This study used S. enterica isolates that had been recovered previously from domestic animals and free-ranging wildlife sampled in selected areas of Zambia. According to the study context described in the manuscript, formal ethical and biosafety approval was not required at the time of sample collection for handling bacterial isolates obtained from domestic animals. Wildlife sampling was conducted under research quotas and permissions previously authorized by the Zambia Wildlife Authority. No experimental infection, invasive animal handling for the purpose of this sequencing work, or additional live-animal intervention was performed as part of the present study. Downstream laboratory procedures were conducted on previously recovered isolates, and all microbiological and molecular analyses were performed in accordance with standard institutional biosafety practices. In particular, biosafety level 2 precautions were followed during isolate handling, deoxyribonucleic acid extraction, PCR setup, amplicon processing, and sequencing workflows, including the use of sterile workspaces, dedicated reagents, and separation of pre- and post-PCR areas to minimize contamination and ensure laboratory safety.

The study was conducted from January 2013 to December 2019. Cattle samples were collected from seven districts in Southern Province: Mazabuka (n = 118), Monze (n = 126), Pemba (n = 126), Choma (n = 125), Kalomo (n = 50), Zimba (n = 28), and Kazungula (n = 59). All faecal samples from horses were collected in Lusaka Province. Wildlife samples were obtained from Kafue National Park (KNP), Lochinvar National Park (LNP), and Central National Park (CNP).

The original samples were collected from free-ranging wildlife and domestic animals across selected areas of Zambia from June 2013 to December 2014 [25]. Sampling and specimen collection were conducted opportunistically, based on the availability of animal species considered potential reservoirs of the pathogen. A total of 1,248 samples from domestic animals (1,008) and wildlife (240) were collected and analyzed for Salmonella [25]. Of the domestic animal samples, 659 (65.4%) were fecal samples from apparently healthy animals (cattle = 632; horses = 27), while 349 (34.6%) were samples from diseased animals submitted for routine microbiological diagnosis. The distribution of samples across the various wildlife sources and geographic locations, including Kafue National Park (KNP), Lusaka Central Park, and Lochinvar National Park (LNP), is presented in Table 1. All samples were characterized through conventional identification methods [26].

Salmonella strains isolated from the range of domestic animals and free-ranging wildlife were examined for the presence of the invA gene [25].

Deoxyribonucleic acid extraction and PCR amplification of invA gene: Genomic DNA was isolated from each Salmonella isolate using a DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) following the manufacturer’s protocol. A single colony was suspended in distilled water (200 μL), lysed with DNAzol (200 μL), and incubated at 25°C for 5 days to ensure complete cell disruption. DNAzol (800 μL) was added, and the mixture was further incubated at 25°C for 30 min. The lysate was clarified by centrifugation at 10,000 × g for 10 min at 4°C, and DNA (750 μL) was precipitated with absolute ethanol (375 μL), mixed thoroughly by vortex. The mixture was centrifuged at 12,000 × g for 10 min at 4°C. After sequential washes with 75% ethanol, the pellet in the microcentrifuge tube was left to dry completely at 25°C for 10 min, then resuspended in 8 mM sodium hydroxide (200 μL). The suspension was incubated for half a day at 4°C, then centrifuged at 12,000 × g for 10 min at 4°C. The solution (200 μL) was neutralized with 2 μL of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Sigma Adrich, Steinheim, Gemerny), and the recovered genomic DNA was stored at −30°C until analysis. To minimize contamination during DNA extraction and downstream molecular procedures, all steps were performed on a sterile workstation using dedicated reagents and separate pre- and post-PCR areas.

Selection of DNA primers and detection of invA gene: The forward and reverse primers (Inqaba Biotechnical Industries (Pty) Ltd, Pretoria, South Africa) at specific molecular sizes were used to detect the presence of a stable genetic region of the invA gene of Salmonella, following the protocol described by Rahn et al. [27]. The working concentration of each primer was 10.0 μM and the oligonucleotide sequences are shown in Table 2.

