Detection of specific Treponema species and Dichelobacter nodosus from digital dermatitis ( Mortellaro ’ s disease ) lesions in Swiss cattle

The aim of this study was to determine the prevalence of the three Treponema species as well as D. nodosus in Digital dermatitis (DD) and slurry of Swiss cattle using PCR. A total of 86 specimens from 24 farms were enrolled in the study. Slurry samples from 21 DD-affected and one unaffected farm were collected to assess the potential of environmental transmission. Nested and real-time PCR were performed from the specimens to detect Treponema species and D. nodosus, respectively. The DD-stages were positive for at least one or more of the DD-associated Treponema species in 50 of 61 cases (82.0%) and in 9 of 25 cases (36.0%) in unaffected animals. Infected animals with small focal active lesions showed a significantly lower prevalence (14.8%) compared to the other DD stages (67.2%; P=0.011). Most prevalent was T. phagedenis (65.1%). D. nodosus was detected in 51.8% of clinical DD lesions and 24.1% in unaffected cases, but its presence was not significantly associated with the various DD-stages. All samples positive for D. nodosus contained the acid protease gene aprB2 but were negative for aprV2, the latter associated with virulence in sheep foot rot. Control farms were negative for all DD-associated Treponema species while positive for aprB2 and negative for aprV2. The presence of aprB2 suggests it is ubiquitous in the animal environment. With respect to the slurry samples, three out of 21 specimens (14.3%) were positive for one or more of the DD-associated Treponema species and eleven out of 21 specimens (52.4%) were positive for aprB2 and negative for aprV2 of D. nodosus. In conclusion, an association was found between the presence of clinical DD and specific Treponema species, while for D. nodosus no such link with DD lesions could be observed.


Introduction
Digital dermatitis (DD) (also known as Hairy Foot Warts, Strawberry Foot Rot, Mortellaro's Disease, Raspberry or Verrucose Dermatitis) is an infectious acute or chronic ulcerative foot disease, initially reported in Italy 1 .DD lesions typically develop on the plantar skin, proximal to the bulb of the heel, or occasionally, within the interdigital cleft.The disease has a big impact on the well-being of animals, as well as the productivity of dairy farms.Losses are caused by high treatment costs including the application of antibiotics, decreased milk production and reduced reproductive efficiency in affected cattle 2,3 .Compared to other foot diseases, DD causes the highest financial impact, as the incidence rate of the clinically relevant M2 stage within a herd is high 4 .The average costs per case of DD were estimated to amount in total to 133 USD, made up of treatment costs (42%), followed by the consequences of decreased fertility (31%) and milk loss (27%) 3 .A more recent study showed that even DD lesions smaller than 2 cm in diameter can cause lameness and production losses 5 .
In recent years, DD has become an emerging issue in dairy herds worldwide with increasing prevalence in many countries 6,7 .A cross-sectional study conducted during routine claw-trimming of 1,449 Swiss dairy cows from June 2010 until February 2011 estimated the prevalence of DD to be 29.1% at the cow and 73.1% at herd level 8 .Poor environmental conditions of housed cattle are likely to result in increased risk of contracting DD, as continuous exposure of feet to moisture and poor hygiene conditions are considered predisposing factors for DD 9,10 .
DD is characterized by an inflammatory dermatitis of the skin with necrosis of infected tissue 11 .Histopathologically, DD lesions show hyperplasia of the epidermis with hyperkeratosis, loss of the stratum corneum and/ or granulosum, necrosis of the epidermis leading to ulceration, colonies of spirochaetal bacteria, and a mixed inflammation of varying severity in the dermis with exocytosis into the epidermis 2,12,13 .The etiology of DD is still not completely resolved.Studies on the pathogenesis of DD support the hypothesis that bacteria from the genus Treponema appear to be the most commonly identified organisms present and involved in DD lesions 14,15 .Treponema species are gram-negative anaerobic spirochaetes that are very difficult to culture 16 .Attempts to isolate Treponema from DD lesions are still mostly unsuccessful due to their fastidious nature and the high level of bacterial contamination of specimens 17 .However, a limited number of Treponema species from DD lesions have been successfully isolated as shown in earlier studies in the United Kingdom 14,18 .Isolates from DD lesions could be attributed to three distinct phylogroups represented by the species Treponema pedis, Treponema phagedenis, and Treponema medium [17][18][19] .The presence of one or any combination of these three species is associated with clinical symptoms of DD as documented by several PCR-based investigations 18,20 .
Identification of the Treponema species associated with DD lesions in Switzerland would enable better understanding of the epidemiology of DD, facilitate efficient treatment campaigns and subsequently control DD.In the present study, we determined the prevalence of T. pedis, T. phagedenis, and T. medium in healthy and DD-affected animals as well as slurry from the environment of DD-positive farms in different regions of Switzerland using PCR-based methods.Farms without clinical findings of DD were included as a control group and examined using the same methodology.

