Chryseobacterium herbae Isolated from the Rhizospheric Soil of Pyrola calliantha H. Andres in Segrila Mountain on the Tibetan Plateau

A non-motile, Gram-staining-negative, orange-pigmented bacterium called herbae pc1-10T was discovered in Tibet in the soil around Pyrola calliantha H. Andres’ roots. The isolate thrived in the temperature range of 10–30 °C (optimal, 25 °C), pH range of 5.0–9.0 (optimum, pH = 6.0), and the NaCl concentration range of 0–1.8% (optimal, 0%). The DNA G+C content of the novel strain was 37.94 mol%. It showed the function of dissolving organophosphorus, acquiring iron from the environment by siderophore and producing indole acetic acid. Moreover, the genome of strain herbae pc1-10T harbors two antibiotic resistance genes (IND-4 and AdeF) encoding a β-lactamase, and the membrane fusion protein of the multidrug efflux complex AdeFGH; antibiotic-resistance-related proteins were detected using the Shotgun proteomics technology. The OrthoANIu values between strains Chryseobacterium herbae pc1-10T; Chryseobacterium oleae CT348T; Chryseobacterium kwangjuense KJ1R5T; and Chryseobacterium vrystaatense R-23566T were 90.94%, 82.96%, and 85.19%, respectively. The in silico DDH values between strains herbae pc1-10T; C. oleae CT348T; C. kwangjuense KJ1R5T; and C. vrystaatense R-23566T were 41.7%, 26.6%, and 29.7%, respectively. Chryseobacterium oleae, Chryseobacterium vrystaatense, and Chryseobacterium kwangjuense, which had 16S rRNA gene sequence similarity scores of 97.80%, 97.52%, and 96.75%, respectively, were its closest phylogenetic relatives. Chryseobacterium herbae sp. nov. is proposed as the designation for the strain herbae pc1-10T (=GDMCC 1.3255 = JCM 35711), which represented a type species based on genotypic and morphological characteristics. This study provides deep knowledge of a Chryseobacterium herbae characteristic description and urges the need for further genomic studies on microorganisms living in alpine ecosystems, especially around medicinal plants.


Introduction
The genus Chryseobacterium is included within the family Weeksellaceae [1], the main bacterial lineage in the phylum Bacteroidetes. It is well known that strains of the plantrelated Chryseobacterium species have the ability to promote plant growth [2][3][4][5][6] and degrade pesticide residues [7,8], and many of them have genes associated with antibiotic resistance [6, [9][10][11][12]. According to the "List of Prokaryotic Names with Standing in Nomenclature" (http://www.bacterio.net/ accessed on 20 May 2023) which interprets the list of all validly published names within this genus, 169 species with validly published names and the mixture was then incubated for thirty minutes at room temperature and in the dark [39]. At 530 nm, color absorbance was measured, and a quantitative analysis was performed by comparing the results to the IAA standard (5-40 mg/L). Chrome azurol S (CAS) agar medium was used to test herbae pc1-10 T 's capacity to produce siderophores [40]. Briefly, in order to examine color change, a 5 µL inoculum of the herbae pc1-10 T was spot plated and stored at 25 • C. To test its nitrogen fixation capacity, we inoculated it on the ASHBY medium by observing if it grows. Additional physiological and biochemical features were determined as follows.
Transmission electron microscopy (TEM) was used to analyze the strain herbae pc1-10 T 's morphological characteristics, including the size and shape of the cells and their surface ornamentation. Gram staining was performed according to Halebian et al. [41]. Filter-paper disks (Hopebio, Qingdao, China) impregnated with a 1% solution of N,N-Dimethyl-p-phenylenediamine dihydrochloride were used to measure the activity of oxidase, a positive test resulted in the production of a blue-purple coating on the filter paper biomass within two minutes. In order to evaluate the effects of starch degradation, plates containing nutrient agar (NA Difco) (0.8%), starch (1%), and agarose (1.5%) were used. These plates were developed by flooding them with iodine solution (1%) after 5 days of incubation, and hydrolysis rings around the strain were then observed to form. According to the manufacturer's (bioMérieux, Marcy-l'Étoile, France) instructions, API ZYM galleries were used to assess the strain herbae pc1-10 T 's enzyme activity. Other biochemical assays were carried out in accordance with Tindall et al.'s instructions, including the methyl red (method 15.2.52) and the Voges-Proskauer (method 15.2.82) tests [42]. The utilization of carbon compounds was measured using API 20NE test strips (bioMérieux) according to the manufacturer's instructions, incubated at 25 • C. The growth of bacteria on various carbohydrates and their derivatives (heterosides, polyalcohols, and uronic acids) was examined using the API 50CH test.
Well-grown cells of strain herbae pc1-10 T grown on tryptic soy agar (TSA) medium for 24 h at 25 • C were used for fatty acid analysis. Separation and identification of whole-cell esters were carried out with the Sherlock Microbial Identification System (MIDI version 3.0) as represented by Vandamme et al. [43]. Polar lipids were extracted using the method of Xu et al. [44] and separated by two-dimensional TLC using silica gel 60 F254 aluminum-backed thin-layer plates (Merck, Darmstadt, Germany) [45]. Isoprenoid quinones of strain herbae pc1-10 T were extracted from freeze-dried cells and analyzed as previously described using LC-MS [46]. Using a molybdophosphoric acid hydrate ethanol solution, the total lipids were identified. Utilizing the ninhydrin reagent, aminolipids were identified. With the help of the Zinzadze reagent, phospholipids were identified, and the naphthol reagent was used to find glycolipids. The data were interpreted as described by Tindall et al. [42].

