Integrated Phenotypic, Molecular, and Genomic Analysis of Antimicrobial Resistance in Yersinia pestis Isolates from Natural Plague Foci of Kazakhstan

Abstract

Plague remains a globally important zoonotic disease maintained in natural foci, with ongoing epizootic activity and periodic human cases reported in several regions of the world. Continuous monitoring of antimicrobial susceptibility in Yersinia pestis is essential because the emergence of resistant strains could compromise the effectiveness of currently recommended therapeutic regimens. In this study, 75 Y. pestis isolates originating from natural plague foci of Kazakhstan were investigated using an integrated approach combining phenotypic susceptibility testing, targeted molecular screening, and whole-genome sequencing (WGS)-based resistome analysis. The collection included historical clinical isolates obtained during plague outbreaks as well as more recent epizootic strains recovered from animal hosts and flea vectors. Phenotypic testing demonstrated uniformly high susceptibility to clinically relevant antimicrobial agents used for plague treatment. Targeted molecular screening by real-time PCR did not detect the analyzed resistance determinants. Genome-wide analysis based on WGS data from NCBI BioProject PRJNA1249055 did not identify acquired antimicrobial resistance genes, major resistance-associated mutations in key chromosomal loci (rpsL, gyrA, and parC), or plasmid-borne resistance determinants. Regulatory loci associated with adaptive responses were highly conserved across the analyzed genomes. The complete concordance between phenotypic, molecular, and genomic findings indicates a stable antimicrobial susceptibility profile of Y. pestis circulating in natural plague foci of Kazakhstan. These results support the continued effectiveness of current therapeutic strategies for plague and highlight the value of integrating genomic surveillance into long-term monitoring programs for this pathogen.

1. Introduction

Plague remains one of the most historically significant and epidemiologically important zoonotic infections caused by Yersinia pestis (Lehmann and Neumann, 1896) [1,2]. Despite major advances in antimicrobial therapy and public health measures, the pathogen continues to persist in natural foci across Central Asia, Africa, and the Americas [3,4]. Transmission is maintained through complex ecological systems involving wild rodents as reservoir hosts and fleas as vectors, resulting in long-term enzootic cycles with periodic epizootic activity [2,5].

Antimicrobial resistance (AMR) is recognized as one of the leading threats to global public health and socio-economic development. According to current estimates, AMR causes approximately 1.27 million deaths annually, and its burden is expected to increase substantially because of the widespread use and misuse of antibiotics [6,7,8,9]. Although antimicrobial resistance in Y. pestis remains uncommon, preserving susceptibility is critically important because effective antibiotic therapy remains the cornerstone of plague treatment and outbreak response.

Kazakhstan represents one of the world’s largest plague-endemic territories, with extensive natural foci distributed across desert, semi-desert, and high-mountain ecosystems [10,11,12]. Historically, plague had a considerable impact on public health in the country, with numerous outbreaks recorded during the twentieth century. Although the last confirmed human case was reported in 2003 [13,14], ongoing epizootic activity in natural reservoirs confirms the continued circulation of Y. pestis and the potential for future re-emergence [15,16].

The emergence of antimicrobial resistance in Y. pestis remains a global concern. Multidrug-resistant strains carrying plasmid-mediated resistance determinants have been reported in Madagascar, whereas resistance in other regions has been associated with chromosomal mutations [17,18]. Although such events remain rare, they emphasize the importance of continuous antimicrobial susceptibility monitoring in countries with extensive natural plague foci.

In Kazakhstan, antimicrobial resistance surveillance is conducted within national programs and international initiatives based on One Health principles, providing an important framework for monitoring emerging resistance threats in both human and animal populations [19,20,21]. In parallel, a recent meta-analysis summarized all reported human plague cases in Kazakhstan between 1926 and 2003, providing an important historical context for contemporary investigations [14]. Although human plague has not been reported in recent decades, the persistence of active natural foci necessitates continued microbiological and genomic surveillance.

Modern surveillance increasingly integrates epidemiological monitoring, laboratory diagnostics, and whole-genome sequencing (WGS). Genome-scale analysis has become an essential tool for investigating pathogen evolution, transmission dynamics, population structure, and antimicrobial resistance determinants [22,23].

