonas johnsoniae should be regarded as a heterotypic synonym of Vreelandella hamiltonii.
Basonym: Halomonas andesensis Guzmán et al. 2010.The description is as given in the proposal of the basonym (Guzmán et al., 2010), with the following addition.The genome size of the type strain is 3.91 Mbp.The DNA G + C content is 52.1 mol%.
Basonym: Halomonas vilamensis Menes et al. 2011.The description is as given in the proposal of the basonym (Menes et al., 2011), with the following addition.The genome size of the type strain is 3.47 Mbp.The DNA G + C content is 55.2 mol%.

Description of
The type strain is YIM M 13059 T = CCTCC AB 2012911 T = DSM 25561 T = JCM 18142 T .

Description of Vreelandella olivaria comb. nov.
Vreelandella olivaria (o.li.va'ri.a.L. fem.adj.olivaria, of or belonging to olives, related to olive-processing effluents from where the type strain was isolated).
Basonym: Halomonas olivaria Amouric et al. 2014.The description is as given in the proposal of the basonym (Amouric et al., 2014), with the following addition.The genome size of the type strain is 5.00 Mbp.The DNA G + C content is 55.3 mol%.
Isolated from salted olive-processing effluents from an evaporation pond (Morocco).
Type strain 16S rRNA gene sequence accession number: DQ645593.The description is as given in the proposal of the basonym (Jiang et al., 2014), with the following addition.The genome size of the type strain is 3.69 Mbp.The DNA G + C content is 59.1 mol%.

Description of Vreelandella salicampi comb. nov.
Vreelandella salicampi (sa.li.cam'pi.L. masc.n. sal, salt; L. masc.n. campus, field; N.L. gen.n. salicampi, of a salt field). Basonym: The description is as given in the proposal of the basonym (Lee et al., 2015), with the following addition.The genome size of the type strain is 3.86 Mbp.The DNA G + C content is 56.2 mol%.

Description of Vreelandella lutescens comb. nov.
Vreelandella lutescens (lu.tes' cens.L. part.adj.lutescens, becoming muddy, related to the muddy color of the mature colony). Basonym: The description is as given in the proposal of the basonym (Wang et al., 2016), with the following addition.The genome size of the type strain is 3.70 Mbp.The DNA G + C content is 56.0 mol%.
Isolated from a sediment sample from Qinghai Lake (China).The type strain is Q1U T = CGMCC 1.15122 T = KCTC 42517 T .Type strain genome sequence accession number: GCA_014640815.1.Type strain 16S rRNA gene sequence accession number: KP259554.The description is as given in the proposal of the basonym (Oguntoyinbo et al., 2018), with the following addition.The genome size of the type strain is 4.93 Mbp.The DNA G + C content is 52.8 mol%.
Type strain 16S rRNA gene sequence accession number: MG030686.The description is as given in the proposal of the basonym (Kämpfer et al., 2018), with the following addition.The genome size of the type strain is 3.61 Mbp.The DNA G + C content is 63.8 mol%.

Description of
Isolated from the rhizosphere of sand dune coastal plant, Coast of Malpe (India).
Basonym: Halomonas piezotolerans Yan et al. 2020.The description is as given in the proposal of the basonym (Yan et al., 2020), with the following addition.The genome size of the type strain is 3.95 Mbp.The DNA G + C content is 57.9 mol%.
Isolated from a deep-sea sediment sample of the New Britain Trench.
Basonym: Halomonas rituensis Gao et al. 2020.The description is as given in the proposal of the basonym (Gao et al., 2020), with the following addition.The genome size of the type strain is 4.47 Mbp.The DNA G + C content is 57.2 mol%.
Vreelandella zhuhanensis (zhu.han.en'sis.N.L. fem.adj.zhuhanensis, pertaining to Zhuhan marsh on the Tibetan Plateau, where the type strain was isolated). Basonym: The description is as given in the proposal of the basonym (Gao et al., 2020), with the following addition.The genome size of the type strain is 3.25 Mbp.The DNA G + C content is 57.1 mol%.
Vreelandella azerica (a.ze'ri.ca.N.L. fem.adj.azerica, pertaining to Azerbaijan, where the Urmia Lake is located and the type strain was isolated).
Basonym: Halomonas azerica Wenting et al. 2021.The description is as given in the proposal of the basonym (Wenting et al., 2021), with the following addition.The genome size of the type strain is 3.42 Mbp.The DNA G + C content is 55.4 mol%.

