Occurrence of Enteropathogenic Bacteria in Urban Pigeons ( Columba livia ) in Italy

Abstract

This study was aimed at evaluating the prevalence of Campylobacter spp., Escherichia coli O157, Salmonella spp., and related virulence factors (the cdt, stx, and eae genes) in urban pigeons of the coastal area of the Campania region (southern Italy). To achieve this goal, cloacal swab samples from a total of 1800 urban pigeons were collected and subjected to culture methods, PCR, and serotyping. The results of the present study showed a prevalence of 48.3% (870/1800), 7.8% (141/1800), and 0.9% (16/1800), for C. jejuni, E. coli O157, and S. Typhimurium, respectively. All C. jejuni isolates (870/870) carried cdt genes, whereas all E. coli O157 isolates carried stx genes, and 14.9% (21/141) carried the eae gene. These findings clearly show that urban pigeons in the coastal area of the Campania region may constitute an environmental reservoir of these pathogens, thus representing a source of infection for other birds, livestock, and humans.

Introduction

Pigeons (Columba livia) are widely distributed in urban areas of Italy, and come into close contact with humans in parks, churches, public gardens, and railroad stations. Their increase has raised public health concerns, and in fact these birds are considered potential reservoirs for several enteropathogenic bacteria, such as Campylobacter spp., Escherichia coli O157, and Salmonella spp. (Lillehaug et al. 2005; Tanaka et al. 2005). Several virulence factor genes carried by these bacteria, such as cytolethal distending toxin (cdt), shigatoxin (stx), and eae genes, are considered to be important for the induction of disease in humans and have been widely studied (Pickett et al. 1996; Gyles 2007).

Few reports of disease transmission between pigeons and humans are available (Haag-Wackernagel and Moch 2004; González-Acuña et al. 2007); however, their close interaction and the role of these birds as vectors of zoonotic agents may pose a public health risk.

In the last 50 years, urban pigeon populations have increased greatly worldwide in most large cities (Magnino et al. 2009), to a level of concern for city administrators and community health officers. Besides being responsible for the massive fouling of buildings and monuments, urban pigeons are considered to be naturally infected with a number of viruses, bacteria, fungi, and protozoa that are pathogenic to humans (Haag-Wackernagel and Moch 2004). In addition, a recent work by Rose and associates (2006) showed that pigeons can cover a maximum distance of 5.29 km, and thus can spread pathogens locally in their environment.

On the basis of these facts, we extensively investigated the prevalence of Campylobacter spp., E. coli O157, Salmonella spp., and related virulence factor genes in urban pigeon populations of the coastal area of the Campania region (southern Italy), which is an area with a high-density pigeon population.

Materials and Methods

Sampling

During the period December 2007–May 2010, cloacal swab samples from a total of 1800 urban pigeons were collected in the coastal area (i.e., the provinces of Caserta, Napoli, and Salerno) of the Campania region in southern Italy. The coastal area of the Campania region includes 60 municipalities, and 30 pigeons from each municipality were captured. It is noteworthy that during the sampling period the Campania region was involved in a waste emergency, and therefore sampling was conducted in municipalities with the waste emergency (n=23), and municipalities without the waste emergency (n=37), for a total of 690 and 1110 pigeons, respectively. Furthermore, in order to evaluate the seasonality of Campylobacter spp., E. coli O157, and Salmonella spp., the sampling period was divided into a warm period (spring/summer) and a cold period (autumn/winter), and we collected 900 pigeons during each period.

Pigeons were trapped by specific pigeon cage traps (The Trap Man, Ormskirk, Lancashire, U.K.) baited with grain. The cages were placed in high-density pigeon areas (i.e., squares, railroad stations, public parks, and seaports). Each pigeon was sampled three times using sterile cotton-tipped swabs, for a total of three swab samples for each pigeon, to detect Campylobacter spp., E. coli O157, and Salmonella spp. In order to avoid to sampling the same animal twice, each pigeon was also marked with a colored plastic ring and then released.

Isolation and characterization of Campylobacter spp.