Although phenotypic antimicrobial susceptibility testing and targeted molecular screening of a subset of these isolates were previously reported [24], no comprehensive genome-wide assessment of antimicrobial resistance determinants has been performed for Y. pestis isolates from Kazakhstan. Consequently, important questions remain regarding the resistome structure, the genetic stability of resistance-associated loci, and the genomic basis of antimicrobial susceptibility in Y. pestis populations circulating in Central Asian natural plague foci. Addressing these knowledge gaps is particularly important in view of the continued epizootic activity of plague in Kazakhstan and the global expansion of antimicrobial resistance among bacterial pathogens.

Therefore, the present study aimed to characterize the antimicrobial susceptibility profile and resistome structure of Y. pestis isolates from natural plague foci of Kazakhstan using an integrated approach combining phenotypic testing, targeted molecular screening, and whole-genome sequencing. By integrating conventional microbiological methods with genome-scale analysis, this study provides new insights into the current antimicrobial susceptibility of Y. pestis circulating in Kazakhstan and contributes to the development of evidence-based surveillance strategies for plague.

2. Materials and Methods

2.1. Bacterial Strains and Study Design

A total of 75 Y. pestis strains were included in the study. The collection comprised 61 clinical isolates obtained from patients and deceased individuals during plague outbreaks in Kazakhstan between 1926 and 2003, as well as 14 isolates recovered from animal hosts and flea vectors in natural plague foci in more recent years. All strains were obtained from the National Working Collection and the Microorganism Depository of the M. Aikimbayev National Scientific Center for Especially Dangerous Infections.

Detailed metadata for all 75 Y. pestis isolates, including laboratory identifiers, geographic origin, natural plague foci, host and vector species, and years of isolation, were reported previously in a comprehensive phenotypic and molecular study and are not reproduced here to avoid duplication [24].

The study was designed as an integrated antimicrobial resistance assessment combining phenotypic susceptibility testing, targeted molecular screening, and whole-genome sequencing (WGS)-based resistome analysis. Phenotypic susceptibility testing and RT-PCR screening results were obtained from our previously published study [24] and were incorporated into the present investigation to enable integrated analysis together with newly performed genome-scale resistome characterization.

The principal novel component of the present study is the comprehensive WGS-based resistome analysis of 75 Y. pestis genomes, including the assessment of acquired antimicrobial resistance genes, resistance-associated chromosomal mutations, and plasmid-mediated resistance determinants. The overall study design integrating phenotypic, molecular, and genomic approaches is presented in Figure 1.

2.2. Culture Conditions and Identification

Y. pestis strains were cultured on Mueller–Hinton agar (HiMedia Laboratories, Mumbai, India) (pH 7.3 ± 0.2) and Hottinger agar (HiMedia Laboratories, Mumbai, India) (pH 7.2 ± 0.1) at incubation temperatures ranging from 28 °C to 37 °C in accordance with standard laboratory procedures for the cultivation of plague bacilli [25].

Taxonomic identification was performed using the automated VITEK^® 2 Compact 30 system (bioMérieux, Marcy-l’Étoile, France). Species identification was based on biochemical profiles generated by the instrument and interpreted according to the manufacturer’s recommendations. All isolates were confirmed as Y. pestis prior to antimicrobial susceptibility testing and molecular analyses.

2.3. Antimicrobial Susceptibility Testing

Antimicrobial susceptibility testing was performed previously and reported in detail by Abdel et al. [24]. The resulting phenotypic data were incorporated into the present study to enable integrated analysis together with whole-genome sequencing (WGS)-based resistome characterization.

Antimicrobial susceptibility was evaluated using two complementary phenotypic methods: the Kirby–Bauer disk diffusion assay and the E-test gradient diffusion method (HiMedia Laboratories, Mumbai, India). Testing was performed in accordance with Clinical and Laboratory Standards Institute (CLSI) guidelines [25,26,27].

Bacterial suspensions were standardized to 0.5 McFarland turbidity (approximately 1.5 × 10⁸ CFU/mL), inoculated onto Mueller–Hinton agar plates (HiMedia Laboratories, Mumbai, India), and incubated at 28 °C. Inhibition zone diameters were measured after 24–48 h of incubation. The antimicrobial agents tested represented the principal classes used in plague therapy and antimicrobial resistance surveillance, including β-lactams, tetracyclines, aminoglycosides, amphenicols, glycopeptides, fluoroquinolones, lincosamides, and macrolides.

