Isolation and characterization of a novel Sphingobium yanoikuyae strain variant that uses biohazardous saturated hydrocarbons and aromatic compounds as sole carbon sources

Background: Green micro-alga, Chlamydomonas reinhardtii (a Chlorophyte), can be cultured in the laboratory heterotrophically or photo-heterotrophically in Tris- Phosphate- Acetate (TAP) medium, which contains acetate as the carbon source. Chlamydomonas can convert acetate in the TAP medium to glucose via the glyoxylate cycle, a pathway present in many microbes and higher plants. A novel bacterial strain, CC4533, was isolated from a contaminated TAP agar medium culture plate of a Chlamydomonas wild type strain. In this article, we present our research on the isolation, and biochemical and molecular characterizations of CC4533. Methods: We conducted several microbiological tests and spectrophotometric analyses to biochemically characterize CC4533. The 16S rRNA gene of CC4533 was partially sequenced for taxonomic identification. We monitored the growth of CC4533 on Tris-Phosphate (TP) agar medium (lacks a carbon source) containing different sugars, aromatic compounds and saturated hydrocarbons, to see if CC4533 can use these chemicals as the sole source of carbon. Results: CC4533 is a Gram-negative, non-enteric yellow pigmented, aerobic, mesophilic bacillus. It is alpha-hemolytic and oxidase-positive. CC4533 can ferment glucose, sucrose and lactose, is starch hydrolysis-negative, resistant to penicillin, polymyxin B and chloramphenicol. CC4533 is sensitive to neomycin. Preliminary spectrophotometric analyses indicate that CC4533 produces b-carotenes. NCBI-BLAST analyses of the partial 16S rRNA gene sequence of CC4533 show 99.55% DNA sequence identity to that of Sphingobium yanoikuyae strain PR86 and S. yanoikuyae strain NRB095. CC4533 can use cyclo-chloroalkanes, saturated hydrocarbons present in car motor oil, polyhydroxyalkanoate, and mono- and poly-cyclic aromatic compounds, as sole carbon sources for growth. Conclusions: Taxonomically, CC4533 is very closely related to the alpha-proteobacterium S. yanoikuyae, whose genome has been sequenced. Future research is needed to probe the potential of CC4533 for environmental bioremediation. Whole genome sequencing of CC4533 will confirm if it is a novel strain of S. yanoikuyae or a new Sphingobium species.

In our research laboratory, we employ the green micro-alga Chlamydomonas reinhardtii, a member of the family Chlorophyceae, as a model system to study oxygenic photosynthesis. C. reinhardtii is grown at the laboratory in the presence (photo-heterotrophically) or absence of light (heterotrophically) in the acetate-containing Tris-Phosphate-Acetate (TAP) medium 13 . TAP contains 0.1% acetate as the sole carbon source 13 . The acetate in TAP medium is used by Chlamydomonas for net biosynthesis of glucose via the glyoxylate/C2 cycle 14 . This allows Chlamydomonas to make sugar heterotrophically in the dark or photo-heterotrophically in the light, without being completely dependent on photosynthesis for glucose biosynthesis 14 . Many bacteria possess glyoxylate cycle and can utilize acetate as a carbon source, like Chlamydomonas 15,16 . Hence, we often encounter bacterial contamination on Chlamydomonas TAP-agar medium culture plates. We observed a yellow-pigmented bacterial contamination on the TAP-agar medium culture plate of the Chlamydomonas wild type strain, CC4533. This bacterium was able to grow on the TAP-agar medium because it was able to use acetate as the carbon source. We named this bacterium as CC4533 because it contaminated the Chlamydomonas wild type strain CC4533. A team of four undergraduate research students from the University of West Georgia and, one high school student from the Carrollton High School in Georgia (USA) were assigned a research project to test four antibiotics that were available in our laboratory to determine which one of these four antibiotics would be suitable for use to minimize Chlamydomonas contamination by CC4533. After the antibiotic-sensitivity of CC4533 was determined, we performed additional microbiological and molecular experiments to better characterize the bacterial strain, CC4533.
We employed a modified version of the Kirby-Bauer (KB) disc diffusion antibiotic susceptibility test to determine antibiotic sensitivity of CC4533. Two different doses of four antibiotics (penicillin, neomycin, chloramphenicol and polymyxin B) were tested in the KB experiment. The results from the KB test show that CC4533 is resistant to penicillin and chloramphenicol but is sensitive to neomycin. Although CC4533 is more sensitive to polymyxin B at higher dose, polymyxin B cannot be used for controlling Chlamydomonas contamination, as at higher dose Chlamydomonas growth is affected. We found that 50 µg/mL of neomycin in TAP medium was potent for eradicating CC4533 contamination on Chlamydomonas media plates in our research laboratory, without hindering Chlamydomonas growth.
We performed standard microbiological tests to biochemically characterize CC4533. These tests show that CC4533 is an aerobic, non-enteric, Gram-negative long rod shaped bacterium (bacillus). CC4533 does not grow on Mannitol Salt Agar and MacConkey agar. CC4533 is starch hydrolysis-negative, alphahemolytic and, is cytochrome c oxidase-positive. CC4533 has a bright yellow pigmentation on Lysogeny Broth (LB) agar and a pale yellow pigmentation on TAP agar. CC4533 grows best at temperatures ranging from 22°C-30°C. CC4533 cannot grow at 37°C on LB agar or on TAP agar medium. Preliminary spectrophotometric analyses of extracted yellow pigment of CC4533 strongly indicate that CC4533 produces β-carotenes. Preliminary results indicate growth medium type (LB Vs. TAP) and, temperature affects β-carotene production in CC4533.
