Detection of Salmonella Reservoirs in Birds of Prey Hosted in an Italian Wildlife Centre: Molecular and Antimicrobial Resistance Characterisation

In the European Union, salmonellosis is one of the most important zoonoses reported. Poultry meat and egg products are the most common food matrices associated with Salmonella presence. Moreover, wild and domestic animals could represent an important reservoir that could favour the direct and indirect transmission of pathogens to humans. Salmonella spp. can infect carnivorous or omnivorous wild birds that regularly ingest food and water exposed to faecal contamination. Birds kept in captivity can act as reservoirs of Salmonella spp. following ingestion of infected prey or feed. In this paper, we describe the isolation of different Salmonella serovars in several species of raptors hosted in aviaries in an Italian wildlife centre and in the raw chicken necks used as their feed but intended for human consumption. Characterisations of strains were carried out by integrating classical methods and whole genome sequencing analysis. The strains of S. bredeney isolated in poultry meat and birds belonged to the same cluster, with some of them being multidrug-resistant (MDR) and carrying the Col(pHAD28) plasmid-borne qnrB19 (fluoro)quinolone resistance gene, thus confirming the source of infection. Differently, the S. infantis found in feed and raptors were all MDR, carried a plasmid of emerging S. infantis (pESI)-like plasmid and belonged to different clusters, possibly suggesting a long-lasting infection or the presence of additional undetected sources. Due to the high risk of fuelling a reservoir of human pathogens, the control and treatment of feed for captive species are crucial.


Introduction
Wild birds, like free-living fauna in general, can serve as reservoirs for numerous pathogen species, including Salmonella spp.and other Enterobacteriaceae, and favour their direct and indirect transmission to other animal species, including humans [1,2].Salmonella spp.and other enterobacteria can become pathogens for wild birds that, in most cases, are healthy carriers and can disseminate pathogens into the environment through contamination of water and feed [1,3].Raptors are at the top of the food chain, and they are considered particularly indicative for monitoring pathogens in the environment [4].In nature, there are more than 500 species of birds of prey, nocturnal and diurnal, found in almost every Microorganisms 2024, 12, 1169 2 of 15 type of habitat; they become infected mainly through predation [4,5].Salmonella spp.often infects wild birds that follow a carnivorous or omnivorous diet, particularly those that feed on the ground or regularly ingest food and water exposed to faecal contamination.For this reason, even birds of prey kept in captivity can act as temporary or permanent reservoirs of Salmonella spp.following ingestion of infected prey or feed [6].Due to their zoonotic potential, these bird species kept in captivity should be monitored to preserve human health, considering humans and animals may share the same sources of exposure to Salmonella spp.[7].At the communitarian and European level, there is no mandatory systematic monitoring of Salmonella spp. in wild species, and the data available belong almost exclusively to research contexts [8].
Salmonella genus is divided into two main species: S. enterica and S. bongori.S. enterica includes 6 subspecies and approximately 2600 known serotypes [9][10][11].Members of the subspecies S. enterica subsp.enterica are primarily responsible for disease in humans (approximately 99% of all cases of salmonellosis), other mammals and birds [12][13][14].Human infections caused by S. enterica subsp.enterica generally result in self-limiting gastroenteritis and do not require antibiotic therapy.Serotypes belonging to this subspecies are present in the intestinal tract of a wide variety of domestic and wild reservoir animals.They can reach human hosts through direct contact or through the consumption of contaminated foods, especially when prepared without respecting basic hygiene guidelines [15].The non-enterica subspecies of S. enterica are more closely related to cold-blooded animals, and their pathogenicity is rather limited.In fact, most human infections from non-enterica subspecies (as S. enterica subsp.diarizonae, S. enterica subsp.arizonae, S. enterica subsp.houtenae and S. enterica subsp.salamae) concern subjects with previous pathologies, immunosuppressed subjects or children; therefore, these Salmonella spp.should be considered as opportunistic pathogens.However, non-enterica subspecies of S. enterica, including S. enterica subsp.diarizonae, have also been isolated in warm-blooded animals, both domestic and wild, such as cattle, pigs, poultry and sheep [14].In 2022, salmonellosis was the second most reported zoonosis in the EU with 65 ′ 208 total human cases (3 ′ 302 In Italy), as well as the most frequent cause of food-borne outbreaks in the EU [16].The serotypes of Salmonella spp.most isolated from human samples were S. enteritidis (54.6%), S. typhimurium (12.1%), monophasic S. typhimurium (1,4,[5],12:i:-) (10.4%), S. infantis (2.3%) and S. newport (1.1%) [16].
Fresh poultry meat and meat products made from poultry meat are the matrices with the highest Salmonella prevalence in EU, while eggs and egg products represent the first matrices most associated with foodborne salmonellosis [16].This trend, confirmed over the years, could have risen by a series of factors, such as the intensive breeding and growth conditions of poultry, the contamination of carcasses during slaughter and hygiene faults in the food preparation processes [17].In addition, the growing and persistent presence of certain Salmonella serovars such as S. infantis has been demonstrated [16].A CTX-M-1 Extended-Spectrum Beta-Lactamase (ESBL)-producing multidrug-resistant (MDR) S. infantis clone carrying the plasmid of emerging S. infantis (pESI) has become widely spread along the broiler meat production chain in Italy and in other European countries [18,19].In recent years, MDR was also found to be extremely/very high in other Salmonella serovars detected in poultry sources, such as S. bredeney [18].
The aim of this study was to investigate the causes of the Salmonella spp.infection found in birds of prey hosted in a wildlife recovery centre, which involved raw chicken meat used as feed, intended for human consumption.Moreover, molecular characterisation, AMR profiles and cluster analysis of Salmonella strains were reported.