To isolate Campylobacter spp., cloacal swab samples were inoculated onto Campylobacter-selective enrichment broth (Oxoid Ltd., Basingstoke, U.K.), and incubated at 42°C for 48 h under microaerobic conditions provided by CampyGen (Oxoid). Subsequently, each sample was streaked onto Campylobacter blood-free selective agar (CCDA; Oxoid). After incubation at 42°C for 48 h under microaerobic conditions, the plates were examined for typical Campylobacter colonies. Suspected colonies were sub-cultured on sheep blood agar (Oxoid), and finally incubated for 24 h at 42°C. Under phase contrast microscopy, colonies comprised of curved or spiral motile rods were presumptively identified as Campylobacter spp. and submitted to a multiplex polymerase chain reaction (PCR). The extraction and purification of DNA from isolated colonies on sheep blood agar was performed using a Bactozol kit (Molecular Research Center, Inc., Cincinnati, OH), as described previously (Khan and Edge 2007). The specific detection of the Campylobacter genus was based on PCR amplification of the cadF gene using the oligonucleotide primers cadF2B and cadR1B, as described by Konkel and colleagues (1999). All DNA extracts were also examined, by a triplex PCR, for the presence of C. jejuni, C. coli, and C. lari species, using oligonucleotide primers ICJ-UP and ICJ-DN, ICC-UP and ICC-DN, and ICL-UP and ICL-DN, respectively, as previously described (Khan and Edge 2007). The PCR conditions were as described by Khan and Edge (2007), and the products were separated by electrophoresis on 1.5% agarose gels (Gibco-BRL, Milan, Italy), stained with ethidium bromide, and visualized under UV light. PCR amplified without the DNA was used as negative control, whereas three reference Campylobacter strains, C. jejuni ATCC 29428, C. coli ATCC 33559, and C. lari ATCC 43675, obtained from LGC Promochem (LGC Promochem, Teddington, Middlesex, U.K.), were used as positive controls. Furthermore, positive samples for C. jejuni and C. coli were also examined for the presence of the cytolethal distending toxin genes (cdtA, cdtB, cdtC, and cdt cluster) using the primers and procedures described by Bang and associates (2003).

Isolation and characterization of E. coli O157

To isolate E. coli O157, cloacal swab samples were inoculated onto 10 mL of modified tryptone soy broth (Oxoid) supplemented with novobiocin (Oxoid). The samples were incubated at 37°C for 12–18 h. One milliliter of each culture medium was added to 20 mL of magnetic beads coated with antibody to O157 (Dynal Biotech ASA, Oslo, Norway), and immunomagnetic separation was performed according to the manufacturer's instructions. The magnetic beads were inoculated onto sorbitol MacConkey agar (Oxoid), supplemented with cefixime-tellurite (Oxoid), and chromogenic E. coli O157 agar (Biolife Italiana S.r.l., Milan, Italy). After incubation at 37°C for 18–24 h, sorbitol-negative colonies were selected and screened for O157 antigen by agglutination with an E. coli O157 latex test kit (Oxoid). Isolated E. coli O157 was subcultured on washed sheep blood plates and incubated overnight at 37°C. All isolates that were confirmed to be E. coli O157 by chromogenic E. coli O157 agar and latex test kit were subjected to a multiplex PCR assay to determine the presence of stx (stx1 and stx2) and the E. coli attaching and effacing (eae). DNA extraction and multiplex PCR assay for stx1, stx2, and eaeA gene amplification were performed as described by Wang and colleagues (2002). The PCR products were separated by electrophoresis on 1.5% agarose gels (Gibco-BRL), stained with ethidium bromide, and visualized under UV light. PCR amplified without the DNA was used as negative control; whereas one reference E. coli O157 ATCC 43894 strain (LGC Promochem) was used as positive control.

Isolation and serotyping of Salmonella spp.

To isolate Salmonella spp., cloacal swabs were inoculated in Buffered Peptone Water (Oxoid) and incubated at 37°C for 18 h. After incubation, the samples were inoculated onto Rappaport-Vassiliadis broth (Oxoid) and incubated at 42°C for 18 h. The cultures obtained were plated onto xylose-lysine-deoxycholate agar (Oxoid), incubated at 37°C, and examined after 24 h. Suspected colonies were inoculated onto a second selective agar, Brilliant Green Agar (Oxoid), and incubated at 37°C for 24 h. All isolates were biochemically identified by using the API20-E system (bioMérieux, Milan, Italy). All strains were stored frozen at −80°C in 20% glycerol. Salmonella isolates were serotyped according to the Kauffman-White scheme. The analyses were carried out in collaboration with the National Reference Laboratory for Salmonella (IZSVe, Legnaro, Italy).