0.1% acetate in the TAP medium can be substituted with alternative carbon sources (e.g. different sugars) to test if these alternative carbon sources can be used by bacteria for energy production. We monitored the ability of CC4533 to grow on Tris-Phosphate (TP) agar medium (lacks acetate) containing monosaccharide (glucose) and two disaccharides (sucrose and lactose). CC4533 was able to use these three sugars as the sole carbon source and was able to ferment these sugars to produce acid.
To determine the taxonomic identity of CC4533, we amplified the 16S rRNA gene partially by DNA polymerase Chain Reaction (PCR). We have submitted this partial 16S rRNA gene sequence to the NCBI GenBank in November 2019, with the definition: Sphingobium yanoikuyae strain PR86 variant; 16S ribosomal RNA gene, partial sequence (Accession number: MN633285.1). Our definition was based on the nearest relative identified by the NCBI-nucleotide BLAST analyses of the partial 16S rRNA gene sequence of CC4533 in 2019, which was Sphingobium yanoikuyae strain PR86 (Accession number: MN232173.1). In April 2020, we found another close relative of CC4533 in the NCBI database based on the partial 16S rRNA gene sequence, which was not detected in our earlier BLAST analyses in 2019. This new relative is Sphingobium yanoikuyae strain NRB095 (Accession number: MK543001.1). 16S rRNA partial gene sequences of these two Sphingobium yanoikuyae strains have identical: score, percentage of sequence identity, E-values and nucleotide changes in the C4 region at identical locations in the 16S rRNA gene, relative to that of CC4533.
Sphingomonas sp. and Sphingobium sp. are of particular interest due to their abilities to degrade cycloalkanes and polycyclic aromatic hydrocarbons (PAH) and polyhydroxyalkanoates (PHA) and their roles in environmental bioremediation  . Hence, we monitored the growth of CC4533 on TP-agar medium plates containing common biohazardous chemicals that are found in our natural environment. We tested the following chemicals: saturated hydrocarbons, cyclohexyl chloride, aromatic acid like benzoate, aromatic ester like phenyl acetate, polyesters (polyhydroxybutyrate, a biodegradable plastic) and poly-cyclic aromatic hydrocarbons. Growth analyses revealed that CC4533 is capable of utilizing these toxic organic compounds as the sole carbon and energy source. In future, we plan to sequence the whole genome of CC4533. This will provide important insights into the metabolic diversity of CC4533 and, will reveal the genes that are recruited by this bacterium to generate biochemical pathways for degradation of xenobiotics. In this article, we present our student-driven research on the isolation of CC4533 (Sphingobium yanoikuyae strain PR86 variant) and its characterizations at the biochemical and molecular level.

Growth media and cultures
Chlamydomonas wild type strain 4A+ (CC-4051 4A+ mt+) was maintained in the lab on TAP agar medium in dim light intensities (15-20 µmol m -2 s -1 ) at 22°C (room temperature). Standard TAP medium recipe can be found at the website of Chlamydomonas Resource Center. Hutner's trace element solution is an ingredient in the TAP medium. Hutner's trace element recipe can also be found on the Chlamydomonas Resource Center website. Our lab's TAP medium recipe is slightly different from the standard one and can be found at https://doi.org/10.17504/protocols. io.bgzujx6w 41 . Liquid 4A+ cultures were grown in liquid TAP medium under low light (50-60 µmol m -2 s -1 ) for 3 days on a New Brunswick Scientific Excella E5 platform shaker (Enfield, CT) at 150 rpm for aeration. CC4533 bacterial strain stock was maintained in the lab under dim light (15-20 µmol m -2 s -1 ) at 22°C on either TAP or LB agar medium. Liquid cultures of CC4533 were grown in culture tubes in 3 mL of TAP medium at 22°C on a MaxQ420HP incubator shaker (Thermo Fisher Scientific, Waltham, MA) at 200 rpm for aeration. Light intensities were measured using a LI-250A light meter (LI-COR, Inc., Lincoln, NE).

Imaging
Images of all media plates used in the experiments were imaged with a Samsung Galaxy S5 cell phone camera. Image cropping and adjustments were made using Photos app in Windows 10. DNA gels were visualized and imaged with a Bio-Rad Molecular Imager Gel Doc XR+ (Bio-Rad, Hercules, CA).
Antibiotic susceptibility test using the modified disc diffusion method Antibiotics that were tested are: penicillin; neomycin, chloramphenicol and polymyxin B. Two doses (50 µg and 100 µg) of each of these four antibiotics were tested in the modified Kirby-Bauer (KB) disc diffusion antibiotic susceptibility tests. KB tests were performed on TAP-agar plates as described in Mitra et al. 2020 [41][42][43] . Antibiotic plates were incubated at 22°C.°C. C. CC4533 plates were imaged after 3 days of incubation and Chlamydomonas plates were imaged after 4 days of incubation. Diameters of zone of inhibitions were measured using a ruler. Statistical analyses 44 and images of all antibiotic plates 45 are available as Underlying data.