Genome Analyses
Genomic DNA was collected according to the method of Marmur [47]. The genomic G+C content and the organism overview of strain herbae pc1-10 T were determined using the RAST Server [48] by genome sequence, which was performed on the Illumina MiSeq platform by the Guangzhou Magigene Company (Guangzhou, China). The reads were put together using the SOAPdenovo version 1 [49]. According to the minimal requirements suggested by Chun et al., the average nucleotide identity (ANI) and in silico DNA-DNA hybridization (DDH) values were determined [50]. The OrthoANIu algorithm (https: //www.ezbiocloud.net/tools/ani accessed on 17 July 2022) was used to determine the ANI value between the two genomes [49]. The Genome-to-Genome Distance Calculator 3.0 (https://ggdc.dsmz.de/ggdc.php# accessed on 17 July 2022) was used to determine the in silico DDH values [51]. Under the accession number JAOAMU000000000, strain herbae pc1-10 T 's Whole-Genome Shotgum project was deposited at GenBank.

Shutgun Proteomics Analyses
We used shotgun proteomics [52][53][54] to detect the global profile of the protein/polypeptide complement within the mixture expression in the cell and culture supernatant of strain herbae pc1-10 T . The cells and the supernatant of herbae pc1-10 T were separated by centrifugation (8000 rpm, 10 min). The cells were washed three times with a 0.85% NaCl, immediately frozen in liquid nitrogen, and transported to Personal Biotechnology Co., Ltd., Shanghai, China at −80 • C until use. The protein sequence database is QLDBPSN025.

Pigment Analyses
We experimentally validate the production of flexirubin by herbae pc1-10 T as described by Siddaramappa et al. [55] by exposing a 2 mL pure culture of strain herbae pc1-10 T , which was cultured in an NA liquid medium for 24 h, to a 500 µL 3% KOH, which resulted in a change from yellow to orange/red if flexirubin pigments were present, followed by a neutralization step with a 500 µL 1.5 N HCl which resulted in a return to yellow pigmentation.

Phylogenetic and Genotypic Analysis
The 16S rRNA gene sequence (1517 bp) of strain herbae pc1-10 T (OP352779) was acquired. Comparative analysis from this 16S rRNA gene sequence showed that the novel strain stands for a member of the genus Chryseobacterium. Strains Chryseobacterium oleae CT348 T , Chryseobacterium vrystaatense RJ-7-14 T , and Chryseobacterium kwangjuense KJ1R5 T , were prominently picked out as reference strains for comparative studies. Phylogenetic analysis grounded on the NJ, ML methods showed that strain herbae pc1-10 T formed a stable subclade with C. oleae CT348 T in the phylogenetic tree (Figures 1 and S3).

