Polymerase chain reaction assay targeting nox gene for rapid identification of Brachyspira canis in dogs.

Genus Brachyspira, as Gram negative anaerobic bacteria, colonize in dogs intestine. The aim of the current study was to determine the prevalence of Brachyspira spp. for the first time in Iran and rapid identification of Brachyspira spp. in dogs by a new designment of a species-specific primer set for B. canis. One hundred fifty-one fecal samples were obtained from dogs by rectal swab. Twenty dogs suffered from diarrhea and 131 of them were healthy. In 9.27% (14/151) of samples, spirochaetes were detected on primary cultures by weak hemolysis and positive Gram staining and then Brachyspira genus was confirmed by NADH oxidase (nox) gene via polymerase chain reaction. Among 14 isolates, twelve isolates were B. canis, one isolate was B. intermedia and another one was non-typeable. From 12 B. canis, only eight isolates were detected by designed specific primers. Ten Brachyspira spp. were isolated from dogs ≤ 1 year old (10/67, 14.92%) and 4 isolates were from > 1 year old dogs (4/84, 4.76%). The isolation rates from healthy and diarrheic dogs were (12/131, 9.16%) and (2/20, 10.00%), respectively. A statistically significant association was observed between the presence of Brachyspira spp. and the age under one year. Based on our findings, the nox gene in B. canis might have more sequence variability compared to other Brachyspira spp.

Introduction

Genus Brachyspira contains several Gram-negative, oxygen-tolerant and anaerobic intestinal spirochete species colonizing in the large intestine of humans, birds (water birds, laying hens and corvid birds) and mammalians (pigs, dogs, and wild rodents). Sixteen different species of Brachyspira have been identified up to now including B. aalborgi, B. alvinipulli, B. canis, B. corvi, B. hampsonii, B. hyodysenteriae, B. ibaraki, B. innocens, B. intermedia, B. murdochii, B. muridarum, B. muris, B. pulli, B. rattus, B. suanatina, and B. pilosicoli.[1]-[4]

Human intestinal spirochetosis (HIS) is spread worldwide and its prevalence (2.50 to 62.50%) is higher among sensitive groups such as male homosexuals and human immunodeficiency virus-positive persons. The HIS is a condition defined by the presence of a layer of spirochetes attached by one cell end to the colorectal epithelium. Although the pathologic significance of HIS is uncertain, it has been linked to weight loss, chronic diarrhea, rectal bleeding, and other abdominal complaints. There are two Brachyspira species in humans including B. pilosicoli and B. aalborgi.[5]-[7]

Brachyspira pilosicoli is a zoonotic agent that was first described as a cause of porcine intestinal spirochaetosis and pigs colitis leading to diarrhea and dysentery.[8]-[10]

The prevalence rate of Brachyspira spp. in dogs is about 5.00% of the canine population and the organisms are transmitted between dogs by the fecal-oral route. Species which have been identified in dogs include B. pilosicoli, B. canis, B. intermedia and B. alvinipulli.[11]-[13]

Reports of the presence of spirochaetes in dog’s feces can be traced back more than 100 years. Canine intestinal spirochetes were first described in mature dogs with normal feces, but Duhamel et al. have later documented intestinal spirochetes in a 3-month-old beagle pup with diarrhea that had concurrent infection with Giardia spp. Likewise, Turek and Meyer have isolated spirochetes from beagle pups.[13]-[16]

Rapid identification of canine intestinal spirochaetes has been accomplished by determination of few biochemical properties[17] and polymerase chain reaction (PCR) based on 16S rDNA sequence.[11],[18] Additional methods used for identification of intestinal spirochaetes from other hosts include phylogenetic analysis based on 16S rDNA sequences, multilocus enzyme electro-phoresis,[11],[18],[19] pulse filed gel electrophoresis,[20] random amplification of polymorphic DNA,[21] duplex PCR[22], [23] and reverse transcription-PCR.[24]

In previous studies, B. pilosicoli was detected by a molecular method with B. pilosicoli-specific primers for nox and 16s rDNA gene,[4],[25] but B. canis-specific primers have not been designed up to this time. The purposes of this study were to isolate and rapid identification of Brachyspira spp. in dogs by designing a species-specific primer set for B. canis and also to discover the prevalence of Brachyspira spp. in dogs for the first time in Iran.

Materials and Methods

Sampling. A total of 151 fecal samples were collected from 151 dogs that were referred to the Veterinary Hospital of the Ferdowsi University of Mashhad, Iran (during 2014 to 2015). The owners were asked to sign consent before sampling. 131 of dogs were apparently healthy (routine checkup) or without any history of diarrhea and antibiotics therapy; however, 20 dogs had diarrhea as a main clinical sign. The samples were obtained directly from the rectum of the dogs using sterile swabs. Samples were collected from 91 female and 60 male dogs aging three months to 11 years old.