Collection of swab samples and DD lesion scoring
A total of 22 farms with clinical DD (14 free-stall and 8 tie-stall) and two farms without clinical DD (one freestall and one tie-stall) were enrolled in the study (Fig. 1) and sampled between November 2015 and June 2018.Twenty two dairy farms and 2 beef cattle farms consisting of different breeds were involved in this study.The cows were allowed daily access to pasture during the grazing season (April to October) and weekly access to an outside yard (mastic asphalt) during the winter season (November to March).
The DD lesions were scored by a trained and experienced study author (MA) according to Döpfer et al. 25 and extended by Berry et al. 26 .Briefly, M1 ("M" refers to Mortellaro's initial description of the lesion) describes infected animals with small (<2 cm diameter) focal active lesions; M2 infected animals with larger (>2 cm diameter) classic active ulcers; M3 healing scabbed lesions, and M4 represents chronic stages of infection, characterized by dyskeratosis or surface proliferation; M4.1 refers to a chronic stage with small active M1 lesions, and M5 describes unaffected animals with no lesions (normal skin).In addition, the farmers were asked about the current status of DD on their farms before the sampling was performed (DD present, DD not present or DD suspected).
Slurry samples (n= 21) from DD-affected farms and one unaffected farm with free-stall accommodation were collected with sterile, dry cotton swabs and pooled from different locations on the floor: (i) from the floor of the drinking area, (ii) from below the cow brush, (iii) from below the concentrate feeding area and (iv) from the milking robot in cases where an automatic milking system was present.In tie-stall farms (data available only from DD-affected farms), the swabs were collected from different locations on the floor at the position of the hind limbs.Slurry samples (but no feet samples) were additionally collected from 8 farms with DD.DD was diagnosed in these farms by clinical inspection of the feet in the claw trimming chute (n= 6) or in the milking parlor (n=2).
The individual swabs were immediately placed into an Eppendorf tube containing 1 ml SV lysis buffer (4 M guanidine thiocyanate, 0.01M Tris-HCl, 1% β-mercaptoethanol) for 2 min with gentle stirring.The swab was then discarded and the remaining lysate transferred to the laboratory for DNA extraction within 1 week of sampling.

Laboratory analyses
DNA extraction DNA extraction was performed from 500 µl lysate according to an adapted protocol of Stäuble et al. 24 using a semi-automated extraction robot (KingFisher™ Duo-Prime, Thermo Fisher Scientific).The DNA was eluted in 60 µl H2O and stored at -20 °C until further processing.

PCR assays
General Treponema and species-specific Treponema PCR.
A general Treponema PCR, as well as specific nested PCR assays for T. medium, T. phagedenis, and T. pedis, were performed according to Evans et al. 14 with the exception of the 16S rRNA gene PCR which was performed utilizing primers 16SUNI-L and 16SUNI-R as published elsewhere 27 .PCR master-mix contained 1× FIREPol ® Master Mix Ready to load with 12.5 mM MgCl2 (Solis Biodyne) and 0.4 mM of each primer.For general Treponema PCR and the initial 16S rRNA gene amplification, 2 µl DNA template was added to 28 µl master-mix.Temperature cycling for these two PCRs entailed an initial denaturation step at 95 °C for 3 min followed by 35 cycles of 95 °C for 30 s, annealing for 30 s at 53 °C and extension at 72 °C for 90 s.A final elongation step at 72 °C for 7 min was included.
The nested species-specific Treponema PCRs were done using 29 µl master-mix and 1 µl PCR product as a template from the initial reaction of 16S rRNA gene amplification (see above).Temperature cycling entailed an initial denaturation step at 95 °C for 3 min, followed by 40 cycles of 95 °C for 30 s, annealing for 30 s either at 68 °C for T. medium and T. pedis primers, or at 64 °C for T. phagedenis primers and extension at 72 °C for 1 min.A final elongation step at 72 °C for 7 min was included.To ensure and to validate each PCR assay, H2O was used as a negative control and the corresponding positive field samples were used as positive controls.PCR products were visualized by gel electrophoresis on submarine 1% agarose gels after staining with ethidium bromide.Positive reactions were determined by the presence of bands of the appropriate sizes using 100 bp ladder as a molecular size marker.