Sample Collection
The samplings were performed by the Italian military police Corp "Raggruppamento Carabinieri Biodiversità Reparto di Lucca" on dropping samples of 18 birds of prey of different species, during routine checks to monitor animal health.The animals were kept in special

Sample Collection
The samplings were performed by the Italian military police Corp "Raggruppamento Carabinieri Biodiversità Reparto di Lucca" on dropping samples of 18 birds of prey of different species, during routine checks to monitor animal health.The animals were kept in special aviaries at the CITES Centre (Convention of International Trade in Endangered Species) of the Natural reserve of Montefalcone, in the Municipality of Castelfranco di Sotto (PI).The CITES Centre of Montefalcone hosts wild birds confiscated and kept in judicial custody in the framework of the official activities against illegal trade of endangered animals.
Dropping samples of the first 10 birds of prey (specimens A1, A2, A3, B1, C1, D1, E1, E2, F1, G1) were conferred at the Istituto Zooprofilattico Sperimentale del Lazio e della Toscana "M.Aleandri" (IZSLT), UOT (Territorial Operative Unit) Tuscany North (Pisa).The samples were collected on the 20 July 2021 inside the aviaries that housed these   Following the communication of positive results to the CITES centre, it was asked to carry out microbiological tests on food stored in their warehouse, utilised to feed the birds.Packs of raw chicken necks (Gallus gallus) and frozen chick carcasses (Gallus gallus) delivered by the same farm were collected.In particular, the chicken neck meat was originally intended for human use, as reported on the labelled packages, indicating it as class A (according to Reg.CEE 1538/91) and to be consumed after cooking.Therefore, on 26 August 2021, the CITES centre staff sampled two packs of chicken necks from the same batch (samples CN1, CN1a) and three chick carcasses (samples CC1, CC2, CC3), all stored frozen in the centre's warehouse.On 7 and 28 September 2021, the CITES centre of Monfalcone, together with the local health authority, collected an additional two packs of frozen chicken necks (CN2, CN3), from the same batch of CN1 and CN1a.The samples were sent to the IZSLT, UOT Tuscany North (Pisa) for microbiological analyses.