Statistical analysis

Data regarding period (warm/cold) and waste emergency (presence/absence) were analyzed by univariate (Pearson's chi-square test for independence) statistical analysis, using each bacteria's (C. jejuni, E. coli O157, and S. typhimurium) status (positive/negative), as dependent variables.

Results

The results of the present study showed a prevalence of 48.3% (870/1800) for Campylobacter spp., with a percentage of isolation ranging between 20.2% for the province of Caserta, and 57.1% for the province of Napoli; a prevalence of 7.8% (141/1800) for E. coli O157, with a percentage of isolation ranging between 1.7% for the province of Caserta, and 10% for the province of Napoli; and a prevalence of 0.9% (16/1800) for Salmonella spp., with a percentage of isolation ranging between 0.8% for the province of Caserta and 1.3% for the province of Napoli (Table 1).

Table 1.

Prevalence of Campylobacter spp., ^a E. coli O157, and Salmonella spp. ^b Isolated From 1800 Pigeons in the Coastal Area of the Campania Region

		Number (%) of positive samples 95% Confidence interval
Province	No. of sites (municipalities)	Campylobacter spp.^a	E. coli O157	Salmonella spp.^b
Caserta	4	25/120 (20.2%)	2/120 (1.7%)	1/120 8 (0.8%)
		95% CI 14.2,29.4%	95% CI 0.3,6.5%	95% CI 0.0,5.2%
Napoli	25	455/750 (57.1%)	75/750 (10.0%)	7/750 (1.3%)
		95% CI 57.1,64.2%	95% CI 8.0,12.4%	95% CI 0.7,2.5%
Salerno	31	390/930 (41.9.%)	64/930 (6.9%)	8/930 (0.9%)
		95% CI 38.7,45.2%	95% CI 5.4,8.7%	95% CI 0.4,1.8%
Total	60	870/1800 (48.3%)	141/1800 (7.8%)	16/1800 (0.9%)
		95% CI 46.0,50.7%	95% CI 6.6,9.2%	95% CI 0.5,1.5%

C. jejuni.

S. Typhimurium.

Specifically, all Campylobacter isolates were identified by triplex PCR as C. jejuni, and all C. jejuni isolated carried cdt genes.

With respect to E. coli O157, all isolates carried the stx1 and stx2 genes, and 14.9% (21/141) carried the eae gene. The results of virulence factors (i.e., cdt, stx1, stx2, and eae genes) of C. jejuni and E. coli O157 are summarized in Table 2.

Table 2.

Prevalence of Virulence Factor Genes (cdtA, cdt B, cdt C, cdt cluster, stx1, stx2, and eae) Carried by C. jejuni and E. coli O157 Isolated from Pigeons in the Coastal Area of the Campania Region

		No. PCR-positive for
Bacteria	No. of isolates	cdtA	cdtB	cdtC	cdt cluster	stx1	stx2	eae
C. jejuni	870	870/870	870/870	870/870	870/870
E. coli O157	141		-			141/141	141/141	21/141

C. jejuni, Campyobacter jejuni; E. coli, Escherichia coli; PCR, polymerase chain reaction.

Regarding Salmonella spp., all isolates were serotyped as S. Typhimurium. Pigeons sampled in the municipalities without waste emergency showed a prevalence of 41.1% (456/1110), 4.6% (51/1110), 0.3% (3/1110), for C. jejuni, E. coli O157, and S. Typhimurium, respectively. The pigeons sampled in the municipalities with waste emergency showed a prevalence of 60.0% (414/690), 13.0% (90/690), and 1.9% (13/690), for C. jejuni, E. coli O157, and S. Typhimurium, respectively. This difference was statistically significant (p<0.0001). Furthermore, there was a highly significant difference related to the sampling period (p<0.0001). In fact, pigeons sampled in the warm period showed a prevalence of 78.2% (704/900), 14.1% (127/900), and 1.8% (16/900), for C. jejuni, E. coli O157, and S. Typhimurium, respectively, whereas pigeons sampled in the cold period showed a prevalence of 18.4% (166/900), 1.6% (14/900), and 0.0% (0/900), for C. jejuni, E. coli O157, and S. Typhimurium, respectively. Results of the statistical analysis are listed in Table 3.

