Acholeplasma testudinis sp. nov., Isolated from the Nares of the Florida Gopher Tortoise (Gopherus polyphemus)

1. Introduction

Gopher tortoises (Gopherus polyphemus) are found in the southeastern United States, with the majority of populations found in Florida. These reptiles are a critical keystone species as their burrows provide food and shelter for an estimated 60 vertebrates and 302 invertebrates, many of which are also considered threatened or endangered [1,2,3,4,5]. Primarily due to significant pressure from anthropogenic habitat loss, the gopher tortoise is listed internationally by the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES; https://cites.org) as vulnerable by the International Union for Conservation of Nature (IUCN; https://www.iucnredlist.org), and regionally threatened by the U. S. Fish and Wildlife Service (https://www.govinfo.gov/content/pkg/FR-2022-10-12/pdf/2022-21659.pdf#page=1). The tortoise is designated as a threatened species in Florida, with varying levels of protection in Alabama, Georgia, Louisiana, Mississippi, and South Carolina. In addition to habitat loss, upper respiratory tract disease (URTD) has been implicated as a contributing factor in die-offs [6,7,8]. URTD refers to a clinical syndrome that occurs in tortoises and is characterized alone or in concert, by a nasal discharge, conjunctivitis, an ocular discharge and palpebral edema and also can alter biological behaviors that could contribute to increased risk of disease transmission and predation [9,10,11]. Isolation of Mycoplasma agassizii from clinically ill tortoises and subsequent transmission studies confirmed this bacterium as an etiological agent of URTD, with M. testudineum associated with milder clinical disease [12,13,14].

As part of an intensive field study of wild gopher tortoise populations in Florida [11], culture and PCR was performed on nasal lavages of tortoises to detect both M. agassizii and M. testudineum. We obtained clinical isolates (N=12) from the nares of tortoises from 7 study sites that, based on 16S rRNA sequence, more closely grouped with Acholeplasma spp. The class Mollicutes is characterized by the loss of a cell wall, significant genome reduction, and the ability to pass through 0.45 um filters [15]. Unlike others within the class Mollicutes, Acholeplasma do not require exogenous cholesterol or serum for growth and also preferentially use the classic UGG rather that UGA as a codon for tryptophan [16,17]. While Mycoplasma spp. are host-associated, cause disease in animals and humans, and do not persist in the environment, Acholeplasma spp. are found in the environment and have been identified in various plant, invertebrate, vertebrate and environmental samples [18,19,20,21,22,23]. The pathogenicity in plants and animals is limited and largely unproven, but in those cases where pathogenicity has been assessed, clinical disease is either absent or much less than for other Mollicutes [24].

Despite the relative ease of growth of Acholeplasma, only 17 species have been formally named and have accessible genomes (Figure 1). Only four new species, including Acholeplasma testudinis, have been described since 2000. Therefore, our primary objective was to validate A. testudinis isolated from the gopher tortoise as a new species. There have been suggested nomenclature changes in the Acholeplasmataceae ([25], with proposed new genera Alteracholeplasma (palmae, parvum), Haploplasma (axanthum, modicum), and Paracholeplasma (brassicae, manati, morum, vituli). However, the list of prokaryotic names with standing in nomenclature (LPSN) [26,27] lists the genus Acholeplasma as correct for all species and is therefore used.

Because of the widespread distribution of the Acholeplasma sp. on multiple sites across Florida and the observance of some tortoises with clinical signs, it was also important to determine if the organism was a pathogen or commensal. Here we provide the foundational support that Acholeplasma testudinis sp. nov. ORD1043^Ts is a new species, isolated from the nasal passages of the environmentally threatened Florida gopher tortoise. Identification of A. testudinis as a new species was made using both classical and molecular methods [28,29,30,31]. The status as a new species was confirmed by 16S rRNA sequence, whole genome sequence, and proteome phylogenies [29,30,31]. Experimental infection studies found limited pathogenic potential for this microbe. A. testudinis is the first described Acholeplasma species isolated from a reptilian host.

