Geographical variation in antimicrobial resistant Salmonella Schwarzengrund from chicken meat in Japan

Abstract

Chicken meat is a major source of foodborne salmonellosis. In Japan, fluoroquinolones and third-generation cephalosporins are the first- and second-choice treatments for Salmonella gastroenteritis, respectively. We investigated the prevalence and antimicrobial resistance of Salmonella in 154 chicken meat products from Hokkaido (42), Tohoku (45), Kanto (5), and Kyushu (62), Japan. Salmonella was isolated from 133 products (86.4%). High resistance rates were observed for streptomycin (56.5%), tetracycline (50.7%), and kanamycin (47.8%), while all isolates were susceptible to cefazolin, cefotaxime, gentamicin, ciprofloxacin, colistin, and chloramphenicol. The most common serovar, Salmonella Schwarzengrund (83.3%), showed clear regional differences in multidrug resistance: 100% in Kyushu, 41.5% in Tohoku, and 0% in Hokkaido. These findings highlight significant geographical variation in antimicrobial resistance among Salmonella Schwarzengrund isolates.

Salmonella spp. are bacterial pathogens that cause foodborne gastroenteritis. According to the World Health Organization, non-typhoidal Salmonella is a global causative agent of diarrhea [24]. Although human non-typhoidal salmonellosis typically causes acute self-limiting enteritis, and antimicrobial therapy is not usually recommended, antimicrobial therapy may be lifesaving in patients with severe symptoms and those who are in health risk groups, such as infants, older individuals, and immunocompromised patients [7, 24]. Chicken meat is the major cause of foodborne salmonellosis and campylobacteriosis worldwide [23, 24]. Numerous studies have investigated the prevalence and antimicrobial resistance of Salmonella in chicken meat in Japan [12, 18, 19, 21, 22]. Surveys conducted in these studies indicate a high prevalence of Salmonella in chicken meat and a wide range of their antimicrobial resistance profiles. This variability in resistance is thought to result from differences in the types and frequency of antimicrobial agents used in chicken farms. According to the Japan Ministry of Agriculture, Forestry and Fisheries (JMAFF), western Japan has a higher number of chicken farms reporting outbreaks of colibacillosis or staphylococcal infections compared to eastern Japan (https://www.maff.go.jp/j/syouan/douei/kansi_densen/kanren_zyouhou.html) (in Japanese). Therefore, compared to poultry farms in eastern Japan, those in western Japan may have a higher frequency of antimicrobial usage and a greater variety of antimicrobials employed. Our recent research found that fluoroquinolone resistance in Campylobacter jejuni isolated from chicken meat in western Japan was higher than in those from eastern Japan [20]. If the antimicrobial resistance profiles of Salmonella in chicken show similar regional differences as those of Campylobacter, the risk of infection with antimicrobial-resistant Salmonella may vary based on the region where the chicken was produced.

This study aimed to determine geographical variation in the antimicrobial resistance of Salmonella isolated from chicken meat. The results of this study may help characterize Salmonella isolated from chicken meat and select antimicrobials for treating human patients with Salmonella enteritis.

One hundred fifty-four chilled chicken meat products consisting of three to five vacuum-packed breast pieces were collected from nine retail shops (six in Hokkaido, one in Tohoku, and two in Kanto regions in Japan) and eight chicken processing plants (plant A in Hokkaido, plant J in Tohoku, and six plants (L to Q) in Kyushu regions in Japan) between February 2023 and November 2024 (Table 1). We collected local meat products that were vacuum-packed at chicken processing plants to eliminate the possibility of post-shipment Salmonella cross-contamination. The product label indicates the name and address of the processing plant and the production lot number. The samples were collected from each production lot. The refrigerated products were sent to the Obihiro University of Agriculture and Veterinary Medical Medicine and stored in the laboratory at 4°C until examination. The samples were examined within 48 hr of purchase.

Table 1. Salmonella prevalence in chicken meat products.