Microbiological Identification
Salmonella spp.identification and isolation were carried out at IZSLT laboratories of Pisa.Droppings and chicken neck samples were tested by cultural examination according to the ISO 6579-1:2017.The frozen chick carcasses were analysed as a whole, and cultural analyses were performed according to "OIE Manual for terrestrial animals 2018" (Chapter 3.9.8par A, B, 2016; Chapter 3.3.11A, B, 2018) [20].The confirmations of characteristic colonies were carried out with biochemical micromethod API 20E (Biomerieux, Paris, France).All available strains were sent to the Enteropathogenic Bacteria Regional Reference Centre (CREP) laboratory at Food Microbiology Unit (IZLST) to proceed with serotyping and molecular analysis.Serotyping was performed according to ISO/TR 6579-3:2014 by seroagglutination using antiserum (Sifin Diagnostics GmbH, Berlin, Germany; SSI Diagnostica A/S, Hillerød, Denmark; Bio-Rad, CA, USA).

Antimicrobial Susceptibility Testing (AST)
AST was performed for all the obtained Salmonella spp.isolates at the National Reference Laboratory for Antimicrobial Resistance (NRL-AR), Department of General Diagnostics (IZSLT), through minimum inhibitory concentration (MIC) determination by broth microdilution, using the EU consensus 96-well microtiter plates (Trek Diagnostic Systems, Westlake, OH, USA).Dilution ranges and interpretation of MIC values were performed as reported in the EU Decision 2020/1729/EU (https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32020D1729,accessed on 19 March 2024) and in the EFSA manual published in 2021 [21], also according to epidemiological cut-offs and clinical breakpoints (when available) of the European Committee on Antimicrobial Susceptibility Testing (EUCAST; http://www.eucast.org,accessed on 19 March 2024).The following drugs were tested: amikacin, ampicillin, cefotaxime, ceftazidime, meropenem, azithromycin, chloramphenicol, nalidixic acid, ciprofloxacin, colistin, gentamicin, sulfamethoxazole, tetracycline, tigecycline and trimethoprim.Escherichia coli ATCC 25922 was used as a quality control strain.

Whole-Genome Sequencing and In Silico Analysis
Whole-genome sequencing analysis were performed on the available strains of S. bredeney and S. infantis isolated from birds of prey and their feed (frozen chicken necks).Genomic DNA was extracted with Bacterial DNA Extraction kit on the automatic extraction system, MagPurix ® (Zinexts Life Science Corp., New Taipei City, Taiwan).Libraries were prepared using Illumina DNA Prep and pair-end (2 × 250 bp) run with a MiSeq sequencer (Illumina Inc., San Diego, CA, USA).Raw reads are stored in the Sequence Read Archive (SRA) at the GenBank database (NCBI) under the BioProject PRJNA1098632, BioSamples from SAMN40917603 to SAMN40917615.The raw reads were processed, and strains were characterised as described in De Bene and colleagues [22].Minimum spanning trees of cgMLST profiles were visualised using the MSTreeV2 algorithm in the GrapeTree (v 1.5.0)software [23].The genetic basis of antimicrobial resistance (AMR) and the presence of plasmid replicons were determined with Staramr (v 0.9.1) [24] based on ResFinder (v 0.8.0.dev2) and PlasmidFinder (v 0.8.0.dev2) databases, respectively [25].The presence of the specific markers of the pESI in the assemblies was determined in silico with the ABRicate tool (v 1.0.1)against a local database [19].The coverage of the whole megaplasmid was determined by mapping the raw-reads against an Italian pESI sequence reference (NZ_OW849779.1)[26].The similarity of the sequences was compared using blast (v 2.10.1+)[27] and represented using BRIG (v 0.95) [28].
The analyses of food samples collected on 26 August 2021 resulted in only one sample of chicken neck (CN1) testing positive for Salmonella spp.and all the others negative (CN1a, CC1, CC2, CC3).The isolated colonies confirmed as Salmonella spp.from the positive sample were serotyped and identified as two serotypes: S. bredeney and S. infantis.The two samples collected on 7 and 28 September 2021 (respectively, CN2 and CN3) were both positive to Salmonella spp.The serotyping analysis on strains from several colonies identified three serotypes: S. bredeney and S. infantis in both samples (CN2, CN3), and S. give