2. Material and Methods

2.1. Study Sites

Clinical isolates of A. testudinis sp. nov. in this study were obtained from 7 study sites in northern and central Florida (Figure 2). Sites included Central Florida private land (CentFL), Flying Eagle (FE), Fort Cooper State Park (FC), Green Swamp West Wildlife Management Area (GSW), Ordway-Swisher Biological Station (ORD), Perry Oldenburg Wildlife Management Area (OLD) and Tenoroc Fish Management Area (TE). Habitat differed among the sites. FE is dominated by Bahia grass (Paspalum notatum); CentFL and TE are previously disturbed sites that are dominated by non-native, ruderal vegetation; ORD, FC, GSW, and OLD are sandhill habitats with varying degrees of management and quality.

2.2. Trapping, Sampling, and Initial Cultures

Gopher tortoises are fossorial reptiles and spend most of their time in underground. Therefore, tortoises were captured opportunistically as encountered or in pitfall traps set directly in front of burrow openings. Pitfall traps consisted of one to five gallon buckets sunk immediately in front of the burrow entrance with a shade cover to prevent heat stress. Traps were checked a minimum of once daily to minimize the time that tortoises were held. For all tortoises captured, comprehensive health assessments were performed, tortoises permanently marked for identification, photodocumented, and morphometric measurements taken, and biological samples collected [11].

Nasal flushes were performed on each captured tortoise after health assessments were completed. The head was first rinsed with water if needed, and then the nares were cleaned with an alcohol swab. A syringe attached to a soft plastic tube (22-25 gauge intravenous catheter with needle stylete removed) was used to flush 0.5-5 mL of sterile 0.9% NaCl solution into the nasal cavity. For adult tortoises, 5 mL was used per nares. During all nasal flushes, an assistant applied pressure to the soft tissue between the ventral mandibles in order to force the tongue into the choanae to prevent aspiration or swallowing of flush. The fluid was collected in a sterile plastic container and SP4 medium [32] was immediately added to the container at a volume of 10% of the flush volume.

Culture techniques on nasal flush samples were similar to methods previously described [33] except as follows. A 400 μL aliquot of the nasal flush sample was inoculated into 4 mL of SP4 broth. Two mL of this broth was filtered using a 0.45 µM filter to remove fungal contaminants. Twenty µL of each broth sample and the nasal lavage fluid was placed on SP4 agar and incubated at 30 °C and 5% CO₂. Broth samples were incubated at 30 °C without CO_2. If a color change occurred in the broth medium and/or colonies were observed on agar, cultures were deemed positive.

2.3. 16S rRNA Sequencing

Clinical isolates (N=12) from 7 different sites (Cent FL, OLD, ORD, FC, TE, GSW, FE) had PCR amplification of a portion of the 16S rRNA gene using generic conserved primers complementary to terminal sequence sense strand nucleotides (nt) 11-30 (5’-AGAGTTTGATCCTGGCTCAGGA-3’) and a mycoplasma genus-specific region anti-sense strand nt 1055 to 1031 (5’-TGCACCATCTGTCACTCTGTTAACCTC-3’) as previously described [34]. The mycoplasma specific primer also amplifies other Mollicutes, including Acholeplasma and Ureaplasma spp. Purified PCR products were submitted to Lonestar Labs (College Station, TX) for Sanger sequencing. Sequence data covered the majority of the conserved and variable regions of the gene, resulting in sequences for 1,000-1,400 nucleotides. A phylogenetic 16S rRNA tree for the 12 clinical isolates and known Acholeplasma spp. Type strains was constructed using Molecular Evolutionary Genetics Analysis (MEGA 12) [35,36]. The evolutionary history was inferred by using the Maximum Likelihood method and Tamura 3-parameter model [37].