Region	Plant	No. of samples	No. of positve samples	Salmonella prevalence (%)	No. of isolates
					Schwarzengrund	Infantis	Manhattan	Untypeable
Hokkaido	A	14	13	92.9	13	0	0	0
	B	8	7	87.5	3	5	0	0
	C	7	6	85.7	5	0	1	0
	D	13	10	76.9	10	0	0	0
Tohoku	E	13	13	100.0	11	3	0	0
	F	2	2	100.0	2	0	0	0
	G	8	8	100.0	8	0	0	0
	H	5	5	100.0	5	0	0	0
	I	8	7	87.5	7	0	0	0
	J	9	7	77.8	8	0	0	0
Kanto	K	5	4	80.0	0	4	0	0
Kyushu	L	13	13	100.0	8	0	6	0
	M	10	4	40.0	4	0	0	0
	N	10	8	80.0	7	0	1	0
	O	9	8	88.9	8	0	0	0
	P	9	9	100.0	8	0	2	0
	Q	11	9	81.8	8	0	0	1
Total		154	133	86.4	115	12	10	1

Each sample consisted of skin (75 g) isolated from more than three pieces of meat and placed in a plastic bag containing 75 mL of buffered peptone water (BPW; Oxoid Ltd., Hampshire, UK). After stomaching, 50 mL of the solution was mixed with 200 mL of BPW and incubated at 37°C for 18 hr for pre-enrichment. After incubation, 0.1 and 1 mL of the culture were added to 10 mL of Rappaport–Vassiliadis broth (Oxoid) and 10 mL of tetrathionate broth (Shimadzu Diagnostic Co., Tokyo, Japan), respectively, and were subsequently incubated at 42°C for 20 hr. After incubation, each culture was streaked onto two selective isolation agar plates: xylose–lysine–deoxycholate agar (Oxoid) and CHROMagar^™Salmonella (CHROMagar, Paris, France) and subsequently incubated at 37°C for 24 hr. A maximum of four suspected Salmonella isolates per sample were biochemically identified as previously described [14]. Isolates were tested for slide agglutination using O antisera (Denka Co., Tokyo, Japan) and tube agglutination using H antisera (Denka). Serovars were identified based on their reaction with O- and H-group antigens, according to the Kauffmann–White scheme [4]. One serovar per sample was suspended in 20% glycerol and stored at −80°C until ready for antimicrobial susceptibility testing.

Following broth microdilution on dried plates (Eiken Chemical, Tokyo, Japan), isolate susceptibility to ampicillin (1–128 mg/L), cefazolin (1–128 mg/L), cefotaxime (0.5–64 mg/L), streptomycin (1–128 mg/L), gentamicin (0.5–64 mg/L), kanamycin (1–128 mg/L), tetracycline (0.5–64 mg/L), nalidixic acid (1–128 mg/L), ciprofloxacin (0.03–4 mg/L), colistin (0.12–16 mg/L), chloramphenicol (1–128 mg/L), and trimethoprim (0.25–16 mg/L) was tested. Escherichia coli ATCC 25922 was used as a quality control strain. The breakpoints for ampicillin (32 mg/L), cefazolin (8 mg/L), cefotaxime (4 mg/L), streptomycin (32 mg/L), gentamicin (16 mg/L), kanamycin (64 mg/L), tetracycline (16 mg/L), nalidixic acid (32 mg/L), ciprofloxacin (1 mg/L), colistin (4 mg/L), chloramphenicol (32 mg/L), and trimethoprim (16 mg/L) were adopted from the Clinical and Laboratory Standards Institute [3] and the Japanese Veterinary Antimicrobial Resistance Monitoring System [15].

All statistical analyses were performed using EZR version 1.68 (Saitama Medical Center, Jichi Medical University, Saitama, Japan) [8], which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics. All P-values were calculated as two-tailed, and a P-value of less than 0.05 was considered statistically significant. Fisher’s exact test was used to evaluate whether there was a statistically significant differences in Salmonella prevalence between broiler flocks shipped to the chicken processing plants at the typical age in Japan and those shipped at 65 days of age. To examine the regional differences in antimicrobial resistance of Salmonella Schwarzengrund (S. Schwarzengrund), we first conducted the Chi-square test or Fisher’s exact test to evaluate whether there were significant differences among three regions (Hokkaido, Tohoku, and Kyushu). If significant differences were observed, post-hoc pairwise comparisons were performed using Bonferroni correction.

