Panophthalmitis in a Gentoo Penguin (Pygoscelis Papua) From The Antarctic Peninsula: Evaluation of Microbiological and Histopathological Analysis Outcomes

In this study, it was aimed to present the results of microbiological, cytological, histopathological, and immunohistochemical analyses of ocular samples from an Antarctic (Ardley Island, King George Island) Gentoo penguin chick (Pygoscelis papua) with pyogranulomatous panophthalmitis in the right eye. Samples were taken from both the healthy left eye and the lesion in the right eye. Conventional culture methods and phenotypic and molecular tests were used for bacterial isolation and identication, respectively. None of the isolates could be identied phenotypically. As a result, four of the ve isolates obtained from the right eye were considered to belong to putative novel bacterial species and taxa as their similarity to GenBank data was below 98.75%. The isolates were considered to be Pasteurellaceae bacterium, Corynebacterium ciconiae, Cardiobacteriaceae bacterium, Actinomyces sp., and Dermabacteraceae bacterium. The only isolate from the left eye was identied as Psychrobacter pygoscelis. The cytological analysis demonstrated cell inltrates composed mostly of degenerate heterophils, reactive macrophages, plasma cells, lymphocytes, and eosinophils. Based on histopathological ndings, the lesion was dened as typical pyogranulomatous panophthalmitis. Immunohistochemistry demonstrated that the granuloma was positive for TNF-α, IL-4, MMP-9, IL-1β, and IL-6. This is the rst documented report of unilateral pyogranulomatous panophthalmitis in a Gentoo penguin chick, living in its natural habitat in Antarctica. This report also describes the isolation of four bacteria from the infected eye, which are considered to belong to novel Genus, species, or taxa. The primary bacterial pathogen that caused the ocular lesion was not able to be detected and remains unclear.

Eyesight is of critical importance for penguins as it aids in migration, orientation, and foraging (Nesterova et al. 2010). Thus, vision disorders adversely affect the capability of these animals to adapt to their physical and social environment. While numerous bacteria, viruses, fungi, and parasites have been isolated from diseases of the avian ocular surface (Griggs 2019), only a few studies are available on the ophthalmic parameters of several penguin species, including ocular bacterial ora, intraocular pressure, and tear production ( Suburo et al. 1988 In general, cultural methods and histopathological, cytological and immunohistochemical analyses are employed to diagnose and determine the aetiology of ocular infections, and most often these different methods are used together. Isolation of bacteria from cases of infection is carried out by cultural methods. The isolates recovered are identi ed at the species level by means of phenotypic tests and molecular analyses (Quinn 2011). It is In situ examination of the right eye demonstrated the presence of a reddish-orange coloured, granulation tissue-like, proliferative and exuberant lesion on the ocular surface. The movement of the nictitating membrane was not observed due to this lesion. The proliferative tissue was determined to occupy the cornea and other parts of the eyeball (Fig. 2). An extensive exuberant mass, which occupied the anterior chamber, made it impossible to identify the intraocular structure. No macroscopic lesion was observed in the left eye.
Using a sterile scalpel, samples were taken from the mass in the right eye into 1 ml of sterile distilled water for microbiological analysis, Cytospin collection uid (Cytobrush, Gynobrush Plus, Germany) for cytological analysis, and 70% ethyl alcohol for histopathological analysis. The left eye of the penguin, which appeared to be healthy, was sampled for microbiological and cytological analyses. The microbiological samples were collected with a swab into Stuart's transport medium, whilst the cytological samples were collected either by brushing into Cytospin collection uid or by the membrane lter technique (cellulose-acetate lter papers; Sartorius, HAWG 047 AO, 3x4 mm in diameter, 0.45-micrometer pore size) into 70% ethyl alcohol (Bolzan et al. 2005; Singh et al. 2005). The swab and brush were gently rolled over the cornea on the limbus, such that the palpebral and forniceal conjunctiva, anterior to the third eyelid, were also sampled.
Filter strips with a pointed tip on one end were placed on the superior, ventral temporal bulbar conjunctiva, near the limbus as described by Bolzan et al. (2005) and Singh et al. (2005) with minor modi cations. The strips were removed with a peeling motion, and the samples were immediately transferred to the laboratory at the Professor Julio Escudero Base, King George Island, Antarctica. The samples were stored at 4°C and examined within 2 weeks.

