R. Baldelli

The northward spread of leishmaniasis in Italy: evidence from retrospective and ongoing studies on the canine reservoir and phlebotomine vectors

Visceral leishmaniasis (VL) incidence has been increased in Italy in humans and dogs since the 1990s, with new foci being detected within traditional boundaries of endemic transmission but also in northern regions previously regarded as non‐endemic. To monitor the putative VL spreading, surveillance was implemented in northern continental Italy comprising: analysis of human cases recorded from 1990 through 2005; retrospective literature analysis of canine leishmaniasis (CanL) and phlebotomine sandfly records through 2002; prospective investigations in dogs from 2003 through 2005 and surveys on sandflies in 2003 and 2004. Two‐hundred‐thirty human cases (11% of Italian cases) were recorded. Their stratification by age and HIV status disclosed a sharp decrease of HIV/VL co‐infections paralleled by concomitant increase of paediatric and HIV‐negative adult patients during the study period. Four patients had no travel history. Seven leishmaniasis foci were retrospectively identified since 1990, whereas prospective investigations in dogs disclosed 47 autochthonous clinical cases and 106 autochthonous seropositives among 5442 dogs (2.1%) from 16 foci of six regions. Parasites were typed as Leishmania infantum MON‐1. Four vector species were identified among 1696 Phlebotomus (Larroussius) collected specimens. Comparisons with historical data showed that P. perniciosus and P. neglectus have increased in density and expanded their geographic range in the study area. Northern continental Italy is now focally endemic for VL and a moderate risk for human disease does exist, although the intensity of transmission seems to be lower than in traditional settings of Mediterranean VL.

Tetracycline-resistant Chlamydia suis isolates in Italy.

ANTIMICROBIAL agents are often used to prevent or treat chlamydial infections. The widespread use of tetracycline antibiotics in this way has encouraged selection for resistant organisms. The aim of this study was to evaluate the sensitivity to tetracycline of some strains of Chlamydia suis isolated

Canine leishmaniasis surveillance in a northern Italy kennel

An epidemiological survey on canine leishmaniasis (CanL) was performed during a 3-year period (2007-2009) in a public kennel of the Bologna province. The presence of the disease was shown in the canine population for the first time in 2007 by indirect fluorescent antibody test (IFAT). The parasite circulation was confirmed also by direct diagnostic tools, as PCR, cytology and cultural method, performed on different bioptic materials. The parasite was isolated and identified as Leishmania infantum zymodeme MON 1. The serological monitoring was performed also in 2008 and 2009 on animals that previously showed negative or uncertain results. The incidence values calculated by significant seroconversions in IFAT titre ≥ 1/160, ranged between 4.9% and 6.6%, indicating a stable focus of leishmaniasis. The entomological survey, performed by sticky and CO(2)-baited traps in 2008, showed the presence of the vector Phlebotomus perfiliewi. This study allowed us to identify a stable focus of CanL in an area that was not considered eco-compatible with the presence of the vector and infection. Our results confirm the northward spread of CanL towards areas not previously affected by autochthonous foci.

In Vitro Activities of Several Antimicrobial Agents against Recently Isolated and Genotyped Chlamydia trachomatis Urogenital Serovars D through K

ABSTRACT A systematic evaluation of the susceptibility of all Chlamydia trachomatis urogenital serovars (D through K) to levofloxacin, erythromycin, doxycycline, clarithromycin, and azithromycin was performed. All C. trachomatis serovars had comparable susceptibilities with respect to the various antimicrobials tested, thus confirming the homogeneous data so far obtained regarding the susceptibility of C. trachomatis to antimicrobial agents.

Serological response to pgp3 protein in animal and human chlamydial infections.