Detection of D. nodosus:
Detection and virulotyping of D. nodosus was done using the competitive rtPCR according to Stäuble et al. 24 .This rtPCR distinguished between the presence of the gene aprV2 encoding the thermostable protease AprV2, and the gene aprB2 coding for the thermosensitive protease AprB2 of D. nodosus.All rtPCR reactions were analyzed in duplicate, with a mean threshold cycle (Ct) value of <40 rated positive.

Statistical analysis
Chi-square test (with Pearson's Chi-Square) was used to investigate associations between the presence of Treponema species, aprB2-positive D. nodosus and the DD status of the animals.In all analyses, an associated probability (p-value) of 0.05 was considered significant.All the data were analyzed using the software package NCSS 10 (NCSS LLC, Kaysville, UT).

DD presence on farms
In the DD positive farms (n=22), 11 farmers were aware of the disease, 8 were completely unaware of the presence of clinical DD on their farm, 2 suspected the presence of clinical DD, but were not sure and information from 1 farm was not documented and is therefore missing.
Chi-square analysis showed that the three DD-associated Treponema species were unequally distributed among the DD stages.Positive samples for aprB2 of D. nodosus were not significantly associated with DD-stages.The two control farms were negative for all DD-associated Treponema species, while aprB2-positive in 77.8%.The aprV2 gene of D. nodosus was not detected in any of the feet investigated.

Slurry samples
Three out of 21 slurry specimens (14.3%) were positive for at least one or more of the DD-associated Treponema species (2/21 [9.5%] for T. medium and 1/21 [4.8%] for T. phagedenis), while 52.4% (11/21) were positive for aprB2 and all were negative for aprV2 of D. nodosus.DD Treponema species and aprB2 positive D. nodosus were not identified in the slurry of the unaffected DD farm.
Of the total 107 samples tested, 87 (81.3%) showed consistent results with the general Treponema PCR and the specific Treponema PCRs.In 10 cases (4 from the same farm with stage M5, 5 from slurry samples and one from a stage M2 animal), the general PCR was positive while all specific Treponema PCRs were negative.In another 10 cases (all clinical stages of DD but no slurry samples were represented) at least one of the specific PCRs was positive whereas the general Treponema PCR was negative.