Bacteriological analysis
The ocular tissue sample taken into distilled water for bacteriological analysis was broken into small pieces by shaking. Slides were prepared from the resulting suspension, and were stained with the Gram and Ziehl-Neelsen techniques. 50 µl of the suspension was directly inoculated onto three plates of blood agar (Blood agar base No: 2, CM0271, Oxoid, Waltham, MA, USA/containing 7% sheep blood) added 2% of NaCl (Merck, Germany). The swab sample taken from the left eye was directly inoculated onto three blood agar plates. The inoculated plates were incubated at 37°C under aerobic, microaerobic and anaerobic conditions for 2-3 days. At the end of the incubation period, the colonies were examined both macroscopically (for size, colour, odour, haemolysis, etc.) and microscopically. Each different type of colony that had grown on the blood agar was subcultured onto sheep blood agar for puri cation. Both phenotypic tests (Gram staining, catalase, oxidase, motility and carbohydrate fermentation tests), as described by Quinn et al. (2011), and molecular analyses (16S rRNA gene sequence analysis) were performed for the identi cation of the recovered isolates at the species or genus level. The recovered isolates are stored in Brucella broth containing 15% glycerine at -84°C, as part of the culture collection of the Microbiology Department of Erciyes University, Faculty of Veterinary Medicine.

Mycological analysis
For the detection of possible fungal agents, samples from both eyes were inoculated onto Sabouraud dextrose agar (SDA, CM0041, Oxoid, United Kingdom). The inoculated plates were incubated at 22°C and 37°C under aerobic conditions for 2-3 days.

DNA extraction
The DNeasy UltraClean Microbial Kit (12224-50, Qiagen, USA) was used for DNA extraction from the bacterial isolates.

16S rRNA sequencing
The species identi cation of the isolates was made by means of 16S rRNA gene sequence analysis. Accordingly, four primers, including 27F and 1492R, and two internal primers 518R: ATT ACC GCG GCT GCT GG and 968F: AAC GCG GAA GAA CCT TAC) were used for the sequencing of the 1465 bp-fragment of 16S rRNA. The PCR mixture used for the ampli cation of this 1465 bp-fragment was 25 µl in volume and contained 12.2 µl of molecular grade deionised water, 2.5 µl 10x PCR buffer, 2.5 µl MgCl 2 (25 mM), 0.5 µl 10mM dNTP, 1 µl of the forward primer (10 pmol), 1 µl of the reverse primer (10 pmol), 0.3 µl Taq DNA polymerase, and 5 µl of the DNA recovered from the isolates (25 ng/µl). Ampli cation was performed as follows: initial denaturation at 94°C for 4 min followed by 35 cycles including denaturation at 94°C for 30 sec, primer annealing at 58°C for 30 sec, primer extension at 72°C for 1 min, and a nal extension at 72°C for 6 min. The PCR products were run on agarose gel electrophoresis and those, which were of the appropriate size (1465 bp) and underwent 16S rRNA sequence analysis. The 16S rRNA gene was manually aligned with type strain sequences retrieved from EzBioCloud database (Yoon et al. 2017 (Edgar, 2004). Maximum likelihood (ML) and maximum parsimony (MP) trees were inferred from the alignment with RAxML (Stamatakis 2014) and TNT (Goloboff et al. 2008), respectively. For ML, rapid bootstrapping in conjunction with the autoMRE bootstopping criterion (Pattengale et al. 2010) and subsequent search for the best tree was used; for MP, 1000 bootstrapping replicates were used in conjunction with tree-bisection-and-reconnection branch swapping and ten random sequence addition replicates. The sequences were checked for a compositional bias using the Χ² test as implemented in PAUP* (Swofford 2002).