Minimum inhibitory concentrations (MICs) were determined using E-test strips (HiMedia Laboratories, Mumbai, India) and interpreted according to CLSI criteria [25,26,27].

2.4. Phenotypic Detection of Resistance Mechanisms

Extended-spectrum β-lactamase (ESBL) production was assessed using standard phenotypic methods based on inhibition zone analysis and confirmatory tests [25,26,27]. Although ESBL production has not been commonly reported in Y. pestis, screening was included because plasmid-mediated β-lactam resistance determinants may theoretically be acquired through horizontal gene transfer from other members of the Enterobacteriaceae. Therefore, ESBL testing was performed as an additional surveillance measure for the detection of potential β-lactam resistance mechanisms.

Quality control strains included Escherichia coli ATCC 25922, Klebsiella pneumoniae ATCC 700603, Staphylococcus aureus ATCC 25923, and Pseudomonas aeruginosa ATCC 27853 [25,26,27].

2.5. Molecular Detection of Resistance Genes (RT-PCR)

Screening for antimicrobial resistance genes was performed previously and reported in detail by Abdel et al. [24]. The resulting molecular data were incorporated into the present study to enable integrated interpretation together with whole-genome sequencing (WGS)-based resistome analysis.

Real-time PCR (RT-PCR) was performed using the BacResista GLA Detection Kit (DNA-Technology LLC, Moscow, Russia). The following groups of resistance determinants were targeted: β-lactam resistance genes (tem, ctx-M-1, shv), carbapenemase genes (kpc, ndm, vim, imp, and oxa variants), glycopeptide resistance genes (vanA, vanB), and the methicillin resistance gene (mecA).

Amplification reactions were monitored using the FAM, HEX, and CY5 fluorescence channels according to the manufacturer’s instructions, and cycle threshold (Ct) values were used for result interpretation [28,29].

2.6. Whole-Genome Sequencing Data and Resistome Analysis

Whole-genome sequencing (WGS) data for 75 Y. pestis strains were obtained from the NCBI BioProject PRJNA1249055. Genome assemblies and annotated genome files (FASTA, GBFF, and FAA formats) were used for downstream resistome analysis. The WGS dataset included isolates originating from multiple natural plague foci, host species, and historical periods. Genome-based analysis was performed using a bioinformatic approach combining annotation screening and sequence-based comparison with established antimicrobial resistance databases.

2.6.1. Detection of Acquired AMR Genes

Screening for acquired antimicrobial resistance genes was performed using sequence similarity-based approaches with reference databases, including the Comprehensive Antibiotic Resistance Database (CARD) and ResFinder. Annotated genomes were screened for known resistance determinants associated with aminoglycosides, β-lactams, carbapenems, tetracyclines, amphenicols, sulfonamides, macrolides, and other clinically relevant antimicrobial classes. Sequence comparison was performed using BLAST-based methods to identify homologs of characterized antimicrobial resistance genes.

2.6.2. Identification of Chromosomal Resistance-Associated Mutations

Key chromosomal loci associated with antimicrobial resistance were analyzed, including rpsL (streptomycin resistance), gyrA and parC (fluoroquinolone resistance), and the regulatory genes pmrA, pmrB, phoP, and phoQ. Amino acid sequences were extracted from annotated genomes and aligned to evaluate sequence conservation and to identify substitutions at canonical resistance-associated positions described in the literature.

2.6.3. Plasmid Analysis

Plasmid content was evaluated using genome assembly annotations to identify the major virulence-associated plasmids of Y. pestis (pCD1, pMT1, and pPCP1) and to screen for potential plasmid-mediated antimicrobial resistance determinants. Additional plasmid sequences were assessed for similarity to known resistance-associated plasmids reported in Enterobacteriaceae and other Gram-negative bacteria.

2.6.4. Comparative Analysis

Comparative genomic analysis was performed to assess the distribution of resistance-associated determinants among isolates originating from different natural plague foci, host species, and periods of isolation. The analysis was designed to evaluate the conservation of resistance-associated loci and to identify potential geographic, ecological, or temporal variation within the study collection.