PCR amplification of the invA gene was carried out following the primer set originally described by Rahn et al. [27]. Each 10 μL reaction consisted of nuclease-free water (6.75 μL), 10× buffer (1.0 μL; Applied Biosystems, Warrington, UK), forward (0.2 μL) and reverse (0.2 μL) invA primers, deoxynucleotide triphosphates (0.8 μL; Takara Bio Inc., Shiga, Japan), Ex Taq polymerase (0.05 μL; Takara Bio Inc.), and 1 μL of DNA template. Reactions were assembled on ice and immediately transferred to a thermocycler (Applied Biosystems AB-Veriti, Foster City, CA, USA) preheated to 95°C. The cycling protocol included an initial denaturation at 95°C for 1 min, followed by 35 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s, with a final extension at 72°C for 5 min before holding at 4°C.

Thirty-four Salmonella strains previously isolated from various domestic animals and free-ranging wildlife samples were molecularly confirmed by detection of the invA gene [25]. All Salmonella isolates were referred to the Deltamune (Pty) Laboratory, Pretoria, South Africa, for serological confirmation and serotyping [25]. Twelve distinct serotypes were identified among the 34 strains of Salmonella isolated from domestic animals and free-ranging wildlife (p = 0.002, 95% confidence interval). The distribution of Salmonella serotypes by animal source is shown in Table 3.

Selection and sequencing of polymerase chain reaction products: In the present study, 12 PCR invA gene PCR amplicon products derived from Salmonella isolates obtained from domestic animals (n = 10) and from two free-ranging wildlife sources – sable antelope and leopard were purposively selected from strains of a previous study [25] and were targeted for PCR product sequencing (Table 4). Among the strains from domestic animals, 8 (80%) were from chickens, while 1 isolate each (10%) was obtained from cattle and a horse. All chicken strains were derived from clinically symptomatic hosts.

Purification of sequencing reactions: Sequencing products were cleaned using an ethanol/ ethylenediamine-tetraacetic acid/sodium acetate precipitation protocol recommended by the manufacturer (Applied Biosystems, Foster City, CA, USA). Each reaction was mixed with 125 mM ethylenediaminetetraacetic acid (2.0 μL) and 3 M sodium acetate (2.0 μL), followed by the addition of absolute ethanol (90 μL) to facilitate DNA precipitation. After a brief incubation in the dark at 25°C for 10 min, samples were pelleted by high-speed centrifugation (20,000 × g for 20 min), washed twice with 200 μL of 70% ethanol, and air dried. The DNA was then dissolved in formamide, denatured at 95°C for 3 min, and subsequently loaded onto a 3500 Genetic Analyzer for capillary electrophoresis.

Sequence assembly and editing: Chromatograms of the nucleotide sequences obtained from Salmonella strains from domestic animals and wildlife (n = 12) were inspected, assembled, and edited using GENETYX version 12 (GENETYX Corporation, Tokyo, Japan). Final consensus sequences were used for downstream analyses. Consensus sequences were compared with reference sequences available in GenBank to identify nucleotide differences using the ClustalW multiple-alignment algorithm (https://www.genome.jp/tools-bin/clustalw accessed on 20 February 2025). Specific nucleotide variations were verified by comparing the raw chromatogram peaks and the consensus sequences generated from each primer. Unreliable regions at the 5′ and 3′ ends of the consensus sequences were trimmed based on the quality of the chromatogram peaks for each primer. Only nucleotide positions that showed consistent, high-quality peaks from both forward and reverse reads were retained. Bases with poor or ambiguous peaks, particularly at the terminal regions of the sequences, were removed during trimming.