Antibacterial susceptibility testing
The susceptibilities of the bacterial isolates to amoxicillin/clavulanic acid (AMC, 30 μg), ampicillin (AM, 10 μg), azithromycin (AZM, 15 μg), enro oxacin (ENR, 5 μg), gentamicin (CN, 10 μg), levo oxacin (LEV, 5 μg), and tetracycline (TE, 30 μg) were determined using the disc diffusion method (Clinical and Laboratory Standards Institute, 2018). The antimicrobial discs were purchased from Bioanalyse, TURKEY. Due to we could not identify the isolates at the species level, the disk diffusion test results of our isolates were evaluated by using the criteria published by the Clinical and Laboratory Standards Institute (CLSI 2018 and CLSI 2019) for Enterobacteriaceae and staphylococci, respectively. Escherichia coli (E. coli) ATCC 25922 and Staphylococcus aureus ATCC 25923 were used as reference strains for the susceptibility tests.

Cytospin Technique
The tissue samples taken from the proliferative mass into Cytospin collection uid for cytological analysis were agitated on a vortex mixer. Cytocentrifuged slides were prepared from this agitated suspension. The brush sample taken from the left eye was also transferred into a xative solution (Cytospin collection uid, Thermo Shandon 6768001, USA).
Wet cytocentrifuged slides were dried and then stained with the Haematoxylin-Eosin (HE) (Scott and Penny 1993) and Diff-Quik methods (Richard Allan Scienti c, Thermo Electron Corporation, 9990700, USA). The slides were mounted with mounting medium (Merck, Germany).

Impression Cytology
This method was performed as described by Bolzan et al. (2005) with minor modi cations. Brie y, the strips were immediately placed in individual Eppendorf tubes containing 95% ethyl alcohol. Next, the strips were stained with Periodic Acid-Schiff (PAS) x Haematoxylin, as described by Nelson et al. (1983) with some modi cations such as the use of a lter with a pore size of 0.45 μm and sodium metabisul te. The samples were hydrated with distilled water for 5 min, immersed in 0.5% periodic acid for 5 min, washed in distilled water for 5 min, immersed in Schiff's reagent for 5 min, washed under running water for 5 min, immersed in sodium metabisul te (three successive immersions, each for 5 min), washed under running water for 10 min, counterstained with Harris' haematoxylin for 1 min, and washed under running water for 5 min. After these steps, the samples were dehydrated in 95% and 100% ethyl alcohol (two successive immersions, each for 1 min) and immersed in xylene (two successive immersions, each for 5 min). Finally, they were mounted onto slides in a way that the cells were faced upwards, and were cover-slipped using mounting medium (Bolzan et al. 2005).

Histopathological analyses
Tissue samples underwent routine processing and were embedded in para n. Nine 5-to 6-μm-thick serial tissue sections, representative of the entire tissue sample, were cut from the para n block and mounted onto glass slides. Each para n section was stained with H&E, and the Brown-Brenn Gram (Pritt et al. 2006), Grocott methamine silver stain (Polyscience) (Wright et al. 2017), acid-fast stain, and immunohistochemical stains. Furthermore, the streptavidin-biotin peroxidase (SABP) complex method (Invitrogen, USA) was performed using ve different antibodies, including those against TNF-α, IL-4, MMP-9, IL-1β, and IL-6. While Chicken Bursa Fabricius tissue sample was used as a positive control, Phosphate buffer solution (PBS) was used instead of primary antibody in the negative control slides. In addition, since there is no commercially produced antibody speci c to penguins, antibodies that are routinely used for immunohistochemical staining in poultry were used in the current study. The primary antibodies used in the present study are shown in Table   S1. Immunoreactions were visualized using aminoethyl carbazole chromogen (AEC, Invitrogen, USA) ( All cytological, histopathological and immunohistochemical (IHC) slides were imaged with a digital camera (Olympus DP71) and digital programmers (DP Controller and the DP Manager) tted to a microscope (BX-51, Olympus, Tokyo, Japan) (minimum of 4 elds per slide, using x10, x40, and x100 objective lenses).
Electron microscopic examination of the isolates All isolates were imaged by the transmission electron microscopy (TEM) and scanning electron microscopy (SEM) for the examination of the individual cell morphologies, at Erciyes University Technology Research and Application Center (ERU-TAUM, Kayseri, Turkiye). For this purpose, 24-48 hour fresh cultures belong to isolates grown in the solid medium were used.