Description of Vreelandella profundi comb. nov.
Vreelandella profundi (pro.fun'di.L. gen.n. profundi, of the depth of the sea). Basonym: The description is as given in the proposal of the basonym (Wang et al., 2022), with the following addition.The genome size of the type strain is 3.60 Mbp.The DNA G + C content is 54.0 mol%.
Isolated from the deep-sea sediment of the Mariana Trench (11.12°N, 142.32°E).
Description of Vreelandella populi comb.nov.
Vreelandella populi (po'pu.li.L. gen.n. populi, of the genus Populus, referring to Populus euphratica). Basonym: The description is as given in the proposal of the basonym (Xu et al., 2021), with the following addition.The genome size of the type strain is 3.80 Mbp.The DNA G + C content is 55.0 mol%.
Isolated from Populus euphratica in Ebinur Lake Wetland Nature Reserve (China).
The description is as given in the original proposal of "Halomonas glaciei" (Reddy et al., 2003), with the following addition.The genome size of the type strain is 4.96 Mbp.The DNA G + C content is 54.4 mol%.
Isolated from the fast ice of Adelie Land, Antarctica.The type strain is DD 39 T = CGMCC 1.7263 T = JCM 11692 T = MTCC 4321 T .
The description is as given in the original proposal of "Halomonas zhaodongensis" (Jiang et al., 2013), with the following addition.The genome size of the type strain is 3.72 Mbp.The DNA G + C content is 53.0 mol%.

Description of Vreelandella lionensis sp. nov.
Vreelandella lionensis (li.on.en'sis.N.L. fem.adj.lionensis, of or belonging to Golfe du Lion [Gulf of Lions], in reference to the origin of the type strain).
The description is as given in the original proposal of "Halomonas lionensis" (Gaboyer et al., 2014), with the following addition.The genome size of the type strain is 3.65 Mbp.The DNA G + C content is 55.9 mol%.
Isolated from the Mediterranean Sea sediment, Gulf of Lions (France).
The description is as given in the original proposal of "Halomonas massiliensis" (Seck et al., 2016), with the following addition.The genome size of the type strain is 3.44 Mbp.The DNA G + C content is 58.4 mol%.
Isolated from the human gut (France).

Description of Vreelandella maris sp. nov.
Vreelandella maris (ma'ris.L. gen.n. maris, of the sea).The description is as given in the original proposal of "Halomonas maris" (Qiu et al., 2021a), with the following addition.The genome size of the type strain is 4.52 Mbp.The DNA G + C content is 54.4 mol%.
Isolated from the deep-sea sediment in the Southwest Indian Ocean (China).
Vreelandella sedimenti (se.di.men'ti.L. gen.n. sedimenti, of sediment, referring to the sediment of the Southwest Indian Ocean, where the type strain was isolated).
The description is as given in the original proposal of "Halomonas sedimenti" (Qiu et al., 2021b), with the following addition.The genome size of the type strain is 5.06 Mbp.The DNA G + C content is 54.3 mol%.
Isolated from the deep-sea sediment in the Southwest Indian Ocean (China).
Description of Bisbaumannia gen.nov.
The DNA G + C content is 67.2 mol%.The genus Bisbaumannia belongs to the family Halomonadaceae.The type species is Bisbaumannia pacifica.
Homotypic synomyms: Halomonas pacifica (Baumann et al. 1972) Dobson and Franzmann 1996; Deleya pacifica (Baumann et al. 1972) Baumann et al. 1983.The description is as given in the original proposal of the basonym (Baumann et al., 1972), with the following addition.The genome size of the type strain is 3.85 Mbp.The DNA G + C content is 67.2 mol%.
Isolated from the seawater off the coast of Oahu (Hawaii, United States), Pacific Ocean.
Description of Billgrantia gen.nov.Billgrantia (Bill.grant'i.a.N.L. fem.n.Billgrantia, named after the microbiologist William [Bill] D. Grant for his great contribution to the study of halophilic microorganisms).
The DNA G + C content ranges between 62.1 and 67.5 mol%.
The genus Billgrantia belongs to the family Halomonadaceae.The type species is Billgrantia desiderata.
Description of Billgrantia desiderata comb.nov.
Basonym: Halomonas desiderata Berendes et al. 1997.The description is as given in the original proposal of the basonym (Berendes et al., 1996), with the following addition.The DNA G + C content is 64.7-64.9mol%.
The type strain is FB2 T = CIP 105505 T = DSM 9502 T = LMG 19548 T .The genome size of the type strain is 4.89 Mbp, and its DNA G + C content is 64.7 mol%.Isolated from a municipal sewage treatment plant in Göttingen (Germany).