Data analyses
Standard deviations and means of the zones of inhibitions from the KB test were calculated using Microsoft Excel. Statistical analyses of the data from the KB disc diffusion test were performed using Microsoft Excels' t-Test: Paired Two Sample for Means tool in the analysis ToolPak.
Standard microbiological tests 1) Growth at different temperatures: CC4533 was streaked on fresh LB and TAP agar medium. Streaked CC4533 media plates were incubated at different temperatures: a) 22°C, b) 30°C, and c) 37°C, for 5 days and then imaged.
2) Growth assays on MacConkey agar (MAC) and Mannitol Salt Agar (MSA): CC4533 was streaked on MAC and MSA plates purchased from Carolina Biological (Burlington, NC). Streaked plates were incubated at 30°C for 3 days. After the incubation period, plates were imaged to monitor CC4533 growth and pH change in the media.
3) Testing CC4533's ability to secrete hemolysins: Growth was monitored at 30°C on tryptic soy agar medium plates containing 5% sheep agar (Carolina Biological; Burlington, NC) for a period of 72 hours. Images were taken after every 24 hours over a period of three days. Classification of hemolysis were assigned according to https://www.asm.org/Protocols/Blood-Agar-Platesand-Hemolysis-Protocols. Images of all tryptic soy blood agar plates are available as Underlying data 46 . 4) Growth assays of CC4533 on Tris-Phosphate (TP)-phenol red-sugar-agar medium: TP medium has all the ingredients of the TAP medium except the acetate (https://doi.org/10.17504/ protocols.io.bgzujx6w) 41 . CC4533 was streaked on three types of TP-phenol red-agar medium containing 1% glucose or 1% sucrose or 1% lactose. pH of the TP medium was 7.2. TP-sugar medium plates were imaged after 5 days of growth at 22°C. Results were interpreted as described in Mitra et al. 2020 41 .

5)
Growth assays of CC4533 on TP agar media plates containing saturated hydrocarbons and aromatic compounds: Growth assays were performed on TP-agar plates coated with different doses of cyclohexyl chloride, polyhydroxybutyrate, phenanthrene, naphthalene, benzoic acid, phenyl acetate and fresh and combusted 10W30 oil, using a modified technique as described in Mitra et al. 2020 41,47 . CC4533 was streaked on the chemicalcoated TP plates and incubated at 22°C and media plates were imaged after two weeks. Images of all media plates are available as Underlying data 48  Pigment analyses LB-grown CC4533 cells and mashed baby carrots were used for pigment extraction. Pigments were extracted with 4 mL of 100% acetone by incubation in dark for 3.5 hours at room temperature. After incubation, the sample tubes were vortexed and then centrifuged at X 4000g for 5 minutes. The yellow supernatant from CC4533 and carrot sample tubes were collected and the cell pellet/tissue debris were discarded. The yellow supernatant was then filtered using a 5 mL syringe fitted to a nylon membrane filters with a cut-off of 0.45 µm. Pigments were analyzed by direct absorbance measurements using the wavelength scan program ranging from 400-600 nm in a Beckman Coulter DU 730 Life science UV/Vis spectrophotometer (Brea, CA). The carrot supernatant absorbance scan was used as a reference overlay against the CC4533 measured scan. To detect carotenoids in the samples, absorption peaks in the 420-490 nm region were monitored (https://assets.publishing.service.gov.uk/media/ 57a08cbae5274a31e00013d4/tech02.pdf) 49,50 .
Genomic DNA isolation and PCR amplification of the partial 16S rRNA gene Genomic DNA was isolated from CC4533 using Qiagen's blood and cell culture DNA mini kit (Qiagen, Valencia, CA) according to the protocol given in the technical manual. Purity of the isolated genomic DNA and its concentration were measured using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). Genomic DNA quality was determined by visualization of the genomic DNA after it was separated by DNA agarose gel electrophoresis.
16S rRNA gene forward (16SF) and reverse (16SR) PCR primers were designed based on primer sequences given in article by Klindworth et al. (2013) 51 . Primer sequences, details about PCR cycling conditions, specific DNA polymerase used in the PCR, separation and visualization of PCR samples can be found in the article by Mitra et al. 2020 41 .
PCR product cloning and DNA sequencing The partial 16S rRNA genomic PCR product was extracted and purified from the DNA agarose gel using the QIAquick Gel Extraction Kit (Qiagen, Valencia, CA). The purified PCR product was cloned as described in Mitra et al. 2020 41 . One clone was sequenced using Sanger Dideoxy sequencing at the UC Berkeley DNA Sequencing Facility. Chromas Lite and nucleotide BLAST program were used to analyze the partial 16S rRNA gene sequences. Raw electropherogram files and DNA sequence text files are available as Underlying data 52 . CC4533 can use acetic acid for growth We found a yellow pigmented-bacterial contamination on a TAP medium plate of a Chlamydomonas wild type strain, CC4533, at our laboratory ( Figure 1A). We purified the bacterial strain from the contaminated TAP-agar medium plate by picking single colonies on fresh TAP agar medium. We picked 40 single colonies and transferred these colonies to fresh LB agar medium. Colony # 28 was selected for our studies ( Figure 1B). Colony # 28 stock was maintained in the lab on LB agar medium at 22°C ( Figure 1C). We named this bacterial strain as CC4533, as the bacterium was isolated from the Chlamydomonas strain, CC4533 TAP-agar medium plate. CC4533 bacterial strain can use acetate as a carbon/energy source. Hence it was able to grow on the TAP-agar medium ( Figure 1A).

