key: cord-0737134-2xmd3jfo
authors: Stefanetti, Valentina; Compagnone, Agnese; Sordini, Chiara; Passamonti, Fabrizio; Rampacci, Elisa; Moscati, Livia; Marenzoni, Maria Luisa
title: Retrospective Biomolecular Investigation of Coxiella burnetii and Leptospira spp. DNA in Cases of Abortion, Stillbirth and Neonatal Mortality in Dogs and Cats
date: 2018-08-20
journal: Top Companion Anim Med
DOI: 10.1053/j.tcam.2018.08.005
sha: 8f78b051ab35efcfb25e72166844b19f8c886bc9
doc_id: 737134
cord_uid: 2xmd3jfo

Abortion and neonatal mortality are events that can occur in breeding bitches and queens. It has been reported that up to 55% and 33% of these cases remain without a known cause, respectively, in canine and feline pregnancies. Unusual abortigenic and potentially zoonotic agents, including Coxiella burnetii and Leptospira spp., may be involved in these cases. C. burnetii is able to cause reproductive disorders in cattle, sheep and goats, and cases of abortion have been observed in dogs and cats. Moreover, several outbreaks of C. burnetii infection in humans have been caused by delivering bitches and queens, and some of these animals experienced abortion. Leptospira interrogans sensu lato is able to cause abortion or stillbirth in several animal species and its abortigenic role has occasionally been described in bitches and queens. The aim of this study was to search for C. burnetii and Leptospira spp. DNA in a retrospective series of 103 cases of canine and feline abortion, stillbirth, and neonatal mortality submitted for the identification of possible infectious agents. One hundred and fifty-one specimens were tested using PCR assays and found negative for C. burnetii and Leptospira DNA. However, in 49 samples (47.6%) other infectious causes of abortion, stillbirth, and neonatal mortality were identified. These results showed that C. burnetii and Leptospira spp. are probably not common abortigenic agents or causes of neonatal deaths in dogs. However, given the potential abortigentic and zoonotic role of these agents, surveillance of canine and feline abortion, stillbirth, and neonatal mortality could be advisable for a systematic investigation of these events.

Abortion, stillbirth, and neonatal mortality are events that can occur in canine and feline reproductive medicine. Their reported incidence ranges from 5% to 35%. The etiology of these losses is complex and can be classically divided into infectious and noninfectious causes. Among the infectious causes, bacterial diseases are reported as a primary cause of mortality in puppies and kittens during the first week of age. 1 Staphylococcus spp., Streptococcus spp., Escherichia coli, Brucella canis, Campylobacter spp., and Salmonella spp., are described as the most frequently isolated pathogens causing pregnancy losses in bitches and queens and canine and feline neonatal mortality. 2À4 Canine distemper virus, canine herpesvirus (CHV-1), canine parvovirus (CPV), canine minute virus, feline herpesvirus (FHV-1), feline panleukopenia virus (FPV), and feline infectious peritonitis, caused by a coronavirus, are reported as the most common viral causes of canine and feline pregnancy loss and neonatal mortality. Toxoplasma gondii and Neospora caninum also are considered as rare causes of abortion and neonatal mortality. 1, 5 Nevertheless, a percentage of these cases of abortion, stillbirth, and neonatal mortality remain without an aetiological identification. For example, it has been reported that 33% of feline neonatal mortality 2 and 55% of canine neonatal deaths 3 were attributed to unknown etiology. In these cases, the involvement of unusual infectious pathogens not routinely investigated in canine and feline abortion, stillbirth, and neonatal mortality cannot be excluded.