2.7. Statistical Analysis

Descriptive statistics were used to summarize phenotypic susceptibility and MIC data, including ranges, mean values, and standard deviations where applicable. Because no resistant isolates were identified and the study was primarily designed as an integrated descriptive surveillance analysis, no inferential group comparisons were performed.

3. Results

3.1. General Characteristics of the Studied Strains

A total of 75 Y. pestis isolates originating from diverse natural plague foci of Kazakhstan were included in the analysis. The collection comprised both historical clinical isolates obtained during plague outbreaks between 1926 and 2003 and more recent isolates recovered from animal reservoirs and flea vectors during epizootological surveillance activities.

Phenotypic antimicrobial susceptibility and molecular screening data for these isolates were reported previously [24] and are incorporated here as part of an integrated assessment combining phenotypic, molecular, and whole-genome sequencing (WGS)-based resistome analyses. Biochemical characterization confirmed that all isolates exhibited phenotypic profiles consistent with Y. pestis, supporting the reliability of subsequent antimicrobial susceptibility testing and genomic investigations.

The geographic, ecological, and epidemiological characteristics of the study collection are summarized in

Table 1. Summary Characteristics of the Yersinia pestis Isolates Included in the Study.

Characteristic	Value
Total number of Y. pestis isolates analyzed	75
Clinical isolates obtained from human cases and fatal outcomes	61
Epizootic isolates recovered from animal hosts and flea vectors	14
Period of isolation	1926–2012
Human-derived isolates (historical epidemic strains)	1926–2003
Animal- and vector-derived isolates	2004–2012
Administrative regions represented	7 (Atyrau, Aktobe, Almaty, Zhambyl, Kyzylorda, Mangystau, and Turkestan)
Natural plague foci represented	17
Ecological zones represented	Desert, semi-desert, and high-mountain plague foci
Principal host species	Rhombomys opimus, Meriones meridianus, Marmota baibacina, and Marmota caudata
Principal sources of isolation	Humans, great gerbils, jirds, marmots, and their ectoparasites

.

The study collection encompasses isolates from both human infections and natural reservoirs, providing broad ecological and temporal representation of Y. pestis populations circulating in Kazakhstan over nearly nine decades.

3.2. Phenotypic Susceptibility to Antimicrobial Agents

Phenotypic testing demonstrated uniformly high susceptibility of Y. pestis isolates to the major classes of antimicrobial agents used for plague treatment and surveillance.

All isolates (100%) were susceptible to β-lactams, tetracyclines, aminoglycosides, amphenicols, glycopeptides, lincosamides, and fluoroquinolones. In contrast, reduced activity was observed for macrolides, with effectiveness ranging from complete inactivity to moderate inhibition (0–58%), which is consistent with the intrinsic low susceptibility of Gram-negative bacteria to this antimicrobial class.

Analysis of inhibition zone diameters revealed a narrow and consistent range across isolates, with no apparent deviations or extreme outliers. The observed homogeneity indicates a stable susceptibility pattern within the study collection.

The overall distribution of phenotypic susceptibility across major antimicrobial classes is summarized in

Table 2. Summary of phenotypic susceptibility of Yersinia pestis isolates to major antimicrobial classes.

Antibiotic Class	Phenotypic Result	Quantitative Summary	Interpretation
β-lactams	100% susceptible	Inhibition zone range 23.2–39.8 mm	High activity preserved
Tetracyclines	100% susceptible	21.0–27.3 mm	Retained activity of first-line drugs
Aminoglycosides	100% susceptible	18.8–27.8 mm	No phenotypic evidence of resistance
Amphenicols	100% susceptible	23.1–26.3 mm	Preserved susceptibility
Glycopeptides	100% susceptible	21.2–25.9 mm	Uniform susceptibility profile
Lincosamides	100% susceptible	21.5–25.9 mm	Uniform susceptibility pattern
Fluoroquinolones	100% susceptible	28.8–36.7 mm	High activity, including ciprofloxacin
Macrolides	Low activity	0.0–58.0% activity	Consistent with expected low efficacy against Gram-negative bacteria

.

Overall, the phenotypic findings indicate preserved susceptibility of Y. pestis isolates across all major antimicrobial classes evaluated. The relatively narrow distribution of inhibition zone diameters and MIC values supports the absence of phenotypic evidence suggesting reduced susceptibility within the study collection.