Specific antibodies to plasmid-encoded protein pgp3 are known to be encountered in human Chlamydia (C.) trachomatis infections. In order to verify whether antibodies to this protein could be developed in animals infected with plasmid-carrying chlamydial strains, 454 animal sera were examined using a home-made pgp3 protein ELISA and Western blots (WB) of recombinant pgp3 protein from Chlamydophila (Cp.) psittaci. Likewise, 50 human sera were tested by ELISA and WB of recombinant pgp3 from C. trachomatis. The reactivity against pgp3 protein was compared to the reactivity against chlamydial elementary bodies (EBs) detected by microimmunofluorescence (MIF) test. The presence of pgp3-specific antibodies was demonstrated in most ducks and pigeons with Cp. psittaci infection detected by MIF, as well as in the majority of symptomatic cats and pigs infected with Cp. felis and C. suis, respectively, which reacted at high titres to Cp. felis and C. suis EBs by MIF. Moreover, most of the sera collected from patients with C. trachomatis culture-confirmed infection and seropositive to C. trachomatis by MIF, presented antibodies specific to C. trachomatis pgp3 recombinant protein. Therefore, pgp3 protein could be a useful marker of chlamydial infections in animals, as well as in humans.

Seroprevalence to chlamydiae in pigs in Italy

ACCORDING to the current taxonomy (Everett and others 1999), four chlamydial species, namely Chlamydia suis (formerly the porcine serovar of Chlamydia trachomatis), Chlamydophila psittaci (formerly avian serovars of Chlamydia psittaci), Chlamydophila abortus (formerly ovine serovars of C psittaci) and Chlamydophila pecorum (formerly Chlamydia pecorum) have been isolated from pigs. Chlamydiae in pigs can cause asymptomatic infections and have been associated with pneumonia, polyarthritis, pericarditis, conjunctivitis, enteritis, reproductive disorders and increased perinatal mortality (Martinov and others 1985, Woolen and others 1990, Rogers and others 1993, Andersen 1994, Zahn and others 1995, Thoma and others 1997, Guscetti and others 2000, Becker and others 2004, Sachse and others 2004). It is widely believed that chlamydiae may act together with other agents in multifactorial infectious diseases (Pospischil 2004). Chlamydial infections in pigs have been reported mainly in eastern European countries, Austria, Germany, Switzerland and Belgium. In Italy, chlamydiae were detected in fetal organs by direct immunofluorescence, and seropositivity was shown by the complement fixation test in 26 of 100 sows with reproductive problems (Sala 2003). This short communication describes a study to determine the seroprevalence of C suis, C pecorum, C abortus and C psittaci by the microimmunofluorescence test (MIF) in 337 pigs from 11 intensive farms located in northern Italy. Serum samples were collected randomly from finishing pigs at an abattoir in 2004. The MIF was performed using purified elementary bodies (EBs) of the four different species as antigens: the Italian C suis isolate MS04, obtained from a conjunctival swab from a pig; the Italian C pecorum isolate PV5268, obtained from a bovine cervical swab; the ovine reference strain S26/3 of C abortus; and the avian reference strain 6BC of C psittaci. MS04 and PV5268 had been characterised by molecular analysis. The EBs were purified by sucrose density-gradient ultracentrifugation by the method of Fukushi and Hirai (1988). The presence of chlamydial antibodies was assessed using fluorescein-conjugated goat anti-pig immuno globulin G serum (Euroclone). The sera were screened