Determination of microbial diversity in the ocular conjunctiva of healthy dogs by total DNA sequencing

Abstract

The conjunctiva plays an essential role in eye health and immunity and acts as a barrier to the entry of microorganisms. Conjunctival infections are common in dogs and result from both the invasion of pathogenic microorganisms and the uncontrolled growth of the existing microbiota. Most of the existing data come from studies based on traditional culture methods. These reports indicate the predominance of gram-positive bacteria, especially Staphylococcus spp. In the present study, we analyzed the microbiota present on the conjunctival surface from a heterogeneous dog population without ophthalmological disorders using DNA sequencing. After a thorough ophthalmological examination, conjunctival swabs were collected from both eyes of 30 dogs. After processing and nucleic acid extraction, the sample pool was subjected to shotgun DNA sequencing through the Illumina platform and analyzed via the Metagenomic Rapid Annotations using Subsystems Technology (MGRAST) server. A predominance of the phylum Proteobacteria and the genera Ralstonia and Burkholderia were identified along with a minority of fungi, whereas viruses were not found. Microbial DNA sequencing has provided new data on this subject, revealing the presence of noncultivable organisms that were previously unknown as part of the ocular microbiome.

Keywords:
bacteria; ophthalmology; eye; microbiome; metagenomics.

Resumo

A conjuntiva desempenha papel fundamental na saúde e imunidade ocular e atua como uma barreira à entrada de microrganismos. As infecções conjuntivais são comuns em cães e resultam tanto da invasão de microrganismos patogênicos quanto do crescimento descontrolado da microbiota existente. A maior parte dos dados existentes provém de estudos baseados em métodos de cultura tradicionais. Esses relatos apontam para o predomínio de bactérias gram-positivas, principalmente Staphylococcus spp. O presente estudo analisou a comunidade microbiana presente na conjuntiva ocular de uma população heterogênea de cães sem distúrbios oftalmológicos por sequenciamento de DNA. Após exame oftálmico minucioso, foram coletados suabe conjuntivais de ambos os olhos de 30 cães. Após processamento e extração de ácidos nucleicos, o pool de amostras foi submetido ao sequenciamento shotgun de DNA por meio da plataforma Illumina e analisado no servidor Metagenomic Rapid Annotations using Subsystems Technology (MG-RAST). Foi identificada uma predominância do filo Proteobacteria e dos gêneros Ralstonia e Burkholderia juntamente com uma minoria de fungos, enquanto vírus não foram encontrados. O sequenciamento do DNA microbiano trouxe novos dados sobre o assunto, revelando a presença de organismos não cultiváveis até então desconhecidos como parte do microbioma ocular.

Palavras-chave:
bactéria; oftalmologia; olho; microbioma; metagenômica

1. Introduction

The conjunctiva plays an essential role in lacrimal function, immune protection of the eye, eye mobility and corneal regeneration, and acts as a barrier to the entry of pathogenic microorganisms. (¹,²) Conjunctival inflammation is one of the most commonly diagnosed eye disorders in veterinary practice; it occurs more rarely in isolation and is more commonly secondary to other inflammatory ocular diseases, such as ulcerative keratitis, tear production deficiencies and glaucoma. In some cases, etiology determination is challenging. (³) Infections can result from both the invasion of a pathogenic microorganism and the uncontrolled growth of the existing microbiota due to poor immunity. (⁴)

Infectious conjunctivitis in dogs may have bacterial, fungal, viral or parasitic origins. (³) The bacteria that have been described in clinical cases are Staphylococcus spp., Streptococcus spp., Bacillus spp., Enterobacter spp., Pseudomonas aeruginosa, Klebsiella spp., and Proteus spp.; (⁵,⁶,⁷) moreover, the fungi that have been described includ Blastomyces dermatitidis and Curvularia spp. (⁸, ⁹) Among the viruses that had been previously reported, canine herpesvirus (CHV-1), canine adenovirus (CAV-1 and CAV-2) and canine distemper virus (CDV) are the main species identified in dogs. (¹⁰,¹¹,¹²)

Regarding the conjunctival microbiota of healthy dogs, no viral species have been described in the studies available in the literature thus far. In contrast, the bacterial microbiota seems to be predominantly composed of Staphylococcus spp. according to traditional cultivation methods. (¹,⁷,¹⁴,¹⁵,¹⁶) Furthermore, the fungi Alternaria spp., Cladosporium spp., Penicillium spp., Aspergillus spp. and yeast Candida species have been detected in the conjunctiva of healthy dogs using mycological culture methods, and are considered to be transitional microbiota.(¹⁶,¹⁷)

With the advent of next generation sequencing (NGS) techniques, knowledge of the uncultivable microbiota from different sites of healthy animals has provided important and more extensive information from those previously known using the culture-dependent methods. To date, only one study has described the ocular microbiota of dogs through NGS, which was specifically accomplished by using 16S rDNA amplicon sequencing. Although the authors demonstrated Firmicutes to be the most prevalent phylum, at the genus level there was a predominance of Bifidobacterium spp., a bacterium that had not been previously described in research. Bifidobacterium spp. were detected in 92.8% of the samples and accounted for 9.1% of all reads. (¹⁸)

When considering the importance of the conjunctiva in eye health and its frequent involvement in infectious conditions, it is extremely important to update the existing data on the ocular microbiota of healthy dogs through more refined molecular techniques, due to the fact that they are currently available. The initial objective of the present study was to determine whether viruses inhabited the ocular conjunctiva of healthy dogs; if so, we aimed to identify these viruses. Furthermore, we aimed to analyze the conjunctival microbiota present at this site.

2. Materials and methods

The methodology that was applied in this study was approved by the Ethics Committee in Animal Use (CEUA) of the Federal University of Rio Grande do Sul (UFRGS) under protocol 35271. This study was also conducted according to the guidelines of the Association for Research in Vision and Ophthalmology (ARVO) on the use of animals in ophthalmic research.