Bacteriological ndings
While the samples were found to be negative for acid-resistant bacteria by Ziehl-Neelsen staining, they were observed to contain numerous Gram-negative coccobacilli and Gram-positive pleomorphic bacilli by Gram staining. The culture method yielded a total of 6 isolates, 5 from the right eye, and one from the left eye of the penguin. The isolates recovered from the right eye were numbered as TAE3-ERU1, TAE3-ERU2, TAE3-ERU3, TAE3-ERU4, TAE3-ERU5, and the isolate from the left eye was numbered as TAE3-ERU6.
The phenotypic and physiological characteristics of the isolates are presented in Table 1.

16S rRNA gene sequence analysis
Based on the BLAST (basic local alignment search tool) and EzBioCloud database searches of the 16S rRNA gene sequencing results of the isolates recovered from the right and left eye demonstrated that 2 of 5 isolates could be done species-level identi cation. Other isolates, which were found below threshold (98.7%) with similarities rate of the type strain, could not be identi ed as species-level. 16S rRNA gene sequence similarity of our isolates is given in Table 2. Each sequence was deposited in the GenBank database with MT444110-MT444114, and MT674807 for TAE3-ERU1 -TAE3-ERU5 and TAE3-ERU6, respectively.

Mycological analysis
All of the samples examined were found to be negative for fungi.

Phylogenetic analysis
According to the phylogenetic analysis result; TAE3-ERU1 was determined to be close to Bisgard Taxon 14 and also this isolate could be a new genus in Pasteurellaceae TAE3-ERU2 was determined to be close to Corynebacterium ciconiae CCUG 47525 type strain.
TAE3-ERU3 was determined to be close to Rappaport and Suttonella in the Cardiobacteriaceae family, TAE3-ERU4, determined to be close the Actinomyces , TAE3-ERU5, found to close in the Devriesea genus in the Dermabacteraceae family TAE3-ERU6 was determined to be close to Psychrobacter pygoscelis I-STPP5b type strain.
Detailed information about phylogenetic analysis was given below phylogenetic trees of isolates in Fig. S1- Furthermore, distinct squamous conjunctival metaplasia, characterized by mucus, keratin, and pyknotic cells with an enlarged cytoplasm and a decreased nucleus/cytoplasm ratio, was observed in some areas. In the impression slides, only abundant heterophil leukocytes and bacteria were observed, and goblet cells were not detected In the cytospin slides prepared from the sample taken from the left eye, columnar epithelial cells, small and round conjunctival cells, and non-corni ed squamous corneal epithelial cells, were common. It was observed that the nucleus/cytoplasm ratio of these eosinophilic cells had not altered and that the cells were similar to each other in shape and size. The examination of the impression slides revealed numerous heterophils, lymphocytes, macrophages, goblet cells and extracellular bacteria. (Fig. S7 (A-I)).