AST
Results of AST were obtained for all the 16 Salmonella spp.isolates.In detail, the AMR phenotypes of S. infantis (n = 5) and S. bredeney (n = 8) are reported in Table 3.All 13 isolates were resistant to at least two antimicrobial classes (tetracyclines and sulphonamides) and most of them (9/13 isolates including four S. bredeney and five S. infantis) were also MDR, being resistant to at least three antimicrobial classes (fluoroquinolones, tetracyclines and sulphonamides).Most of the MDR S. infantis were microbiologically (and clinically for betalactams and ciprofloxacin) resistant to five (beta-lactams, fluroquinolones, trimethoprim, sulphonamides, tetracyclines, n = 2 isolates) or six (beta-lactams, phenicols, fluroquinolones, trimethoprim, sulphonamides, tetracyclines, n = 2 isolates) antimicrobial classes.Of them, two S. infantis, both isolated from birds of prey (A1 and D2), were also Extended Spectrum Cephalosporin Resistant (ESC-R), displaying MIC values of 4 mg/L and >4 mg/L for ceftazidime and cefotaxime, respectively.Differently, the two S. enterica subsp.diarizonae IIIb and the one S. give isolates resulted fully susceptible.

Whole-Genome Sequencing and In Silico Analysis
The 13 isolates identified as S. infantis and S. bredeney were successfully sequenced (isolate ID reported in Table 3).The strains were in silico serotyped and multilocus sequence typed (MLST).The strains previously identified as S. bredeney found in raptors A2, A3, B1, D1, C1, and chicken necks CN1, CN2 and CN3, resulted in having in silico predicted profile S. bredeney, 4:l,v:1,7, and all belonging to the ST897.The strains identified as S. infantis found in raptors A1 and D2, and in chicken neck CN1, CN2 and CN3, resulted in having in silico predicted profile S. infantis, 7:r:1,5, and belonging to the ST32, the most common ST for this serovar (Enterobase).The analysis of cgMLST (Figure 2) showed that the isolates of S. bredeney formed a unique cluster cgST 201416 with isolates from birds of prey (A2, A3, B1, D1 and C1) and chicken necks (C1, CN2 and CN3).The pairwise allelic distances were from 0 to 4. The same analysis for the isolates identified as S. infantis produced three clusters: cgST 145963 (isolated from droppings of A1 and D2), the cluster cgST 93308 (CN2 and CN3) and cgST 254768 (CN1).The phenotypic resistance patterns were confirmed by the presence of the corresponding AMR genes in almost all isolates (Table 3).In detail, all sulphonamide-resistant isolates harboured the corresponding sulfamethoxazole resistance genes sul1 (S.Infantis) or sul2 (S.Bredeney), all tetracycline-resistant isolates harboured tetA (S.Infantis) or tetB (S.Bredeney), and the two chloramphenicol-resistant S. Infantis harboured cmlA1.As for beta-lactam resistance in S. Infantis, ampicillin-resistant isolates harboured the blaTEM-1B gene, while the two ESC-R isolates harboured the ESBL gene blaCTX-M1.Fluoroquinolone resistance in all S. Infantis isolates was conferred by chromosomal point mutations in both gyrA (D87G or D87Y) and parC (T57S), while in the four fluoroquinolone-resistant S. Bredeney, the corresponding genetic background included in 3 isolates both the transferable qnrB19 gene and parC mutation (T57S).One fluoroquinolone-resistant S. Bredeney presented only the parC mutation (T57S).The strains of S. Bredeney all belonging to the The phenotypic resistance patterns were confirmed by the presence of the corresponding AMR genes in almost all isolates (Table 3).In detail, all sulphonamide-resistant isolates harboured the corresponding sulfamethoxazole resistance genes sul1 (S. infantis) or sul2 (S.Bredeney), all tetracycline-resistant isolates harboured tetA (S. infantis) or tetB (S. bredeney), and the two chloramphenicol-resistant S. infantis harboured cmlA1.As for beta-lactam resistance in S. infantis, ampicillin-resistant isolates harboured the bla TEM-1B gene, while the two ESC-R isolates harboured the ESBL gene bla CTX-M1 .Fluoroquinolone resistance in all S. infantis isolates was conferred by chromosomal point mutations in both gyrA (D87G or D87Y) and parC (T57S), while in the four fluoroquinolone-resistant S. bredeney, the corresponding genetic background included in 3 isolates both the transferable qnrB19 gene and parC mutation (T57S).One fluoroquinolone-resistant S. bredeney presented only the parC mutation (T57S).The strains of S. bredeney all belonging to the same cluster, shared almost the same resistance genes pattern and the three isolates positive for qnrB19, co-harboured on the same contig, the Col(pHAD28) plasmid replicon.For S. infantis, the resistance genes pattern was shared within strains belonging to the same cgMLST, and all were positive for the IncFIB (pN55391) plasmid replicon.One of the cgST 145963 strains was also positive for the Col(pHAD28) plasmid replicon.The schematic results of resistance genes and plasmids found are reported in Table 3.