at 1:32 dilution in phosphate-buffered saline supplemented with 2 per cent fetal calf serum. The test was performed according to the method of Wang and Grayston (1986). Serial twofold dilutions of the sera that tested positive at 1:32 dilution were tested to determine the antibody titre. The reciprocal of the highest serum dilution showing an apple-green fluorescence of EBs was considered to be the antibody titre. Antibodies to C suis were detected in 214 of the 337 (63·5 per cent) samples tested, with antibody titres ranging from 32 to 512. Seropositivity to C suis was observed in pigs from all 11 farms investigated, ranging from 20 per cent to 100 per cent (mean 61 per cent). Only a few sera with high antibody titres to C suis reacted weakly at 1:32 dilution with the other chlamydial species. No antibody titres above 32 were detected in any sera to C pecorum, C abortus or C psittaci. The high chlamydial seroprevalence is consistent with the results of serological surveys performed in other countries. Wendt and others (1998) reported a chlamydial seropositivity ranging from 4 per cent to 72 per cent in breeding sows in Germany, in a study that used an ELISA for antibodies to the chlamydial-specific lipopolysaccharide (LPS) antigen; significantly higher percentages of seropositive sows were found in herds with reproductive disorders. Vanrompay and others (2004) reported seropositivity in 96·5 per cent of 258 closed pig breeding farms in Belgium, tested by an indirect ELISA using C psittaci recombinant major outer membrane protein as antigen. Camenisch and others (2004) reported that 61·7 per cent of 193 sera taken from Swiss herds of sows with or without reproductive problems were positive for antibodies to LPS of Chlamydiaceae, with no significant difference between the two groups of herds. In those studies, family-specific antibodies were detected. In the present study, the MIF performed with EBs of the different chlamydial species allowed the evaluation of species-specific seroreactivity and showed a high seroprevalence for C suis. Since antichlamydial vaccines are not administered in pig herds, this seropositivity suggests that the pigs had extensive contact with C suis. The association of C suis with the porcine intestinal tract (Schiller and others 1997a, Hoelzle and others 2000) and its shedding in faeces could increase its spread on farms. C suis is associated with conjunctivitis, enteritis and pneumonia (Rogers and Andersen 2000, Merialdi and others 2003, Sachse and others 2004); it has also been detected in fetal organs from porcine abortions together with C pecorum (Schiller and others 1997b), and in cervical swabs of sows with reproductive problems in association with C abortus (Hoelzle and others 2000). In the present study, the history of the herds described respiratory and reproductive problems on only two farms and on these farms seroprevalences of 75 per cent and 66 per cent against chlamydiae were detected. Since these farms did not test for antibodies to other infectious agents, the association between the C suis seropositivity and the clinical signs is not clear. The remaining nine farms, which showed seropositivity ranging from 20 per cent to 100 per cent (mean 59 per cent), did not report clinical signs suggestive of chlamydiosis. The seropositivity on these farms could be due to asymptomatic infections; for example, intestinal chlamydiosis seems to be a mainly subclinical condition (Nietfeld and others 1997). Further investigations are needed to assess the role of C suis as a bacterial pathogen in pigs.