2.4. Whole Genome Sequencing

Based on the results of the 16S rRNA phylogeny [29], A. testudinis ORD1043 was chosen as the Type strain and used for whole genome sequencing. The isolate was filter cloned three times and grown in SP4 medium without serum at 30˚C. For genomic DNA extraction, 50 mL of a mid-log phase culture was spun at >12,000 x g for 60 minutes. Pellets were washed in 1X phosphate buffered saline (PBS) three times. After resuspension in 2mL PBS, samples were aliquoted into sterile 1.5 mL tubes and an appropriate volume of TELT buffer (50 mM Tris pH 8.0, 62.5 mM EDTA pH 8.0, 2.4 M LiCl, 4% Triton X-100) was added until pellets were dissolved via gentle rocking. An equal volume of phenol-chloroform isoamyl was added, and the sample spun at >12,000 xg for 10 minutes. The aqueous phase was removed, and an equal volume of ice-cold isopropanol was added to precipitate total nucleic acid. The precipitate was pelleted as before and washed in 70% ethanol. Tubes were air dried for 3 minutes and total nucleic acid was resuspended in nuclease free water. All reagents and kits were obtained from Thermo Fisher Scientific, Waltham, MA. The DNA was quantified via Qubit^TM Broad Range Assay and submitted to SeqCenter (Pittsburgh, PA), who performed RNAse treatment, and performed Illumina sequencing. Quality control and adapter trimming was performed with bcl2fastq (https://support.illumina.com/sequencing/sequencing_software/bcl2fastq-conversion-software.html). Short read assembly was performed with Unicycler [38]. Assembly statistics were recorded with QUAST [39]. Bioinformatics provided included annotation with Prokka [40]; structural variant detection [41], Minimap analysis [42], and SAMtools [43]. The A. testudinis ORD1043 genome was annotated with the Prokaryotic Genome Annotation Pipeline (PGAP) prior to submission to GenBank [44,45,46].

2.5. Comparative Genomics and Proteomics

Whole genome sequences are available for 17 of the 18 named species within the Acholeplasmataceae. Because the 16S rRNA sequence of A. testudinis^Ts ORD1043 and 10 clinical isolates were 99.2 ± 0.13%, this met the minimum standards for genome speciation [29] and therefore only the Type strain was fully sequenced. The 16S rRNA sequence was used to find the nearest phylogenetic neighbors. The genome sequence of A. testudinis was compared to the genomes of A. equifetale, A. equirhinis, A. granularum, A. hippikon, A. laidlawii, A. manati, A. multilocale, A. oculi, and A. vituli. Genomes were compared for average nucleotide identity (ANI) using OrthoANI [47,48], predicted digital DNA:DNA hybridization (dDDH), and proteome comparison using the Type (Strain) Genome Server [49,50,51]. Websites for the programs used in these analyses are https://www.ezbiocloud.net/tools/orthoaniu and https://tygs.dsmz.de. Similar genome finder data was generated using the Bacterial and Viral Bioinformatics Resource Center, https://www.bv-brc.org [52].

Comparison of selected proteins (Der, ribosome biogenesis GTPase; Eno, phosphopyruvate hydratase; FtsY, signal recognition particle-docking protein; Gap, type I glyceraldehyde-3-phosphate dehydrogenase; MiaA, tRNA (adenosine(37)-N6)-dimethylallyltransferase; MutL, DNA mismatch repair endonuclease; MutM, DNA-formamidopyrimidine glycosylase; ParC, DNA topoisomerase IV subunit A; PolC, PolC-type DNA polymerase III; RecA, recombinase RecA; RpoB, DNA-directed RNA polymerase subunit beta; RsmD, 16S rRNA (guanine(966)-N(2))-methyltransferase; RsmG, 6S rRNA (guanine(527)-N(7))-methyltransferase; SecG, preprotein translocase subunit; SpeB, agmatinase; TpiA, Triose-phosphate isomerase; TruA, tRNA pseudouridine(38-40) synthase; Tuf, elongation factor Tu; UvrA, excinucleaseABC_subunit) to determine percent identify was performed using Clustal (v.1.2.4) [53].