2.1 Ophthalmic examination and sample collection

The first stage of the study was conducted at the Veterinary Clinic Hospital of UFRGS in Porto Alegre, RS, Brazil, and included both eyes of 30 healthy dogs of different sexes, breeds, and ages that were domiciled in different environments. A description of the sampled population is available in Table 1.

Table 1
Sample composition of the dogs that were used in the study.

Dog	Sex	Age (years)	Breed	Home Environment	Reason for being taken to the hospital
1	M	8	Lhasa Apso	House with garden	Ophthalmic check-up
2	F	6	Mixed breed	House with garden	Ophthalmic check-up
3	F	10	Yorkshire Terrier	Apartment	Clinical check-up and collection of blood tests
4	F	12	Poodle	Concrete yard house	Ophthalmic check-up
5	M	1	Pug	Apartment	Pre castration evaluation
6	F	5	Boxer	House with garden	Cardiac evaluation
7	M	10	Mixed breed	Farm	Clinical check-up and collection of blood tests
8	M	12	Mixed breed	House with garden	Ophthalmic check-up
9	F	16	American Cocker Spaniel	Apartment	Clinical check-up and collection of blood tests
10	F	11	Dachshund	Apartment	Cardiac evaluation
11	F	6	Mixed breed	House with garden	Ophthalmic check-up
12	F	11	Dalmatian	House with garden	Ophthalmic check-up
13	F	1	Yorkshire Terrier	House with garden	Pre castration evaluation
14	M	2	Shih Tzu	Apartment	Pre castration evaluation
15	M	7	Golden Retriever	House with garden	Ophthalmic check-up
16	M	3	English Bulldog	Apartment	Ophthalmic check-up
17	F	2	Shih Tzu	Apartment	Pre castration evaluation
18	M	12	Mixed breed	Concrete yard house	Cardiac evaluation
19	F	1	French Bulldog	Apartment	Pre castration evaluation
20	F	5	Shih Tzu	Apartment	Ophthalmic check-up
21	M	4	Lhasa Apso	House with garden	Ophthalmic check-up
22	M	13	Shih Tzu	Apartment	Ophthalmic check-up
23	M	6	Yorkshire Terrier	House with garden	Clinical check-up and collection of blood tests
24	F	11	Mixed breed	Farm	Cardiac evaluation
25	M	3	Miniature Pinscher	Apartment	Neurological evaluation
26	F	1	Shih Tzu	House with garden	Pre castration evaluation
27	F	12	Mixed breed	House with garden	Ophthalmic check-up
28	M	10	Golden Retriever	House with garden	Clinical check-up and collection of blood tests
29	F	6	Yorkshire Terrier	Apartment	Clinical check-up and collection of blood tests
30	M	7	Mixed breed	Apartment	Ophthalmic check-up

• ^*
  M = male. F = female

All of the animals underwent ophthalmic evaluation performed by a professional ophthalmologist prior to sample collection, including the Schirmer tear test (Schirmer Lacrimal Test, Ophthalmos, São Paulo, Brazil), slit lamp biomicroscopy (SL15, Kowa Company, Nagoya, Japan), a fluorescein test (fluorescein sodium 1%, Ophthalmos, São Paulo, Brazil) and rebound tonometry (Tonovet®, Tiolat, Helsinki, Finland). The exclusion criteria previously used by Ledbetter were used in the current study. (¹²) The absence of eye diseases was noted. After this evaluation, a drop of anesthetic eye drop (1% tetracaine hydrochloride and 0.1% phenylephrine hydrochloride, Allergan, São Paulo, Brazil) was instilled onto the ocular surface. Under gentle manual restraint, conjunctival samples were collected with a sterile cotton swab (Absorve®, Jiangsu Suyun Medical Materials, Lianyungang, China) that touched the lower and upper palpebral conjunctiva and bulbar conjunctiva without contacting the outer surface of the eyelids.

The swabs were placed in DNase/RNase-free microtubes. No later than two hours after collection, the swabs were taken to the laboratory. Inside of a biosafety cabinet, each sample was diluted in 250 µl of ultrapure water plus 250 µl of DNA stabilization Buffer AS (Qiagen, Germantown, USA) to provide sample stability, after which samples were kept at -80 °C until processing.

2.2 Processing of the samples and sequencing of the nucleic acids

Nucleic acid extraction procedures were performed with care to reduce cross contamination. All of the samples were thawed at room temperature. After vortex homogenization, a pool with 200 µl of each sample was formed. This pool was filtered through a 0.22 µm sterile syringe filter (Merck Millipore, Darmstadt, Germany) to remove impurities and cellular debris. Afterwards, the pool was ultracentrifuged on a 25% sucrose mattress at 4 °C (27,000 rpm) for two hours. The resulting pellet was treated with DNAse (2 U) and RNAse (5 µl, 20 mg/ml) enzymes (¹⁹). Subsequently, DNA extraction was performed with the phenol/chloroform protocol, and RNA extraction was performed with TRIzol® LS reagent (Life Technologies, Carlsbad, USA) according to the manufacturer›s instructions. The nucleic acids were subsequently enriched with DNA and RNA amplification kits (Sigma Aldrich, Saint Louis, USA). The nucleic acids were then subjected to purification by using the PureLink® PCR Purification Kit (Thermo Fisher Scientific, Waltham, USA). Its quality and quantity were evaluated by using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, USA) and a Qubit fluorometer (Invitrogen, Carlsbad, USA), respectively.

A DNA shotgun metagenomic library was prepared with 50 ng of purified nucleic acids by using a Nextera DNA Preparation Kit (Illumina, San Diego, USA) according to the manufacturer’s instructions. Library sequencing was performed with an Illumina® sequencer (Illumina, San Diego, USA) by using the MiSeq v2 300 platform (2x150 cycles).