Histopathological ndings
Histopathological examination showed that the proliferative mass, characterized by necrotic debris, cell in ltration, and numerous pleomorphic and lamentous branching bacilli, and sulfur granules, had replaced all layers of the anterior chamber of the eye. It was observed that the surface epithelium of the cornea had completely disappeared and the stroma was necrotic. Dense in ammatory cell in ltration had extended to the deep stromal layer of the cornea. The anterior chamber was completely displaced by necrotic and pyogranulomatous lesion. Furthermore, broblasts and brocytes were found at signi cantly higher numbers in the anterior chamber of the eye. Histopathologically, the reddish-orange coloured proliferative mass was a typical pyogranuloma. The granuloma was mature and displayed extensive coagulation necrosis at the centre with multinucleated giant cells palisading along the necrotic line. These giant cells were surrounded by an in ltration of numerous macrophages, heterophils, lymphocytes and broblasts, and few plasma cells and eosinophils ( Fig. 3 A-E). Brown-Brenn staining demonstrated the clustering of club-like organisms comprised of numerous pleomorphic and lamentous Gram-positive bacteria within sulphur granules, which were surrounded by neutrophils and lymphocytes. Also, few small Gram-negative coccobacilli were also observed.
Grocott-Gomori methenamine-silver staining revealed the presence of branched agents, characteristic of actinomycosis (Fig. S8). Fungi, and based on the results of Ziehl-Neelsen staining, acid-fast bacteria were detected neither within nor in the periphery of the granuloma. Furthermore, dysplastic alterations, including large hyperchromatic nuclei and binucleated cells with an increased nucleus/cytoplasm ratio and an irregular nucleus, were also observed. A histopathological section representative of the proliferative mass is shown in Fig. 3 (F-Q).
In the conjunctival epithelial and stromal cells of the cornea, strong IL-4 immunoexpression was detected (Fig.  3T). Cytoplasmic immunoreactivity for IL-4 and MMP-9 was observed to be very strong in the regions of the conjunctiva where macrophages and giant cells and necrosis were intense. (Fig. 3Y, Z, X).
Moderate cytoplasmic immunoexpression of MMP-9 was detected in the epithelial cells of the cornea. While the granuloma, which was composed of heterophils, macrophages, and multinucleated giant cells, showed positive immunoreactivity for IL-1β, TNFα, and IL-6, the immunoreactivity for these proin ammatory cytokines was moderate in the epithelial cells of the cornea and sclera (Fig. 3R, S; U-Y, AA, AB). Immunostained slides prepared from the brush sample of the left eye were used as a control for the evaluation of MMP-9, TNF-α, IL-4, IL-1B, and IL-6 positivity. The immunochemical staining results of the slides prepared from the right and left eyes are shown in Fig. 3R-AB.

Morphological ndings of the isolates
The morphological appearance of the bacterial isolates by light microscope, Transmission electron microscope, and Scanning electron microscopy SEM are presented in Fig. 4.