Description of Billgrantia campisalis comb. nov.
Billgrantia campisalis (cam.pi.sa'lis.L. masc.n. campus, field, plain; L. masc.n. sal, salt; N.L. gen.n. campisalis, of the plain of salt, of the salt plain). Basonym: The description is as given in the original proposal of the basonym (Mormile et al., 1999), with the following addition.The genome size of the type strain is 4.27 Mbp.The DNA G + C content is 66.3 mol%.
Isolated from a soil sample collected from a salt flat south of Alkali Lake, Washington State (USA).
The type strain is 4A T = ATCC 700597 T = CIP 106639 T = DSM 15413 T .
Basonym: Halomonas gudaonensis Wang et al. 2007.The description is as given in the original proposal of the basonym (Wang et al., 2007), with the following addition.The genome size of the type strain is 4.17 Mbp.The DNA G + C content is 64.9 mol%.
Isolated from saline soil contaminated by crude oil, Gudao (China).

Description of Billgrantia kenyensis comb. nov.
Billgrantia kenyensis (ke.ny.en'sis.N.L. fem.adj.kenyensis, Kenyan, of Kenya, the region of isolation). Basonym: The description is as given in the original proposal of the basonym (Boltyanskaya et al., 2007), with the following addition.The genome size of the type strain is 4.42 Mbp.The DNA G + C content is 63.8 mol%.
Isolated from sediments from soda lakes (Kenya).
Basonym: Halomonas endophytica Chen et al. 2018.The description is as given in the original proposal of the basonym (Chen et al., 2018), with the following addition.The genome size of the type strain is 4.98 Mbp.The DNA G + C content is 62.1 mol%.
Isolated from liquid in the stems of Populus euphratica in Xinjiang (China).
Type strain 16S rRNA gene sequence accession number: MF850257.The description is as given in the original proposal of the basonym (Lu et al., 2020), with the following addition.The genome size of the type strain is 4.79 Mbp.The DNA G + C content is 62.9 mol%.

Description of
Isolated from hypersaline Lake Pengyanco on the Tibetan Plateau (China).

Description of
Basonym: Halomonas lactosivorans Ming et al. 2020.The description is as given in the original proposal of the basonym (Ming et al., 2020), with the following addition.The genome size of the type strain is 4.36 Mbp.The DNA G + C content is 66.7 mol%.
Basonym: Halomonas pellis Li et al. 2020.The description is as given in the original proposal of the basonym (Li et al., 2020), with the following addition.The genome size of the type strain is 4.35 Mbp.The DNA G + C content is 63.6 mol%.
Billgrantia azerbaijanica (a.zer.bai.ja'ni.ca.N.L. fem.adj.azerbaijanica, pertaining to Azerbaijan, a region in the north-west of Iran, where Urmia Lake is located and from which the type strain was isolated).
Basonym: Halomonas azerbaijanica Kazemi et al. 2021.The description is as given in the original proposal of the basonym (Kazemi et al., 2021), with the following addition.The genome size of the type strain is 4.58 Mbp.The DNA G + C content is 67.5 mol%.
Billgrantia diversa (di.ver'sa. L. fem. part. adj. diversa, different, distinct). Basonym: The description is as given in the original proposal of the basonym (Wang et al., 2021), with the following addition.The genome size of the type strain is 4.49 Mbp.The DNA G + C content is 62.9 mol%.
Isolated from the deep-sea sediment of the Pacific Ocean.The type strain is D167-6-1 T = KCTC 72441 T = MCCC 1A13316 T .Type strain genome sequence accession number: GCA_014931605.1.