In total, 154 products were packaged in 17 plants (A–Q). Among the 17 plants, 4, 6, 1, and 6 were located in the Hokkaido, Tohoku, Kanto, and Kyushu regions, respectively (Table 1). Salmonella was isolated from 133 (86.4%) products packaged in all plants. Salmonella prevalence in products packaged in each of the plants other than plant M was 89.6%. Salmonella prevalence in the products packaged at plant M was 40.0%. All 10 broiler flocks slaughtered at plant M were 65 days old. The average age of the broiler flocks slaughtered at other seven plants (A, J, L, N, O, P, and Q) was 48.2 days (range, 44–51 days), which represented the typical age at which broiler flocks were shipped to the chicken processing plants in Japan. Salmonella prevalence was significantly higher in the seven plants (88.0%) than in plant M (40.0%) (P<0.01). The most frequent serovar isolated in this study was S. Schwarzengrund (115 isolates, 83.3%), followed by S. Infantis (12 isolates, 8.7%), S. Manhattan (10 isolates, 7.2%), and untypeable Salmonella (1 isolate, 0.7%). Two serovars were isolated from the five products. S. Schwarzengrund was isolated from products of plants other than plant K in Kanto. S. Infantis was isolated from Plant B in Hokkaido, Plant E in Tohoku, and Plant K in Kanto, Japan. S. Manhattan was isolated from the products of plant C in Hokkaido and plants L, N and P in Kyushu.

The resistance rates to ampicillin, streptomycin, kanamycin, tetracycline, nalidixic acid, and trimethoprim were 0.7%, 56.5%, 47.8%, 50.7%, 17.4%, and 26.1%, respectively (Table 2). All the isolates were susceptible to cefazolin, cefotaxime, gentamicin, ciprofloxacin, colistin, and chloramphenicol. Among the S. Schwarzengrund isolates, the rates of resistance to streptomycin, kanamycin, and tetracycline were significantly higher (P<0.05) in Kyushu than in Hokkaido or Tohoku isolates. The rates of resistance to nalidixic acid and trimethoprim were significantly higher (P<0.05) in Kyushu than in Hokkaido. Moreover, none of the isolates from Hokkaido were multidrug-resistant, defined as resistance to two or more classes of antimicrobials. The rates of multidrug resistance in isolates from Tohoku and Kyushu were 41.5% (17/41) and 100.0% (43/43), respectively (Table 3). The multidrug resistance rate was significantly (P<0.01) higher in Kyushu than in Hokkaido and Tohoku.

Table 2. Antimicrobial resistance of Salmonella isolated from the chicken meat products.

Serovar		No. of isolates	No. of resistant isolates (%)
			ABPC	CEZ	CTX	SM	GM	KM	TC	NA	CPFX	CL	CP	TMP
Schwarzengrund
	Hokkaido	31	0 (0.0)	0 (0.0)	0 (0.0)	4 (12.9)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
	Tohoku	41	0 (0.0)	0 (0.0)	0 (0.0)	19 (46.3)*	0 (0.0)	25 (61.0)*	15 (36.6)*	4 (9.8)	0 (0.0)	0 (0.0)	0 (0.0)	8 (19.5)
	Kyushu	43	0 (0.0)	0 (0.0)	0 (0.0)	40 (93.0)*,**	0 (0.0)	39 (90.7)*,**	42 (97.7)*,**	17 (39.5)*,**	0 (0.0)	0 (0.0)	0 (0.0)	26 (60.5)*,**
Infantis
	Hokkaido	5	0 (0.0)	0 (0.0)	0 (0.0)	2 (40.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
	Tohoku	3	0 (0.0)	0 (0.0)	0 (0.0)	3 (100.0)	0 (0.0)	0 (0.0)	3 (100.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
	Kanto	4	0 (0.0)	0 (0.0)	0 (0.0)	2 (50.0)	0 (0.0)	2 (50.0)	2 (50.0)	2 (50.0)	0 (0.0)	0 (0.0)	0 (0.0)	2 (50.0)
Manhattan
	Hokkaido	1	0 (0.0)	0 (0.0)	0 (0.0)	1 (100.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
	Kyushu	9	0 (0.0)	0 (0.0)	0 (0.0)	7 (77.8)	0 (0.0)	0 (0.0)	8 (88.9)	1 (11.1)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
Untypeable
	Kyushu	1	1 (100.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
Total		138	1 (0.7)	0 (0.0)	0 (0.0)	78 (56.5)	0 (0.0)	66 (47.8)	70 (50.7)	24 (17.4)	0 (0.0)	0 (0.0)	0 (0.0)	36 (26.1)