Physiological, Biochemical, and Chemotaxonomic Analysis
Cells (0.9-2.4 µm long and 0.5-0.7 µm wide) were observed to be aerobic ( Figure S1), Gram-staining-negative, rod-shaped, and non-motile. Colonies on nutrient agar were detected as orange-pigmented, entire, convex, and circular ( Figure S4). On the NA agar, a colony with a size of 0.5-1 mm was seen for 5 days at 25 • C. At pH 5.0-9.0 and 10-30 • C (optimal: 25 • C), cells grew; pH 6.0 was optimum. Cells were found to grow most effectively in the absence of NaCl yet tolerated 1.8% NaCl. Catalase and oxidase, on the other hand, were determined to be negative. The strain herbae pc1-10 T significantly grew well and was observed to generate a halo of organophosphorus-dissolving circles but no degradant halo of phosphate on their respective media. The strain was cultured in an NA-tryptophan liquid medium for 2 days and an IAA yield was detected to be 3.53 mg/L ( Figure S4, Table S1). Unfortunately, the strain herbae pc1-10 T did not grow on the CAS medium (with nutrient agar providing the essential elements for growth) and the ASHBY medium. In the pigment detection experiment, after adding the KOH solution to the bacterial suspension, the color changes from yellow to orange-red, and the addition of HCl does not change color ( Figure S7), which can prove its production of the flexirubin-type pigment.

Physiological, Biochemical, and Chemotaxonomic Analysis
Cells (0.9-2.4 μm long and 0.5-0.7 μm wide) were observed to be aerobic ( Figure S1), Gram-staining-negative, rod-shaped, and non-motile. Colonies on nutrient agar were detected as orange-pigmented, entire, convex, and circular ( Figure S4). On the NA agar, a colony with a size of 0.5-1 mm was seen for 5 days at 25 °C. At pH 5.0-9.0 and 10-30 °C (optimal: 25 °C), cells grew; pH 6.0 was optimum. Cells were found to grow most effectively in the absence of NaCl yet tolerated 1.8% NaCl. Catalase and oxidase, on the other hand, were determined to be negative. The strain herbae pc1-10 T significantly grew well and was observed to generate a halo of organophosphorus-dissolving circles but no degradant halo of phosphate on their respective media. The strain was cultured in an NA-tryptophan liquid medium for 2 days and an IAA yield was detected to be 3.53 mg/L (Figure Acid phosphatase, Naphthol-AS-BI-phosphohydrolase, α-Galactosidase, β-Glucosidase, N-Acetyl-β-glucosaminidase, α-Fucosidase. However, other negative results for Valine arylamidase, Cystine arylamidase, Trypsin, Chymotrypsin, β-Galactosidase, β-Glucuronidase, α-Glucosidase, and α-Mannosidase were obtained. Strain herbae pc1-10 T showed negative results for Nitrate reduction, D-Glucose fermentation, Arginine dihydrolase, Urease, β-Galactosidase, Arabinose, N-Acetyl-D-glucosamine, D-Maltose, Gluconate, Capric acid, Adipic acid, Malic acid, Citrate, p-hydroxy-Phenylacetic acid and positive results for β-Glucosidase, Indole production, Gelatinase and weakly positive for Glucose, Mannitol, mannose in the API 20NE strip test. Table 2 compares strains from the genus Chryseobacterium that are phylogenetically linked to strain herbae pc1-10 T in terms of how well they thrive on various carbohydrates and their derivatives.