Description of
The DNA G + C content is 64.2-64.3mol%.The type strain is Y3S6 T = KACC 21536 T = NBRC 114315 T = TBRC 15164 T .The genome size of the type strain is 4.39 Mbp, and its DNA G + C content is 64.3 mol%.Isolated from surface seawater, Busan (Korea).
Type strain 16S rRNA gene sequence accession number: MN625868.The description is as given in the original proposal of "Halomonas chromatireducens" (Shapovalova et al., 2009), with the following addition.The genome size of the type strain is 3.97 Mbp.The DNA G + C content is 62.8 mol%.
The description is as given in the original proposal of "Halomonas aerodenitrificans" (Wang and Shao, 2021), with the following addition.The genome size of the type strain is 5.08 Mbp.The DNA G + C content is 64.0 mol%.
Isolated from coastal water, 5 m depth, from the Taiwan Strait (China).
Billgrantia ethanolica (e.tha.no'li.ca. N.L. neut. n. ethanol, ethanol; L. fem. adj. suff.-ica, suffix used with various meanings; N.L. fem.adj.ethanolica, belonging to ethanol, in reference to the ability of the species to utilize ethanol as a substrate for growth).
The description is as given in the original proposal of "Halomonas ethanolica" (Wang and Shao, 2021), with the following addition.The genome size of the type strain is 4.57 Mbp.The DNA G + C content is 64.5 mol%.
Isolated from the sediment from shrimp culture pond, Zhangzhou (China).
The description is as given in the original proposal of "Halomonas sulfidoxydans" (Wang and Shao, 2021), with the following addition.The genome size of the type strain is 4.49 Mbp.The DNA G + C content is 66.0 mol%.
Isolated from surface sediments in the coastal sea at Taiwan Strait (China).
Billgrantia tianxiuensis (tian.xiu.en'sis.N.L. fem.adj.tianxiuensis, pertaining to the Tianxiu Hydrothermal Field, on the Northwest Indian Ridge, from where the type strain was isolated).
The description is as given in the original proposal of "Halomonas tianxiuensis" (Wang and Shao, 2021), with the following addition.The genome size of the type strain is 5.02 Mbp.The DNA G + C content is 63.9 mol%.
Isolated from sulfide from Tianxiu hydrothermal vents, 3,440 m depth, Northwest Indian Ocean.
The description is as given in the original proposal of "Halomonas zhangzhouensis" (Wang and Shao, 2021), with the following addition.The genome size of the type strain is 4.41 Mbp.The DNA G + C content is 63.3 mol%.
Isolated from the sediments from a shrimp culture pond in Zhangzhou (China).
The DNA G + C content ranges between 63.8 and 64.5 mol%.
The genus Franzmannia belongs to the family Halomonadaceae.The type species is Franzmannia pantelleriensis.(Romano et al., 1996), with the following addition.The genome size of the type strain is 4.40 Mbp.The DNA G + C content is 63.9 mol%.

Description of
Isolated from the hard sand of Venere Lake, Pantelleria Island (Italy).
The type strain is AAP T = ATCC 700273 T = CIP 105506 T = DSM 9661 T = LMG 19550 T .
Basonym: Halomonas qiaohouensis Wang et al. 2015.The description is as given in the proposal of the basonym (Wang et al., 2014), with the following addition.The genome size of the type strain is 4.65 Mbp.The DNA G + C content is 64.5 mol%.
Isolated from the salt mine soil in southwest China.The type strain is YIM QH88 T = ACCC 60021 T = CCTCC AB 2012965 T = DSM 26770 T .
The description is as given in the original proposal of "Halomonas salipaludis" (Xing et al., 2021), with the following addition.The genome size of the type strain is 5.48 Mbp.The DNA G + C content is 63.8 mol%.
The DNA G + C content ranges between 58.5 and 62.2 mol%.The genus Litchfieldella belongs to the family Halomonadaceae.The type species is Litchfieldella anticariensis.
Basonym: Halomonas anticariensis Martínez-Cánovas et al. 2004.The description is as given in the proposal of the basonym (Martínez-Cánovas et al., 2004), with the following addition.The genome size of the type strain is 5.07 Mbp.The DNA G + C content is 58.5 mol%.
Isolated from a soil sample from Fuente de Piedra, a salinewetland wildfowl reserve in Málaga (Spain).