Bacterial culture. Fecal specimens were cultured on the selective tryptose soy agar (Merck, Darmstadt, Germany) supplemented with 5.00% sheep blood and the antibiotic supplement was composed of three antibiotics of the following final concentrations: spectinomycin 400 μg mL-1, vancomycin 25.00 μg mL-1 and colistin 25.00 μg mL-1.[26] Sheep blood agar plates were incubated in an anaerobic jar at 37 ˚C for five days. Plates being weakly beta-hemolytic were checked for the presence of Brachyspira spp. by Gram staining. Two or three passages were done to obtain pure culture in positive samples.

DNA extraction and PCR. The PCR preparation kit (Denazist Asia, Mashhad, Iran) was used for DNA extraction according to the manufacturer’s instructions. Briefly, a single colony from pure culture was chosen and then subjected to DNA extraction procedure. At first, a genus-specific primer pair was used for confirmation by targeting conserve segment of the NADH oxidase (nox) gene.[27] The nox gene was chosen as a target gene because its sequence is less conserved than 16S or 23S rDNA gene sequences. The DNA extracts were used for species-specific primer sets targeting B. pilosicoli, B. intermedia, B. innocence, B. merdochii and B. hyodysenteriae (Table 1).[27]

Table 1

Oligonucleotide primers used for genus- and species-specific Brachyspira polymerase chain reaction

Name	Sequence 5' to 3'	Product size	Temperatures (˚C )	Comment
BrNOX2-F BrNOX2-R	C(AT)GGTTCTTGGCCTGTAACT(CG)CCATAACTCT(AT)GGCA(AT)AGC	250bp	55	Detects allBrachyspira species
Hyo-F1 Hyo-R1	AGGTGACTGTGCTACTGTAACGTCTGCTGCCTTCTT	320bp	62	Specific forB. hyodysenteriae
Pilo2004-F1 Pilo2004-R1	TGAAATCTTCTAAAGATGAGTAGCTAAAGCAATATATTCA	96bp	55	Specific forB. pilosicoli
Interm2004-F2 Interm2004-R2	TTGCCTAGAGTTATGGGTAATGACATAACTACCATATCTACT	182bp	55	Specific forB. intermedia
Innoc-F1 Innoc-R1	ATGGTGCTATAAAAGTAGACACCAACCAGTAGAAGCCATG	249bp	55	Specific forB. innocens
Murd-F1 Murd-R1	GAATACTGCGGTACTCAAGGAGAGAATGCGTGAATAGCTTCG	260bp	55	Specific forB. murdochii
Canis-F Canis-R	AGGAAATCATTGATGAGTTGATATAGTCCTGAGCTCCG	439bp	56	Specific forB. canis

-specific PCR designment. The only two nox gene sequences belonged to B. canis obtained from public databases with the accession number of EF436590 and EU819071 and both of them were analyzed by CLC Genomics workbench (version 7.0; CLC Bio, Aarhus, Denmark). Based on multiple sequence alignments, highly conserved regions of nox gene were chosen and PCR primers were obtained from these regions. Oligonucleotide primers were synthesized by Macrogen Inc., Seoul, Korea. The PCR primers used in this study are listed in Table 1.

The PCRs were run in total reaction volumes of 25.00 µL containing the primers. The materials used in the PCR reaction were provided by Ampliqon (Odense, Copenhagen, Denmark). Amplification reactions were carried out in a 50 µL reaction volume containing 5 µL 10x PCR buffer, 5 mM dNTPs, 25.00 mM MgCl2, 5.00 U of Taq DNA polymerase and the primers (Table 1), at the given concentrations. The thermocycler conditions were an initial denaturation for 15 min at 95 ˚C and the reaction mixture was subjected to 40 cycles of heat denaturation at 94 ˚C for 30 sec. The temperatures for each primer annealing are shown in Table 1 which were 30 sec for all the primers, and DNA extension was done at 72 ˚C for 1 min which was completed by a final extension of further 10 min at 72 ˚C.[27] Following PCR, 10.00 µL of the amplification products were electrophoresed in a 2.00% Tris acetate EDTA–agarose gel. The gel was stained with ethidium bromide and the bands were visualized under ultraviolet light. All PCR reactions were carried out in duplicate with negative and positive controls. For positive controls, DNA extracts which have been confirmed by 16S rDNA sequencing and biochemical scheme analysis in previous studies were used.[4],[28],[29]

Evaluation of sensitivity and specificity of the specific primers. To determine the primers specificity for B. canis, PCR assay with specific primers B. canis forward and reverse (BCF) and Brachyspira canis reverse (BCR) was performed on the DNA extracted from B. pilosicoli, B. intermedia, B. innocens, B. murdochi, Clostridium perfringens, C. difficile and Escherichia coli. In order to obtain the lowest amount of DNA for PCR primers amplification and detection, concentration of a sample of extracted DNA was determined by spectrophotometer device ranging from 10-2 to 10-7 copies/reaction. The PCR assay was eventually performed on each sample.