Genome Analyses
The genome of strain herbae pc1-10 T was 5,142,603 bp long including 36 contigs with an N50 value of 306,357 and a genome coverage of 81.0×. The DNA G+C content was 37.94 mol%, the mean length was 142,836.5 bp, and 71 tRNA genes, 3 rRNA genes, and 1 tmRNA were predicted. The strain herbae pc1-10 T full-length 16S rRNA scaffold was retrieved from the genome assembly to confirm that the genomic data is legitimate and uncontaminated, and one 16S rRNA sequence was received. The result was consistent with that of the PCR sequence. The OrthoANIu values between strains C. herbae pc1-10 T ; C. oleae CT348 T ; C. kwangjuense KJ1R5 T ; and C. vrystaatense R-23566 T were 90.94%, 82.96%, and 85.19%, respectively. The in silico DDH values between strains herbae pc1-10 T ; C. oleae CT348 T ; C. kwangjuense KJ1R5 T ; and C. vrystaatense R-23566 T were 41.7%, 26.6%, and 29.7%, respectively.
The herbae pc1-10 T genome's RAST annotation showed 264 subsystems with 4733 genes and an 18% subsystem coverage, totaling 264 subsystems (see Figure 2 below for details). It provides initial annotations of gene functions for herbae pc1-10 T 's genomes, such as producing siderophores, plant hormones, beta-lactamase, etc. Furthermore, we annotated the genome of herbae pc1-10 T through Proksee (https://proksee.ca/projects/new accessed on 21 May 2023) (Figure 2). Antibiotic resistance gene predictions showed that herbae pc1-10 T contained two antibiotic resistance genes: IND-4, which is a beta-lactamase found in Chryseobacterium indologenes, and AdeF, which Furthermore, we annotated the genome of herbae pc1-10 T through Proksee (https: //proksee.ca/projects/new accessed on 21 May 2023) (Figure 2). Antibiotic resistance gene predictions showed that herbae pc1-10 T contained two antibiotic resistance genes: IND-4, which is a beta-lactamase found in Chryseobacterium indologenes, and AdeF, which is the membrane fusion protein of the multidrug efflux complex AdeFGH; the proteincoding information is provided by CARD [59] (https://card.mcmaster.ca/ accessed on 21 May 2023). According to a study, bacteria that live in soil or water naturally have ARGs that they can use to compete with other organisms that share the same habitat [60]. We used CRISPRCasFinder [61] to identify CRISPR (clustered regularly interspaced short palindromic repeats) arrays; 5 of the 23 analysis sequences were found to contain CRISPR but no Cas genes were found in the vicinity of CRISPRs. Components of these biological systems can be used in multiple applications in genetic engineering [62,63].
Secondary metabolites (Table 3) were identified using antiSMASH version 7.0.0 (https://antismash.secondarymetabolites.org/, accessed on 14 May 2023). One gene cluster with arylpolyene and resorcinol biosynthesis domains was found in strain herbae pc1-10 T , and it showed a 75% similarity to a known biosynthetic gene cluster generating flexirubin, the major pigment of Chryseobacterium [64]. Strain herbae pc1-10 T additionally had a gene cluster with NI-siderophore biosynthetic domains (Figure 3d) showing a 33% similarity to fulvivirgamide A2 biosynthetic gene cluster from Fulvivirga marina [65] and lanthipeptide class I (Figure 3e) biosynthetic domains showing a 17% similarity to pinensins biosynthetic gene cluster from Chitinophaga pinensis DSM 2588 [66], which is the first antifungal lantibiotics [67]. Other gene clusters containing microviridin (Figure 3b), hglE-KS, T1PKS, terpene (Figure 3c), and lanthipeptide class I (Figure 3a) biosynthetic were observed with no similarity to known clusters. The serine proteases chymotrypsin, trypsin, and elastase are just a few of the ones that microviridins are effective and selective inhibitors of [68]. Furthermore, it was discovered that microviridin J was poisonous to Crustacean Daphnia [69].