The sequencing generated a total of 371,504 high-quality reads. These reads were initially trimmed with the FASTQ app and then submitted to a de novo assembly tool by using the SPAdes version 3.9 program. The 8,593 generated contigs were compared to the GenBank virus database through Blast2GO version 5.2. Afterward, with the aid of Geneious Prime version 9.1.1 and the BLASTx tool, the results were manually confirmed. During this process, only bacteriophage-like DNA sequences were found, but no other viruses were identified. However, when the reads were analyzed in the MG-RAST 4.0.3 server (https://www.mgrast.org) a large number of microorganisms were identified, and thereby considered in the analysis. The same contigs that were formed via SPAdes were subjected to this analysis, and the platform was able to recognize all of the sequences, totaling 5,228,105 base pairs (bp), with an average of 608 bp per sequence. The taxonomic distribution was determined based on the lowest common ancestor (LCA) algorithm.

3. Results

According to the MG-RAST analysis the majority of the detected sequences belonged to the Bacteria domain, followed by the Eukaryota and Archaea domains. Some sequences were not identified, and a minority corresponded to the virus domain (bacteriophages), as shown in Figure 1.

Figure 1
Taxonomic distribution in the ocular conjunctiva of healthy dogs. The domains were predicted by using the MG-RAST server. The abundance of domains is related to the total number of analyzed contigs. Abundance percentage values are indicated.

A rarefaction analysis was performed to estimate the microbiological diversity identified in the sequence set. This curve became stable when it reached a value close to 8000 contigs, which indicates that the identification of new organisms is unlikely after this plateau is reached, even if further sequencing is performed (Figure 2). Therefore, the rarefaction curve showed that the depth of sequencing was sufficient to identify the microbial diversity of the sample.

Figure 2
The rarefaction curve shows the depth of sequencing. The curve shows the relationship between the number of recovered species (x-axis) and the total number of analysed contigs (y-axis).

Among the bacteriophages, MG-RAST identified only eleven sequences, most of which were from the bacteria present in this analysis, such as Burkholderia spp., Ralstonia spp., Pseudomonas spp., Staphylococcus spp. and Bacillus spp. (Table 2). In the Archaea domain, all of the sequences correspond to the phylum Euryarchaeota, but no further classification has been determined.

Table 2
Bacteriophages identified through the MG-RAST method.

Phage	Identification	Sequence
Burkholderia phage KS9	NODE_2318_length_665_cov_1.71747_1_665	RefSeq284c8c671eaf2852cf01728d79c372f7
Burkholderia phage BcepF1	NODE_3356_length_567_cov_1.06591_117_561	RefSeq82fb224beff6a95296b46f9516b0bcfd
Burkholderia phage BcepF1	NODE_439_length_1177_cov_1.43143_1_273	RefSeq9573319e900ee58963ad81118b706841
Burkholderia phage phiE12-2	NODE_2099_length_692_cov_2.06372_108_692	RefSeqc4b3ce5a51d01dcc981d710d57bb50e4
Ralstonia phage RSM3	NODE_81_length_1941_cov_2.56229_148_1182	RefSeq80ea0d09ce3f9846bfd764144c593dd
Staphylococcus phage 11	NODE_3771_length_540_cov_2.33898_1_540	RefSeq2f60c05664cce42f18d2babc8fcea5f5
Staphylococcus phage CNPH82	NODE_2237_length_675_cov_1.79015_1_675	RefSeqa86769730bc8ced98a9b4f165bd9fd76
Pseudomonas phage F116	NODE_3261_length_575_cov_2.77679_1_575	RefSeqc2c61a51fe247b405dec5f72d8c66ba0
Bacillus phage phi29	NODE_5659_length_450_cov_18.8947_1_450	RefSeqce41329b343832940da794a173da94cf
Bacillus phage phi29	NODE_8338_length_262_cov_103.704_1_262	RefSeqce41329b343832940da794a173da94cf
Sinorhizobium phage PBC5	NODE_4358_length_505_cov_1.21693_1_505	RefSeq4e81e8bdb137d9d08e725074393ec2f9

In the Eukaryota domain, two fungal phyla (Basidiomycota and Ascomycota) and a Chordata phylum, which includes vertebrates, were found. Figure 3 illustrates this classification.

Figure 3
Composition of the Eukaryota domain detected in the ocular conjunctiva of healthy dogs. Data from the MG-RAST server, including the percent value of each classification were obtained. The percentages are relative to the total number of analysed contigs via the server.

Within the Bacteria domain, which represented 94.49% of the analyzed sequences, the most prevalent phyla were Proteobacteria and Bacteroidetes, accounting for 97.4% of the total bacteria that were found. The other phyla that were present were Actinobacteria, Firmicutes and Verrucomicrobia, among others. When this domain was divided into classes, Betaproteobacteria (86.73%), Alphaproteobacteria (7.47%), Gammaproteobacteria (2.24%), Sphingobacteria (0.96%), and Actinobacteria (0.87%) accounted for 98.27% of the analyzed sequences. When considering the taxonomic order, 97.89% of the sequences corresponded to Burkholderiales (89.97%), Sphingomonadales (3.38%), Rhizobiales (2.57%), Sphingobacteriales (1.02%), and Pseudomonadales (0.95%). At the family level, 96.98% of the sequences belonged to Burkholderiaceae (90.32%), Sphingomonadaceae (2.80%), Bradyrhizobiaceae (1.67%), Comamonadaceae (1.12%), and Sphingobacteriaceae (1.07%). Figure 4 demonstrates this classification.

Figure 4
The bacterial community composition in the ocular conjunctiva of healthy dogs (a) at the phylum level, thus demonstrating the predominance of the Proteobacteria phylum; (b) at the class level, wherein Betaproteobacteria is indicated; (c) at the order level, wherein Burkholderiales predominates, followed by Sphingomonadales; and (d) at the family level, wherein Burkholderiaceae is the most represented family.

Most of the sequences corresponded to Ralstonia (52.35%) and Burkholderia (39.58%), representing 91.93% of the total. Figure 5 shows the fifty genera that MG-RAST was able to identify from the analysed sequences. This chart shows only the genera that were represented by more than 10 sequences. Interestingly, the genera Staphylococcus and Bacillus were identified in only five sequences each, totaling only 0.18% each. The names of all of the bacterial genera that were found in the analysis are available in Figure 6.