Results
CC4533 is resistant to polymyxin B, penicillin and chloramphenicol and is sensitive to neomycin We performed two different sets of experiments for testing antibiotic-sensitivity. In the first experiment, we determined the antibiotic-sensitivity of CC4533 and Chlamydomonas 4A+ wild type strain to identify a suitable antibiotic and the required dose that would inhibit growth of CC4533 but will not affect the growth of Chlamydomonas. In the second set of experiments, we streaked Chlamydomonas and bacterium CC4533 together Table 1. Mean diameters of zones of inhibitions obtained using the disc-diffusion antibiotic susceptibility test. Zones of inhibitions induced by four different antibiotics were studied for Chlamydomonas reinhardtii and the bacterial strain, CC4533. Grey and white rows represent 50 µg and 100 µg dose of each antibiotics applied on the filter paper discs, respectably. Three biological replicates with three internal replicates were used to calculate the mean and standard deviations shown in the table. Statistical analyses 44 and images of all antibiotic plates 45 are available as Underlying data.

C. reinhardtii CC4533
Penicillin 0 mm ± 0 0 mm ± 0 0 mm ± 0 0 mm ± 0 Polymyxin B 8.5 mm ± 0.2 8.7 mm ± 0.1 9.6 mm ± 0.4 10.4 mm ± 0.4 Neomycin 9.5 mm ± 0.5 13.5 mm ± 0.5 11.1 mm ± 0.1 14.8 mm ± 0.2 Chloramphenicol 0 mm ± 0 0 mm ± 0 0 mm ± 0 0 mm ± 0 on the TAP-agar medium plate containing the antibiotic at the proper dose that we determined based on the results obtained from the first experiment set. We used two different doses (50 µg and 100 µg) of each of the following four antibiotics: penicillin, chloramphenicol, neomycin and polymyxin B. Mean diameter of the zone of inhibition for each antibiotic dose with the standard deviations are shown in Table 1. Detailed statistical analyses of the data from three biological replicates (each of which had three internal replicates) are available as Underlying data 44 . Images of TAP antibiotic-agar plates from the KB experiments are available as Underlying data 45 .
At 50 µg dose, bacterium CC4533 was more sensitive to polymyxin B than Chlamydomonas and, both Chlamydomonas and CC4533 were sensitive to the 100 µg dose of polymyxin B (Table 1; Underlying data 44,45 ). Statistical analyses supported that the 50 µg dose of polymyxin B is not very effective in inhibiting growth of CC4533, without drastically affecting Chlamydomonas growth (Underlying data 44,45 Table 1).
Chlamydomonas and CC4533 were both resistant to the 50 µg and 100 µg of chloramphenicol as no zone of inhibition was obtained in the KB test (Underlying data 44,45 ; Table 1).
Chlamydomonas was less sensitive to both 50 µg and 100 µg dose of neomycin than bacterium CC4533 (Underlying data 44,45 ; Table 1) and the sensitivity was statistically highly significant (Underlying data 44,45 ; Table 1). Taken together, our KB test results showed that neomycin would be the best antibiotic choice to eliminate CC4533 contamination of Chlamydomonas. Polymyxin B at a higher dose (see results) could be the second choice of antibiotic for minimizing CC4533 contamination, but Chlamydomonas growth will be also affected at the higher dose of polymyxin B.
In the second experiment, we tested combined growth of CC4533 and the wild type Chlamydomonas strain 4A+ on TAP agar plates containing 50 µg and 100 µg of polymyxin B and neomycin per mL of the TAP medium ( Figure 2). CC4533 was able to grow along with Chlamydomonas on TAP media plates containing 50 µg and 100 µg of polymyxin B per mL of the TAP medium ( Figure 2A and Figure 2B). CC4533 did not grow on TAP plates containing 50 µg and 100 µg of neomycin per mL of the TAP medium ( Figure 2C and Figure 2D). Chlamydomonas grew slowly on the TAP medium plate containing 100 µg neomycin/mL of TAP but grew at a normal rate on the TAP medium plate containing 50 µg neomycin/mL of TAP ( Figure 2C and Figure 2D). Hence neomycin at a concentration of 50 µg/mL in the TAP medium is most effective in inhibiting the growth of bacterium CC4533 on Chlamydomonas culture plates without affecting the growth of Chlamydomonas.

Gram staining
Gram staining of CC4533 revealed that CC4533 is a Gramnegative bacillus. Cells are straight long rods which join to form chains ( Figure 3).
C4533 grows best at 22-30°C but fails to grow at 37°C We have monitored the growth of CC4533 on LB-agar and TAP-agar over 5 days at different temperatures namely 22°C, 30°C and 37°C. CC4533 grew well on LB and TAP-agar at 22°C and 30°C but could not grow at 37°C (Figure 4). Yellow pigmentation of CC4533 was visibly reduced on TAP medium compared to that on the LB medium. Growth at 30°C reduced pigment accumulation compared to that at 22°C on both LB and TAP medium ( Figure 4).