Table 3.

Prevalence of C. jejuni, E. coli O157, and S. Typhimurium Isolated from 1800 Pigeons in the Coastal Area of the Campania Region Stratified by Period (Warm/Cold), and Waste Emergency (Presence/Absence)

	Campylobacter jejuni				Escherichia coli O157				Salmonella Typhimurium
Bacteria Environment	No. of positive pigeons/tested	%	95% CI	p^a	No. of positive pigeons/tested	%	95% CI	p^a	No. of positive pigeons/tested	%	95% CI	p^a
Period
Warm	704/900	78.2	75.3–80.8	<0.0001	127/900	14.1	11.9–16.6	<0.0001	16/900	1.8	1.0–2.9	<0.0001
Cold	166/900	18.4	16.0–21.2		14/900	1.6	0.9–2.7		0/900	0.0	0.0–0.5
Waste emergency
Presence	414/690	60.0	56.2–63.7	<0.0001	90/690	13.0	10.7–15.8	<0.0001	13/690	1.9	1.0–3.3	<0.0001
Absence	456/1110	41.1	38.2–44.0		51/1110	4.6	3.5–6.0		3/1110	0.3	0.1–0.9
Total	870/1800	48.3	46.0–50.7		141/1800	7.8	6.6–9.2		16/1800	0.9	0.5–1.5

By chi-square testing.

CI, confidence interval.

Discussion

A review drawn by Haag-Wackernagel and Moch (2004), showed that feral pigeons harbor a total of 60 different human pathogenic organisms, 5 of which are viruses, 9 are bacteria, 45 are fungi, and 1 is a protozoan.

The findings of the present survey have demonstrated the occurrence of C. jejuni, E. coli O157, and S. Typhimurium in urban pigeons, with a prevalence of 48.3%, 7.8%, and 0.9%, respectively.

The prevalence rate for C. jejuni in the present study is in accord with a recent study conducted in Spain by Vázquez and associates (2010), who reported a prevalence for C. jejuni of 69.1%. In contrast, our values are higher than those reported in Croatia by Vučemilo and colleagues (2003), in Chile by Fernandez and associates (1996), and in Oslo by Lillehaug and co-workers (2005), who recorded prevalences of 8.1%, 6.7%, and 3.0%, respectively. We can explain these differences of prevalence rates by taking into account the different diagnostic techniques used. In fact, the studies mentioned above used biochemical identification tools, whereas in the present study we used PCR, which has been shown to be a highly sensitive tool for C. jejuni detection (Persson and Olsen 2005). However, different urban environments, climates, and other factors could be involved.

Moreover, with respect to cdt genes, our results confirm the high prevalence of the cdtA, cdtB, and cdtC genes in C. jejuni, which is consistent with the results of previous studies (Pickett et al. 1996; Martínez et al. 2006). Little is known about the frequency of these toxin genes among Campylobacter isolated from pigeons. To our knowledge, the results presented here are the first report on the prevalence of cdt genes carried by C. jejuni isolated from pigeons, and thus suggest that the cdt genes may frequently be present in C. jejuni also isolated from this avian species.

With respect to E. coli O157, the prevalence rate of the present study is slightly higher than those reported by Abulreesh (2011), who recorded a prevalence of 2.5% from Saudi Arabia, and by Wani and associates (2004), who recorded a prevalence of 4% from India. In contrast, our results are consistent with those reported by Kobayashi and colleagues (2002), and Morabito and co-workers (2001), who reported prevalence rates for shigatoxin-producing E. coli of 7.5% and 10.7%, respectively, even though none of these isolates belonged to the O157 serogroup.