As shown in Figure 2, inhibition zone diameters demonstrated low variability across all antibiotic classes, indicating a homogeneous susceptibility profile among the analyzed isolates.

Boxplots represent the median, interquartile range, and overall distribution of inhibition zone diameters for each antibiotic class. The relatively narrow distribution of inhibition zone diameters and MIC values indicates a stable susceptibility profile within the study collection, with no phenotypic evidence of reduced susceptibility.

3.3. Quantitative Assessment of Antibiotic Activity (MIC Analysis)

Minimum inhibitory concentration (MIC) analysis confirmed the high sensitivity of the strains to key antimicrobial agents.

Field isolates demonstrated:

−

MIC values as low as 0.023 µg/mL (moxifloxacin);

−

upper MIC values up to 4 µg/mL (amikacin);

−

mean MIC ≈ 1.06 µg/mL.

Some isolates demonstrated slightly greater MIC variability; however, all values remained within established susceptibility thresholds.

These findings support the continued effectiveness of first-line and reserve antimicrobial agents used in plague treatment, including streptomycin, gentamicin, doxycycline, ciprofloxacin, and chloramphenicol.

3.4. Phenotypic Evidence of Resistance Mechanisms

No phenotypic evidence of extended-spectrum β-lactamase (ESBL) production was identified among the tested isolates. These findings further support the conclusion that the studied population does not exhibit clinically relevant antimicrobial resistance phenotypes.

3.5. Molecular Screening of Antibiotic Resistance Genes (RT-PCR)

Molecular analysis using real-time PCR did not detect any of the targeted resistance genes in the examined Y. pestis strains.

The following resistance determinants were absent in all isolates:

−

β-lactam resistance genes: tem, ctx-M-1, shv;

−

carbapenemases: kpc, ndm, vim, imp, oxa variants;

−

glycopeptide resistance genes: vanA, vanB;

−

methicillin resistance gene: mecA.

No amplification signals were observed in diagnostic channels, while internal controls confirmed the validity of the assay.

These findings indicate that no targeted horizontally acquired antimicrobial resistance determinants were detected in the analyzed strain collection and are consistent with the observed phenotypic susceptibility profiles (Table 2) and molecular screening results summarized in

Table 3. Summary of molecular screening and preliminary resistome findings in 75 Yersinia pestis isolates.

Determinant/Feature	Method	Result	Interpretation
tem	RT-PCR	Not detected	No evidence of common acquired β-lactam resistance determinant
ctx-M-1	RT-PCR	Not detected	No ESBL-associated signal
shv	RT-PCR	Not detected	No ESBL-associated signal
oxa-type targets	RT-PCR	Not detected	No carbapenemase-associated signal in screened panel
imp	RT-PCR	Not detected	No metallo-β-lactamase signal
kpc	RT-PCR	Not detected	No carbapenemase signal
ndm	RT-PCR	Not detected	No carbapenemase signal
vim	RT-PCR	Not detected	No carbapenemase signal
vanA/B	RT-PCR	Not detected	No glycopeptide resistance determinant detected
mecA	RT-PCR	Not detected	No methicillin resistance determinant detected
ESBL phenotype	Phenotypic confirmatory testing	Not detected	No phenotypic evidence of extended-spectrum β-lactamase production
Acquired AMR plasmids	WGS-based plasmid analysis	Not identified	No additional plasmid replicons associated with antimicrobial resistance were identified at the assembly level

.

The molecular screening results further support the phenotypic findings and indicate that none of the targeted antimicrobial resistance determinants were detected in the analyzed strain collection. Together, these data demonstrate concordance between phenotypic susceptibility profiles and molecular screening results.

To further validate these observations at the genomic level, whole-genome sequencing (WGS) data were analyzed.

3.6. Whole-Genome Sequencing (WGS) and Resistome Analysis

To further validate the phenotypic and molecular findings at the genomic level, whole-genome sequencing (WGS) data were analyzed to assess the presence of acquired antimicrobial resistance genes and chromosomal mutations associated with antibiotic resistance. Whole-genome sequencing data for 75 Y. pestis isolates were obtained from the NCBI BioProject PRJNA1249055, originally generated for studies on genetic diversity and biovar classification of Central Asian Y. pestis isolates [30].

A comprehensive genome-wide screening approach was applied to all annotated GBFF files to identify both acquired antimicrobial resistance determinants and chromosomal mutations associated with antibiotic resistance.