CC4533 can utilize and ferment different sugars for growth
We grew CC4533 on TP (lacks the carbon source, acetate) agar medium containing three different types of sugars (glucose, sucrose and lactose) and the pH indicator, phenol red ( Figure 5). Figure 5A, Figure 5C and Figure 5E represent control TP + 1% glucose, TP + 1% sucrose and TP + 1% lactose, plates, respectively. CC4533 grew very well on all sugar supplemented TP agar plates ( Figure 5). It fermented sugars on all TP-sugar medium plates to produce acid which lowered the pH in the TP medium. The drop in pH, changed the color of phenol red from light red color to a yellow color ( Figure 5B, Figure 5D and Figure 5F).
CC4533 fails to grow on MacConkey Agar (MAC) and on Mannitol Salt Agar (MSA) CC4533 was unable to grow on MAC and E. coli, a Gramnegative enteric bacterium ( Figure 6A), was able to grow on MAC. MAC contains lactose as the carbon source. The pink color of E. coli on MAC plate, indicated that it can ferment the sugar lactose present in the MAC medium to produce acid as the acidic pH (pH below 6.8) changed the color of neutral red to pink ( Figure 6B). The drop in the pH in the MAC medium around the growth of E. coli, precipitated the bile salts out of the MAC medium which caused a hazy pink zone to develop around the E. coli growth ( Figure 6B). CC4533 can use lactose as a carbon source ( Figure 5F). CC4533 cannot grow on MAC because it is sensitive to bile salts and crystal violet in the MAC medium ( Figure 6A). Our results indicate that CC4533 is a non-enteric bacterium unlike E. coli.
CC4533 fails to grow on MSA ( Figure 7A). Staphylococcus aureus grew on MSA and fermented the sugar-alcohol, mannitol, present in the MSA medium to produce acid, which changed the phenol red's color from red to yellow ( Figure 7B). Our results show that CC4533 is salt-sensitive as MSA contains about 7.5-10% of NaCl, which inhibits growth of many gram-negative bacteria and, allows selection of high salt-tolerant grampositive bacterium like Staphylococcus.
CC4533 is alpha-hemolytic CC4533 did not show any hemolysis or discoloration of blood agar medium after 24 hours growth (Underlying data 46 ). After 48 hours, a dark brown discoloration around the cell growth was observed (Underlying data 46 ), which became more pronounced after 72 hours of growth ( Figure 8A), indicating CC4533 is 6-glucosidase to hydrolyze amylose and amylopectin (starch). We did not have a starch hydrolysis-positive strain in our lab to use as a positive control in this experiment.
CC4533 uses cytochrome c oxidase in the respiratory electron transport chain Aerobic, facultative anaerobic or microaerophilic bacteria that uses cytochrome c oxidase in the electron transport chain associated with cellular respiration can be identified by the oxidase test. We conducted the oxidase test on CC4533 and on a yellow pigmented-Microbacterium sp., using a disposable slidethat contains a film coated with oxidase reagent tetramethylp-phenylenediamine (TMPD). CC4533 oxidized the TMPD to indophenols, a purple colored product, within 5-10 seconds. CC4533 is oxidase-positive (Figure 10; left). Microbacterium sp. is oxidase-negative as it failed to change the color of TMPD to purple within 5-10 seconds. (Figure 10; right). We grow CC4533 under aerobic conditions in our lab (Figure 1-Figure 9). Our results show that CC4533 is an aerobic bacterium that uses cytochrome c in the respiratory electron transport chain.  alpha-hemolytic. Figure 8B shows that S. aureus is beta-hemolytic as complete hemolysis can be seen around the S. aureus growth after 24 hours of growth. After 72 hours, the clearing on the blood agar medium plate is more pronounced. Images of tryptic soy blood agar plates are available as Underlying data 46 and in Figure 8B.

CC4533 is unable to hydrolyze starch
We performed starch hydrolysis test on CC4533 and E. coli on Mueller-Hinton agar medium, which contains 0.15% starch (Figure 9). Background and scientific basis of the starch hydrolysis test can be found in Mitra et al. 2020 41 . Both CC4533 ( Figure 9B) and E. coli ( Figure 9D

CC4533 synthesizes β-carotene
It is known that many bacteria accumulate carotenoids, which gives them orange to yellow pigmentation. As CC4533 is yellow-pigmented, we tested for the presence of carotenoids in CC4533, grown under normal room light (30-40 µmol m -2 s -1 ). We monitored the absorption spectrum of the acetone-extracted CC4533 pigment using wavelength scan program ranging from 400-600 nm in a UV-Vis spectrophotometer. Carotenoids absorb strongly in the visible light range from 400-495 nm, with absorption peaking near 450 nm (https://assets.publishing.service.gov. uk/media/57a08cbae5274a31e00013d4/tech02.pdf) 49,50 . We found two major absorption peaks in the 400-500 nm region of the spectrum (Figure 11) 49,50 . The absorption peak with the absorbance reading (0.294) is at 453 nm and the one with the absorbance reading of 0.253 is at 480 nm ( Figure 11B and Figure 11C). β-carotene shows two major absorption peaks that range between 451 nm -454 nm and 477 nm -480 nm 49,50 . The observed two absorption peaks in CC4533 pigment extract strongly indicates the presence of β-carotene in CC4533.