pESI-like Megaplasmid Presence in S. Infantis Isolates
The analysis in silico of the markers of pESI in S. infantis assemblies (Figure 3) detected the presence of the backbone of pESI, the truncated oriV from IncP (replicate origin associated a IncP plasmids; AM261769), the qacE∆ gene (encoding resistence to quaternary ammonium compounds) and the fim gene (encoding a fimbria protein) in all 5 isolates.The K88 gene (a gene included in a chaperon-usher fimbria operon) was found only in three of the isolates.After mapping the quality-trimmed reads of the WGS of the S. infantis isolates against CTX-M-1 (NZ_OW849779.1),the coverage of the plasmid sequence found was 99.19% (A1), 95.76% (CN1), 98.51% (CN2), 98.50% (CN3) and 99.64% (D2).
same cluster, shared almost the same resistance genes pattern and the three isolates positive for qnrB19, co-harboured on the same contig, the Col(pHAD28) plasmid replicon.For S. Infantis, the resistance genes pattern was shared within strains belonging to the same cgMLST, and all were positive for the IncFIB (pN55391) plasmid replicon.One of the cgST 145963 strains was also positive for the Col(pHAD28) plasmid replicon.The schematic results of resistance genes and plasmids found are reported in Table 3.

pESI-like Megaplasmid Presence in S. Infantis Isolates
The analysis in silico of the markers of pESI in S. Infantis assemblies (Figure 3) detected the presence of the backbone of pESI, the truncated oriV from IncP (replicate origin associated a IncP plasmids; AM261769), the qacEΔ gene (encoding resistence to quaternary ammonium compounds) and the fim gene (encoding a fimbria protein) in all 5 isolates.The K88 gene (a gene included in a chaperon-usher fimbria operon) was found only in three of the isolates.After mapping the quality-trimmed reads of the WGS of the S. Infantis isolates against CTX-M-1 (NZ_OW849779.1),the coverage of the plasmid sequence found was 99.19% (A1), 95.76% (CN1), 98.51% (CN2), 98.50% (CN3) and 99.64% (D2).The blast alignment of the assemblies against the reference pESI-CTX-M-1 (NZ_OW849779.1)(Figure 3) evinced that some contigs of the S. infantis genome could be identified as part of the pESI plasmid.Those contigs contained dfrA1, sul1 and tetA in all 5 S. infantis, dfrA14 in all but in CN1, aph(3 ′′ ) only in CN2 and CN3 and bla CTX-M-1 only in A1 and D2.All assemblies also carried resistance genes towards heavy metals (merA) and disinfectants (qacE∆).Although, plasmidFinder identified IncFIB, described as the origin of replication for pESI, it was missing in the blast CN3 alignment.Together, those evidences indicated that all the 5 S. infantis isolates harboured pESI.