ABPC: ampicillin; CEZ: cefazolin; CTX: cefotaxime; SM: streptomycin; GM: gentamicin; KM: kanamycin; TC: tetracycline; NA: nalidixic acid; CPFX: ciprofloxacin; CL: colistin; CP: chloramphenicol; and TMP: trimethoprim. *: statistical difference compared with Hokkaido. **: statistical difference compared with Tohoku.

Table 3. Antimicrobial resistance profiles of Salmonella Schwarzengrund isolated in this study.

Region	Antimicrobial resistance profile	No. of isolates
Hokkaido	SM	4
	Susceptible	27
Tohoku	SM+KM+TC+NA+TMP	1
	SM+KM+TC+NA	1
	SM+KM+TC+TMP	6
	SM+KM+TC	3
	SM+TC	4
	SM+KM	4
	KM+NA	2
	KM	8
	TMP	1
	Susceptible	11
Kyushu	SM+KM+TC+NA+TMP	12
	SM+KM+TC+NA	4
	SM+KM+TC+TMP	11
	KM+TC+NA+TMP	1
	SM+KM+TC	8
	SM+KM+TMP	1
	KM+TC+TMP	1
	SM+TC	4
	KM+TC	1
Total		115

SM: streptomycin; KM: kanamycin; TC: tetracycline; NA: nalidixic acid; and TMP: trimethoprim.

According to the Statistical Survey on Livestock of the Japanese Ministry of Agriculture, Forestry, and Fisheries (JMAFF) (https://www.maff.go.jp/j/tokei/kouhyou/tikusan/index.html), the Hokkaido, Tohoku, Kanto, and Kyushu regions produced 5.3%, 23.2%, 3.8%, and 50.6%, respectively, of the total number of broilers in Japan in 2023. A national survey conducted between 2007 and 2008 reported that the prevalence of Salmonella in broiler flocks was 86.1% [16]. Momose et al. [11] examined Salmonella presence in the cecal contents of 75 broiler flocks in Japan between 2017 and 2022, reporting Salmonella prevalence of 88.0%. Although the reason remains unclear, Salmonella prevalence in Japanese broilers tends to be higher than that in other countries [1, 2, 9, 13]. In this study, we used breast skin as a specimen for Salmonella isolation. This was based on our 2022 survey, in which Salmonella presence was examined in breast meat products produced from Salmonella-infected broiler flocks using breast skin as a specimen [21]. The results of that study showed that Salmonella was isolated from 96.3% (26/27) of the skin samples. The high Salmonella contamination rate in chicken meat in the present study could be attributed to the high Salmonella prevalence in broiler flocks. Salmonella prevalence in chicken meat products packaged at the plant M (40.0%), was lower than those packaged at other plants, and the age of all five broiler flocks at the time of slaughtering was 65 days. We previously collected cecal content from five broiler flocks at plant M in 2022 and isolated Salmonella from two (40%) broiler flocks, both of which were 65 days old [21]. While the parents of the broilers slaughtered at 65 days were the same as those of ordinary broilers, chicken meat products derived from broilers slaughtered at 65 days had a special trade name to distinguish them from ordinary chicken meat products. We previously reported that Salmonella prevalence in Jidori flocks was 28.6% (6/21) [17]. The age of the slaughtered Jidori flocks ranged from 81 to 93 days. Salmonella isolation rates may differ between ordinary chicken meat products and other chicken meat products, and one of the reasons may be the length of the chicken flock’s raising period.