Description of Litchfieldella qijiaojingensis comb. nov.
Litchfieldella qijiaojingensis (qi.jiao'jing.en'sis.N.L. fem.adj.qijiaojingensis, pertaining to Qijiaojing Lake, Xinjiang Province, north-west China, where the sample from which the type strain was isolated and was collected).
Basonym: Halomonas qijiaojingensis Chen et al. 2012.The description is as given in the proposal of the basonym (Chen et al., 2011), with the following addition.The genome size of the type strain is 4.77 Mbp.The DNA G + C content is 60.8 mol%.
Description of Onishia gen.nov.
Onishia (O.ni'shi.a.N.L. fem.n.Onishia, named after Dr. Hiroshi Ōnishi [Japan], in recognition of his many contributions to the study of halophilic microorganisms).
The DNA G + C content ranges between 61.1 and 62.3 mol%.
The genus Onishia belongs to the family Halomonadaceae.The type species is Onishia taeanensis.
Basonym: Halomonas taeanensis Lee et al. 2005.The description is as given in the proposal of the basonym (Lee et al., 2005), with the following addition.The genome size of the type strain is 3.76 Mbp.The DNA G + C content is 62.3 mol%.
Isolated from the soil from a solar saltern (Korea).The type strain is BH539 T = CIP 109003 T = DSM 16463 T = KCTC 12284 T .
Type strain 16S rRNA gene sequence accession number: AY671975.The description is as given in the proposal of the basonym (Diéguez et al., 2020), with the following addition.The genome size of the type strain is 3.68 Mbp.The DNA G + C content is 61.1 mol%.
The DNA G + C content ranges between 59.1 and 67.4 mol%.
The genus Modicisalibacter belongs to the family Halomonadaceae.The type species is Modicisalibacter tunisiensis.The description is as given in the proposal of the basonym (Heyrman et al., 2002), with the following addition.The genome size of the type strain is 4.14 Mbp.The DNA G + C content is 61.9 mol%.
Isolated from microbial biofilms colonizing the walls and murals of the Saint-Catherine chapel, Castle Herberstein (Austria).
The type strain is R-5058 T = CIP 108825 T = DSM 14789 T = LMG 20969 T .
Basonym: Halomonas lutea Wang et al. 2008.The description is as given in the proposal of the basonym (Wang et al., 2008), with the following addition.The genome size of the type strain is 4.53 Mbp.The DNA G + C content is 59.1 mol%.
Basonym: Halomonas ilicicola Arenas et al. 2009.The description is as given in the proposal of the basonym (Arenas et al., 2009), with the following addition.The genome size of the type strain is 3.96 Mbp.The DNA G + C content is 63.2 mol%.
Type strain 16S rRNA gene sequence accession number: EU218533.The description is as given in the proposal of the basonym (Zhao et al., 2012), with the following addition.The genome size of the type strain is 4.36 Mbp.The DNA G + C content is 61.3 mol%.

Description of
Isolated from saline soil contaminated with crude oil from Xianhe, Shandong Province (China).
Type strain 16S rRNA gene sequence accession number: EF421176.The description is as given previously (Xu et al., 2013;Navarro-Torre et al., 2020), with the following addition.The genome size of the type strain is 3.55 Mbp.The DNA G + C content is 64.4 mol%.