2.6. Biochemical Testing

In addition to the genomic characterization of A. testudinis as a new species, we also incorporated recommended biochemical tests [28]. All clinical isolates had a classic fried egg appearance on solid medium. Absence of a cell wall was confirmed by resistance to penicillin for all clinical isolates and by electron microscopy for the Type strain. Prior to testing, all isolates were filtered through a 0.45 µM filter three times. All tests included A. laidlawii and M. mycoides capri (formerly large colony), which have similar growth rates to the clinical isolates, as controls. Lack of incorporation of cholesterol in the membrane was determined by insensitivity to digitonin lysis as well as by the ability to grow after three successive passages in serum-free medium. For biochemical characterization, the presence of enzymes associated with common metabolic processes were tested using the Becton Dickinson & Company Baltimore Biological Laboratories (BD BBL™) Crystal Bacterial Identification kit for Gram positive microbes. The tests were performed according to manufacturer’s protocol, and results were read at 24 hours and 48 hours. To determine temperature growth range, isolates were serially diluted in 96 well culture plates containing 180 $\mathsf{\mu }$L of SP4 medium without serum and incubated at 4, 25, 30, 35, 40, or 45˚ C. Plates were read at 12, 24, and 48 hrs and checked for color change indicative of growth. For electron microscopy, broth culture was dropped on a filter membrane (Cytiva Whateman Nuclepore Track-Etch membrane, 0.2 µM), air dried for 30 min, and fixed with 2.5% glutaraldehyde for 1 hr. The membrane was then rinsed with PBS and re-fixed with 1% OsO4 for 1 hr. After washing with PBS, the membrane was dehydrated by 30%, 50%, 70%, 80%, 90% and 100% ethanol. To take the SEM pictures, the membrane was sputter-coated for 1 min and image was captured using a Quanta FEG 650 scanning electron microscope.

2.7. Experimental Infection Studies

All study protocols for animal use were approved by the University of Florida Institutional Animal Care and Use Committee (IACUC) A753 and performed under FWC research permit # WX02160d. Nine tortoises were inoculated intranasally with 10⁸ CFU of A. testudinis ORD1043; seven control tortoises were inoculated with sterile SP4 broth as previously described [12]. Animals were evaluated for clinical signs and changes in appetite or activity daily by trained animal care technicians during feeding. Tortoises were evaluated by a veterinarian (LDW) at a minimum of every other day for the first week post infection, and twice weekly thereafter. At each veterinary monitoring period, animals received a general physical examination and were evaluated for the following clinical signs of URTD: nasal and ocular discharge, palpebral edema, and conjunctivitis. A photographic record of each tortoise was made before inoculation and when clinical signs were observed. Clinical signs were graded individually on a scale of 0-3, respectively none, mild, moderate, or severe. Visual grading of signs was confirmed by independent observation of the photographic record as deemed necessary. Any other pertinent physical examination findings or behavioral changes (i.e., observed changes in appetite, activity level, basking, etc.) were noted. At 91 days post infection (PI), tortoises were euthanized by administration of ketamine intramuscularly at a dose of 60 to 80 mg/kg, followed by a concentrated barbiturate solution (pentobarbital) administered intravenously at a dose of 390 mg/kg. A complete necropsy was performed, including serology, cultures, PCR, and histopathology. Lesion scoring has been described in detail [12,54], and briefly is as follows: 0 = Normal; 1 = Mild lesions (mild leukocyte infiltration, mild edema and minimal changes in the mucosal epithelium); 2-3 = Moderate lesions (moderate leukocyte infiltration, moderate edema, and proliferative/dysplastic changes to the mucosal epithelium); and 4-5 = Severe lesions (diffuse infiltrate, marked edema, necrosis of the epithelium with erosions, and dysplastic and metaplastic changes in the olfactory epithelium).

2.8. Statistical Analysis

All statistics were conducted using Prism 11.0.2 (GraphPad Software, San Diego, CA). For clinical signs over time, data were analyzed by two-way ANOVA, with Geisser-Greenhouse’s epsilon correction factor. Days with clinical signs were analyzed by Fisher’s exact test. Lesion severity data was analyzed by the Wilcoxon test. A p-value < 0.05 was considered statistically significant.