In the present study, S. Schwarzengrund was the most frequent serovar, accounting for 83.3% of Salmonella isolates. The Tokyo Metropolitan Institute of Public Health has monitored serovars and antimicrobial resistance profiles of Salmonella in chicken meat and reported that S. Infantis was the most frequent serovar until 2015, but S. Schwarzengrund has been the most frequent since 2016 [22]. Other Japanese reports also showed that the most frequent serovar in Salmonella isolated from chicken meat collected around 2015 was S. Infantis [6, 12]. Moreover, one farm in eastern Japan was investigated for Salmonella prevalence twice: the first investigation was conducted between 2017 and 2018, in which 13 isolates were isolated from 14 broiler flocks and 9 were S. Infantis; the second investigation was conducted between 2021 and 2022, in which 10 isolates were isolated from 11 broiler flocks, all of which were S. Schwarzengrund [11]. The national survey conducted between 2007 and 2008, mentioned above, reported that the top three serovars were S. Infantis, S. Manhattan, and S. Schwarzengrund [16]. Notably, no S. Schwarzengrund was isolated from broiler flocks in the Tohoku and Kanto regions at that time [16]. However, in the present study, S. Schwarzengrund was found in chicken meat products packaged at all six plants in the Tohoku region. While the exact reason for this shift remains unclear, it is possible that S. Schwarzengrund has spread to broiler farms in the Tohoku region over the past decade.

Among S. Schwarzengrund isolates, resistance rates to streptomycin, kanamycin, and tetracycline were significantly higher in Kyushu than in Hokkaido or Tohoku. No multidrug-resistant S. Schwarzengrund was isolated in Hokkaido, although the rates of multidrug resistance in isolates from Tohoku and Kyushu were 41.5% and 100.0%, respectively. The use of antimicrobials may be more common in broilers from Kyushu than in those from Hokkaido or Tohoku. Data on the usage and sales volume of antimicrobial substances has been published by the JMAFF by animal species, but it does not include usage or sales volumes by region (https://www.maff.go.jp/nval/yakuzai/yakuzai_p3_6.html).

Almost all S. Schwarzengrund isolated from Japanese chicken meat are classified as ST241 by multilocus sequence typing [6, 10]. Furthermore, Ikeuchi et al. [5] recently suggested that ST241 has evolved from ST96, a globally prevalent sequence type of S. Schwarzengrund, and subsequently spread to broiler farms across Japan. In Japan, a small number of grandparent-stock companies import broiler grandparent flocks from overseas. The chickens born from these grandparent flocks are sold as breeding stock to integrated chicken producers and day-old-chick suppliers. These producers and suppliers raise the chickens on breeder farms, where eggs are sent to hatcheries. The day-old chicks hatched at the hatcheries are then supplied to their own or contracted broiler farms. Veterinarians working at grandparent-stock companies, chicken producers, and day-old-chick suppliers prescribe antimicrobials according to each company’s treatment policies. The results of this study suggest that S. Schwarzengrund may be transmitted through this chain of chicken transportation and acquire antimicrobial resistance. To validate this hypothesis, it is essential to conduct Salmonella examinations at each stage, analyze the genetic characteristics of the isolates, investigate antimicrobial resistance, and collect data on antimicrobial use at each step.

In this study, we were unable to purchase chicken products from Kyushu at retail stores in Hokkaido; therefore, we collected chicken products from six poultry processing plants in Kyushu. In contrast, we were able to purchase chicken products from Hokkaido, Tohoku, Kanto, and Kyushu at retail stores in the Kanto region. The regional variation in the distribution of chicken meat products, there may be similar regional variations in the antimicrobial resistance profiles of Salmonella isolated from human patients with salmonellosis.

In conclusion, the contamination rate of Salmonella in chicken meat products is very high; therefore, consumers should thoroughly cook chicken meat before consumption to prevent Salmonella infection. Moreover, efficient strategies for Salmonella management on broiler farms and chicken processing plants should be developed. In Japan, fluoroquinolones and third-generation cephalosporins (TGCs) are recommended as the first- and second-choice antimicrobials for patients with Salmonella enteritis, respectively [7]. All Salmonella isolates in this study were susceptible to ciprofloxacin and cefotaxime, although geographical variation in the antimicrobial resistance of Salmonella in chicken meat products was clearly observed. Thus, administration of fluoroquinolones or TGCs is an effective treatment option for patients with Salmonella enteritis caused by the consumption of contaminated chicken meat. In particular, there are likely to be many options for antimicrobials that can be administered to patients with Salmonella enteritis caused by the consumption of chicken produced in Hokkaido.

CONFLICT OF INTEREST

The authors have no conflict of interest.