Description of
Isolated from deep-sea sediment from the South Atlantic Ocean.The type strain is B6 T = CGMCC 1.12450 T = JCM 18472 T .Type strain genome sequence accession number: GCA_000731955.1.
The description is as given in the proposal of the basonym (Navarro-Torre et al., 2020), with the following addition.The genome size of the type strain is 4.65 Mbp.The DNA G + C content is 64.9 mol%.
The description is as given in the original proposal of "Halomonas coralii" (Vidal et al., 2019), with the following addition.The genome size of the type strain is 4.44 Mbp.The DNA G + C content is 66.3 mol%.

:
Halomonas titanicae Mann et al. 2010.The description is as given in the proposal of the basonym (Sánchez-Porro et al., 2010), with the following addition.The genome size of the type strain is 5.34 Mbp.The DNA G + C content is 54.6 mol%.Isolated from the rusticles of the RMS Titanic wreck.The type strain is BH1 T = ATCC BAA-1257 T = CECT 7585 T = DSM 22872 T = JCM 16411 T = LMG 25388 T .
Modicisalibacter xianhensis, andModicisalibacter zincidurans, respectively.Therefore, it has been necessary to extend the description of the genus Modicisalibacter, as indicated below.
Vreelandella stevensii comb.nov.stevensii (ste.ven'si.i.N.L. gen.n. stevensii, of Stevens, named after Dr. David A. Stevens, a physician/epidemiologist who isolated and characterized the first strains).Basonym: Halomonas stevensii Kim et al. 2010.The description is as given in the proposal of the basonym (Kim K. K. et al., 2010), with the following addition.The genome size of the type strain is 3.69 Mbp.The DNA G + C content is 60.3 mol%.Isolated from the blood of a renal care patient at California, San Jose, Santa Clara Valley Medical Center (United States).
Vreelandella jeotgali comb.nov.Vreelandella nanhaiensis (nan.hai.en'sis.N.L. fem.adj.nanhaiensis, pertaining to Nanhai, a sea in South China where the sample was collected).Basonym: Halomonas nanhaiensis Long et al. 2013.The description is as given in the proposal of the basonym (Long et al., 2013), with the following addition.The genome size of the type strain is 4.03 Mbp.The DNA G + C content is 54.4 mol%.
Billgrantia bachuensis comb.nov.bachuensis (ba.chu.en'sis.N.L. fem.adj.bachuensis, pertaining to Bachu, north-western China, where the strain was isolated).Basonym: Halomonas bachuensis Xiao et al. 2021.The description is as given in the original proposal of the basonym (Xiao et al., 2021), with the following addition.The genome size of the type strain is 4.70 Mbp.The DNA G + C content is 63.6 mol%.Billgrantia antri (an'tri.L. gen.n. antri, of a cave, referring to the location of the isolate).Basonym: Halomonas antri So et al. 2022.
Franzmannia pantelleriensis comb.nov.pantelleriensis (pan.tel.le.ri.en'sis.L. fem.adj.pantelleriensis, pertaining to Pantelleria Island [the place of isolation] in the south of Sicily, Italy).Basonym: Halomonas pantelleriensis corrig Romano et al. 1997.The description is as given in the proposal of the basonym (Amjres et al., 2011)varro- Torre et al., 2020)ensis(xin.ji.ang.en'sis.N.L. fem.adj.xinjiangensis,pertaining to Xinjiang, a region of China, from where the type strain was isolated).Basonym:Halomonas xinjiangensis Guan et al. 2010.The description is as given previously(Guan et al., 2010;Navarro- Torre et al., 2020), with the following addition.The genome size of the type strain is 3.79 Mbp.The DNA G + C content is 60.7 mol%.Isolated from a soil sample from Lop Nur salt lake (4 ° 23′ N 9 ° 18′ E, 778 m altitude) in Xinjiang Province, north-west China.Litchfieldella rifensis (ri.fen'sis.N.L. fem.adj.rifensis,pertaining to the Rif Mountains in northern Morocco, where the strain was isolated).Basonym: Halomonas rifensis Amjres et al. 2011.The description is as given in the proposal of the basonym(Amjres et al., 2011), with the following addition.The genome size of the type strain is 4.83 Mbp.The DNA G + C content is 62.2 mol%.Isolated from a solar saltern in the Rif Mountains (Morocco).The type strain is HK31 T = CECT 7698 T = LMG 25695 T .