The nucleotide sequences obtained from Salmonella strains from domestic animals (n = 10) and wildlife (n = 2) were assembled and edited using the GENETYX version 12 software. Multiple alignments between nucleotide and amino acid sequences of Salmonella were performed based on the ClustalW method (https://www.genome. jp/tools-bin/clustalw accessed on 20 February 2025). Reference sequences of Salmonella were also downloaded from the National Center for Biotechnology Information GenBank (n = 8) and compared with those from domestic animals and free-ranging wildlife. Further reference sequences that were shorter than the study strains were removed from the analysis.

The evolutionary history of nucleotides and amino acids in the strains was analyzed using maximum-likelihood methods. The model was selected based on the Bayesian information criterion. Based on this criterion, the Tamura–Nei evolutionary model had the lowest Bayesian information criterion score and was thus selected as the best model for the trees. The trees were constructed using MEGA7 version software [28]. The reliability of the branching order of the trees was determined using 1000 bootstrap replicates. Twenty invA gene sequences from local strains (n = 12) and international reference strains (n = 8) were included in the phylogenetic tree (Figure 3).

The nucleotide sequences of the invA gene of Salmonella isolates from domestic animals and free-ranging wildlife were deposited in GenBank under accession numbers PP921856–PP921862.

At the time of sample collection, ethical and biosafety approval were not required for handling Salmonella isolates from domestic animals. Wildlife samples were collected according to research quotas previously authorized by the Zambia Wildlife Authority [25]. Nevertheless, all biosafety level 2 procedures were strictly followed during the downstream handling and processing of the samples.

Sequence alignments were saved in ClustalW (.aln) format, and the resulting phylogenetic trees were stored as MEGA tree (.mts) files for future accessibility.

Diversity in invA sequences from both the intra- and inter-serovar Salmonella enterica subsp. enterica serotypes were observed when 12 S. enterica strains that were sequenced in this study were aligned with eight reference sequences of the invA gene available from the National Center for Biotechnology Information (NCBI) (Figure 1 and Table 5).

The comparison revealed substitution differences mainly in the region between 40 bp and 250 bp positions of the nucleotide sequence of the invA-284 bp gene sequenced in this study. A substitution was observed where one of these bases, adenine, cytosine, guanine, and thymine, was replaced by another. Further analysis of the invA nucleotide sequence differences revealed six distinct sequence types among the 12 strains from domestic animals and free-ranging wildlife strains. Of these, sequence type I was the most predominant, with 50% (6/12) of the strains being 100% identical. These strains were indistinguishable from the reference international strains with accession numbers (S. enterica subsp. enterica KJ718884.1) and S. enterica subsp. enterica serovar Houtenae ST22 (DQ644627.1) from Egypt and the Centers for Disease Control and Prevention, respectively. Sequence variation was classified according to the position and type of single-nucleotide polymorphisms identified between positions 40 bp and 250 bp within the 280 bp invA gene fragment of S. enterica.

On the other hand, 50% (6/12) of S. enterica serovars revealed sequence variations and belonged to sequence types II–VI. This study revealed that nucleotide substitution in the invA gene was more predominant between 40 bp and 250 bp positions. Further, it was observed that S. enterica strains isolated from chickens were represented in five sequence types (I, II, IV, V, and VI). Among these, types IV, V, and VI were represented by single strains with accession numbers S. enterica serovar Ruanda strain P13e (PP921854), S. enterica serovar Stockholm strain B5 (PP921857), and S. enterica serovar Chardan strain B7 (PP921858), respectively. Furthermore, two strains from free-ranging wildlife were observed to be unique in the sense that each belonged to a different sequence type.

This study further showed transition substitution along the sequence chains of S. enterica serovars at base pair positions C133T (S. enterica serovar Stockholm strain B5 [PP921857]), T220C (S. enterica serovar Heidelberg strain 670 [PP921861], S. enterica serovar Hadar strain B1 [PP921856], S. enterica serovar Stockholm strain B5 [PP921857], and S. enterica serovar Chardan strain B7 [PP921858]), C238T (S. enterica serovar Pomona strain SI-1 [PP921852]), and G250A (S. enterica serovar Heidelberg strain 670 [PP921861]). On the other hand, transversion substitution was identified in two S. enterica serovars at base pair positions C40G (S. enterica serovar Stockholm strain B5 [PP921857]), A210C (S. enterica serovar Ruanda strain P13e [PP921854]), and T211A (S. enterica serovar Ruanda strain P13e [PP921854]).