Figure 5
The most abundant bacterial genera that were identified in the ocular conjunctiva of healthy dogs. The x-axis shows the most prevalent genera matched in the samples. The y-axis shows the percentage of contigs for each identified genus.

Figure 6
Bacterial genera composition from the ocular conjunctiva of healthy dogs. The analysis was performed with the MG-RAST server, which identified the 50 most prevalent bacterial genera in the analyzed samples.

4. Discussion

In the present study, we investigated the conjunctival microbial diversity of dogs without ophthalmological disorders through metagenomic DNA sequencing (a shotgun approach) and observed a predominance of the phylum Proteobacteria and the genera Ralstonia and Burkholderia. A previous study in which amplicons of the 16S rRNA V3-V4 hypervariable region gene were sequenced showed a predominance of the Firmicutes phylum, with Proteobacteria being among the most prevalent in the ocular surfaces of dogs. (¹⁸) Similarly, Proteobacteria is also the most commonly found phylum in the conjunctiva of healthy humans. (²⁰,²¹,²²) These microorganisms likely interact with each other and with the host’s immune system, allowing constant vigilance of the ocular surface microenvironment.

Previous studies have been conducted to analyse the ocular microbiota of dogs using microbiological cultivation techniques. (¹,⁷,¹⁴,¹⁵,¹⁶) Unlike our findings, these authors found a predominance of Staphylococcus spp. However, data from such different methodologies cannot be compared because NGS analyses are highly sensitive and can identify microorganisms that are unidentified in conventional culture analyses. (²³)

There is considerable variability in the species and amount of bacteria that make up the microbiota of a given site, which is influenced by geographic location, nutrition, and climate. (²⁴) With a goal of achieving wide variability, our study used 30 dogs from different dwellings that were fed in different ways.

The initial objective of the present study was to determine whether there were viruses inhabiting the ocular conjunctiva of healthy dogs; if so, we aimed to determine which viruses were present. The methodology was based on other studies that also sought to identify the virome of various sites from clinical samples, such as feces, (²⁵) serum, (²⁶) organs, (²⁷) and secretions. (²⁸) The virion particle size of the viruses of veterinary interest is extremely variable, with an average variation between 17 and 300 nm in diameter. (²⁹) The 0.22 µm filter is capable of filtering cell debris and bacteria and allows for virus passage, which is a suggested step in processing samples for virus detection. (²⁶) Although the processing method was based on the literature, only 11 sequences were identified as bacteriophages from Ralstonia spp., Burkholderia spp., Staphylococcus spp., Pseudomonas spp., Bacillus spp., and Sinorhizobium spp., but no other viruses were found. This result indicates that viruses may not inhabit the eye conjunctiva of dogs without ophthalmological disorders. Therefore, when other veterinarians or researchers identify viruses in samples from diseased eyes, they should consider them as potential pathogenic agents of the clinical condition in question.

Even with a potential loss of bacteria through a 0.22 µm filter, a wide range of bacteria were detected, and the rarefaction curve was stable. We highlight that this scenario can introduce bias in the recovered sequences. However, as the identified metagenomes were similar to those of previous ocular metagenome analyses, we can explore the data. In addition, an important number of eukaryotic organisms of different sizes have been found; we believe that this is due to plasma membrane lysis generated by manipulation, which allows nucleic acids to pass through the filter. Due to the fact that the utilized technique detects genetic material and not viable cells, it was possible to recover a wide range of bacteria, as well as some fungal sequences. Therefore, the most abundant microorganism sequences (mainly consisting of bacteria) were carefully analyzed.

Among the eukaryotic organisms, we identified sequences of fungi of the phyla Basidiomycota and Ascomycota. In a previous study, Suchodolski et al. (³⁰) identified fungal DNA in small bowel biopsies of 64 healthy dogs and 71 dogs with enteropathies by using PCR. All 51 identified phylotypes belonged to the Ascomycota (32 phylotypes) or Basidiomycota (19 phylotypes) phyla. Several years later, Foster et al. (³¹) determined the stool microbiome of 12 healthy dogs and 7 dogs with acute diarrhea through NGS, and the same phyla (Ascomycota and Basidiomycota) were observed to be the most abundant, in addition to being found in more than 50% of the dogs in both groups. Due to orofecal contact and the habit of smelling and even ingesting fecal material, the presence of enteric fungi in the ocular microbiota of dogs is justifiable and expected.

The ocular microbiota of dogs has been the subject of several studies using traditional microbiological culture techniques in recent decades. (¹,¹⁴,³²) These reports indicate the predominance of gram-positive bacteria, especially Staphylococcus spp., belonging to the Firmicutes phylum. (¹,⁷,¹⁵,³²) Staphylococcus spp. have also been isolated from the skin and hair of healthy and dermatopathic dogs. (³³,³⁴) The eyelids are ocular structures that are internally delimited by the eyelid conjunctiva and externally delimited by skin and fine hairs. (¹) Thus, Staphylococcus spp. may be present on the outside of the eyelids of dogs with or without conjunctivitis and blepharitis. In addition, the bacterium is also present on healthy human skin, including the region of the hands. (³⁵) However, in the present study, this bacterium was identified in a very small number of sequences (0.18%), as has been observed in other NGS studies, (¹⁸) which can be explained by the extreme care to prevent touching of the eyelids and the eyelid margins of the dogs during collection, as well as by the previous washing of hands, the use of gloves during the collection procedure and the subsequent manipulation of the samples in the laboratory. Furthermore, the high detection of Staphylococcus spp. in previous studies with traditional culture methods may be related to its possible inhibitory effect on other bacteria in culture. (³⁶)