80% of the carotenoids in carrots is β-carotenes 53,54 . We extracted pigments from carrots and measured its absorption peaks ( Figure 12). We used the carrot pigments' absorption curve as a reference scan overlay against the measured scan of CC4533 pigment extract (Figure 12). We found that the extracted-carrot pigments exhibited three major peaks that are representative of β-carotene absorption at 429 nm, 451 nm and 477 nm (https://assets. publishing.service.gov.uk/media/57a08cbae5274a31e00013d4/ tech02.pdf; Figure 12A, Figure 12C, Figure 12D; Figure 12E). Two of these peaks (451 nm and 477 nm) are very close to that based on the partial 16S rRNA gene sequence. Both strains of Sphingobium yanoikuyae detected by our NCBI-BLAST analyses had identical scores, E-values and sequence identity (including the same nucleotide substitutions at the identical locations within the 16S rRNA gene) when their partial 16S rRNA gene sequences were compared against that of CC4533.
CC4533 AKA Sphingobium yanoikuyae strain PR86 variant can use biohazardous saturated hydrocarbons and aromatic compounds as the sole carbon source for growth Sphingobium sp. are known to use alkanes, polycyclic aromatic hydrocarbons (PAH), polyhydroxyalkanoates like polyhydroxybutyrate (PHB) and other aromatic compounds as alternative carbon sources for growth  . Hence, we tested if CC4533 can utilize saturated hydrocarbons, PAH, other biohazardous aromatic compounds and PHB as carbon sources for growth. These growth analyses were performed as described in Mitra et al. 2020 41 . Different doses of the chemicals that were tested can be also found in Mitra et al. 2020 and are also available as Underlying data 48 .
CC4533 was streaked on a TP-agar medium plate, which lacks a carbon source, as a negative control to show that CC4533 does not grow on a TP medium plate in the absence of a carbon source ( Figure 14A). CC4533 was able to grow on TP medium containing all doses of 1% cyclohexyl chloride (a mono chlorinated hydrocarbon) ( Figure 14B) and 1% PHB (belongs to the class of polyhydroxyalkanoates that are used as bio-derived and biodegradable plastics) ( Figure 14C). Our results show that CC4533 can use both these organic compounds as sole carbon sources for growth.
Motor oils contain petroleum-based hydrocarbons which contain between 18 and 34 carbon atoms per molecule, poly-alpha olefins or their mixtures in different ratios. We monitored the growth of CC4533 on TP medium containing 2% (v/v) fresh 10W30 car motor oil ( Figures 15A-B) and 2% (v/v) combusted 10W30 car motor oil ( Figures 15C-D). C4533 grew on TPagar containing all doses of 2% fresh ( Figure 15A-B) and 2% of absorption peaks (453 nm and 480 nm) of CC4533 pigment extract ( Figure 11B, Figure 11C, Figure 12B, Figure 12D and Figure 12E). The 429 nm peak in the pigment extract of carrot was not prominently visible in the pigment extract of CC4533 ( Figure 12C). Our preliminary data strongly indicates the presence of β-carotene in CC4533.
Partial 16S rRNA gene sequence of CC4533 has 99.55% sequence identity with that of Sphingobium yanoikuyae The full length 16S rRNA gene is 1541 bp long ( Figure 13A; based on E. coli 16S rRNA gene). There are nine hypervariable (V1-V9) and nine conserved regions (C1-C9) in the 16S rRNA gene [55][56][57] . 11 nucleotides (788-798) within the C4 conserved region are totally conserved in bacteria 58 . This super-conserved region is represented in Figure 13 as a black box within the C4 region 58 . Forward and reverse PCR primers are represented by black arrows in the schematic of the 16S rRNA gene ( Figure 13A). PCR amplification of the partial 16S rRNA gene of CC4533 generated an amplicon of approximately 460 bp in size ( Figure 13B). The amplicon shown in Figure 13B was sequenced to determine the nearest relative of CC4533.
In fall 2019, NCBI-nucleotide BLAST analyses identified the nearest relative of CC4533 as Sphingobium yanoikuyae strain PR86 (GenBank Accession #: MN232173.1) based on the partial 16S ribosomal RNA gene sequence. This hit has a score of 802; zero E-value and percent identity of 99.55%. Figure 13C shows two nucleotide substitutions (transitions) that are present in CC4533 16S rRNA partial gene sequence relative to that in Sphingobium yanoikuyae strain PR86. These two nucleotide substitutions are in the 11 bp super-conserved sub-region within C4 region ( Figure 13C). We deposited in November 2019, the partial 16S rRNA sequence of CC4533 in GenBank with the definition: Sphingobium yanoikuyae strain PR86 variant, 16S ribosomal RNA gene, partial sequence (Accession number: MN633285.1). DNA sequencing data of the 16S rRNA gene of CC4533 are available as Underlying data 52 . In 2020, NCBI-BLAST analyses revealed another close relative of CC4533: Sphingobium yanoikuyae strain NRB095 (Accession number: MK543001.1) combusted 10W30 motor oil (Underlying data 48 ; Figure 15C-D), indicating that it can utilize petroleum-derived hydrocarbons as carbon sources.