Discussion
Our study reports the infection of birds of prey hosted in a wildlife recovery centre with two potentially pathogenic serovars of Salmonella spp.(S. bredeney and S. infantis).Since 2018, the wildlife recovery centre has sent samples regarding the health monitoring of hosted animals to IZSLT.To our knowledge, the described events in the study were the first occurrence of Salmonella spp.detection.The infection of S. bredeney in the five raptors was assessed as associated and genetically linked to the raw chicken necks used to feed the individuals.The cluster analysis of the core genome confirmed the chicken necks as the source of this contamination.The S. infantis infection of three birds of prey was reconducted genetically to a common source.However, the source was not identified in the samples of raw chicken necks collected in the warehouse of CITES centre, since no genetic correlation was found between strains isolated from raptors and chicken necks used as feed.The serovar S. infantis was strictly related to broiler sources (95.6%) according to the latest EU One Health 2022 Zoonoses Report [16], leading to the hypothesis that also for these raptors, the source of contamination may likely have been raw chicken meat, probably used as feed.
In the literature, there are no studies reporting bacterial contamination specifically in chicken neck meat, while there are several ones that focus on bacteriological analyses, including the search of Salmonella spp., on chicken neck skin.This matrix is particularly exposed to great faecal contamination during the slaughter process, where the animals are positioned upside down, favouring the flow of the washing liquid towards the neck [29].In the cited studies, the Salmonella spp.recorded wide prevalence ranges (0-70%) in the sampled slaughterhouses [15,17,30].Regarding the analysis of serotypes, S. infantis and S. bredeney were often isolated in these studies.Moreover, in Italy, S. infantis was the most isolated serotype within broiler farms (50.6%) and was frequently found in humans, representing a serotype of public health concern [31].Although much less frequently, S. bredeney and S. give can also cause outbreaks of human salmonellosis, mainly causing gastrointestinal symptoms [32,33].S. enterica subsp.diarizonae, being part of the nonenterica subspecies, is usually considered an opportunistic pathogen; however, it has been associated with cases of gastroenteritis, especially in children.Beside S. enterica subsp.diarizonae, other non-enterica subspecies (S. enterica subsp.arizonae, S. enterica subsp.houtenae and S. enterica subsp.salamae) were associated with human diseases in the last 20 years [14,34].Over the last 20 years, some studies have been carried out on birds of prey present in wildlife recovery centres located in Spain: prevalence rates of infection with Salmonella spp.ranged around 5-10%.The most identified serotypes were S. bredeney, S. enteriditis, S. typhimurium and S. havana [4,35,36].Also, in a study carried out in Southern Italy on carcasses of wild birds of prey, a similar prevalence of infection with Salmonella spp. was observed (6.8%), in particular, S. salamae, S. napoli and S. typhimurium [37].
Captivity plays a fundamental role in influencing the oral and intestinal microbiome of hosted birds of prey and is also associated with high rates of antibiotic resistance, compared to free-living birds [38].This change can already occur after a month of direct contact between animals and humans, and the diet, especially if based on raw food, represents the first determining causal factor [39][40][41].Some studies have shown that birds of prey fed poultry meat develop a wider range of Gram-negative bacterial flora [38]; in particular, a study on falcons has shown that the diets most commonly fed to these birds increase the levels of Salmonella in the intestinal flora [42].
To avoid the transmission of Salmonella and other bacteria from hosted birds to humans, it is essential to train animal care personnel on the potential zoonotic risks of these pathogens, through appropriate adoption of adequate hygiene and personal protection measures.Increasing biosecurity is certainly the most effective method to prevent outbreaks of infectious diseases within captivity facilities [38].
Three serovars of potentially pathogenic Salmonella were found (S. bredeney and S. infantis in samples of chicken neks and raptor stools, and S. give, only in chicken necks).In the literature, phylogenetic analyses of S. bredeney partitioned the serovar in two ST (ST241 and ST897) [43].In particular, the ST897 is associated also with the serovars S. kimuenza, an infrequent serotype found in humans, livestock and poultry [44][45][46].Furthermore, the strains of S. infantis all belonged to ST32, considered the dominant MLST type of this serovar [47].A recent study based on genomes deposited in a public database revealed that, in Europe, this ST represents more than 97% of strains and 99% worldwide [47].
For S. infantis and S. bredeney, it is noteworthy that most of the isolates (9/13) were MDR.In particular, 4/8 S. bredeney were MDR, showing resistance to tetracyclines, sulphonamides and fluoroquinolones.In three isolates, resistance to (fluoro)quinolones, a Highest Priority Critically Important antimicrobial (HPCIA) [48] class, was conferred by the presence of qnrB19 located in Col(pHAD28) plasmids, as previously observed for other Salmonella serovars [49].The same isolates showed the concomitant presence of the T57S parC mutation.Although the role of the ParC T57S substitution conferring (fluoro)quinolone resistance is still controversial [50], we detected in one fluoroquinoloneresistant isolate (CN2 with MIC values of 0,12 and 16 mg/L for ciprofloxacin and nalidixic acid, respectively) only the T57S parC mutation.As for the five S. infantis isolates, all were MDR showing microbiological resistance to five or six antimicrobial classes.Two of them, A1 and D2, were also ESBL-producers (CTX-M-1 type), thus including resistance to two antimicrobial classes (3rd and 4th generation cephalosporins and fluoroquinolones) classified as HPCIAs [48].The increasing prevalence of pESI(like)-positive, MDR S. infantis in Europe is of major concern.In all the S. infantis assemblies, pESI markers have been identified, and the presence of an elevated proportion of the plasmid sequence has been assessed by mapping of the raw reads.This suggests the presence of this megaplasmid in S. infantis isolated from chicken meat and in stool from birds of prey.Moreover, the results of the molecular characterisation of the isolated pointed out that the resistance genes, including bla CTX-M-1, in the two CTX-M-producing S. infantis, were located in the pESI megaplasmid.The importance of this particular plasmid is that it is a mosaic plasmid of around 300 kbp with an elevated capacity to acquire AMR, virulence, fitness genes and toxin/antitoxin systems that enhance its persistence in the S. infantis host [19,26].Indeed, as observed in the S. infantis herein described, in the CTX-M-1 producing S. infantis clone circulating in Italy and other European countries, this megaplasmid typically carries together with the ESBL gene bla CTX-M-1 , also tet(A), sul1, dfrA1 or dfrA14 and, in some cases, aadA1 (conferring resistance to tetracycline, sulfamethoxazole, trimethoprim and streptomycin, respectively), as well as resistance genes towards heavy metals (merA) and disinfectants (qacE∆) [19].