The only study that has ever been performed using NGS sequencing to determine the ocular microbiome of healthy dogs, but with a different method than ours, demonstrated a predominance of the Firmicutes phylum. However, the most prevalent genus was Bifidobacterium spp., which belongs to the Actinobacteria phylum. (¹⁸) In our study, the phylum Actinobacteria was identified; however, the phylum Proteobacteria was the most prevalent phylum, which is in agreement with recent data on the ocular microbiome of other species, p>such as felines, (³⁷) equines, (⁹) and human. (²⁰,²¹,²²)

Ralstonia spp. was the most frequently observed genus in our analysis (Figure 4). It consists of Betaproteobacteria that inhabit the environment and are mainly found in water and soil, with some opportunistic pathogenic species. (³⁸) Hoffmann et al. (³⁹) evaluated the skin microbiome of healthy and allergic dogs, including ocular surface collection and other mucosal sites, via pyrosequencing. The genus Ralstonia was the most abundant genus in the healthy dog samples and comprised 35% of the total bacteria found in the eye conjunctiva. The proportion of Ralstonia spp. found in allergic dogs was significantly lower than that in healthy dogs. These authors concluded that Ralstonia spp. originated from the environment, given the frequent interaction of dogs with the outdoor environment. (³⁹) Our study corroborates this previous consideration, as the main habitats of most of the sampled dogs were yards or garden houses. However, considering that this was the second identification of Ralstonia spp. using metagenome sequencing and that this was the predominant bacterium in the eye conjunctiva, it may not only be a transient environmental bacterium, but may also be a permanent part of the eye microbiota of healthy dogs. A recent NGS study in horses also demonstrated that the eye microbial community was dominated by the phylum Proteobacteria and the genus Ralstonia. (⁴⁰)

In the current study, the genus Burkholderia was the second most common genus in the sample, accounting for with Ralstonia spp. almost 92% of the resulting sequences (Figure 4). This bacterium has been found to be associated with ocular melioidosis in humans and felines, (⁴¹,⁴²) as well as with systemic diseases in dogs. (⁴³) Although the present study did not reach the taxonomic level of species, the observation of Burkholderia spp. on the ocular conjunctiva of dogs warrants further research to better clarify this phenomenon.

Although it had a much smaller occurrence than the first two genera, Cupriavidus spp., which is a bacterium from the Burkholderiaceae family, was the third most frequent genus and was found in 1.79% of the sequences (Figure 4). There have been reports of the isolation of Cupriavidus spp. in diverse environments, including soil and water, (⁴⁴,⁴⁵) as well as in humans with systemic diseases. (⁴⁶)

In our study, the presence of Sphingomonas spp., which is an Alphaproteobacteria from the family Sphingomonadaceae, was verified in 1.11% of the sequences. It has already been isolated from air and dust samples, (⁴⁷) as well as from dog skin microbiota. (⁴⁸) The human ocular microbiome has been examined through NGS, and Sphingomonas spp. were identified among the most prevalent bacteria, representing 1% to 10% of all of the detected genera. (²⁰,²²) Recently, it was also identified at an abundance of 7.2% in the ocular microbiome of healthy horses. (⁴⁹) All of these results led us to believe that the results of the current study indicate actual colonization of the ocular conjunctiva of dogs by Sphingomonas spp.

To date, this is the first study to identify Pseudomonas spp. on the ocular conjunctiva of healthy dogs by using metagenomic DNA sequencing (Figure 4). This bacterium was identified in the conjunctival sac of healthy dogs and dogs with ulcerative keratitis in China through bacterial culture methods. (¹⁵) Pseudomonas spp. are essentially opportunistic in cases of eye diseases in dogs, whereby they cause tissue destruction and lead to blepharitis, conjunctivitis, keratitis, scleritis, and endophthalmitis, among other diseases. (⁶) Pseudomonas spp. is also part of the human eye microbiome, which has been confirmed via NGS. (²¹)

A specific limitation of this study was the lack of differentiation between resident and transient microbiota found in the canine eye conjunctiva. However, it is still difficult to perform resident/transient microorganism differentiation at permanently exposed mucous sites with direct air contact. Although transient microbiota may exist, their identification is relevant, as transient microorganisms can become pathogenic depending on the conditions that they encounter. Previous studies have also been unable to perform this differentiation. (⁷,¹⁵,¹⁶,¹⁸) New studies are necessary to better understand the relationship between the conjunctival microbiota and cases of clinical disease. It would be interesting to replicate our methodology using samples from dogs with clinical conjunctivitis.

5. Conclusion

In this study, we mainly observed Ralstonia spp., Burkholderia spp., Cupriavidus spp., Sphingomonas spp., and Pseudomonas spp. inhabiting the eye conjunctiva of dogs without ophthalmological disorders. Although fungi were identified in low percentages, eukaryotic viral agents were not identified; however, some sequences related to bacteriophages were found. These findings suggest that these bacteria may compose the conjunctival microbial diversity, which is relevant for maintaining eye surface health. The shotgun DNA sequencing approach has provided new data on this subject, thus demonstrating the presence of bacteria previously unknown as being part of the ocular environment. It is important to identify the microbiota of the eye to understand the infectious processes that affect it and to rationally direct the appropriate utilized treatments.

Declarations

None.

Ethics approval

This study was approved by the Ethics Committee in Animal Use (CEUA) of the Federal University of Rio Grande do Sul (UFRGS) under protocol 35271.

Data availability statement

All of the data that are relevant to the study are included in the article or uploaded as supplementary information.

• Funding

  This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) - Finance Code 001, Conselho Nacional de Pesquisa - Brazil (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul - Brazil (FAPERGS), and Propesq/UFRGS.