2.9. Data Availability

The genome sequence for Acholeplasma testudinis sp. nov. Type strain ORD1043 is available as BioProject: PRJNA923665, BioSample: SAMN32730483, JAQLOJ000000000. Genome sequences used for A. equifetale (GCF_900660755.1), A. equirhinis (GCF_017052655.1), A. granularum (GCF_000526235.1), A. hippikon, A. laidlawii (GCF_000018785.1), A. manati (GCF_025742995.1), A. multilocale (GCF_000483165.1), A. oculi (GCF_900444665.1), A. vituli (GCF_025446935.1), and Acholeplasma sp. SMAG U10445 (DATGRG010000001-DATGRG010000107) are available from GenBank. A. testudinis sp. nov. Type strain ORD1043 is available from the authors (MBB) and will be deposited in ATCC.

3. Results

3.1. Acholeplasma Testudinis Is a New Species, with Strain ORD 1043 as the Type Strain

Twelve clinical isolates of Acholeplasma were isolated from the nares of wild gopher tortoises on 7 study sites in north central Florida (Figure 1). 16S phylogenetic analysis (Figure 3) confirmed that 11 isolates were predicted to be a new species, with A. equirhinis as the most closely known relative. Based on PCR amplification and sequencing of the 16S rRNA gene, 11 clinical isolates had a 99.2 ± 0.13% identity of 16S rRNA (Figure 3B) and were considered to be the same species [29]. A. testudinis ORD1043 was chosen as the Type strain. One clinical isolate (FE1018) did not cluster with the others and based on 16S homology, was considered to be A. hippikon. A. testudinis is only the fourth new formally named new species in the past two decades and only the eighteenth overall (Figure 1). A. testudinis^Ts ORD1043 and all clinical isolates exhibited colonies with a typical fried egg appearance (Figure 3C) and electron microscopy confirmed the absence of a cell wall (Figure 3D).

3.2. Biochemical Characteristics

Classical methods for minimum standards [15,19,28] for determination of a new species were performed on 11 clinical isolates. A. testudinis^Ts ORD1043 and 10 clinical isolates were filterable through a 0.45 $\mu$m filter and did not require serum for growth. All isolates were resistant to penicillin, insensitive to digitonin lysis, and produced acid from glucose and trehalose. All isolates failed to hydrolyze urea or esculin. The tortoise pathogens M. agassizii and M. testudineum grow very slowly, requiring 4-6 weeks in culture, but the A. testudinis isolates grew within one week of primary isolation. All isolates grew well in SP4 medium at a growth range of 30-45 °C, with optimal growth at 35-3 °C.

3.3. Comparison of 16S rRNA Sequences

The evolutionary relationship (Figure 3A) showed that A. testudinis and 10 clinical isolates formed a distinct cluster, with an average nucleotide identity of >99% (Figure 3B). The most closely known relative for A. testudinis is A. equirhinis. Of note is that one clinical isolate (FE1018) had >99% identity with A. hippikon; both were 95 and 96% identical with A. testudinis and A. equirhinis, respectively. The full 16S rRNA nucleotide identities for all known Acholeplasma species and the clinical isolates is shown in Figure 3B.

3.4. Whole Genome Sequencing (WGS) of A. Testudinis^ts ORD1043, GenBank PRJNA923665

Molecular methods for minimum standards [29,30,31] for determination of a new species were performed. Illumina WGS sequencing of A. testudinis^Ts ORD1043 resulted in 30 contigs, 12 of which were >50,000 bp. The largest contig was 231,374 bp. The genome size of 1,472,290 bp and 31.8% CG content is consistent with other Acholeplasma spp. A total of 1,373 genes were annotated, including 1,325 protein coding, 3 rRNA (5S, 16S, 23S), 35 tRNAs, and 3 ncRNA genes. Seven pseudogenes were annotated, 6 of which were incomplete and 1 of which contained an internal stop codon. Because the WGS is available for named members of Acholeplasmatacea, we were able to compare A. testudinis^Ts ORD1043 with other phylogenetically related Acholeplasma spp. (Figure 4). This rigorous molecular comparison [29,31,55] of A. testudinis with the closest type strain genomes [49,56] at the 16S (Figure 4A), WGS (Figure 4B), and proteome levels (Figure 4C) all confirmed that A. testudinis^Ts ORD1043 is a new species with A. equirhinis as the closest relative. The accepted species boundary for average nucleotide identity (ANI) and digital DNA:DNA hybridization (dDDH) values are 95-96% and 70%, respectively. The values obtained in comparison of A. testudinis^Ts ORD1043 with other Acholeplasma spp. (Table 1) clearly met and far exceeded the species boundaries.

