Occurrence of Salmonella Typhimurium and its monophasic variant (4, [5],12:i:-) in healthy and clinically ill pigs in northern Italy

Background The serovar Typhimurium (4, [5],12:i:1,2), is the most frequently isolated serovar in case of salmonellosis in pigs in Europe and its monophasic variant (4, [5],12:i:-) has been increasingly responsible for Salmonella outbreaks in humans. A total of 25,215 samples were collected, during the years 2002–2017, from 1359 pig farms located in Northern Italy. Samples were collected from different material sources including fecal samples, rectal swabs, gut content and different organs. Results Salmonella was isolated in 15.80% of samples and, among the isolates, 733 were typed as Salmonella Typhimurium (ST) or its monophasic variant (MST). Over time, there was an increase of isolation of MST which outnumbered ST. Most of the strains were isolated in animals during the weaning stage and the growing – fattening period whereas the clinical cases were mainly present in young pigs after weaning. Conclusions This study confirms the presence of ST and MST in pig farms although, considering the total of isolated serotypes, with lower percentages than previously reported. In the last few years, ST has increasingly been replaced by MST suggesting that MST has a competitive advantage over ST, probably due to its different antigenicity and pathogenicity which renders the infection stealthier to recognize and control.

The monophasic variant of Salmonella Typhimurium (4, [5],12:i:-) has been increasingly responsible for Salmonella outbreaks in humans, being the third (after the serovars Salmonella Enteritidis and Salmonella Typhimurium) most commonly reported serovar in the EU in 2012 [6], and frequently reported across the world [7,8]. This serovar, strongly associated with swine food chain, especially in Europe [9], was rarely identified before the mid-1990s but its isolation in both animals and humans, has increased in the last 20 years [4,[10][11][12].
Salmonella prevalence varies widely among farms and at different growth stages within the same farm and, due to the high number of factors and the complex relationships among pathogen and host, definitive understanding of the transmission, shedding and carrier states of salmonellae are still difficult [1].
The aim of this study is to describe and evaluate the occurrence, over a 15-year period, of Salmonella Typhimurium (ST) and its monophasic variant Salmonella enterica subsp. enterica 4, [5],12:i:-(MST) in pigs and its association with clinical conditions.

Strains isolation
A total of 25,215 samples were collected, during the years 2002-2017, from 1359 pig farms located in Northern Italy where clinical enteric forms or cases of on-farm mortality occurred. In particular, in each farm around 20 samples per year were collected either from living animals (fecal samples or rectal swabs) or from carcasses (gut content and different organs like spleen, liver and lymph nodes collected at necropsy) and sent to our lab by farm vets.
The isolation and identification of Salmonella isolates were carried out always by the same lab, initially in accordance with

Strain serotyping
The complete serological characterization of Salmonella was performed by slide agglutination for the determination of somatic antigens, while, for the determination of flagellar antigens, the method of tube agglutination was followed according to the Spicer [13] technique, modified by Morris et al. [14].
In particular, the characterization of monophasic variant of Salmonella Typhimurium, was performed through two consecutive phase inversions by passaging through a U-shaped glass tube containing semisolid agar with H:i antiserum. The isolates that still did not display the second phase after the first and the second passage were considered, phenotypically, monophasic.

Phage-typing
Phage-typing was performed at the Italian National Reference Centre for Animal Salmonellosis according to Anderson et al. [15]. Typing of MST strains started in 2011 and for this reason only 225 isolates were typed. The total number of typed ST strains was 235.

Clinical case definition
We considered a "clinical case" (C) as an illness of variable severity manifested by enteric signs with presence of Salmonella and in absence of isolation of other enteric pathogens. When Salmonella isolation occurred and no enteric signs were shown the condition was referred as a "non-clinical case" (NC).

Statistical methods
Statistical analysis was performed using GraphPad 6.0 for MAC OS X (GraphPad Software Inc.; San Diego; CA) and SAS (SAS Institute, Inc., Cary, North Carolina) softwares. Differences in proportions were estimated using Fisher's exact test. Differences in ST and MST prevalence among years were evaluated through Chisquared test. A P value less than 0.05 was considered statistically significant while a value between 0.05 and 0.1 will be defined as trend.

Serotyping results
Salmonella was isolated from 3983 out of 25,215 fecal samples (15.80%). Among the isolates, 246 were typed as Salmonella Typhimurium and 487 as its monophasic variant. Two hundred eighty-five out of 1359 farms (20.97% of the total) resulted positive for the serovars investigated in this study. The distribution of the isolates, during the considered period, is showed in Table 1 and Fig. 1 Five hundred seventeen out of 733 collected isolates came from pigs whose different ages were known: amongst them, most of the MST and ST strains were isolated in animals during the weaning (since 30 days of age till 25/30 kg weigh) stage (n = 311; 60.15%) and the growing (25/30 kg -60 kg weigh)fattening (60 kg weigh to slaughtering) period (n = 177; 34.24%) whereas the number of isolates from breeders (n = 11; 2.13%) as well as from suckling piglets (n = 18; 3.48%) was low. Similar patterns of distribution were observed considering ST and MST separately (Fig. 2).
Clinical signs were associated with 114 out of 246 (46.34%) isolates of ST and 184 out of 487 (37.78%) isolates of MST (Table 2). Although it seems that the association of MST to clinical signs is lower than in ST, the difference only approached the statistical significance (P = 0.08). Figure 3 shows the percentage of strains (on the total of MST and ST strains) associated with clinical illness referred to the single production stage. Most of the clinical cases were present in young pigs after weaning, while in fatteners and breeders the occurrence of clinical signs were lower. When infected with MST, clinical signs were present in 45.17% (103/228) and 25.95% (34/ 131) of pigs, in weaners and growers-fatteners respectively, while in case of isolation of ST, clinical signs were present in 53.01% (44/83) and 43.48% (20/46) of pigs, respectively.