Discussion
Digital dermatitis is most commonly diagnosed by clinical foot examination 28 .Laboratory investigations are often not conclusive due to the unclear etiology of DD and difficulties when culturing Treponema.Fluorescent in situ hybridization (FISH) has already been applied to visualize and localize Treponema species 13 .Results of the FISH analysis indicated that Treponema species were the predominant bacteria in the deep part of the lesions at border between affected and healthy tissue 11 .
The use of PCR-based techniques has helped to overcome these problems and certain Treponema species have recently been shown to be involved in the etiology of DD.To the best of our knowledge, our study is the first to investigate the prevalence of three Treponema species and D. nodosus in DD lesions and environmental samples in Switzerland.
In addition, the present study found only two control farms without any clinical signs of DD lesions.This demonstrates that DD prevalence on farms is underestimated.Eight out of 21 farmers were not aware of the presence of clinical DD on their farm.This can be explained by the findings of Berry et al. 29 who reported that only certain stages of DD cause pain and thus early growth stages of DD do not always present with lameness.Furthermore, detection of slight lameness is challenging for farmers.Multiple studies have associated Treponema species with DD lesions in cattle 22,25,30 .Evans et al. 14 reported that T. phagedenis-like, T.medium/T.vincentii-like and T. putidum/T.denticola-like were present in DD lesion biopsies with a prevalence of 98%, 96.1% and 74.5%, respectively.In our study, we used swab specimens and demonstrated the presence of at least one of the DD-associated Treponema species in DD lesions (82.0%) and healthy tissue (36%) using the same PCR-based approach.Swab sampling unlike the biopsy technique, does not require local anesthesia, but may only be sensitive to Treponema that are present on superficial tissues.Treponema species like T. medium and T. phagedenis have been reported to be found deep inside lesions 31 .This may explain negative PCR results even in the presence of typical DD lesions as bacteria in the deep part of the lesions cannot be sampled by the swab method.Moter et al. 31 reported that the presence of certain Treponema species may correlate with the invasiveness of the disease.More recently, Krull et al. 23 reported an increase in Treponema prevalence from 0.0% in healthy foot skin to 94.3% in DD lesions using deep sequencing analysis and compared to a novel scoring system based on lesion morphology.Sullivan et al. 32 reported that in beef cattle, at least one of the three DD-associated Treponema species were positive in all isolated DD lesions, while no DD-associated Treponema DNA was amplified from healthy foot tissues.Similar to our study, there was at least one Treponema species in a statistically significant proportion of DD lesions ((M1 to M4.1) as compared to healthy foot tissues (M5).
Finally, the prevalence was lower in swabs of M1 lesions as compared to the other lesion stages (M2 to M4.1).This is also in agreement with Krull et al. 23 , who reported relatively low abundance of Treponema species in the early stages of the lesions as compared to the advanced lesions, and the presence of certain Treponema species may correlate with the invasiveness of the disease 31 .
There was good consistency between the general and the specific Treponema PCR with results showing more than 80% agreement.The fact that some samples representing all clinical stages were only positive with the specific but not the general Treponema PCRs is most probably due to the higher sensitivity of the nested specific PCRs compared to the conventional one-step general PCR.On the other hand, samples solely positive with the general Treponema PCR could indicate the presence of other Treponema species than the ones covered by the specific PCR in those samples.As these included mainly samples from slurry and stage M5 animals from a single farm, it suggests that these Treponema are most probably not associated with DD lesions.D. nodosus was recently shown to cause interdigital dermatitis in cattle and is considered to be a major player in the pathogenesis and polymicrobial character of DD lesions 22,33 .We did not detect a difference in D. nodosus prevalence in DD lesions versus healthy foot skin specimens, which challenges the current view on the relevance of this species in DD pathogenesis.The lack of a statistical difference of D. nodosus in healthy and DD lesions is most likely due to its presence in all stages of lesions.This is in agreement with Krull et al. 23 and Zinicola et al. 34 who found no difference in the prevalence of D. nodosus between DD lesions and healthy skin specimens.Sullivan et al. 32 reported that D. nodosus were present in 68% and 26% of beef cattle DD lesions and clinically healthy feet of beef cattle, respectively, comparable with our findings (51.8% and 24.1% in DD and healthy skin specimens, respectively).The established real-time PCR by Stäuble et al. 24 allowed the virulotyping of D. nodosus and confirmed the absence of acid protease AprV2-positive D. nodosus in all samples collected in our study.
The slurry samples showed a low prevalence of DD-associated Treponema species (14.3%).Similarly, a work by Klitgaard et al. 35 , using high-throughput sequencing, identified a small amount of bacterial DNA from DD-associated Treponema species in environmental samples (e.g., manure slurry) collected from dairy farms.Although the DD Treponema species were present in very low abundance in environmental samples, their presence may act as a source of infection 23 .Evans et al. 36 used quantitative PCR to test environmental and animal associated samples for the three Treponema species identified in DD lesions.In their study, they were unable to identify the three DD-associated Treponema species in environmental or fecal samples; however, several samples from the bovine rectal mucosal junction and gingiva were positive.

Fig. 1 :
Fig.1: Geographical distribution of farms with clinical Digital Dermatitis (DD) (black squares) and farms without clinical DD (white circles) in Switzerland.The map was created with QGIS v2.8.1 (https://www.qgis.org/de/site/)using postal codes of farm locations.Postal code 6166 represents 3 farms and 4466 corresponds to 2 farms.

Table 1 :
PCR detection of T. medium, T. phagedenis, T. pedis and aprB2-positive D. nodosus in DD specimens swabbed from lesions in Swiss cattle.