References

• ¹
  Andrade AL, Stringhini G, Bonello FL, Marinho M, Perri SHV. Conjunctival microbiota of healthy dogs in Araçatuba city (SP) [Conjunctival microbiota of healthy dogs in Araçatuba city (SP)]. Arq Bras Oftalmol [Internet]. 2002 Jun [cited 2019 May 5];65:323-326. Available from: https://doi.org/10.1590/S0004-27492002000300008 Portuguese.
  » https://doi.org/10.1590/S0004-27492002000300008
• ²
  Galera PD, Ávila MO, Ribeiro CR, Santos FV.. Estudo da microbiota da conjuntiva ocular de macacos-prego (Cebus apella-LINNAEUS, 1758) e macacos bugio (Alouatta caraya-HUMBOLDT, 1812), provenientes do reservatório de Manso, MT, Brasil. Arq Inst Biol [Internet]. 2002 Apr [cited 2020 May 12];69:33-36. Available from: http://www.biologico.agricultura.sp.gov.br/uploads/docs/arq/V69_2/galera.pdf Portuguese.
  » http://www.biologico.agricultura.sp.gov.br/uploads/docs/arq/V69_2/galera.pdf
• ³
  Maggs DJ. Eyelids. In: Slatter’s Fundamentals of Veterinary Ophthalmology. 4. ed. St Louis: Saunders Elsevier; 2008. p. 107-134.
• ⁴
  Armstrong RA. The microbiology of the eye. Ophthalmic Physiol Opt. 2000;6:429-441. https://doi.org/10.1111/j.1475-1313.2000.tb01121.x
  » https://doi.org/10.1111/j.1475-1313.2000.tb01121.x
• ⁵
  Murphy JM, Lavach JD, Severin GA. Survey of conjunctival flora in dogs with clinical signs of external eye disease. J Am Vet Med Assoc. 1978;172:66-68. https://europepmc.org/article/med/624664
  » https://europepmc.org/article/med/624664
• ⁶
  Ledbetter EC, Mun JJ, Kowbel D, Fleiszig SMJ.. Pathogenic phenotype and genotype of Pseudomonas aeruginosa isolates from spontaneous canine ocular infections. Invest Ophthalmol Vis Sci. 2009;50:729-736. https://doi.org/10.1167/iovs.08-2358
  » https://doi.org/10.1167/iovs.08-2358
• ⁷
  Oria AP, Pinna MH, Furtado MA, Pinheiro ACO, Gomes Junior DC, Costa Neto JM.. Microbiota conjuntival em cães clinicamente sadios e cães com ceratoconjuntivite seca. Ciênc Anim Bras. 2013;14:495-500. https://doi.org/10.5216/cab.v14I4.19210
  » https://doi.org/10.5216/cab.v14I4.19210
• ⁸
  Buyukmihci N. Ocular lesions of blastomycosis in the dog. J Am Vet Med Assoc [Internet]. 1982 Feb [cited 2021 Mar 19];180:426-431. Available from: https://escholarship.org/uc/item/8t04b5vh
  » https://escholarship.org/uc/item/8t04b5vh
• ⁹
  Scott EM, Carter RT. Canine keratomycosis in 11 dogs: a case series (2000-2011). J Am Anim Hosp Assoc. 2014;50:112-118. https://doi.org/10.5326/JAAHA-MS-6012
  » https://doi.org/10.5326/JAAHA-MS-6012
• ¹⁰
  Decaro N, Campolo M, Elia G, Buonavoglia D, Colaianni ML, Lorusso A, et al.Infectious canine hepatitis: an “old” disease reemerging in Italy. Res Vet Sci. 2007;83:269-273. https://doi.org/10.1016/j.rvsc.2006.11.009
  » https://doi.org/10.1016/j.rvsc.2006.11.009
• ¹¹
  Almeida DE, Roveratti C, Brito FL, Godoy GS, Duque JCM, Bechara GH, et al. Conjunctival effects of canine distemper virus-induced keratoconjunctivitis sicca. Vet Ophthalmol. 2009;12:211-215 https://doi.org/10.1111/j.1463-5224.2009.00699.x
  » https://doi.org/10.1111/j.1463-5224.2009.00699.x
• ¹²
  Ledbetter EC, Hornbuckle WE, Dubovi EJ. Virologic survey of dogs with naturally acquired idiopathic conjunctivitis. J Am Vet Med Assoc. 2009;235:954-959. https://doi.org/10.2460/javma.235.8.954
  » https://doi.org/10.2460/javma.235.8.954
• ¹³
  Ledbetter EC, Kim SG, Dubovi EJ, Bicalho RC. Experimental reactivation of latent canine herpesvirus-1 and induction of recurrent ocular disease in adult dogs. Vet Microbiol. 2009;138:98-105. https://doi.org/10.1016/j.vetmic.2009.03.013
  » https://doi.org/10.1016/j.vetmic.2009.03.013
• ¹⁴
  McDonald PJ, Watson ADJ. Microbial flora of normal canine conjunctivae. J Small Anim Pract. 1976;17:809-812. https://doi.org/10.1111/j.1748-5827.1976.tb06947.x
  » https://doi.org/10.1111/j.1748-5827.1976.tb06947.x
• ¹⁵
  Wang L, Pan Q, Zhang L, Xue Q, Cui J, Changming Q. Investigation of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Beijing, China. Vet Ophthalmol. 2008;11:145-149. https://doi.org/10.1111/j.1463-5224.2008.00579.x
  » https://doi.org/10.1111/j.1463-5224.2008.00579.x
• ¹⁶
  Santos LGF, Almeida ABPF, Silva MC, Oliveira JT, Dutra V, Souza VRF. Microbiota conjuntival de cães hígidos e com afecções oftálmicas [Conjunctival microorganisms from healthy dogs and with eye diseases]. Acta Sci Vet [Internet]. 2009 [cited 2022 May 18];37(2):165-169. Available from: https://www.redalyc.org/pdf/2890/289021830008.pdf
  » https://www.redalyc.org/pdf/2890/289021830008.pdf
• ¹⁷
  Verneuil M, Durand B, Marcon C, Guillot J. Conjunctival and cutaneous fungal flora in clinically normal dogs in southern France. J Mycol Med. 2014;24:25-28. https://doi.org/10.1016/j.mycmed.2013.11.002