Discussion
In the present study, an ocular lesion occupying the right eye of a Gentoo penguin chick, living in its natural habitat (Antarctic Peninsula), was de ned as pyogranulomatous panophthalmitis, based on cytological, histopathological and immunohistochemical analyses. Bacteria, which were unable to be identi ed at species level, but classi ed under 5 different genera, and considered belonging to putative novel taxa, genera and species, were isolated from the ocular lesion. Furthermore, Psychrobacter pygoscelis was isolated from the seemingly healthy left eye of the same animal. To our knowledge, in respect of its particularities described, In the present study, based on histopathological examination (Fig. 3), the lesion in the right eye of the penguin was de ned as pyogranulomatous panophthalmitis. To the authors' knowledge, there is no previous report on the diagnosis of panophthalmitis in captive or non-captive penguins. However, reports have been published on Pseudomonas aeruginosa-induced cases of unilateral panophthalmitis in turkeys ( Nakamura and Abe 1987; Aziz and Barnes 2001), and E. coli-induced cases of colisepticaemia-related panophthalmitis in chickens. The cellular nature of the panophthalmitis cases detected in chickens was described as being granulomatous (Gross 1957). Similarly, in the present study, based on the cellular characteristics determined by histopathological examination, panophthalmitis detected in the right eye of the penguin was de ned as being pyogranulomatous. Granulomas are generally characterized by well-delineated focal lesions that develop in the event of the immune system's failure to eliminate infectious microorganisms. These lesions primarily contain mononuclear phagocytic cells, macrophages (histiocytes), epithelioid cells and multinucleated giant cells (Williams and Williams 1983).While bacterial and fungal granulomatous lesions associated with Mycobacteria spp., Salmonella spp., Escherichia coli, Nocardia spp., Actinomyces spp. and Aspergillus spp. have been detected in various systems of avian species, those associated with Nocardia spp. and Actinomyces spp. have been reported only rarely (Park and Jaensch 2009). In the current study, while Brown-Brenn staining of the ocular tissue sections prepared from the lesion demonstrated the presence of many Gram-positive lamentous pleomorphic bacteria, and Gram-negative coccobacilli and pleomorphic bacteria, GMS staining displayed the presence of branching lamentous bacteria surrounded by intense in ammatory cell in ltration, typically referred to as sulphur granules (Fig. S8). The culture method enabled the isolation of bacteria from the ocular lesion, which were determined as belonging to 5 different genera, and included Pasteurellaceae, Corynebacterium ciconiae, Cardiobacteriaceae, Actinomyces sp., and Dermabacteraceae. Although sulfur granules have been reported to occur in infections caused by agents such as Nocardia spp., Staphylococcus spp., and Streptomyces spp., (Wang et al. 2012), these agents could not be isolated in this study. In addition, lamentous structures characteristic of Actinomyces were detected in sulfur granules in the GMS staining. So these agents (Nocardia spp., Staphylococcus spp., and Streptomyces spp.) have been eliminated as possible aetiological agents based on these results. Also, no fungal growth having been observed upon mycological analysis, and the absence of acid-resistant bacteria having been con rmed by Ziehl-Neelsen staining in the histopathological and microbiological analyses eliminated the possibility of Mycobacteria spp. and Aspergillus spp. being involved in the aetiology of the granuloma. The bacteria, which were isolated by the culture method and identi ed as Corynebacterium ciconiae, Actinomyces sp. and Dermabacteraceae bacterium, all belong to the class Actinobacteria and cause granulomatous in ammation. Therefore, the granulomatous in ammation detected in the right eye of the penguin is considered most probably to have been caused by these particular organisms. There is only one previous study reporting granulomatous in ammation in penguins, which describes adenitis of the salt gland in blue penguins (Eudyptula minor) inhabiting New Zealand (Suepaul et al. 2010). Thus, to the authors' knowledge, this is the rst study to report unilateral pyogranulomatous panophthalmitis in a Gentoo penguin chick. Many different bacteria, viruses, fungi and parasites have been isolated from cases of ocular surface disease in various avian species  ( Shoshani et al. 2005;Cole et al. 2005;Kaiser and Stäheli, 2013). Similar to previous reports in other avian species (Kaiser and Stäheli, 2013), it was also determined in the Gentoo penguin that pyogranulomatous panophthalmitis was associated with high level of MM-9 secretion and strong TNFα, IL-1β, IL-6 and IL-4 immunoreactivity in response to the in ammatory stimulus (Fig. 3). In the present study, based on strong TNFα, IL-1β, IL-6, IL-4 and MMP-9 immunopositivity having been detected in the in ammatory cells, it was considered that the indicated cytokines and enzyme could have accompanied the development of the granuloma in the eye diagnosed with mixed bacterial infection.