Finally, the prevalence rate for Salmonella spp. in the present study is lower than those reported by Tanaka and associates (2005) from Japan, and by Dovc and colleagues (2004) from Slovenia, who reported a prevalence for Salmonella spp. of 3.9% and 5.7%, respectively. The low prevalence of Salmonella spp. recovered in the present study may be explained taking into account that pigeons intermittently shed the bacteria (De Herdt and Devriese 2000).

The results reported in the present study clearly showed that urban pigeons in the coastal area of the Campania region may constitute an environmental reservoir of Campylobacter spp., E. coli O157, and Salmonella spp., and may pose a risk to other birds, livestock, and humans. Today, the pigeon feeders are mainly responsible for the establishment of large pigeon populations in our cities, and a supplemental food source for them is provided by rubbish and seasonally-occurring natural food, such as grass and tree seeds in parks and gardens (Magnino et al. 2009). Unfortunately, in the last 5 years the Campania region was spotlighted by the media because of its waste emergency, which was an important cause of the increase in pigeons and related pathogens isolated in the present study, as supported by our results. In fact, pigeons sampled in the municipalities with the waste emergency showed a prevalence of infection significantly higher than that seen in pigeons sampled in the municipalities without the waste emergency. In addition, pigeons sampled during the warm period displayed a significantly higher positive rate for the bacteria investigated than those sampled during the cold period.

In light of these facts, specific measures for the prevention of urban pigeon-related cases of infections in humans should be adopted at different levels. Educational initiatives to communicate the health risks and recommendations for minimizing these risks should be primarily directed at children and immunocompromised individuals. Children should be warned not to handle sick or dead pigeons, whereas immunocompromised individuals should be educated to carefully limit their contact with urban pigeons and enforce strict hygienic procedures when dealing with the birds.

A reduced and healthier population of urban pigeons should be included among the aims of administrators and health officers in many European towns and cities, as a general intervention for preserving urban hygiene. As suggested by Magnino and colleagues (2009), the management of feral pigeon populations in the urban environment is a complex issue that requires careful planning. Before any intervention, an evaluation of the local situation regarding the number of birds and their aggregation sites is mandatory. Fencing of buildings with pigeon-deterring systems such as net-like structures and other mechanical devices represents a first-line intervention for preventing fouling, as well as the administration of contraceptive drugs to reduce the bird population, coupled with other measures, in particular a feeding ban. Pigeon feeders should be encouraged to stop or limit their activity by enforcing a feeding ban in urban areas that are close to hospitals, railway stations, parks, and sanctuaries, where the avoidance of pigeon aggregation is a top priority.

Footnotes

Acknowledgments

The study was funded by the Campania region under the project “Riduzione numerica mediante trattamento con Nicarbazina di animali sinantropi (Columba livia) in aree ad alta urbanizzazione.”

The authors wish to acknowledge Dr. Paolo Sarnelli, director of Regional Veterinary Services, for his support, Dr. Antonia Ricci, director of the National Reference Laboratory for Salmonella (IZSVe, Legnaro, Italy), and all other IZSVe personnel for serotyping of Salmonella strains. The authors are grateful to Dr. Laura Rinaldi for her support with the statistical analysis.

Author Disclosure Statement

No competing financial interests exist.

References

Abulreesh

. Free living rock pigeon (Columba livia) as an environmental reservoir of enteric bacterial pathogens resistant to antimicrobial drugs in Saudi Arabia. Curr Res Bacteriol, 2011; 4:28–33.

Bang

, Nielsen

, Scheutz

, et al. PCR detection of seven virulence and toxin genes of Campylobacter jejuni and Campylobacter coli isolates from Danish pigs and cattle and cytolethal distending toxin production of the isolates. J Appl Microbiol, 2003; 94:1003–1014.

De Herdt

, Devriese

. Pigeons. In: Avian Medicine. Tully

, Lawton

MPC

, Dorrestein

, eds. United Kingdom: Butterworth-Heinemann Ltd., 2000:320–322.

Dovc

, Zorman-Rojs

, Vergles Rataj

, et al. Health status of free-living pigeons (Columba livia domestica) in the city of Ljubljana. Acta Vet Hung, 2004; 52:219–226.