Worldwide, C burnetii and Leptospira interrogans sensu lato are bacterial agents well known for their potential abortigenic role in many animal species. C. burnetii is the causative agent of an important and underdiagnosed zoonosis with worldwide distribution, Q fever. Livestock species have traditionally been considered as reservoirs of infection, with microorganisms found in high concentrations in the placenta and reproductive tracts of cattle, sheep, and goats. However, the potential role of other animal reservoirs, including cats and dogs, has been supposed. 6 Antibodies against C. burnetii have been detected in the serum of cats worldwide. 7 Moreover, it has been hypothesized that cats have been responsible for transmission of C. burnetii to humans in Q fever outbreaks due to exposure to parturient queens, in which various degrees of seroprevalence have been detected. 7À10 Nagaoka et al. 11 provided the first evidence of active infection, isolating 9 of 29 cases of C. burnetii from vaginal swabs of domestic queens, and C. burnetii DNA was amplified through PCR assay from 3 out of 37 uterine tissues from cats with and without a history of reproductive abnormalities. 12 In canine species, the epidemiological role of C. burnetii infection is still debated; however, seropositivity has been reported in dogs, and human outbreaks associated with parturient bitches have been described. 13, 14 C. burnetii DNA has also been recently detected in 4 of 54 canine placentas originating from aborting animals in The Netherlands. 15 Leptospirosis is a worldwide re-emerging disease caused by a gram-negative bacterium of the genus Leptospira. There are over 250 pathogenic serovars that are adapted to different wild or domestic animal reservoir hosts, and serovars are further grouped into antigenically related serogroups. Infection with pathogenic leptospires can cause a wide range of clinical manifestations from subclinical to severe and potentially lethal disease. 16 However, there are only a few reports of reproductive disorders in dogs related to leptospiral infection; abortion has occurred in dogs after transplacental spread of the serovar Buenos Aires, 17 and a report suggested that abortion was associated with serovar Bratislava infection. 18 Some commercial companies that promote diagnostic devices currently suggest including leptospirosis testing in cases of abortion in dogs. In feline species, seroprevalence of Leptospira spp. varies from 0% to 35% depending on the geographical area and the diagnostic methods used 16 ; however, despite serological evidence, there is little information about clinical leptospirosis in cats. A recent report described 3 confirmed and naturally infected clinical cases of feline leptospirosis presenting with renal failure. 19 Stillbirths, abortions, and retained placenta due to Leptospira spp. also have been reported in queens. 20 The aim of this study was to search for C. burnetii and Leptospira spp. DNA in a retrospective series of cases of canine and feline abortion, stillbirth, and neonatal mortality submitted to a diagnostic laboratory of infectious diseases.

Considering a normal gestation period of 57-72 days for dogs and 52-74 days for cats, the cases were classified as follows: abortion was defined as fetal loss during the second half of pregnancy and characterized by the expulsion of a dead conceptus or a living one incapable of independent life; stillbirth was defined as puppies or kittens reported as dead at birth; neonatal mortality was divided into early neonatal mortality if puppy or kitten death occurred within 7 days after birth and late neonatal mortality if the death occurred 7-28 days after birth.

According to this classification, abortions, stillbirth cases, and dead neonatal dogs and cats were collected between 2008 and 2015 by clinicians specializing in small animal reproduction. The samples were promptly sent to the infectious disease laboratory, Department of Veterinary Medicine (University of Perugia) to search for at least one of the most common infectious agents of abortion or neonatal mortality, both bacterial and viral, including CHV-1 and CPV for dogs and FPV and FHV-1 for cats. The pathological materials arrived at the laboratory within 24 hours of the event (abortion, stillbirth, or neonatal mortality) and were maintained refrigerated at 4°C until processing. When the veterinarians were not able to send the pathological material within 24 hours, it was stored frozen and sent afterward.

While a biomolecular method was used to detect the viral agents (CHV-1 and CPV for dogs; FPV and FHV-1 for cats), a standard bacteriological examination was performed. These examinations were not carried out systematically, but only when required by the referring veterinarian and only when the samples were suitable for the test required (for example, only refrigerated samples for the bacteriological examination). For the bacteriological examination, samples were inoculated onto blood agar, mannitol-salt agar, and MacConkey's agar plates. The plates were incubated for a minimum of 24 hours at 37°C under aerobic and anaerobic conditions. Isolation of a pure or predominant bacterial species from all the specimens originating from the same puppy or kitten, was considered a positive result. The identification of bacteria was performed by examining colony characteristics, gram staining, and use of biochemical tests and commercially available API kits (bioMerieux, Etoile, France). Breed and exposure to risk factors for infectious diseases were not recorded or investigated.