Seroepidemiologic Survey for Chlamydia suis in Wild Boar (Sus scrofa) Populations in Italy

We used serology to estimate the prevalence of exposure to chlamydiae in Italian populations of wild boars (Sus scrofa). Sera from 173 hunter-killed wild boars harvested during the 2006–2009 hunting seasons in three Italian regions were tested for antibodies to Chlamydia suis, Chlamydophila pecorum, Chlamydophila abortus, and Chlamydophila psittaci by the microimmunofluorescence test. Antibody titers to chlamydiae ≥1:32 were detected in 110 of the 173 samples tested (63.6%). Specific reactivity could be assessed only in 44 sera with antibody titers to C. suis that were two- to threefold higher than antibody titers against the other chlamydial species; the other 66 sera had similar reactivity against all the chlamydia species tested. Antibody to C. suis was detected in sera from wild boar populations with rare or no known contact with domestic pigs. These results suggest that the wild boar could be a chlamydia reservoir and may acquire chlamydiae independent of contacts with the domestic pig.

Antibody-neutralizing activity against all urogenital Chlamydia trachomatis serovars in Chlamydia suis-infected pigs

It is known that neutralizing species-specific or serovar-specific antibodies are produced in response to chlamydial infection in humans and in some animal species. In a previous study, a strong in vitro neutralizing activity to Chlamydia suis in 80% of sera from C. suis-infected pigs had been observed. In view of the close relationship between C. suis and Chlamydia trachomatis, in the present study, the neutralizing activity against D-K C. trachomatis and C. suis purified elementary bodies (EBs) in sera collected from C. trachomatis-infected patients and C. suis-infected pigs was evaluated. A neutralizing activity of 50-70% was observed in the human sera against the homologous serovar and one to five heterologous C. trachomatis serovars. These sera were also able to neutralize C. suis EBs. The pig sera showed a strong neutralizing activity (70-100%) against C. suis EBs and all eight urogenital C. trachomatis serovars. These results suggested the presence of common immunogenic antigens in C. trachomatis and C. suis. Immunoblot analysis, performed to elucidate the target of this neutralizing activity, showed a clear reactivity in human and pig sera against two proteins of 150 and 40 kDa MW, when tested either with C. trachomatis or with C. suis EBs.

Serosurvey for CPV-2, distemper virus, ehrlichiosis and leishmaniosis in free-ranging dogs in Italy

ROUGH estimates of the number of free-ranging dogs in Italy (dogs that are allowed to move freely in and out of the villages and in the countryside) suggest that the population has increased from nearly 800,000 (Boitani and Fabbri 1983) to approximately 1·2 million dogs (Genovesi and Dupre 2000) over the past 20 years. In southern Italy, free-ranging dogs mainly belong to the ‘guard dog’ or shepherd dog breeds (Maremmano phenotype), and usually live outdoors with their flocks or herds throughout the year. Owners support them with scrap food two to three times a week, so that the dogs have to find other sources of food, such as temporary dumps or offal. These dogs are generally unvaccinated, are not treated against parasites and are free to reproduce. It has been suggested that free-ranging dogs represent a reservoir of infection for both wild canids and pets living in the same habitat. This short communication describes a study to evaluate the seroprevalence of canine parvovirus type 2 (CPV-2), canine distemper virus (CDV), Ehrlichia canis and Leishmania infantum in a population of free-ranging shepherd dogs, which shared the same environment as a small wolf population with a positive serology (Italian National Wildlife Institute, unpublished data). The study area was located in the Pollino National Park in the Basilicata and Calabria regions of southern Italy. The park has an area of 1925 km2 and a mean altitude of 807 m above sea level (range 100 to 2200 m). The park is divided into three main geographical areas: the massif of Pollino in the centre of the park, the massif of Orsomarso at the south west of the park and the Ionic area located in the east. The climate ranges from the continental climate recorded in the massif of Pollino to the extreme Mediterranean type in the Ionic area. A specific survey estimated that there were approximately 2000 free-ranging dogs in the whole of the