Phage typing results
The isolates belonged to 23 different phage types (Table 3), and eight of them were common to both serovars. The most frequently isolated phage types were DT193 (62 isolates) and DT120 (47 isolates), followed by U311 (45 isolates) and U302 (29 isolates). Other types found were DT104 (18 isolates), DT20A (16 isolates) and DT208 (11 isolates). DT193 was the most frequent type among MST strains while U302 was the most common among ST strains. Forty-nine isolates showed a pattern which did not conform (RDNC) to any defined pattern, and 149 could not be phage typed (NT).

Discussion
The results presented here were obtained from farms, located in Northern Italy in a high-density pig population area, and checked when clinical enteric forms or cases of on-farm mortality occurred. We focused on ST and MST, considering the prominent role of these two serovars in the pig population [4,12].
ST and MST represented 12.23% and 6.18, respectively of the Salmonella serovars isolated, low percentages compared to those from other reports. Indeed, recent studies suggested that ST and MST represent between 40 and 50% of the Italian isolates, with MST increasing from 9.66 to 46.34% in the last 10 years [16,17]. A similar increasing has also been reported in other countries [12]. Although it may seem that, in this study, a lower prevalence than those available in the literature to date were recorded, a comparison can not actually be made because in this study we used a clinical sampling with the presence of the enteric clinical form in the herd as inclusion criterion. Therefore, the fact that our sampling was not randomly performed, but following clinical criteria, does not make possible a comparison with other studies. The distribution of the studied serovars, during the considered period, highlighted a predominance of ST on MST in the first period and, since 2008, a reversion of this tendency with MST becoming more predominant over the following years. Indeed after 2009, MST subsequent recorded peaks were likely related to cyclic outbreaks and to the hypothesized pigs' role of reservoir for this serovar [18]. The data reported here are in accordance with recent reports where the increasing prevalence of MST is well documented, in particular in the United Kingdom, Poland and Malta [6,19,20]. In the United Kingdom, MST represented 60.7% of the Salmonella isolates obtained from a surveillance program in pigs in 2015 [21].
It is conceivable to hypothesize that MST has a selective advantage over ST. It was suggested that a number of factors (i.e. involvement of prophages and antigenic changes) can cause a reduced immune response to MST in herds when compared to ST [4]. More recently, a comparative whole-genome sequencing and   phylogenomic analysis of MST isolates from the United Kingdom and Italy during the period 2005-2012, revealed a high level of microevolution that may affect antigenicity, pathogenicity, and transmission [22]. Although the increase of the number of the MST isolates coincides with the revision of the isolation method, in order to rule out any influence of the testing method upon the results, it is important to highlight that the analyses were performed according to ISO methods: the change of the enrichment media can not affect the suitability of detecting any of the known motile Salmonella serovars [23]. The replacements of two enrichment broth media with a semi-solid medium leads, on the other hand, to a better selection on the background flora able to grow in the broths and potentially able to mask the isolation of Salmonella strains [23].
When considering the production stage, both ST and MST showed their highest presence in the weaning and growing period as reported previously [24,25]. A comparison between the prevalence of ST and MST in different production stages showed no significant differences. Overall, these findings suggest a higher level of susceptibility in younger pigs, irrespective of the serovars involved. Pigs can become infected at any production stage but the decline of maternal antibodies after weaning makes younger pigs more susceptible to the infection [26].
When considering the association between clinical signs and isolates, we observed that clinical signs were associated more to ST than to MST and that most of clinical cases were present in young pigs after weaning. These data, although only approaching significance, are supportive of a competitive advantage of MST over ST.
The phage-typing highlighted the prevalence of four types representing about 70% of the typed isolates (DT193, DT120, U311, U302) and this has been a common feature of European isolates for the last 20 years [11,27,28]. DT193 has to be regarded as an important phage type also for ST, considering its increase in Europe in the last years [29] and its role in human cases of salmonellosis.

Conclusions
In conclusion, this study confirms the presence of ST and MST in pig farms although, considering the total of isolated serotypes, with lower percentages than previously reported.
In the last few years, ST has increasingly been replaced by MST suggesting that MST has a competitive advantage over ST, probably due to its different antigenicity and pathogeneicity which renders the infection stealthier to recognize and control. More detailed studies should be undertaken to assess the mechanisms underpinning the competitive advantage of MST over ST in pigs.