DNA extraction from placenta (when available) and a pool of lung, liver, and spleen from the canine or feline foetuses or neonates was performed using the GenElute Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich, St. Louis). The concentration and purity of the extracted DNA were quantified using a NanoDrop spectrophotometer (NanoDrop 2000, Thermo Fisher Scientific, Milan, Italy). The extracted DNA was stored at ¡20°C until use. A PCR protocol described by Berri et al. 21 amplifying a fragment of 687 bp of the IS1111 insertion sequence was used to detect C. burnetii DNA. The previously reported detection limit of the C. burnetii PCR assay was 10 ¡2 dilution of C. burnetii positive control DNA (comparable to 12 pg/ml of DNA), 12 but considering that the mean number of IS1111 copies varies between 7 and 110, 22 no exact limit was defined.

A nested PCR protocol for Leptospira spp. was developed to increase the sensitivity of a conventional PCR previously described, 23 amplifying a preserved fragment of the 16S rRNA gene present in both saprophytic and pathogenic Leptospira spp. The first PCR was performed using the previously published primer pair, resulting in a 330 bp product, whereas an internal pair of primers (FN: 5 0 -CATG-CAAGTCAAGCGGAGTA-3 0 ; RN: 5 0 -GGCTCATCTCCGAGCAATAA-3 0 ) was designed using the Primer3 software (http://bioinfo.ut.ee/primer3-0.4.0/primer3/), yielding a fragment of 166 bp. Serial 10-fold dilutions of the Leptospira controls, ranging from 5 ng/mL to 0.05 pg/mL, were carried out to determine the increase in sensitivity of the nested protocol in the biological samples compared with the single protocol. Both saprophytic and pathogenic Leptospirae, kindly provided by the National Reference Centre for Animal Leptospirosis (L. interrogans serovar pomona, type Mezzano I; L. interrogans serovar hardjo, type Hardjoprajitno; L. interrogans serovar Bratislava, type Riccio 2; L. kirschneri serovar grippotyphosa, type Moskva V; L. biflexa serovar patoc, type Patoc), and unrelated bacteria (Streptococcus spp.; Eschericia coli; Klebsiella pneumoniae; Pseudomonas aeruginosa; Staphylococcus intermedius; Proteus vulgaris; Enterococcus faecalis), as well as DNA from bovine and caprine faeces and clinical specimens from animals with previously diagnosed Leptospira infection, were used to test the specificity of the protocol.

A total of 25 mL of reaction mixture contained 10 £ buffer, 3 mM MgCl 2 , 200 mM each deoxyribonucleotide triphosphate, 1 mM each primer (Sigma-Genosys), 0.5 U Taq DNA polymerase (Microtech, Italy). A fixed volume of DNA was used, 5 mL for the first cycle and 1 mL for the nested protocol. The cycling conditions were as follows: 95°C for 5 minutes, 35 cycles of 94°C for 30 seconds, 62°C for 30 seconds for Leptospira and 59°C for 30 seconds for C. burnetii, and 72°C for 45 seconds, and 72°C for 10 minutes. In each set of reactions, a positive control (L. interrogans serovar Bratislava, type Riccio 2; C. burnetii 9 mile/I/EP1, ATCC VR-615), and a negative control (negative sample), as well as a negative reaction mix control (containing the reagents and water instead of DNA), were included in each run.

A previously published PCR protocol targeting the LipL32 gene of pathogenic Leptospira spp. was planned to confirm any case of positivity for Leptospira DNA. 24 

A total of 151 specimens were examined in this study, originating from 103 cases in dogs (n = 94) and cats (n = 9), among which there were 5 abortions (4.8%), 3 stillbirths (2.9%), 24 cases of early neonatal mortality (23.3%), 18 cases of late neonatal mortality (17.5%), and 53 cases in which no age was recorded (51.4%).