Shut-Gun Proteomics Results
The qualitative statistical results of proteins are shown in the attached Table S2. Finally, we identified 2455 (cell) and 2650 (supernatant) proteins and 19,568 (cell) and 23,825 (supernatant) peptides from the two samples, respectively (see Supplementary Materials Table S3). first antifungal lantibiotics [67]. Other gene clusters containing microviridin (Figure 3b), hglE-KS, T1PKS, terpene (Figure 3c), and lanthipeptide class I (Figure 3a) biosynthetic were observed with no similarity to known clusters. The serine proteases chymotrypsin, trypsin, and elastase are just a few of the ones that microviridins are effective and selective inhibitors of [68]. Furthermore, it was discovered that microviridin J was poisonous to Crustacean Daphnia [69].  The TonB-dependent siderophore receptor (MCT2563047.1), the biopolymer transporter ExbD (MCT2562352.1, MCT2562353.1, MCT2563473.1), and the MotA/TolQ/ExbB proton channel family protein (MCT2562351.1) were detected in both cell and supernatant, which means that herbae pc1-10 T could uptake siderophore from the outside environment [70]. The pinensin family lanthipeptide (MCT2564031.1) was detected in the supernatant, and it was regarded as the first lantibiotics isolated from a Gram-negative native producer [67]. It is famous for its significant resistance to some fungi; however, we found that herbae pc1-10 T only had a weak inhibitory effect on the growth of pathogenic fungi (Magnaporthe oryzae) of rice blast ( Figure S6). Terpene synthase family protein (MCT2561927.1) was detected in the supernatant and cell. Penicillin-binding transpepti-dase domain-containing protein (MCT2563304.1), the TolC family protein (MCT2563484.1, MCT2561713.1, MCT2564157.1), nitroreductase (MCT2564731.1) [71], and the efflux transporter outer membrane subunit (MCT2563723.1) were detected in cell samples. These proteins are closely related to antibiotic resistance. However, microviridin was not detected in the culture environment and cells, inferring that the predicted result is unreliable, or that the gene is not expressed in our culture environment, or the expression level does not reach the proteome detection threshold, etc.
The type strain's genomic DNA has a G+C concentration of 37.9 mol%. Genome sequence accession number: JAOAMU000000000. 16S rRNA gene accession number: OP352779. The type strain, herbae pc1-10 T (=GDMCC 1.3255 T = JCM 35711 T ), was isolated from the root system soil of Pyrola calliantha H. in Tibet.

Discussion
Rhizosphere microbial colonization often has plant genotype specific selectivity [72]. With technological progress, genomics and proteomics provide more sufficient technical support for microbial molecular level research [54]. The strain herbae pc1-10 T initially showed the function of dissolving organophosphorus, siderophore-mediated iron acquisition systems, and producing indole acetic acid (3.53 mg/L in 2 days), which may have a positive effect on plant growth. Moreover, the genome of strain herbae pc1-10 T harbors two antibiotic resistance genes (IND-4 and AdeF) encoding a β-lactamase, which is responsible for β-lactam antibiotic resistance, and encoding the membrane fusion protein of the multidrug efflux complex AdeFGH, respectively. They were confirmed in Shut-gun proteomic testing. Due to the fact that bacteria that live naturally in soil or water have inbuilt ARGs to combat the chemical substances created by rivals living in the same environment, we speculate that the antibiotic resistance gene possessed by strain herbae pc1-10 T is caused by related chemicals present in the rhizosphere of Pyrola calliantha H. Andres. In a word, analyzing the internal characteristics of microorganisms in special habitats may provide new ideas for future production practices and scientific research.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms11082017/s1, Figure S1: Transmission electron micrograph of the cell of strain herbae pc1-10 T . Bar, 500 nm. Figure S2: The polar lipids of strain herbae pc1-10 T . Figure S3: Maximum-likelihood phylogenetic tree of strain herbae pc1-10 T and its relatives based on the comparison of the 16S rRNA gene sequences. GenBank accession numbers given in parentheses; Figure S4: Growth morphology of strains on NA plates. Figure S5: The indole acetic acid (IAA) standard (5-40 mg/L) curve. Figure S6: Antagonistic experiment of against pathogenic fungi (Magnaporthe oryzae) of rice blast. Figure S7: Reversible colour change in pigment of strain herbae pc1-10 T . Figure S8: Pie-diagram showing the genes of Chryseobacterium herbae pc1-10 T genome distributed in various subsystem categories. Data for the subsystem categories were obtained by SEED viewer in the RAST server. Table S1: The standard curve raw data and OD 530 values of the strain herbae pc1-10 T to be tested. 16S rRNA gene sequence of Chryseobacterium herbae sp. nov. (from the genome). Table S2: Statistics of protein identification results.   Data Availability Statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.