The nucleotide substitution may result in significant protein structural changes in some strains and, in others, lead to silent protein structures (Figure 2 and Table 6). Interestingly, six sequences belonging to sequence type I with accession numbers S. enterica serovar Garoli strain 3 (PP921851), S. enterica serovar Enteritidis strain B10 (PP921862), S. enterica serovar Enteritidis strain WFI.3 (PP921853), S. enterica spp. strain P19 (PP921855), S. enterica serovar Sendai strain P21 (PP921859), and S. enterica serovar Ruanda strain P12 (PP921860) were not susceptible to substitution and hence maintained a similar amino acid sequence pattern arginine/proline/ arginine/arginine/cysteine/alanine/arginine/glutamine (RPRRCARQ) at positions 14, 30, 41, 45, 71, 74, and 84, respectively.

Homology percentage of nucleotide sequences of the invA gene of S. enterica serovars revealed nucleotide similarities of 98%–100% among the 12 strains isolated from domestic animals and free-ranging wildlife (Table 7). While sequence comparison of the invA gene of local S. enterica strains with eight reference strains (S. enterica subsp. enterica [KJ718884.1], S. enterica subsp. enterica serovar Houtenae [DQ644627.1], S. enterica subsp. enterica serovar Choleraesuis [1GHPS1], S. enterica subsp. enterica serovar Typhimurium [M90846.1], S. enterica subsp. enterica [KJ718878.1], S. enterica serovar Typhimurium [CP012144.1], S. enterica subsp. enterica serovar Bredeney [KP279306.1], and S. enterica serovar Enteritidis [CP007323.1]) from GenBank showed 74.0%–100.0% identity.

To further amplify the relationship among invA sequences at the serotype level and source, a phylogenetic tree was constructed (Figure 3). The phylogenetic analysis of the invA gene sequences revealed two major lineages (Clade A and Clade B), with Clade A displaying a further eight subgroupings (a–h). All 12 sequences from domestic animals and free-ranging wildlife of Zambia belonged to Clade A (Figure 3). Clade B was represented by a single reference strain S. enterica serovar Enteritidis strain EC20110222 (CP007323.1) isolated from Canada and formed an out-group lineage.

S. enterica serotypes are a common cause of gastroenteritis and bacteremia infection worldwide [29–32]. The invA virulence gene is the most common and significant genetic marker in most, if not all, S. enterica subsp. enterica strains that cause salmonellosis globally [33]. In addition, the invA gene is an important marker in the identification of the Salmonella genus. Previous studies attempted to collate and compare Salmonella invA gene variation from a wide range of potential animal and human sources [13, 24].

Similarly, the present study investigated the molecular diversity of the invA virulence gene in S. enterica strains isolated from domestic animals and free-ranging wildlife in selected areas of Zambia. In the present study, we observed that all serotypes were stable in the regions between positions 1–39 bp and 251–284 bp, suggesting that these regions are highly conserved within the invA 284 bp gene. This may underscore the significant role of the invA gene in ensuring that polymerase chain reaction-based diagnostic tests are broadly applicable across S. enterica serotypes [11–14]. However, the region between 40 bp and 250 bp in the nucleotide sequence of the invA 284 bp gene showed a likelihood of being a hotspot.