Statistical analysis. Data analysis was performed using SPSS software (version 21.0; IBM, Chicago, USA). The relationships between bacteria and the presence of diarrhea, bacteria, and gender, bacteria and age were assessed by Pearson Chi-square. A p-value < 0.05 was considered significant.

Results

A total of 151 fecal samples were collected throughout the study, 131 from clinically healthy non-diarrheic and 20 from diarrheic dogs. From these samples, 14 (9.27%) spirochaetes were detected on primary cultures by by weak hemolysis and positive Gram staining and then Brachyspira genus was confirmed by NADH oxidase (nox) gene PCR. The median age of non-diarrheic and diarrheic dogs included in this research was not statistically different which was two years (from 1 month to 10 years) and 2.50 years (from 5 months to 7 years). Eighty-one female and 50 male dogs were in the non-diarrheic group. Diarrheic group consisted of 10 female and 10 male dogs. Ten Brachyspira spp. isolates were from dogs ≤ 1 year old (10/67, 14.92%) and four isolates from > 1-year-old dogs (4/84, 4.76%). The isolation rate from healthy and diarrheic dogs was 9.16% (12/131) and 10.00% (2/20) respectively. Out of 91 female dogs, 11 (12.00%) and from 60 male dogs, three dogs (5.00%) were positive in culturing identification method and PCR.

A total of 14 isolates in species-specific PCR of Brachyspira had shown nox gene with 250 bp (Fig. 1) and in order to identify the Brachyspira species (such as B. pilocicoli, B. intermedia, B. innocence, B. merdochii, B. hyodysnteriae and designed B. canis; Fig. 2), species-specific PCR primer was performed. Among these 14 isolates, 12 isolates were B. canis, one isolate was B. intermedia and another one was non-typeable due to loss of DNA extract.

Fig. 1

The nox gene polymerase chain reaction. M: 100 bp DNA marker; C+: Positive control 250 bp band; C-: Negative control; Lanes 1-11: Brachyspira genus isolates

Fig. 2

Brachyspira species- specific primer. M: 100 bp DNA marker; C1+: Control positive for B. intermedia; C2+: Control positive for B. canis; C1-: Control negative for B. intermedia; C2-: Control negative for B. canis; Lane 1: B. intermedia isolate; Lanes 2-4: B. canis isolates

From 12 B. canis, only eight isolates were detected by specific primers designed in this study, while the other four isolates were confirmed by the School of Veterinary and Life Sciences, Murdoch University, Perth, Australia after full sequencing of 16S rDNA and NADH oxidase. However, one isolate was found to be negative by genus-specific 16S rDNA and genus-specific nox PCRs in the study conducted by Murdoch University. This could be due to the fact that the sample had insufficient DNA. Accession numbers of four isolates based on 16S rDNA gene deposited in the GenBank are as follows: KT381873, KT381872, KT381871, and KT381870.

No significant association was observed between the presence of Brachyspira spp. with diarrhea and between the presence of Brachyspira spp. with age (p > 0.05), but a statistically significant association was observed between the presence of Brachyspira spp. and ≤ 1-year-old dogs (p < 0.05). Characterization of Brachyspira spp. isolated from dogs is shown in Table 2. Sensitivity assay showed a range threshold for detection of microbial DNA at 10-2 to 10-7 copy number (B. canis).

Table 2

Characterization of Brachyspira spp. isolated from dogs

Brachyspira spp.	Gender	Age (months)	Presence of diarrhea
B. canis	Female	12	+
	Female	36	+
	Female	5	-
	Female	12	-
	Female	12	-
	Female	12	-
	Female	12	-
	Female	12	-
	Female	24	-
	Female	36	-
	Male	12	-
	Male	5	-
B. intermedia	Female	12	-
Non-typable	Male	24	-

Discussion

Genus Brachyspira, as anaerobic intestinal spiro-chaetes, is known worldwide to colonize in the dogs intestine, which has been seen in dogs feces and colons for decades; however, inconsistent reports regarding its role in the emergence of the diseases have made it difficult to determine its pathogenetic significances. The B. pilosicoli is a zoonotic agent with a human public health risk. There have been previous studies reporting the isolation of B. pilosicoli from dogs and human confirming the possibility of inter-species transfer in various geographical areas.[23],[30]

There has been no report regarding the distribution or prevalence of Brachyspira spp., although more people tend to keep dogs as pets in their homes in Iran. The current study was performed to determine the prevalence of cultivable anaerobic intestinal spirochetes in dogs population.