3.5. Comparison of Selected Proteins

Because of the rapid expansion of metagenomic studies of noncultivated bacterial populations, we also used the similar genome finder to expand and identify other potential Acholeplasma spp. [52] The phylogenetic tree (Figure 5A) found a highly similar genome SMAG-U10445 from the metagenomic analysis of soil samples obtained in Utah [22]. Analysis of the genome sequence revealed that only partial 16S rRNA sequences were available and were insufficient to generate a match during BLAST searches among the known species. However, genes encoding 12 of the 19 proteins analyzed (Figure 5B) were present in SMAG-U10445; 9 of the encoded proteins shared 100% identity with A. testudinis. This provides strong support for our hypothesis that A. testudinis was likely acquired by gopher tortoises during burrowing activities and interaction with soils. The percentage identity heat map (Figure 5B) with other known Acholeplasma spp. provides additional support that A. testudinis is a new species.

3.6. A. Testudinis^Ts ORD1043 has Limited Pathogenic Potential

For the most part, Acholeplasma spp. are not associated with pathogenicity and are primarily considered as commensals or transient flora from environmental sources [24]. In an experimental infection, we confirm that A. testudinis has limited to no pathogenic potential. Because clinical signs may be intermittent, animals were monitored frequently during the 91 day infection period. The two most common clinical signs associated with URTD are nasal and ocular discharge. The clinical presentation in individual animals over time is shown in Figure 6. For both nasal and ocular discharge, the clinical signs were extremely mild and highly unlikely to be biologically significant (Figure 6). Although there was a minimal increase in severity of nasal discharge seen in experimentally infected tortoises, the effect of treatment was not statistically significant (P=0.09, Figure 6); similar findings occurred with nasal discharge (P=0.09, Figure 6). An interaction between time and treatment was seen with ocular discharge, with ocular discharge seen more commonly in the first few weeks post infection (P=0.008, Figure 6). For both nasal and ocular discharge, the greatest source of variation was the individual animal (P<0.0002). Two tortoises (Tortoise 6, infected group; Tortoise 9, control group) had more severe and frequent clinical signs. Although all animals had tested negative prior to the infection study, at necropsy these two tortoises were culture/PCR positive for M. agassizii, the etiologic agent of URTD. Because results from these tortoises were confounded by the presence of a known pathogen and were identified statistically as anomalies, they were excluded from other analyses but are presented in Figure 6 for completeness of the dataset.

The mean clinical signs for each individual tortoise, excluding the two M. agassizii positive animals are shown in Figure 7 A, B and demonstrate the individual animal variation. We next determined if, excluding the two M. agassizii positive animals, there was a difference in the total number of days when clinical signs, regardless of severity, was present (Table 2). There were 27 monitoring times for each tortoise over the 91 day infection study. Nasal discharge occurred more frequently (P<0.0001, Fisher’s exact test) in infected (11.11% of days) than in control (0.53% of days) animals. However, all animals in both groups experienced only mild clinical signs that were deemed not to be biologically significant. The number of total days with clinical signs of ocular discharge was not different between control (29.63% of days) and infected (29.17% of days) tortoises, P>0.9. However, moderate signs were seen on 6 of 63 (9.5%) clinical day events in infected animals whereas no control animals had moderate clinical signs reported.