Fernández

, Gesche

, Montefusco

, et al. Wild birds as reservoir of thermophilic enteropathogenic Campylobacter species in southern Chile. Mem Inst Oswaldo Cruz, 1996; 91:699–700.

González-Acuña

, Silva

, Moreno

, et al. Detection of some zoonotic agents in the domestic pigeon (Columba livia) in the city of Chillan, Chile. Rev Chilena Infectol, 2007; 24:199–203. (In Spanish).

Gyles

. Shiga toxin-producing Escherichia coli: an overview. J Anim Sci, 2007; 85:E45–E62.

Haag-Wackernagel

, Moch

. Health hazards posed by feral pigeons. J Infect, 2004; 48:307–313.

Khan

, Edge

. Development of a novel triplex PCR assay for the detection and differentiation of thermophilic species of Campylobacter using 16S-23S rDNA internal transcribed spacer (ITS) region. J Appl Microbiol, 2007; 103:2561–2569.

10.

Kobayashi

, Pohjanvirta

, Pelkonen

. Prevalence and characteristics of intimin- and Shiga toxin-producing Escherichia coli from gulls, pigeons and broilers in Finland. J Vet Med Sci, 2002; 64:1071–1073.

11.

Konkel

, Gray

, Kim

, Garvis

, Yoon

. Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadF virulence gene and its product. J Clin Microbiol, 1999; 37:510–517.

12.

Lillehaug

, Monceyron Jonassen

, Bergsjø

, et al. Screening of feral pigeon (Columba livia), mallard (Anas platyrhynchos) and graylag goose (Anser anser) populations for Campylobacter spp., Salmonella spp., avian influenza virus and avian paramyxovirus. Acta Vet Scand, 2005; 46:193–202.

13.

Magnino

, Haag-Wackernagel

, Geigenfeind

, et al. Chlamydial infections in feral pigeons in Europe: Review of data and focus on public health implications. Vet Microbiol, 2009; 135:54–67.

14.

Martínez

, Mateo

, Churruca

, et al. Detection of cdtA, cdtB, and cdtC genes in Campylobacter jejuni by multiplex PCR. Int J Med Microbiol, 2006; 296:45–48.

15.

Morabito

, Dell'Omo

, Agrimi

, et al. Detection and characterization of Shiga toxin-producing Escherichia coli in feral pigeons. Vet Microbiol, 2001; 82:275–283.

16.

Persson

, Olsen

. Multiplex PCR for identification of Campylobacter coli and Campylobacter jejuni from pure cultures and directly on stool samples. J Med Microbiol, 2005; 54:1043–1047.

17.

Pickett

, Pesci

, Cottle

, et al. Prevalence of cytolethal distending toxin production in Campylobacter jejuni and relatedness of Campylobacter sp. cdtB gene. Infect Immun, 1996; 64:2070–2078.

18.

Rose

, Nagel

, Haag-Wackernagel

. Spatio-temporal use of the urban habitat by feral pigeons (Columba livia). Behav Ecol Sociobiol, 2006; 60:242–254.

19.

Tanaka

, Miyazawa

, Watarai

, et al. Bacteriological survey of feces from feral pigeons in Japan. J Vet Med Sci, 2005; 67:951–953.

20.

Vázquez

, Esperón

, Neves

, et al. Screening for several potential pathogens in feral pigeons (Columba livia) in Madrid. Acta Vet Scand, 2010; 52:45.

21.

Vučemilo

, Vlahović

, Dovč

, et al. Prevalence of Campylobacter jejuni, Salmonella typhimurium, and avian Paramyxovirus type 1 (PMV-1) in pigeons from different regions in Croatia. Z Jagdwiss, 2003; 49:303–313.

22.

Wang

, Clark

, Rodgers

. Detection in Escherichia coli of the genes encoding the major virulence factors, the genes defining the O157:H7 serotype, and components of the type 2 Shiga toxin family by multiplex PCR. J Clin Microbiol, 2002; 40:3613–3619.

23.

Wani

, Samanta

, Bhat

, et al. Investigation of shiga toxin-producing Escherichia coli in avian species in India. Lett Appl Microbiol, 2004; 39:389–394.