national park, with a mean density of 1·07 dogs/km2 (95 per cent confidence interval [CI] 0·95 to 1·13 dogs/km2) (Crispino and Ciucci 2002). During October 2001, and April and September 2002 a total of 131 free-ranging dogs were sampled. The dogs were randomly selected from a list of livestock owners provided by the Local Forestry Corp, which is in charge of issuing grazing permits. Illegal grazing is relatively common, so the number of livestock owners is likely to be underestimated. Because the expected prevalence was unknown, the sampling intensity was designed to detect a 50 per cent prevalence with a maximum error of 10 per cent in a population of 2000 dogs (Barnett 1991). The sex of the sampled dogs was recorded in the field together with information on the altitude and geographical area, whereas information regarding age, vaccinations and treatment against parasites was provided directly by the owners. CPV-2 and CDV seroprevalences were detected by a commercial ELISA kit (Ingezim, Spain). Antibodies against E canis and L infantum were assayed by indirect immunofluorescence tests, as described by Ristic and others (1972) and Gradoni and Gramiccia (2000), respectively. Multiple logistic regression was used to evaluate which of the independent variables (age, sex, altitude and geographical area) were associated with seropositivity (odds ratio). Independent variables were included in the model by the forward likelihood ratio selection (with a cut-off value for inclusion of 0·05). The fit of the final model was determined by the Hosmer-Lemeshow test statistic (Hosmer and Lemeshow 2000). The strengths of association between the dependent and independent variables were estimated by the odds ratios with a CI of 95 per cent, while the variability explained by the final model was quantified by the Nagelkerke R2. Statistical analyses were performed using SPSS 11.0. The age of the sampled dogs ranged from two months to 11 years, with a mean (sd) age of three (2·6) years; the male to female ratio was 1·08. According to the information provided by the owners, 14 dogs were vaccinated against CDV and CPV-2. Therefore, only the 117 unvaccinated dogs were analysed for these infections. Out of the 117 dogs, 97 were positive for CPV-2 (82·9 per cent, sd 4·1), and 94 were positive for CDV (80·3 per cent, sd 4·3). Thirty dogs out of 131 had anti-E canis antibodies (22·9 per cent, sd 4·8), with titres ranging from 1:80 to 1:10,240 (geometric mean 2178), nine dogs had anti-L infantum antibodies (6·9 per cent, sd 2·9), with titres ranging from 1:80 to 1:2560 (geometric mean 403). The results of the logistic regression are shown in Table 1. Age was positively associated with the seroprevalence of CPV-2 and CDV. Other variables (sex, altitude and geographical area) were not significantly associated with CPV-2 or CDV seropositivity. Altitude was the only variable associated with L infantum seroprevalence (higher altitudes corresponded to decreasing prevalence). Altitude and the massif of Pollino region were associated with E canis seroprevalence. The high seroprevalence and association with age for both CPV-2 and CDV (odds ratio: CPV-2 1·703 and CDV 1·445), irrespective of the sampled areas, suggest that these two viruses are widespread; findings such as these are often associated with endemic situations (Anderson and May 1991). The seroprevalence of ehrlichiosis was similar to that in urban and rural dogs sampled in kennels and in veterinary clinics (Baneth and others 1996, Piergili Fioretti and others 1996, Sainz and others 1996). The negative association of altitude with seroprevalence could be due to the low density of Rhipicephalus sanguineus, the vector of E canis at higher altitudes. Longevity and fertility of the ticks are reduced by the thermal excursion at high altitudes (Sweatman 1967). The higher seroprevalence of E canis in the massif of Pollino was likely to be due to the continental climate in this area, which allows the focal diffusion of the vector. Leishmania seroprevalence was lower than that previously observed in the same region but in different habitats (Casalinuovo and others 1996). The relatively high mean altitude of the park does not represent an optimal habitat for the vector (mainly Phlebotomus perniciosus) (Schrey and others 1989, Alves-Pires and Ribeiro 1991) and, accordingly, the few positive cases in this survey were recorded at the lowest altitudes in the park. The logistic Veterinary Record (2007) 160, 91-92