  » https://doi.org/10.1016/j.mycmed.2013.11.002
• ¹⁸
  Leis ML, Costa MO. Initial description of the core ocular surface microbiome in dogs: bacterial community diversity and composition in a defined canine population. Vet Ophthalmol. 2019;22:337-344. https://doi.org/10.1111/vop.12599
  » https://doi.org/10.1111/vop.12599
• ¹⁹
  Lima DA, Cibulski SP, Finkler F, Teixeira TF, Varela APM, Cerva C, et al. Faecal virome of healthy chickens reveals a large diversity of the eukaryote viral community, including novel circular ssDNA viruses. J Gen Virol. 2017;98:690-703. https://doi.org/10.1099/jgv.0.000711
  » https://doi.org/10.1099/jgv.0.000711
• ²⁰
  Dong Q, Brulc JM, Iovieno A, Bates B, Garoutte A, Miller D, et al. Diversity of bacteria at healthy human conjunctiva. Invest Ophthal Vis Sci. 2011;52:5408-5413. https://doi.org/10.1167/iovs.10-6939
  » https://doi.org/10.1167/iovs.10-6939
• ²¹
  Huang Y, Yang B, Li W. Defining the normal core microbiome of conjunctival microbial communities. Clin Microbiol Infect. 2016;22:643-647. https://doi.org/10.1016/j.cmi.2016.04.008
  » https://doi.org/10.1016/j.cmi.2016.04.008
• ²²
  Ozkan J, Nielsen S, Diez-Vives C, Coroneo M, Thomas T, Willcox M. Temporal stability and composition of the ocular surface microbiome. Sci Rep. 2017;7:1-11. https://doi.org/10.1038/s41598-017-10494-9
  » https://doi.org/10.1038/s41598-017-10494-9
• ²³
  Vaz-Moreira I, Egas C, Nunes OC, Manaia CM. Culture-dependent and culture-independent diversity surveys target different bacteria: a case study in a freshwater sample. Antonie Van Leeuwenhoek. 2011;100:245-257. https://doi.org/10.1007/s10482-011-9583-0
  » https://doi.org/10.1007/s10482-011-9583-0
• ²⁴
  Carter GR, Cole JR. Classification, normal flora and laboratory safety. In: Diagnostic procedures in veterinary bacteriology and mycology. 2. ed. San Diego: Academic Press; 1990. p. 1-8. https://doi.org/10.1016/B978-0-12-161775-2.50005-0
  » https://doi.org/10.1016/B978-0-12-161775-2.50005-0
• ²⁵
  Kluge M, Campos FS, Tavares M, Amorim DB, Valdez FP, Giongo A, et al. Metagenomic survey of viral diversity obtained from faeces of Subantarctic and South American fur seals. PLoS One [Internet]. 2016 Mar [cited 2020 Aug 12];11:e0151921. Available from: https://doi.org/10.1371/journal.pone.0151921
  » https://doi.org/10.1371/journal.pone.0151921
• ²⁶
  Weber MN, Cibulski SP, Olegario JC et al. Characterization of dog serum virome from Northeastern Brazil. Virology. 2018;525:192-199. Available from: https://doi.org/10.1016/j.virol.2018.09.023
  » https://doi.org/10.1016/j.virol.2018.09.023
• ²⁷
  Alves CDBT, Budaszewski RF, Cibulski SP, Silva MS, Puhl DE, Mósena ACS, et al. Mamastrovirus 5 detected in a crab-eating fox (Cerdocyon thous): Expanding wildlife host range of astroviruses. Comp Immunol Microbiol Inf Dis. 2018;58:36-43. https://doi.org/10.1016/j.cimid.2018.08.002
  » https://doi.org/10.1016/j.cimid.2018.08.002
• ²⁸
  Liu Z, Yang S, Wang Y, Shen Q, Yang Y, Deng X, et al. Identification of a novel human papillomavirus by metagenomic analysis of vaginal swab samples from pregnant women. Virol J. 2016;13:1-7. https://doi.org/10.1186/s12985-016-0583-6
  » https://doi.org/10.1186/s12985-016-0583-6
• ²⁹
  Quinn PJ, Markey BK, Leonard FC et al. Nature, structure and taxonomy of viruses. In: Veterinary Microbiology and Microbial Disease. 2. ed. West Sussex: Wiley-Blackwell;2011. p. 505-508.
• ³⁰
  Suchodolski JS, Morris EK, Allenspach K, Jergens AE, Harmoinen JA, Westermarck E, et al. Prevalence and identification of fungal DNA in the small intestine of healthy dogs and dogs with chronic enteropathies. Vet Microbiol. 2008;132:379-388. https://doi.org/10.1016/j.vetmic.2008.05.017
  » https://doi.org/10.1016/j.vetmic.2008.05.017
• ³¹
  Foster ML, Dowd SE, Stephenson C, Steiner JM, Suchodolski JS. Characterization of the fungal microbiome (mycobiome) in faecal samples from dogs. Vet Med Int [Internet]. 2013 Apr [cited 2023 July 09]:e658373:1-8. Available from: https://doi.org/10.1155/2013/658373
  » https://doi.org/10.1155/2013/658373
• ³²
  Prado MR, Rocha MF, Brito ÉH, Girão MD, Monteiro AJ, Teixeira MFS, et al. Survey of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Fortaleza, Ceará, Brazil. Vet Ophthalmol. 2005;8:33-37. https://doi.org/10.1111/j.1463-5224.2005.04061.x
  » https://doi.org/10.1111/j.1463-5224.2005.04061.x
• ³³
  Allaker RP, Lloyd DH, Simpson AI. Occurrence of Staphylococcus intermedius on the hair and skin of normal dogs. Res Vet Sci. 1992;52:174-176. https://doi.org/10.1016/0034-5288(92)90006-N
  » https://doi.org/10.1016/0034-5288(92)90006-N