Staphylococcus spp., Streptococcus spp., Clostridium spp., Erysipelothrix spp., Mycoplasma spp., Moraxella spp., Chlamydophila psittaci, Salmonella spp., Bordotella spp., Escherichia coli, Pseudomonas spp., Actinobacillus spp., Pasteurella spp. and Mycobacterium spp. Today, the identi cation of microbial agents isolated from cases of infection by the culture method and molecular techniques is particularly signi cant for the establishment of treatment protocols (Miller et al. 2009). While the similarity of 16S rDNA gene sequences between two strains serves as a simple yet reliable criterion for the identi cation of novel strains, phylogenetic analysis has become the key to elucidate the evolutional relationship between related taxa ( Ludwig and Klenk 2001; Kim and Chun 2014).
Six bacteria, referred to as TAE3-ERU1/TAE3-ERU6, were isolated from samples taken from the right and left eyes of the penguin using the culture method. These bacteria underwent detailed investigation, including phenotypic analysis such as Gram staining, as well as molecular analysis and electron microscopic examination.
TAE3-ERU1, which was recovered from the right eye and identi ed as belonging to the family Pasteurellaceae, was 98.65% genetically similar to Bisgaard Taxon 14 according to the Genbank Database, and was found to be positioned between Bisgaard Taxon 14 and Bisgaard Taxon 32, based on phylogenetic analysis (Fig. S1). Based on EzBioCloud database, 16S rRNA similarity of TAE3-ERU1 was found to Pasteurella testudines (94.22%). For describing of a novel genus has to be as an approximately 5-6% divergence in 16S rRNA gene sequence from its closest genus in taxonomy studies (Qin et al. 2014). Due to the similarity value of TAE3-ERU1, we concluded that this isolate belonged to a novel genus in Pasteurellaceae family.
The family Pasteurellaceae includes 30 genera and 94 validly named species in current use (Bisgaard's taxa 2020; ICSP 2020; LPSN 2020a; Pasteurellacea 2020). This family also includes unidenti ed 55 taxa, known to have been isolated from various sources, which have been enumerated from 1 to 55 under the Bisgaard taxon (Bisgaard's taxa, 2020). Bisgaard Taxon 14 and Bisgaard Taxon 32 are known to be avian taxa (Christensen et al. 2003). Taxon 14 is reported to have been isolated from lesions of the upper respiratory tract in ducks, turkeys, pigeons, geese and peacocks ( Bisgaard and Mutter, 1986;Bisgaard, 1993) and from cases of blepharoconjunctivitis in turkeys (Günther et al. 2006). Furthermore, Taxon 32 has been isolated from cases of pneumonia and conjunctivitis in pigeons and buzzards (Christensen et al. 2003).
TAE3-ERU2 was identi ed as a Corynebacterium ciconiae with 99.05%. The genus Corynebacterium comprises 125 species (LPSN, 2020b) of medical, veterinary or biotechnological importance, which have been isolated from various sources (Dorella et al. 2006). Members of this genus are not only part of the normal microbiota of the skin and mucosa in a variety of hosts, but can also be found in soil, water and the environment (Ott, 2018).
TAE3-ERU3 was found to show 94.70% similarity to the feline oral taxon of Cardiobacterium sp., according to the Genbank database. Also, Based on the EzBioCloud database, this isolate was close to Pasteurella testudines (94.22%) and Rappaport israeli (93.92%) and was determined a member of the Cardiobacteriaceae family. Therefore, based on phylogenetic analysis, it was determined that isolate TAE3-ERU3 could belong to a novel genus under the family Cardiobacteriaceae (Fig. S3). The family Cardiobacteriaceae includes three genera, namely, Cardiobacterium, Dichelobacter and Suttonella. In 2017, the species Rappaport israeli was isolated from human blood culture and was classi ed under a novel genus belonging to this family. However, no information is available on this novel genus in the LPSN database.
16S rRNA similarity of TAE3-ERU4, recovered from the right eye of the penguin, was found to Actinomyces liubingyangii with 97.40%. Phylogenetic analysis and similarity value of isolate are revealed isolate TAE3-ERU4 to be a novel Actinomyces species (Fig. S4). There are almost 29 validly named Actinomyces species. Actinomyces species are found on the mucosal surfaces of many animals, and access deeper tissues by means of trauma, surgical procedures, and foreign substances that disrupt the mucosal barrier (Park and Jaensch 2009). Actinomyces sp. has also been isolated from a case of canine ulcerative keratitis (Sherman et al. 2013).