Prevalence of antibodies to Bartonella henselae in dogs in Italy

Bartonella species are Gram-negative, haemotropic, usually vector-borne bacteria, many of which have been identified as emerging pathogens in a wide range of domestic and wild mammals. Among the 11 species or subspecies known or suspected to be pathogenic for human beings, six have been isolated from pet dogs and cats (Chomel and others 2006). In terms of human health, one of the most important Bartonella species is Bartonella henselae, the agent of cat-scratch disease and bacillary angiomatosis. Several studies in various countries have identified cats as the primary reservoir of B henselae, and the domesticated species is most often associated with transmission of the infection to people by scratches or bites (Boulouis and others 2005). The most probable route for cat-to-cat transmission is via intradermal inoculation of infected flea faeces (Chomel and others 1996). Bartonella species was first isolated from a dog in 1993, from a case of endocarditis (Breitschwerdt and others 1995). Dogs have been shown to be infected by several zoonotic Bartonella species: Bartonella clarridgeiae, Bartonella elizabethae, B henselae, Bartonella vinsonii subspecies berkhoffii, Bartonella washoensis and Bartonella quintana (Chomel and others 2006). The transmission of B henselae from dogs to human beings has been occasionally suggested. B henselae was found to be the aetiological agent in a case of human osteomyelitis following a dog scratch (Keret and others 1998). Lymphadenopathy has been described in a dog owner showing seroreactivity for B henselae, and Bartonella species DNA was amplified from gingival swabs of his dogs (Tsukahara and others 1998). Unlike cats, in which Bartonella species infection is rarely symptomatic, dogs develop clinical signs that are very similar to those observed in human beings, such as peliosis hepatis; therefore, dogs may be an important sentinel and a comparative model for human infection. At present, there is little information about the prevalence of B henselae infection in dogs. Serosurveys carried out on dogs in Hawaii, Japan and the UK revealed seroprevalences of 6·5 per cent (Demers and others 1995), 7·7 per cent (Tsukahara and others 1998) and 3 per cent (Barnes and others 2000), respectively. In 2004, a study of dogs from south-east USA described a total B henselae seroprevalence of 23·5 per cent, and statistically significant differences between clinically healthy dogs with no history of exposure to ticks or fleas and dogs with clinical signs compatible with vector-borne diseases (Solano-Gallego and others 2004). These results suggest that ticks or fleas may have a role in the transmission of B henselae to dogs; it is also possible that cats could infect dogs via a scratch or bite. In Italy, B henselae infection has been described in cats (Cabassi and others 2002, Ebani and others 2002, Fabbi and others 2004), but there has been no survey of the epidemiology of Bartonella species epidemiology in dogs. This short communication describes a study to assess the seroprevalence of B henselae in dogs in northern Italy. A total of 381 serum samples, taken from dogs being tested for leishmaniosis between January 2004 and February 2005 at the Department of Veterinary Public Health and Animal Pathology in Bologna, northern Italy, were tested; 275 of the dogs lived outdoors in a rural area, and 106 lived indoors. The sera were tested by an indirect fluorescent antibody test (IFAT), using B henselae type 1 (Houston 1 CP 103737) as antigen. B henselae was cocultivated with antibiotic-free Vero cell cultures, by the method of Zbinden and others (1995). Vero cells were grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10 per cent fetal calf serum (Gibco). B henselae was grown on brain heart infusion agar (Difco) supplemented with 1 per cent haemin solution (Sigma) and 5 per cent sheep blood for five days at 37°C in 5 per cent carbon dioxide. The Vero cells were suspended in DMEM and mixed with the B henselae. The suspension, containing 1 x 105 Vero cells and 1 x 107 bacteria per ml, was seeded into wells of tissue culture chamber slides (0·2 ml) (Lab-Tek; Nunc). After 48 hours, the medium was discarded, and the slides were fixed with acetone and ethanol (1:1) and used for the IFAT. Sera were screened at 1:64 dilution in phosphate-buffered saline (PBS) and incubated with infected cells for 30 minutes at 37°C. The slides were washed three times with PBS and dipped once in distilled water before air drying. Commercial fluorescein-conjugated goat anti-dog immunoglobulin G serum (KPL) was added and the slides were incubated for 30 minutes at 37°C and washed as described above. The slides were viewed with a fluorescent microscope at x 40 magnification. The reciprocal of the highest serum dilution showing bright fluorescence of the bacteria was considered to be the antibody titre. Antibodies to B henselae were detected in 23 of the 381 (6 per cent) samples tested, with antibody titres ranging from 64 to 512. Eight sera were positive at a titre of 64, six at 128, eight at 256 and one at 512. Twenty-two of these dogs were apparently healthy and seronegative for leishmaniosis; one dog with a Bartonella titre of 128 had systemic lymphadenopathy and dermatitis, and was seropositive for leishmaniosis (IFAT titre 1280). Nineteen Bartonella-positive dogs were sporting dogs living in a rural area, which were probably exposed to tick bites; four dogs lived indoors, and no data were available about the presence of cats in their environment. The results of this preliminary study show a seroprevalence of Bartonella of 6 per cent. As cross-reactivity between Bartonella species or mixed infections have been suggested (Chomel and others 2003, MacDonald and others 2004) it is not possible to be certain that this seropositivity is B henselae-specific. Further focused investigations are needed to assess the epidemiological relevance of B henselae infection in dogs in Italy.