Key chromosomal loci known to be involved in antimicrobial resistance in Y. pestis and related Enterobacteriaceae were systematically extracted and compared across all genomes, including rpsL, gyrA, parC, pmrA, pmrB, phoP, and phoQ. Amino acid sequences were aligned to assess conservation and to detect substitutions at canonical resistance-associated positions.

In parallel, genome annotations were screened for the presence of acquired antimicrobial resistance genes representing major clinically relevant classes, including aminoglycoside resistance genes (strA, strB, aadA, aac, aph), tetracycline resistance genes, chloramphenicol resistance genes (cat-family), sulfonamide resistance genes (sul1/sul2/sul3), β-lactamases (bla-family), carbapenemases (kpc, ndm, vim, imp, oxa-type), as well as other resistance determinants (vanA, vanB, mecA, qnr, dfrA, erm, and mph). Additionally, plasmid content was evaluated to identify the presence of known virulence-associated plasmids and to screen for potential plasmid-mediated antimicrobial resistance determinants.

Genome-wide screening did not identify any known acquired antimicrobial resistance determinants across the analyzed isolates. In particular, no genes associated with resistance to aminoglycosides (strA, strB, aadA, aac, aph), tetracyclines, amphenicols, sulfonamides (sul1/sul2/sul3), or β-lactams were detected. Likewise, no genes encoding carbapenemases (kpc-, ndm-, vim-, imp-, oxa-type) or other clinically relevant resistance determinants (qnr, dfrA, erm, and mph) were identified. The only annotation related to the sul family corresponded to the intrinsic sulA gene, which is not interpreted as an acquired sulfonamide resistance determinant.

Analysis of chromosomal loci associated with antimicrobial resistance demonstrated a high level of conservation across all isolates. No substitutions were observed at canonical resistance-associated positions in rpsL (Lys43 and Lys88) or gyrA (Ser83 and Asp87), indicating the absence of mutations linked to streptomycin and fluoroquinolone resistance, respectively. The parC gene was also fully conserved among the analyzed genomes. Regulatory loci involved in adaptive responses, including pmrA, pmrB, phoP, and phoQ, were highly conserved across all analyzed genomes, with no evidence of resistance-associated variation.

Plasmid analysis confirmed the presence of the core virulence plasmids (pCD1, pMT1, and pPCP1) in all 75 isolates. The cryptic plasmid pCKF was detected in three isolates, while no additional plasmid replicons associated with antimicrobial resistance were identified.

A summary of WGS-based resistome findings is presented in

Table 4. Whole-genome sequencing (WGS)-based resistome analysis of Yersinia pestis isolates (n = 75).

Category	Feature	Result	Interpretation
Acquired AMR genes	Aminoglycoside resistance genes (strA, strB, aadA, aac, aph)	Not detected	No evidence of acquired aminoglycoside resistance
	Tetracycline resistance genes (tet-family)	Not detected	No acquired tetracycline resistance
	Chloramphenicol resistance genes (cat-family)	Not detected	No acquired amphenicol resistance
	Sulfonamide resistance genes (sul1/sul2/sul3)	Not detected	No acquired sulfonamide resistance
	β-lactamase genes (bla-family)	Not detected	No acquired β-lactam resistance
	Carbapenemases (kpc, ndm, vim, imp, oxa-type)	Not detected	No carbapenem resistance determinants
	Other AMR genes (qnr, dfrA, erm, mph)	Not detected	No additional resistance determinants
Chromosomal loci	rpsL (Lys43, Lys88)	No mutations	No streptomycin resistance-associated substitutions
	gyrA (Ser83, Asp87)	No mutations	No fluoroquinolone resistance-associated substitutions
	parC	No variation	Conserved across isolates
	pmrA, pmrB, phoP, phoQ	No variation	Conserved regulatory loci with no evidence of resistance-associated variation
Plasmid content	Core virulence plasmids (pCD1, pMT1, pPCP1)	Detected in 75/75 isolates	Typical plasmid profile of Y. pestis
	Cryptic plasmid (pCKF)	Detected in 3/75 isolates	Not associated with AMR
	MDR-associated plasmids	Not detected	No plasmid-mediated resistance

. Overall, the genomic findings are consistent with the phenotypic susceptibility profiles and molecular screening results, indicating the absence of acquired antimicrobial resistance genes, major resistance-associated chromosomal mutations, and plasmid-mediated resistance mechanisms in the analyzed Y. pestis population.