• ³⁴
  Fazakerley J, Nuttall T, Sales D, Schmidt V, Carter SD, Hart CA, et al. Staphylococcal colonization of mucosal and lesional skin sites in atopic and healthy dogs. Vet Dermatol. 2009;20:179-184. https://doi.org/10.1111/j.13653164.2009.00745.x
  » https://doi.org/10.1111/j.13653164.2009.00745.x
• ³⁵
  Watanabe K, Nakaminami H, Azuma C, Tanaka I, Nakase K, Matsunaga N, et al. Methicillin-resistant Staphylococcus epidermidis is part of the skin flora on the hands of both healthy individuals and hospital workers. Biol Pharm Bull [Internet]. 2016 Aug [cited 2021 Apr 21]:b16-00528. Available from: https://doi.org/10.1248/bpb.b16-00528
  » https://doi.org/10.1248/bpb.b16-00528
• ³⁶
  Barrow GI. Microbial antagonism by Staphylococcus aureus. Microbiol. 1963;31:471-481. https://doi.org/10.1099/00221287-31-3-471
  » https://doi.org/10.1099/00221287-31-3-471
• ³⁷
  Weese SJ, Nichols J, Jalali M, Litster, A. The oral and conjunctival microbiotas in cats with and without feline immunodeficiency virus infection. Vet Res. 2015;46:1-11. https://doi.org/10.1186/s13567-014-0140-5
  » https://doi.org/10.1186/s13567-014-0140-5
• ³⁸
  Ryan MP, Adley CC. Ralstonia spp.: emerging global opportunistic pathogens. Eur J Clin Microbiol & Infect Dis. 2014;33:291-304. https://doi.org/10.1007/s10096-013-1975-9
  » https://doi.org/10.1007/s10096-013-1975-9
• ³⁹
  Hoffmann AR, Patterson AP, Diesel A, Lawhon SD, Ly HJ, Stephenson CE, et al. The skin microbiome in healthy and allergic dogs. PloS one [Internet]. 2014 Jan [cited 2022 Jan 08]:9:e83197. Available from: https://doi.org/10.1371/journal.pone.0083197
  » https://doi.org/10.1371/journal.pone.0083197
• ⁴⁰
  LaFrentz S, Abarca E, Mohammed HH, Cuming R, Arias CR. Characterization of the normal equine conjunctival bacterial community using culture-independent methods. Vet Ophthalmol. 2020;23:480-488. https://doi.org/10.1111/vop.12743
  » https://doi.org/10.1111/vop.12743
• ⁴¹
  Chen KJ, Sun MH, Hou CH, Sun CC, Chen TL. Burkholderia pseudomallei endophthalmitis. J Clin Microbiol. 2007;45:4073-4074. https://doi.org/10.1128/JCM.01467-07
  » https://doi.org/10.1128/JCM.01467-07
• ⁴²
  Parkes HM, Shilton CM, Jerrett IV, Benedict S, Spratt BG, Godoy D, et al. Primary ocular melioidosis due to a single genotype of Burkholderia pseudomallei in two cats from Arnhem Land in the Northern Territory of Australia. J Feline Med Surg. 2009;11:856-863. https://doi.org/10.1016/j.jfms.2009.02.009
  » https://doi.org/10.1016/j.jfms.2009.02.009
• ⁴³
  Lloyd JM, Suijdendorp P, Soutar WR. Melioidosis in a Dog. Aust Vet J. 1988;65:191-192. https://doi.org/10.1111/j.1751-0813.1988.tb14300.x
  » https://doi.org/10.1111/j.1751-0813.1988.tb14300.x
• ⁴⁴
  Makkar NS, Casida-Junior LE. Cupriavidus necator gen. nov., sp. nov.; a nonobligate bacterial predator of bacteria in soil. Int J Syst Evol Microbiol. 1987;37:323-326. https://doi.org/10.1099/00207713-37-4-323
  » https://doi.org/10.1099/00207713-37-4-323
• ⁴⁵
  Ray J, Waters RJ, Skerker JM, Kuehl JV, Price MN, Huang J, et al. Complete genome sequence of Cupriavidus basilensis 4G11, isolated from the Oak Ridge field research center site. mBio [Internet]. 2015 May [cited 2021 Oct 23]:3:e00322-15. Available from: https://doi.org/10.1128/genomeA.00322-15
  » https://doi.org/10.1128/genomeA.00322-15
• ⁴⁶
  Langevin S, Vincelette J, Bekal S, Gaudreau C. First case of invasive human infection caused by Cupriavidus metallidurans. J Clin Microbiol. 2011;49:744-745. https://doi.org/10.1128/JCM.01947-10
  » https://doi.org/10.1128/JCM.01947-10
• ⁴⁷
  Andersson AM, Weiss N, Rainey F, Salkinoja-Salonen MS. Dust-borne bacteria in animal sheds, schools and children’s day care centres. J Appl Microbiol. 1999;86:622-634. https://doi.org/10.1046/j.1365-2672.1999.00706.x
  » https://doi.org/10.1046/j.1365-2672.1999.00706.x
• ⁴⁸
  Cuscó A, Belanger JM, Gershony L, Islas-Trejo A, Levy K, Medrano JF, et al. Individual signatures and environmental factors shape skin microbiota in healthy dogs. Microbiome. 2017;5:1-15. https://doi.org/10.1186/s40168-017-0355-6
  » https://doi.org/10.1186/s40168-017-0355-6
• ⁴⁹
  Scott EM, Arnold C, Dowell S, Suchodolski JS. Evaluation of the bacterial ocular surface microbiome in clinically normal horses before and after treatment with topical neomycin-polymyxin-bacitracin. PloS one [Internet]. 2019 Apr [cited 2020 Mar 03]:14:e0214877. https://doi.org/10.1371/journal.pone.0214877
  » https://doi.org/10.1371/journal.pone.0214877

Publication Dates

• Publication in this collection
  09 Sept 2024
• Date of issue
  2024

History

• Received
  16 Oct 2023
• Accepted
  26 Apr 2024
• Published
  24 July 2024