All the samples tested were negative for Leptospira and C. burnetii DNAs. Detection limits of the Leptospira spp. PCR assay were 0.05 ng/ mL in the first cycle and 0.5 pg/ mL in the nested cycle. The nested protocol increased the sensitivity of the test 100 times. Considering that the genome of a single Leptospira, corresponding to 4.3 £ 10 6 bp in L. interrogans and 3.6 £ 10 6 bp in L. biflexa, weights 5-10 fg, the protocol was able to detect 50-100 bacteria/mL. All the tested Leptospira strains used as controls were positive, whereas no amplification was obtained from bacteria other than Leptospira spp. or bovine and caprine faeces, confirming the specificity of the test.

Although not systematically investigated, in 49 cases (47.6%) other likely causes of abortion and neonatal mortality were identified. Overall, CHV-1 was detected in 7.8% cases (6/77), and one of these showed a simultaneous coinfection with a beta-haemolytic strain of E. coli. FHV-1 was detected in 11% (1/9) and FPV in 9% (1/11) of the cases. Bacteriological examination was performed in 62 cases, 38 (61.3%) of which were positive. In particular, E. coli (20/ 

The role of C. burnettii and Leptospira spp. in canine and feline abortion, stillbirth, and neonatal mortality is generally minimally investigated. In this study, no evidence of C. burnetii and Leptospira spp. DNA was found in cases of canine and feline abortion, stillbirth, and neonatal mortality submitted for the investigation of infectious causes; in fewer than the half of the cases had an infectious agent already been identified as a possible cause. The presence of C. burnetii and Leptospira has been previously reported in the same geographical area and time frame in which this study was conducted, although in different animal species; leptospiral DNA was found in equine abortion and stillbirth, 25 whereas C. burnetii has been found in sheep in the same area. 26 Moreover, diagnoses of cases of leptospirosis in dogs are reported regularly. 27 According to these previous studies, it might be supposed that our negative results could be related to the fact that cases in dogs and cats not exposed to risk factors for Leptospira spp. and C. burnetii infections were analysed in this study. Indeed, Q fever and coxiellosis are traditionally associated with contact with cattle, sheep, and goats, which are considered the main reservoir of infection. In a previous study, which tested canine sera to detect antibodies for C. burnetii, dogs that had contact with ruminants were 10 times more likely to be positive. 28 On the other hand, it has been observed that dogs living in kennels had a higher prevalence of Leptospira spp. when compared with client-owned dogs, and that, despite vaccination, risk factors for exposure are relevant for the infection. 29 Similarly, in cats, a significantly higher C.burnetii antibody positivity rate was found in stray cats than in pets, 30 and leptospiral antibody prevalence is higher in outdoor cats, as well as in cats living in urban areas and cats that are known hunters. 31 It has been suggested that the active predatory behaviour of some cats increases the chance of infection for both C. burnetii and Leptospira spp., due to contact with reservoirs such as rodents. 7, 16 Unfortunately, in the current study, a serological investigation to establish the exposure to these pathogens was not performed in the bitches and queens, and no indication of breed or lifestyle was specifically requested upon admission of the specimens. Further studies targeting especially populations at risk for these infections should be designed, using animals exposed to water sources or environments contaminated by faeces, urine, or parturition products of animal reservoirs in order to increase the likelihood of finding positive results and determining whether C. burnetii and Leptospira spp. are responsible for canine and feline abortion and neonatal losses.

Overall, it is possible that only a specific subpopulation of dogs and cats, especially pedigree cats and pets, was submitted for diagnostic examination in cases of abortion and neonatal mortality in the current study. A previous retrospective analysis of feline neonatal mortality in the United Kingdom suggested that pedigree cat breeders generally investigate the reasons for such deaths 2 and it is possible that infectious diseases associated with environmental exposure are underrepresented in this group of animals.