16S rRNA similarity of TAE3-ERU5 was found to belong to Dermabacteraceae, and taxonomically, was determined to show 96.88% similarity to Devriesa agamarum. Classi ed under the Actinobacteria family, Devriesa agamarum has been reported to be associated with dermatitis and septicaemia in reptiles. Phylogenetic analysis demonstrated that isolate TAE3-ERU5 could belong to a novel genus in the Dermabacteraceae family, however, it is a different clade from Devriesea, Helcobacillus, and Dermabacter genus. (Fig. S5).
TAE3-ERU6 recovered from the healthy left eye of the penguin in the present study was identi ed as Psychrobacter pygoscelis on the basis that it showed 99.11% similarity to this species upon undergoing 16S rRNA sequence analysis. The number of newly described Psychrobacter species has increased in parallel with the increased number of discoveries related to polar and marine habitats. Today, the genus Psychrobacter includes 41 validly named species (Genus Psychrobacter 2020).
Isolates TAE3-ERU2, TAE3-ERU4 and TAE3-ERU5 (Corynebacterium ciconiae, Actinomyces sp., Dermabacteraceae), also recovered from the right eye, were determined to belong to the family Actinobacteria. While marine environments are known to be a rich source of Actinobacteria, bacteria in this family are reported to be capable of producing various active secondary metabolites ( 2015), who isolated Actinobacteria in Chile, investigated the potential of these bacteria producing antibacterial substances and reported that Actinobacteria in the marine sediment could be cultured and were richly diverse. Undabarrena et al. (2016) reported to have detected bacteria belonging to the families Dermabacteraceae and Corynebacteriaceae, under which the isolates TAE3-ERU5 and TAE3-ERU2 recovered in the present study were classi ed, among the Actinobacteria they isolated from marine sediment in Chile. Furthermore, other members of the family Actinobacteria, including Arthrobacter spp. Brevibacterium spp. and Corynebacterium spp. have also been reported to exist in the dry valleys of Antarctica (Cary et al. 2010).
All of the 5 isolates recovered in the present study were determined to be susceptible to each of the 7 antibiotics tested. Only the isolate belonging to the family Dermabacteraceae was found to be resistant to gentamicin. This high antibacterial susceptibility was attributed to penguins not being exposed to antibiotics. Today, only very few geographical regions, such as the Antarctic, are free from direct and indirect contact with human populations and domestic animals (Barbosa and Palacios 2009). In a recent study, in which bacteria were isolated from freshwater samples pertaining to different regions of the Antarctic Peninsula that were either subjected to or free from human contact and effect, Jara et al. (2020) assessed the antibacterial susceptibility of these isolates. Accordingly, these researchers reported that the number of antibiotic-resistant bacteria and antibiotic resistance genes were higher in the regions that were affected by humans and wildlife, compared to the regions that were far from human and animal populations.
In the present study, bacteria of 5 different genera were isolated from the right eye of the penguin. Thus, it was not possible to identify the primary causative agent of the pyogranulomatous in ammation in the right eye. Cytological and histopathological analyses demonstrated pyogranuloma as a characteristic nding of Actinomyces sp. infection, and the extra-and intracellular presence of various other Gram-positive and Gramnegative bacteria was also determined in the ocular lesion.
[2020] included in the Norwegian Polar Institute's Quantarctica package, using the Sentinel Hub (Fig. 1). We thank Professor Dr. Ayhan Duzler for his assistance in taking electron microscope pictures of bacteria.
We also thank the Rectorate of Erciyes University for funding the bacteriological, cytological, histopathological, and immunohistochemical analyses of the ocular samples examined in this study. Finally, we extend our gratitude to the Scienti c and Technological Research Council of Turkey (TUBITAK; project number: 119Z847) for nancially supporting the 16S rRNA gene sequence analyses of the bacteria isolated in this study.

Data availability
The data that support the ndings of this study are available from the corresponding author upon reasonable request.

Con ict of Interest
The authors declare that they have no con ict of interest.