3.7. Histological Changes Support Limited Pathogenicity (Figure 7C)

No animals developed detectable antibody levels to A. testudinis. In contrast, in a previous study [12], tortoises experimentally infected with a significantly lower infectious dose of M. agassizii seroconverted by 4 weeks post-infection. Although differences in lesion severity between treatment group did not achieve statistical significance, three animals that were infected with A. testudinis had histopathological upper respiratory tract lesions compatible with those seen in URTD (Figure 7C); the M. agassizii-positive tortoise #9 was excluded. For most infected tortoises, the lesions were mild, consisting of mild diffuse increases in lymphocytes and heterophils in the respiratory epithelium. No other consistent lesions were observed. However, three infected tortoises had more substantive changes to the respiratory and olfactory epithelium, including lymphocytic and heterophilic infiltrates, some loss of architecture, erosion, and perivascular cuffing. These animals also had moderate clinical signs at least once during the study period. All control tortoises had mild or no lesions.

4. Discussion

Within the Mollicutes, the monophyletic family Acholeplasmataceae contains the genus Acholeplasma and the provisional taxon “Candidatus Phytoplasma”. Species within both genera share similar characteristics of reduced genome size, loss of a cell wall, and loss of metabolic pathways [57,58,59]. A notable difference is that while acholeplasmas can be cultivated, to date phytoplasmas are refractory to culture from either their phloem-restricted plant hosts or their vector transmission hosts [60,61,62,63,64]. In spite of this limitation, there are currently genomes available for >70 Phytoplasma spp. while the A. testudinis genome reported here is one of only 14 genomes for Acholeplasma spp. Another striking difference is the virulence potential associated with the two genera. Ca. Phytoplasma spp. are major phytopathogens, causing significant worldwide disease in vegetable crops as well as in ornamental horticulture plants[60,61,62]. In direct contrast, Acholeplasma spp. are found in the environment and isolated or detected by molecular methods from soil, water, plants, invertebrates, and vertebrates, but they are rarely associated with clinical disease [24,65]. Acholeplasma spp. are, however, notorious as tissue culture contaminants and are included among microbial agents that must be monitored during industrial production [66,67,68,69,70,71].

Because our original isolation of A. testudinis was from the nares of a gopher tortoise with mild nasal discharge, we felt that it was important to confirm if this microbe had pathogenic potential. The 10⁸ CFU infectious dose chosen for our study was purposefully high, as we reasoned that even limited pathogenicity could be detected at that infectious dose. We were guided by results from previous experimental infections with M. agassizii strain 723 in gopher tortoises [12] and M. agassizii strain PS6 in desert tortoises [13]. Intranasal inoculation with 10⁸ CFU M. agassizii strain 723 produced both ocular and nasal discharge in 7 of 9 tortoises at 4 weeks and 8 of 9 tortoises at 8 weeks post infection (PI). Mean nasal and ocular discharge clinical scores in these tortoises were >1 from 4 weeks PI until the end of the observation period at 16 weeks PI [12]. In contrast, infection with A. testudinis in our current study did not result in significant clinical signs. Two animals (1 in control group and 1 in infected group) that tested positive for M. agassizii at necropsy were removed from the dataset as potential confounders but data from clinical signs in these animals is presented for completeness.