Few studies have been conducted on Brachyspira spp. in Iran,[4],[28],[29] the focus is usually on poultry and related industry. This study was the first investigation of Brachyspira spp. in dogs population in Iran. The 9.00% prevalence rate obtained in this study was almost similar to the 11.00% prevalence rate of Brachyspira spp. of a recently reported research on dogs in Thailand.[31] On the other hand, this rate is less than some other studies including 18.70% in Australia,[31]-[32] 17.60% in Swedish pet dogs[17] and 40.80% in pet shop puppies in Australia.[11]

In a study performed in Australia, of 40.80% (n = 20) Brachyspira spp. isolated from pet shop puppies by primary culture, but in PCR method, 14.20% (n = 5) were B. canis and 4.10% (n = 2) were B. pilosicoli. Major of isolates were B. canis (5/7, 71.42%). All B. canis were isolated from healthy dogs.[30] In Thailand, 10.59% (16/151) Brachyspira spp. were isolated. The isolation rate from puppies (< 1 year) was 13.180% (12/91) and the rate from adults was 6.70% (4/60). There was no significant association between Brachyspira spp. and the presence of diarrhea.[20]

Šperling et al. had only reported 10 Brachyspira spp-positive out of 1139 dogs being studied in the Czech Republic, 9 isolates were B. pilosicoli and one isolate was B. hyodysenteriae,23 which was not compatible with the results obtained in our study.

In Spain, three species of Brachyspira were identified in 13.29% (41/311) dogs. The highest prevalence belonged to B. canis (8.00%, 25/311). The prevalence rates of B. pilosicoli and B. intermedia were 4.82% (15/311) and 0.30% (1/311), respectively.[25] In that study, a statistically significant association between B. pilosicoli and the presence of diarrhea was established in dogs for the first time.

Our study was similar to this study in terms of B. canis (12/151, 7.90%) and B.intermedia (1/151, 0.60%) while there was no significant association between Brachyspira spp. and diarrhea in current study.

In a recent study in Thailand, out of 151 dogs, 17 (11.30%) had fecal shedding of Brachyspira spp. including B. canis (9/151, 6.00%), B. pilosicoli (4/151, 2.60%), B. pulli (3/151, 2.00%) and B. intermedia (1/151, 0.70%). Spirochaetes were isolated from 6 of the 47 (12.80%) healthy dogs and from 11 of the 104 (10.60%) dogs with diarrhea. The Brachyspira spp. were more common among the dogs younger than one year old (12/91, 13.20%) compared to adults (5/60, 8.30%), but this difference was not significant. In the current study, the percentages of isolation of B. canis and B. intermedia as well as the isolation of Brachyspira spp. from dogs ≤ 1-year-old were similar; however, we found a significant association between Brachyspira spp. and age.

The majority of the isolates (12/14; 85.70%) were identified as B. canis, which has already been reported in other studies in different geographical regions such as Australia, Thailand and Spain. In another word, B. canis is considered as the most common Brachyspira spp. in dogs stool.[11],[25],[31] The B. canis is a common commensal species of dogs.[31]

In the current study, B. intermedia contamination rate was very low (1/151, 0.60%), which was similar to the studies conducted in Spain and Thailand. The prevalence of B. intermedia was 0.60% and 0.30% in Thailand and Spain, respectively.[25],[31] Sampled dogs from Iran and Spain were healthy, but Thai dogs had bloody diarrhea. As B. intermedia could be generally found in chicken colon, dogs infected by B. intermedia had probably consumed contaminated chicken meat or chicken carcasses.[31],[33]

Unlike other studies, we could not find any B. pilosicoli isolate. The small number of samples with diarrhea in our study might be the reason for low levels of B. pilosicoli detection because a recent study has demonstrated a significant relationship between diarrhea and the presence of B. pilosicoli.[25]

To the best of our knowledge, there are no available species-specific PCR assays for B. canis identification.[25] Detection of B. canis based on biochemical test and specific gene sequencing is time-consuming and it is not cost-effective. In our study, 8 of 12 (67.00%) B. canis were identified using species-specific primers.

Continued research in detecting target genes and designing specific primers of B. canis appears fully justified so that the researchers and clinicians could differentiate and determine pathogenic species of diarrhea in dogs from non-pathogenic ones. Although the designed primer seems to have proper specificity and sensitivity to detect B. canis, the failure to identify four isolates and the importance of full sequencing of nox gene and 16S rDNA have indicated that firstly, designing primer with high detection range for this species requires more data and secondly, unlike other Brachyspira spp. of birds and other animals, the nox gene in B. canis has more sequence variability. Relatively, the B. canis-specific primers introduction and detection require further investigations using of modified oligonucleotides.