In this region, substitutions of single-nucleotide polymorphisms were observed, resulting in distinct sequence types while maintaining the unique nucleotide sequence targeted to identify bacteria of the Salmonella genus. These hotspots are genomic regions characterized by high rates of recombination, mutation, or integration of foreign DNA [34, 35]. Within Salmonella pathogenicity island 1, this region facilitates lateral gene transfer and adaptive evolution by promoting the integration of foreign genetic elements such as plasmids, transposons, or bacteriophages. Therefore, the detection of multiple invA sequence types, including unique variants not previously reported in GenBank, underscores the dynamic evolution of S. enterica in mixed-animal ecosystems.

These findings expand the current understanding of invA gene plasticity and provide a genetic basis for refining molecular surveillance and diagnostic assays in sub-Saharan Africa. For instance, the polymorphic regions identified in the invA nucleotide sequences of domestic and wild animal isolates could help design Zambia-adapted PCR primers, reducing the likelihood of false-negative detections due to sequence variability. The submission of GenBank sequences PP921851–PP921862 adds valuable regional reference data for global Salmonella databases and bioinformatics benchmarking.

Further, this study reports invA sequence variations among intra- and inter-serovar S. enterica strains, although partial sequencing of the full-length 2176 bp invA gene was performed. Interestingly, sequences of S. enterica strains from chickens were more diverse. This can be attributed to sampling bias and the discriminatory use of antibiotics in the poultry industry, as well as among backyard poultry keepers, which may select for resistant S. enterica strains and subsequently impact the invA gene [36].

Furthermore, the study demonstrated that nucleotide substitutions in this region resulted in significant amino acid changes in some strains from domestic animals and free-ranging wildlife. For instance, nucleotide substitution in S. enterica serovar Heidelberg strain 670 (PP921861) at position 220 bp resulted in amino acid changes from alanine-to-arginine at position 74 (Figure 2 and Table 6). This phenomenon was observed in at least 45% (9/20) of sequences from both local and reference strains, suggesting that alanine is likely to be substituted by arginine at position 74. The findings of this study indicate that substitution of alanine with arginine at amino acid position 74 in the invA gene may be associated with structural alterations in the protein, potentially affecting the functionality of the type III secretion system or contributing to host adaptation [37], particularly if the residue resides within a transmembrane helix or a tightly packed structural domain.

Topology analysis, secondary structure prediction, and mutation stability assessments were not conducted because they were beyond the scope of the present study. The invA gene encodes invasion protein A, a key component of the Salmonella type III secretion system, which facilitates host cell invasion and colonization [15]. Therefore, substitution of alanine by arginine in the variable region of the invA gene could influence traits such as the ability to evade host innate immune responses or adapt to different host niches [38–40]. However, whether the genetic variations observed in some S. enterica strains in the present study indicate increased or reduced virulence remains unknown. Functional studies are therefore required to determine whether these variations affect virulence, host specificity, or environmental persistence.

In contrast, the present study shows that glutamine substitution at position 84 leads to silent substitution changes (serine). In Egypt, a similar study showed multiple nucleotide substitutions in the sequence chain of S. enterica serovar Typhimurium strains between positions 208 bp and 1813 bp of the 2058 bp invA gene [13]. That study revealed amino acid substitution from serine to phenylalanine at residue number 530. From these studies, it may be noted that pathogenic bacteria can evolve over time to develop strategies that facilitate survival or propagation within a host, causing infection and spreading disease [41, 42].

Factors associated with rapid generation rates of substitution in amino acid sequences may include selection for antigenic diversity to escape host immune responses [43]. Low or no nucleotide substitution was observed in some S. enterica strains from Zambia, such as S. enterica serovar Enteritidis strain B10 (PP921862), S. enterica serovar Sendai (PP921859), S. enterica serovar Garoli strain 3 (PP921851), S. enterica spp. (PP921855), and S. enterica serovar Ruanda (PP921860), despite differences in serotype and host source.

The findings of the present study demonstrate a close genetic relationship between isolates originating from domestic and wildlife hosts, suggesting the presence of potential bidirectional zoonotic transmission pathways within Zambia’s ecosystem. The nucleotide diversity of S. enterica isolates from clinically symptomatic chickens in Zambia (74%–100%) aligns closely with reports from clinically symptomatic chickens in China (72.9%–97.6%) [44] but is notably lower than that observed in poultry samples from Egypt (99.2%–99.4%) [13].