Phenanthrene is a PAH composed of three fused benzene rings. Napthalene is a PAH consisting of two fused benzene rings. CC4533 was streaked on a TP medium plate which lacks a carbon source and this plate served as the negative control in the experiment ( Figure 16A). We monitored the growth of CC4533 on TP medium containing 1% phenanthrene ( Figure 16B) and 1% naphthalene (Figures 16C-D). CC4533 grew on TP-agar containing all doses of 1% phenanthrene and 1% naphthalene (Underlying data 48 ; Figures 16B-D).
Benzoic acid is an aromatic carboxylic acid with a single benzene ring. Benzoic acid and sodium benzoate are commonly used as food preservatives and these preservatives are major environmental pollutants. Phenyl acetate is the ester of phenol and acetyl chloride. Phenylacetate is a common environmental pollutant and is a central intermediate in the pathways that degrade aromatic chemicals (e.g. phenylalanine, phenylacetaldehyde, lignin-related phenylpropane units, environmental contaminants like styrene and ethylbenzene) 59 . CC4533 was able to utilize all tested doses of benzoic acid (Figures 17A-B) and phenyl acetate (Figures 17C-D) in the TP medium as the sole carbon sources (Underlying data 48 ).

Discussion
CC4533 is a mesophilic, yellow pigmented, Gram-negative rod (Figure 1-Figure 4). It can ferment glucose, sucrose and lactose ( Figure 5). It is a non-enteric, salt-sensitive, alpha hemolytic, starch hydrolysis-negative, oxidase-positive bacterium ( Figure 6-Figure 10). NCBI-nucleotide BLAST analyses of the partial 16S rRNA gene sequence of CC4533 revealed that the best match is to that of Sphingobium yanoikuyae strain PR86 (GenBank Accession #: MN232173.1) and to that of Sphingobium yanoikuyae strain NRB095 (Accession number: MK543001.1) with a sequence identity of 99.55% and zero E-value, indicating strongly that CC4533 is a new strain of Sphingobium yanoikuyae.
Whole genome sequencing can confirm if CC4533 is a new species of Sphingobium or a new strain of Sphingobium yanoikuyae (see discussion end).
Sphingobium sp. are chemoorganotrophs and have been isolated from soils, freshwater and marine habitats, activated sludge, phyllosphere and rhizosphere. The isolation source for Sphingobium yanoikuyae strain PR86 (GenBank Accession #: MN232173.1) is sunflower endo-phyllosphere and that for Sphingobium yanoikuyae strain NRB095 (Accession number: MK543001.1) is surface disinfested root of the tropical Koronivia grass Brachiaria humidicola. There are various examples of plant growth-promoting organisms within the Sphingomonadaceae family but the genus Sphingobium is largely limited to members that degrade xenobiotic compounds [61][62][63][64][65] . Exception is the Sphingobium sp. strain AEW4, isolated from the rhizosphere of the beach grass, Ammophila breviligulata 61 . This strain has plant growth promoting properties via production of siderophores and indole-3-acetic acid and induces root growth 61 . Sphingobium paulinellae sp. nov. and Sphingobium algicola sp. nov are isolated from Paulinella chromatophora, a freshwater filose amoeba with photosynthetic endosymbionts (chromatophores) of cyanobacterial origin 66 . In nature, C. reinhardtii is predominantly found in temperate, nutrient-rich, cultivated field soils in Northern America and Japan 67,68 . We isolated CC4533 (Sphingobium yanoikuyae strain PR86 variant) in our laboratory from TAP agar medium plates of the micro-green-alga C. reinhardtii.
CC4533 is resistant to the β-lactam group of drugs like penicillin, and to the broad-spectrum antibiotic, chloramphenicol (Table 1). It is resistant to both 50 µg and 100 µg doses of the cationic antimicrobial polypeptide, polymyxin B, which is one of the most effective drugs against Gram-negative bacteria 69 (Table 1; Figure 2, Underlying data 44,45 ). CC4533 is sensitive to neomycin (Underlying data 44,45 ; Table 1; Figure 2). The diversity and antibiotic resistance patterns of Sphingomonadaceae isolates from drinking water show that the highest antibiotic resistance prevalence values were in members of the genera Sphingomonas and Sphingobium, especially in tap water and in water from     We have shown in this work that CC4533 (Sphingobium yanoikuyae strain PR86 variant) can utilize PAH, chlorinated alkanes, petroleum derivatives in 10W30 car motor oil, PHB and benzoic acids and phenyl acetate as sole carbon source in TP medium ( Figure 14-Figure 17). Recently we isolated and characterized a novel bacterial strain, LMJ/Bacterium strain clone LIB091_C05_1243 variant 16S ribosomal RNA gene, partial sequence (Accession number: MN633292.1) 41 . The nearest relative of LMJ with a genus name is Acidovorax sp. strain A16OP12 (Accession #: MN519578.1) 41 . LMJ can grow on TP-medium containing saturated hydrocarbons, PHB and PAH like CC4533 41 . LMJ hardly grows on TP medium containing phenyl acetate and benzoate unlike CC4533 41 . CC4533 strain grows more robustly than LMJ on every tested toxic aromatic compound and hydrocarbon containing plates at all doses 41 . Hence, CC4533 displays a higher potential for environmental bioremediation than the bacterial strain LMJ strain 41 .