These findings provide genome-scale evidence supporting the preserved susceptibility profile observed in the analyzed Y. pestis collection. The concordance between phenotypic, molecular, and genomic data indicates remarkable stability of resistance-associated loci despite long-term circulation of the pathogen in diverse natural plague foci of Kazakhstan. The implications of these findings in the context of global antimicrobial resistance trends are discussed below.

4. Discussion

Monitoring antimicrobial susceptibility remains a critical component of plague surveillance and public health preparedness, particularly in countries where extensive natural plague foci support the long-term circulation of Y. pestis. In addition to conventional phenotypic susceptibility testing, contemporary surveillance increasingly incorporates molecular and genomic approaches that enable the detection of resistance determinants and provide insights into pathogen evolution, adaptation, and transmission dynamics [31,32,33].

The principal novelty of the present study lies in the integration of phenotypic susceptibility testing, targeted molecular screening, and whole-genome sequencing (WGS)-based resistome analysis within a unified analytical framework. Although phenotypic susceptibility profiles and selected molecular findings for part of this strain collection were previously reported [24], no comprehensive genome-scale assessment of antimicrobial resistance determinants has been performed for Y. pestis isolates from Kazakhstan. The present investigation therefore provides the first integrated resistome characterization of a large collection of Kazakhstani Y. pestis isolates representing multiple natural plague foci, ecological zones, host species, and historical periods.

The phenotypic results demonstrated uniformly high susceptibility of all examined isolates to the major classes of antimicrobial agents currently recommended for plague treatment, including aminoglycosides, tetracyclines, fluoroquinolones, β-lactams, and amphenicols. These findings indicate that first-line therapeutic agents such as streptomycin, gentamicin, doxycycline, ciprofloxacin, and chloramphenicol remain effective against Y. pestis strains circulating in Kazakhstan. The observed susceptibility patterns are consistent with reports from other plague-endemic regions and support the continued effectiveness of currently recommended treatment regimens [34,35,36].

Despite the generally favorable susceptibility profile of Y. pestis worldwide, resistant strains have occasionally been documented. The first multidrug-resistant (MDR) Y. pestis strain was identified in Madagascar in 1995 and carried a transferable plasmid conferring resistance to streptomycin, tetracycline, and chloramphenicol [37,38]. Subsequent studies demonstrated that resistance may also arise through chromosomal mutations. For example, streptomycin resistance associated with mutations in the rpsL gene has been described in isolates from China [39]. These observations confirm that Y. pestis possesses the capacity to acquire resistance through both horizontal gene transfer and spontaneous chromosomal mutation, although such events remain uncommon.

Recent international reviews indicate that only a limited number of resistant Y. pestis isolates have been reported worldwide during the last three decades [39,40]. Consequently, antimicrobial resistance in plague remains sporadic rather than widespread. The findings obtained in the present study are therefore consistent with the current global epidemiological situation, which is characterized by preserved susceptibility among most known Y. pestis populations despite occasional reports of resistant strains.

A major strength of the present investigation is the incorporation of WGS data to validate phenotypic and molecular findings. Genome-wide screening of 75 isolates did not identify any acquired antimicrobial resistance genes, including determinants associated with aminoglycoside, tetracycline, amphenicol, sulfonamide, β-lactam, or carbapenem resistance. Likewise, no plasmid-borne resistance determinants were detected. These results substantially strengthen the conclusions derived from phenotypic testing by providing genome-scale evidence supporting the observed susceptibility patterns.

Analysis of chromosomal loci known to be associated with antimicrobial resistance further demonstrated a high degree of genetic conservation among the studied isolates. No substitutions were identified at canonical resistance-associated positions in rpsL and gyrA, which are commonly implicated in resistance to streptomycin and fluoroquinolones, respectively. Similarly, parC, pmrA, pmrB, phoP, and phoQ were highly conserved across the collection, with no evidence of resistance-associated variation. Overall, the genomic findings indicate remarkable stability of resistance-associated loci within Y. pestis populations circulating in Kazakhstan.