In the same study, a higher homology percentage similarity (99.2%–99.4%) for nucleotide sequences and (99.6%–99.9%) for amino acid sequences were recorded between Egyptian local isolates and other published sequences in GenBank [13]. These comparisons highlight both regional variations and unique evolutionary patterns of S. enterica lineages across different ecological and geographic contexts.

In the present study, the international reference S. enterica serovar Enteritidis strain (CP007323.1) from Canada showed more nucleotide substitutions than local S. enterica serovar Enteritidis strains from horse (PP921862) and chicken (PP921853). All local invA nucleotide sequences obtained from domestic and wildlife isolates showed ≤98% similarity to one or more known international GenBank reference sequences and were considered putative novel invA allelic variants.

The detection of putative novel invA allelic variants underscores the dynamic evolution of S. enterica within diverse ecological and host settings. Such genetic changes may influence virulence and transmission potential while undermining the sensitivity of invA-based diagnostic assays widely used in surveillance programs. Sustained genomic monitoring and routine validation of diagnostic tools are therefore critical to ensure early detection, accurate outbreak tracing, and effective public health response to emerging Salmonella strains.

Further, the study observed that all Zambian isolates clustered within Clade A, indicating localized lineage evolution potentially influenced by antibiotic selection pressures or ecological interactions between domestic animals and wildlife. Three sequences from domestic animals (S. enterica serovar Ruanda strain P13e [PP921854], S. enterica serovar Stockholm strain B5 [PP921857]) and wildlife (S. enterica serovar Pomona [PP921852]) each represented distinct groups, suggesting that they were genetically not closely related to either local or reference strains.

Bacteria can adapt to various adverse environmental conditions, thereby facilitating their survival. Adaptations enable bacteria to overcome challenges and evolve as successful pathogens, potentially resulting in severe disease outcomes [38, 45]. This study observed sequence diversity in the invA gene across hosts, highlighting shared and divergent genomic signatures that may underpin cross-species transmission dynamics.

Therefore, understanding these molecular variations provides critical insight into how ecological overlap and pathogen exchange at the wildlife–livestock interface contribute to the broader zoonotic Salmonella reservoir, informing integrated surveillance, biosecurity, and public health interventions under the One Health framework. To the best of our knowledge, this is the first report from Zambia demonstrating nucleotide sequence variation in the invA gene of S. enterica strains isolated from both domestic and wildlife hosts within a single molecular dataset using short-target phylogenetic inference based on a 284 bp invA gene fragment.

Our findings show that even a short gene fragment can provide substantial phylogenetic resolution, enabling inference of relationships among isolates and comparison of nucleotide diversity with global reference strains. Therefore, the present study provides a foundational dataset to support regional molecular surveillance through full genome sequencing of Salmonella virulence genes in southern Africa, bridging wildlife ecology and livestock epidemiology.

This study has several limitations. First, only a fragment of the invA gene was sequenced, which may under-represent total nucleotide variation and limits phylogenetic resolution and detection of variants outside the amplified region. Second, the number of wildlife-derived S. enterica isolates was small, and sampling was geographically and temporally constrained, reducing statistical power and generalizability. Third, primer choice and amplification in the PCR may have introduced bias against divergent alleles.

Additionally, comprehensive phenotypic and epidemiologic metadata linking sequence variation to virulence or resistance were lacking. Although standard laboratory controls were applied, sequencing or amplification artifacts cannot be completely excluded. These limitations highlight the need for expanded sampling and validation through full-gene or whole-genome sequencing with paired phenotypic characterization in future studies.

Future studies should include whole invA or whole-genome sequencing, increase wildlife sample diversity, include multiple seasons and locations, and integrate genetic data with epidemiologic metadata.