In future we plan to optimize the process of uniformly coating the TP medium surface with chemicals as we have noticed that the tested hydrophobic chemicals were unevenly deposited after coating, when the solvent chloroform evaporates. We think that this uneven coating can affect proper utilization of these aromatic chemicals by CC4533. Our TP medium is a nutritionally stringent minimal medium compared to the traditional M9 minimal medium used for bacterial growth 41 . The final concentrations of phosphate, nitrogen, magnesium and carbon in the M9 medium is approximately, 70-fold, 2.5-fold, 4-fold and 5-fold higher than that present in our lab's TAP medium, respectively 41 . We would like to test if CC4533 can grow on the M9 medium without an added carbon source like a sugar or acetic acid. If CC4533 fails to grow on the M9 medium plates without a carbon source, we will compare the growth of CC4533 in M9 medium containing different PAHs or other aromatic compounds as alternative carbon sources against that in TP medium.
Additionally, we need to collaborate with a research lab that can test the concentration of these toxic organic chemicals in TP medium before and after the CC4533 growth, to get additional evidence that CC4533 is degrading PAH, PHB and other hydrocarbons in the TP medium.
Bacterial heterocyclic aromatic compound degradation pathways mainly involve the oxidation reactions such as angular dioxygenation of carbazole and dibenzofuran, lateral dioxygenation of dibenzothiophene in Kodama pathway, and S-oxidation of dibenzothiophene in 4S pathway 34, 71,72 . A consequence of the oxidation reaction is the formation of reactive oxygen species (ROS), which can damage DNA, proteins, and membranes 73,74 . Carotenoids present in a wide variety of bacteria, algae, fungi, and plants are the most prominent membraneintegrated antioxidants 75 . Sphingobium yanoikuyae XLDN2-5, a yellow pigmented PAH-degrading strain, synthesizes zeaxanthin from β-carotene through β-cryptoxanthin via the carotenoid biosynthetic pathway 49 . In Sphingobium yanoikuyae XLDN2-5, there is a direct correlation between the increase in the amount of zeaxanthin and the enhancement of hydrogen peroxide production during the biodegradation of heterocyclic aromatic compounds 49 . High levels of carotenoids in this Sphingobium strain were consistent with the enhanced transcription of the gene encoding phytoene desaturase, one of the key enzymes for carotenoid biosynthesis 49 .
Pigment analyses of CC4533 strongly indicate the presence of β-carotene in the CC4533 pigment extract (Figure 11; Figure 12). We plan to confirm our results by conducting HPLC analyses of the CC4533 pigment extract and a commercial pure sample of β-carotene. Carotenoid production can be measured in CC4533 grown on LB medium that contain the ROS species like hydrogen peroxide or contains photo-sensitizers like Rose Bengal that generates the ROS, singlet oxygen, in the presence of light and oxygen 76 . Bleaching herbicides, such as norflurazon, interferes with the carotenoid biosynthetic pathway 77 . Norflurazon blocks the enzyme Phytoene desaturase (PDS), which converts phytoene (colorless carotene) to red-colored lycopene 78 . ROS sensitivity of CC4533 can be tested in the presence and absence of the inhibitor of PDS. Carotenoids are used as regulators for membrane fluidity by Staphylococcus xylosus 79 . For two Staphylococcus xylosus strains there was an increase in staphyloxanthin and other carotenoids when grown at 10°C but no carotenoids could be detected when grown at 30°C 79 . CC4533 cannot grow at 37°C and appears lot less yellow-pigmented at 30°C compared to when it is grown at 22°C (Figure 4). Pigment reduction is more pronounced in CC4533 on TAP medium than on LB (Figure 4). Quantitative carotenoid assays at different temperatures can be performed to study the temperature effect on pigment production.
Partial 16S rRNA gene sequence of CC4533 shows two transitional nucleotide changes when compared to that of the Sphingobium yanoikuyae strain PR86 (GenBank Accession #: MN232173.1) and to that of Sphingobium yanoikuyae strain NRB095 (Accession number: MK543001.1). These two nucleotide substitutions are in the 11 bp invariable sub-region (788 bp -798 bp) within C4 region ( Figure 13C). These results show that conserved regions of the 16S rRNA are not truly "conserved". Conserved regions of the 16S rRNA gene exhibit considerable variations that need to be considered when using this gene as a biomarker 80 .
Sphingomonads have attracted the attention of microbiologists and biotechnologists due to their biodegradative and biosynthetic capabilities, and have been utilized for a wide range of biotechnological applications from bioremediation of contaminants to production of extracellular polymers 28 . We have shown in this study that CC4533 is a Sphingobium yanoikuyae strain and has traits that can be exploited for bioremediation upon further research. We found in the NCBI database, 20 genome assemblies of Sphingobium yanoikuyae (Representative genome: Sphingobium yanoikuyae ATCC 51230; ID: 3110) and 70 genome assemblies of uncharacterized environmental isolates. Because of funding limitations, we could not sequence the whole genome of CC4533 at the time of this manuscript submission. But we will have funds in fall 2020 to sequence the whole genome of CC4533 using the Pacific Biosciences technology. Whole genome sequencing will allow us to: 1) confirm if CC4533 is a novel species of Sphingobium or a new strain of Sphingobium yanoikuyae and, 2) will reveal genes in CC4533 that are recruited for degradation of xenobiotics and contribute to its metabolic diversity, relevant to environmental bioremediation and industrial biotechnology.