Figure 1 .
Figure 1.Timetable of sampling activities linked to the investigations carried out.Figure 1. Timetable of sampling activities linked to the investigations carried out.

Figure 1 .
Figure 1.Timetable of sampling activities linked to the investigations carried out.Figure 1. Timetable of sampling activities linked to the investigations carried out.

Figure 2 .
Figure 2. The minimum spanning tree (MST) of the cgMLST of S. Bredeney isolates (A) and S. Infantis isolates (B).The numbers on the branches represent the allelic distances (amount of different alleles) between isolates.

Figure 2 .
Figure 2. The minimum spanning tree (MST) of the cgMLST of S. bredeney isolates (A) and S. infantis isolates (B).The numbers on the branches represent the allelic distances (amount of different alleles) between isolates.

Figure 3 . 3 .
Figure 3. Graphical representation of the similarity of some contigs of the S. Infantis assemblies isolate from bald eagle (A1, in fuchsia), chicken neck (CN1, CN2, CN3) and Crested caracara (D2; in Figure 3. Graphical representation of the similarity of some contigs of the S. infantis assemblies isolate from bald eagle (A1, in fuchsia), chicken neck (CN1, CN2, CN3) and Crested caracara (D2; in yellow), when compared with the reference pESI plasmid (NZ_OW849779.1).Annotation of the genes is based on the sequence of NZ_OW849779.1.

Table 1 .
Table1for raptor specimens and Table2for food samples.Description of sampling scheme of collected droppings of birds of prey.For each sampling date, the results of Salmonella spp.detection are reported [(−) negative, (+) positive, (/) if not sampled], and the serovar identified (n/a if not applicable).

Table 2 .
Description of sampling scheme of the feed and food matrix collected (chicken necks and chicks).For each sampling date, the results of Salmonella spp.detection are reported [(−) negative, (+) positive, (/) if not sampled], and the serovar identified (n/a if not applicable).

Table 3 .
Summary of mainly strains features, presence (+, highlighted in light grey) or absence (−) of resistance genes and plasmid replicons for S. bredeney and S. infantis isolates.