The complete concordance between phenotypic susceptibility testing, RT-PCR screening, and WGS-based resistome analysis provides a high level of confidence in the results. Agreement across independent methodological approaches substantially reduces the likelihood that clinically significant resistance determinants were overlooked and demonstrates the value of combining conventional microbiological techniques with modern genomic surveillance tools.

The lack of detectable resistance determinants in the investigated strains may be explained by several ecological and epidemiological factors. First, antibiotic selective pressure within natural plague ecosystems is likely limited compared with clinical environments where antimicrobial exposure is frequent. Second, the relative ecological isolation of many plague reservoirs may reduce opportunities for horizontal transfer of resistance genes from other bacterial species. Finally, long-term implementation of plague surveillance and control programs in Kazakhstan may have contributed to maintaining genetically stable pathogen populations without substantial acquisition of resistance determinants.

From a public health perspective, these findings are encouraging because they support the continued use of standard therapeutic regimens for plague and reinforce preparedness strategies for potential outbreaks. Nevertheless, previously documented resistant Y. pestis strains from other regions demonstrate that resistance emergence remains possible. Continuous surveillance therefore remains essential, particularly in the context of the growing global burden of antimicrobial resistance among other members of the Enterobacteriaceae and related bacterial pathogens.

The incorporation of genomic approaches into routine surveillance systems represents an important advancement in plague monitoring. Whole-genome sequencing not only facilitates early detection of resistance determinants but also provides valuable information regarding pathogen evolution, population structure, and transmission pathways. Within the framework of modern biosafety and One Health strategies, genomic surveillance can therefore serve as an important component of integrated early-warning systems designed to identify emerging threats before they become clinically significant.

Several limitations of the present study should be acknowledged. First, although the analyzed collection represents one of the largest systematically characterized datasets of Y. pestis currently available from Central Asia, the study was geographically restricted to Kazakhstan. Therefore, the findings should be interpreted primarily in the context of Central Asian plague foci and may not fully reflect the global diversity of Y. pestis populations. Second, the predominance of historical clinical isolates reflects the epidemiological situation in Kazakhstan, where no human plague cases have been reported since 2003. Consequently, more recent isolates were primarily obtained from animal reservoirs and flea vectors during routine epizootological surveillance. Nevertheless, the inclusion of both historical and contemporary isolates enabled assessment of antimicrobial susceptibility patterns over an extended temporal period and provided insight into the long-term stability of susceptibility profiles within natural plague foci. Third, the genomic analysis focused on currently recognized antimicrobial resistance determinants and canonical resistance-associated mutations. As a result, previously undescribed resistance mechanisms, rare adaptive variants, regulatory alterations, or other cryptic resistance determinants may not have been detected. Finally, functional validation experiments and transcriptomic analyses were beyond the scope of the present study. Future investigations integrating transcriptomic, proteomic, and experimental approaches may provide additional insight into adaptive responses and potential cryptic resistance mechanisms in Y. pestis.

5. Conclusions

This study provides the first integrated assessment of antimicrobial susceptibility and resistome structure in a large collection of Y. pestis isolates originating from natural plague foci of Kazakhstan. By combining phenotypic susceptibility testing, targeted molecular screening, and whole-genome sequencing (WGS)-based resistome analysis, the study offers a comprehensive evaluation of antimicrobial resistance-associated determinants in Y. pestis populations circulating in Central Asia.

No acquired antimicrobial resistance genes, major resistance-associated chromosomal mutations, or plasmid-mediated resistance determinants were identified in the analyzed isolates. The strong concordance between phenotypic, molecular, and genomic findings indicates a preserved susceptibility profile across both historical and contemporary Y. pestis strains collected over nearly nine decades.

These findings support the continued effectiveness of currently recommended antimicrobial agents for plague treatment in Kazakhstan and are consistent with the broader global pattern of generally preserved susceptibility reported for Y. pestis. At the same time, the sporadic occurrence of resistant strains documented elsewhere highlights the importance of maintaining continuous surveillance programs capable of detecting emerging resistance at an early stage.

The integration of whole-genome sequencing into routine plague surveillance represents an important advancement for monitoring pathogen evolution and antimicrobial resistance. Within the framework of modern biosafety and One Health strategies, combining epidemiological, ecological, microbiological, and genomic data provides a robust foundation for evidence-based risk assessment and long-term preparedness against plague and other zoonotic infections.