The failure to confirm the presence of C. burnetii and Leptospira spp. in this study could also be attributed to the timing of sample collection and to the nonoptimal storage conditions that can occur, according to the specific infection. 32 Additionally, in this study placentas were very rarely submitted for analysis because bitches and queens commonly ingest them. Examination of the placenta and umbilical cord is recognized as an important means of diagnosing leptospirosis in some species. 33 In addition, there is strong evidence that the placenta represents the critical tissue also for the detection of C. burnetii, because up to 10 9 C. burnetii per g of placental tissue are released in some species at the time of delivery. 34 These results showed that C. burnetii and Leptospira spp. are probably not common agents of canine abortion or neonatal deaths, whereas a higher number of feline cases is needed to arrive at conclusions. However, considering that reproductive failure can occur following C. burnetii and Leptospira spp. infection, these 2 agents should always be considered in the differential diagnosis in any case of abortion, stillbirth, and neonatal mortality, especially when none of the usually suspected pathogens causing reproductive abnormalities has been identified. 12, 35 Moreover, due to the zoonotic potential of these aetiological agents and the close relationship between companion animals and humans, an active surveillance programme encompassing a larger number of canine and feline abortions/stillbirths/neonatal deaths, including a standardized form for recording signals and investigating some risk factors, could be useful to identify these uncommon agents of abortion and neonatal losses and improve the epidemiological knowledge of them.

Fetal loss in the dog and cat

Kitten mortality in the United Kingdom: a retrospective analysis of 274 histopathological examinations

A review of the fading puppy syndrome (also known as fading puppy complex)

Bacterial and protozoal causes of pregnancy loss in the bitch and queen

Molecular characterization of Canine minute virus associated with neonatal mortality in a litter of Jack Russell terrier dogs

Investigating Coxiella burnetii infection in a breeding cattery at the centre of a Q fever outbreak

Exposure to parturient cats: a risk factor for acquisition of Q fever in Maritime Canada

Truckin' pneumoniaÀan outbreak of Q fever in a truck repair plant probably due to aerosols from clothing contaminated by contact with newborn kittens

An outbreak of cat-associated Q fever in the United States

Isolation of Coxiella burnetii from the vagina of feline clients at veterinary clinics

Evaluation of associations among Coxiella burnetii and reproductive abnormalities in cats

A dog-related outbreak of Q fever

Draft genome sequence of Coxiella burnetii Dog Utad, a strain isolated from a dog-related outbreak of Q fever

Search for possible additional reservoirs for human Q fever

European consensus statement on leptospirosis in dogs and cats

Buenos Aires, a new Leptospira serovar of serogroup Djasiman, isolated from an aborted dog fetus in Argentina

Clinical leptospirosis in three cats

Feline stillbirths associated with mixed Salmonella typhimurium and Leptospira infection

The detection of Coxiella burnetii from ovine genital swabs, milk and fecal samples by the use of a single touchdown polymerase chain reaction

Saint Girons I. Polymerase chain reaction for detection of Leptospira spp. in clinical samples

Detection of pathogenic Leptospira spp. through TaqMan polymerase chain reaction targeting the LipL32 gene

Causes of equine abortion, stillbirth and neonatal death in central Italy

Investigation of coxiella burnetii occurrence in dairy sheep flocks by molecular analysis in Umbrian area: a pilot study

Serological surveillance of Leptospirosis in Italy: two-year national data

Survey of seroprevalence of Q fever in dogs in the southeast of France

Serological survey of leptospiral infection in kennelled dogs in Italy

Seroprevalence of Coxiella burnetii infections among cats in different living environments

Serologic and urinary PCR survey of leptospirosis in healthy cats and in cats with kidney disease

Canine and feline abortion diagnostics

Funisitis associated with leptospiral abortion in an equine placenta

Q fever: a zoonosis

Canine leptospirosisÀdo we have a problem?

The authors would like to thank Carlo Sanesi for his helpful technical support. We also gratefully acknowledge Dr Silvia Tagliabue and the National Reference Centre for Animal Leptospirosis (Istituto Zooprofilattico Sperimentale Lombardia ed Emilia Romagna) for kindly providing the leptospiral serovars used as positive controls.