Ocular discharge was not different between control and infected animals, occurring in 29% of observational days in both groups. Interestingly, one of the M. agassizii-infected tortoises that was excluded from the study had an ocular discharge for 26 of the 27 (96%) observational days. While the majority of the ocular discharge was mild, moderate clinical signs were observed on 4 separate days in the tortoise known to be co-infected with both A. testudinis and M. agassizii. Moderate ocular discharge was also observed for 1 day (N=3 tortoises) and 3 days (N=1 tortoise) in the infected group. No moderate nasal discharge was seen in the control tortoises with the exception of the M. agassizii positive animals (N= 2 days). Nasal discharge was observed in 24 of the 216 observational days for infected and only 1 of 189 observational days for control animals. While statistically significant, this likely is not biologically significant as all clinical signs were mild. Additionally, with the exception of 1 tortoise that had clinical signs for 14 of the 27 observational days, all other tortoises had clinical signs at only 4 (N=2), 2 (N=1), 1 (N=1) or none (N=3) of the 27 observational days per tortoise. No nasal discharge was observed in 5 of 8 tortoises in the first 4 weeks of infection. For the overall study period, nasal discharge was observed at more than 15% of observational days for only one tortoise (48% of days with clinical signs) and the M. agassizii-infected tortoise (63%) that was excluded from the study. Control animals experienced nasal discharge for 0.5% of observational days. Therefore, we cannot rule out on that A. testudinis might impact limited clinical responses or exacerbate existing infections with other pathogens. However, none of the tortoises had a serological response to A. testudinis. In previous experimental infections with 10^1, 10³, and 10⁵ CFU M. agassizii strain 723, an antibody response was seen by week 6 PI, with linear increases thereafter [12]. Taken together, our data suggests that A. testudinis is not a significant pathogen and most likely a commensal.

Because gopher tortoises are fossorial and spend 90% of their time in burrows [72], the most likely source of A. testudinis is soil. As herbivores, plant materials are also a potential source. However, given the habitat differences on the multiple sites where A. testudinis was isolated from tortoises, we feel that this is a less likely source than soil. Even more compelling was the discovery of a noncultivated Acholeplasma sp. SMAG U10445 from a metagenomic study of soils in Utah that had 100% identity with 9 housekeeping proteins (Figure 5B) and clustered with A. testudinis in a similar genome phylogenetic analysis (Figure 5A).

Another potential implication of our study is the role that A. testudinis as well as other potential new species might have on development of pathogen discovery and diagnostics for wildlife diseases. With the increased use of molecular techniques, the field of pathogen discovery in wildlife is increasing dramatically [73,74,75,76]. It is not unlikely that, especially for viruses, future studies to isolate as well as to elucidate pathogenic potential may require development of specific new cell lines from relevant hosts. Because of the known potential for both Acholeplasma spp. and Mycoplasma spp. to both contaminate and induce cytopathic changes in cell lines, it is important that newly developed cell lines be rigorously screened for these agents [77,78,79,80].

5. Conclusions

A. testudinis is one of only 4 new species of Acholeplasma that has been formally describes in the past two decades and to our knowledge, the only species isolated from as reptile. Based on WGS analysis as well as conventional 16S rRNA sequencing and biochemical testing, A. testudinis^Ts ORD1043 clearly meets the standards for new species status. In a recent study [81],the ribosomal rescue system and conserved genome synteny was described in Acholeplasma. Similar to what was found in five Acholeplasma spp.[81], in A. testudinis smpB (PG914_04165) also is flanked by rnr (PG914_04160) upstream and immediately downstream by a patatin-like phospholipase family protein (PG914_04170) followed by a phosphatase PAP2 family protein (PG914_04180). We also found evidence of a downstream DDE-type integrase/transposase/recombinase (PG914_04200) and IS3 family transposase (PG914_04205). The species description follows.

Description of Acholeplasma testudinis sp. nov.

[tes.tu.di.nis. L. n. testudo tortoise; L. gen. n. testudinis of the tortoise]

Lineage: Bacteria; Terrabacteria group; Tenericutes; Mollicutes; Acholeplasmatales; Acholeplasmatacea; Acholeplasma

Cells lack a cell wall. Cells were filterable through a 0.45 μM filter. Isolates grew in SP4 broth containing no serum or added cholesterol. They grew between 30˚C and 45˚C, with optimal growth at 35-37 °C. Isolates fermented glucose, 16S rRNA and whole genome sequencing analysis showed these isolates $\le$94% similar to reference Acholeplasma strains. Isolated from the nares of wild gopher tortoises (Gopherus polyphemus). Limited pathogenicity based on experimental infection study. The ribosomal rescue system and conserved genome synteny was present [81]. The 16S rRNA and whole genome sequence (PRJNA923665, BioSample: SAMN32730483) is distinct, and the Type strain is ORD1043.