Widespread anticoagulant resistance in house mice (Mus musculus musculus) linked to the Tyr139Phe mutation in the Czech Republic

Abstract

Despite the widespread use of anticoagulant rodenticides in baits for controlling commensal rodent pests, their application is problematic due to secondary intoxication and increasing resistance. In contrast to studies on Western European house mice (Mus musculus domesticus), few resistance studies have focused on Eastern European house mice (M. musculus musculus), which have a western distribution boundary in the Czech Republic. This study newly analysed the VKORC1 gene in M. m. musculus field populations from Czech farms and grain stores and identified a nonsynonymous mutation Tyr139Phe. This mutation was common throughout the Czech Republic and was present in 80.2% of the 86 individuals sampled. Additionally, all individuals exhibited a genotype with three synonymous mutations specific to the subspecies M. m. musculus. The functional (mortality–survival) response of the Tyr139Phe mutation was validated in a laboratory choice feeding test using bromadiolone-based bait, where all resistant homozygous individuals survived, while all susceptible mice died, with a mean survival of 6.9 days.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-85447-8.

Introduction

Anticoagulant rodenticides (ARs) have been the preferred and widely used tool for commensal rodent pest control since the 1950s¹. All anticoagulant active substances share a common mode of action. They disrupt the vitamin K cycle in the liver, which results in fatal haemorrhage and causes the death of rodents within 3–15 days after bait ingestion. Although the times necessary to kill individual sensitive rodents do not differ profoundly among various ARs (e.g. ^2–4), ARs vary in their potency and toxicity to both target and nontarget organisms⁵.

Shortly after warfarin, the first AR active ingredient, was introduced to the market, the first resistant populations of rodents emerged in Europe (in the United Kingdom): the Norway rat Rattus norvegicus in Scotland⁶, the house mouse Mus musculus in England⁷, and the black rat R. rattus in England⁸. The occurrence of resistant populations led to the development of other active substances, namely, first-generation anticoagulant rodenticides (FGARs) and later more potent second-generation anticoagulant rodenticides (SGARs), which were also effective against resistant individuals.

Currently, the resistance of commensal rodents to all FGAR and some SGAR substances is documented worldwide and is widespread in many countries where anticoagulants are used intensively^9,10. The genetic mechanism of resistance to anticoagulants in rodents was described in 2005¹¹. The authors demonstrated that resistance is associated with mutations in the VKORC1 (vitamin K epoxide reductase complex subunit 1) gene, which encodes a vitamin K epoxide reductase enzyme. Mutations are frequently SNPs at different sites in a gene. To date, dozens of different mutations that spread across species and geographical areas have been described, and the process is still ongoing^9,12–14. Extensive geographic mapping of VKORC1 mutations is ongoing in several European countries (e.g., Belgium, France, Germany, the Netherlands and the United Kingdom). However, there are still many geographic gaps in our understanding of resistance, including in the Czech Republic¹⁰.

The study of resistance to anticoagulants and their geographical distribution in rodent populations has multiple dimensions. These most commonly include the species or subspecies of rodent, the type of chemical used as a rodenticide, the extent of physiological resistance, and the genetic background of this resistance. Resistance has been confirmed for all early rodenticides (FGARs), such as warfarin, chlorophacinone, diphacinone, and coumatetralyl, as well as for less potent SGARs, such as bromadiolone and difenacoum. However, resistance has not yet been described for the three most potent SGARs – brodifacoum, flocoumafen, and difethialone⁹. Nevertheless, some recent studies have indicated that higher doses of brodifacoum may be necessary to ensure the death of targeted rodents^15–19.

The geographic documentation of resistance has not been uniformly explored among various species and subspecies of commensal rodents in Europe. Resistance is most frequently monitored in Norway rats, house mice are a less extensively studied species, while black rats receive the least attention due to their limited distribution⁹. In the case of European house mice, the majority of published studies are from western and southern countries, where the subspecies of Western European house mouse (M. musculus domesticus) is widely distributed (e.g.^14,20–23). In addition to common SNPs, resistance in the VKORC1 gene as a result of hybridization of two mouse species (M. m. domesticus and the Algerian mouse M. spretus) was also identified. Introgression of the VKORC1 gene from M. spretus results in a group of four SNPs in the M. m. domesticus genotype in Western Europe^20,24,25.

On the other hand, populations of Eastern European house mice (M. m. musculus) currently distributed across central, northern and eastern Europe have rarely been studied^26–28. The published studies from Serbia, Russia and Finland documented the presence of the known VKORC1 mutations that are frequently occurring in the M. m. domesticus subspecies and also identified new mutations that have not been previously described in M. musculus populations^26–28. The Czech Republic marks the western edge of the M. m. musculus European distribution, including the small territory of the musculus-domesticus hybrid zone, which is located at the Czech-Germany border²⁹. There is only limited evidence of the presence of resistant rodent populations in the Czech Republic from the past. Until the late 1970s, warfarin resistance was not documented and populations of warfarin-sensitive house mice and Norway rats were still prevalent³⁰. In the early 1990s, the first warfarin resistant population of mice³¹ and rats³² were recorded. With regard to the history of anticoagulant use in the Czech Republic, warfarin concentrated dust was a frequently used ingredient in mixed baits until 2005. Then, it was replaced relatively quickly by ready-to-use baits containing SGARs. Following the complete ban of warfarin dust in 2010, bromadiolone and brodifacoum became the major active rodenticide substances used in the Czech farms (with respective usage rates of approximately 70% and 30%, respectively) (J. Plachý, co-owner of the largest Czech company distributing rodenticidal products, pers. comm.).

In the last decade, we have received occasional reports from national pest control operators indicating that some SGAR preparations have low efficacy during rodent control operations. Additionally, our laboratory efficacy tests of anticoagulant baits carried out in the past have demonstrated prolonged survival of tested mice in some cases. Between 2016 and 2018, we conducted a series of no-choice feeding trials with various anticoagulant baits in wild-derived populations of house mice. Some individuals survived the no-choice feeding test with the bromadiolone-based bait (Frankova, Aulicky, Stejskal unpublished data), which led to the suspicion that resistance may be present and their tissue samples were therefore taken and stored. To verify this suspicion regarding anticoagulant resistance in house mice in our country, we decided to screen the VKORC1 gene for mutations in the following years.

This study focused on the understudied subspecies M. m. musculus in terms of resistance to ARs. The specific aims were to: (i) sequence the VKORC1 gene of mice that survived the anticoagulant feeding test and identify any mutations associated with anticoagulant resistance, (ii) estimate the frequency and geographical distribution of VKORC1 mutations conferring anticoagulant resistance in house mouse populations in the Czech Republic and (iii) perform a validation standard rodenticide toxicological test on susceptible and resistant mice.

Materials and methods

Rodent sampling for sequencing

The initial stage of the study comprised the VKORC1 screening of tissue samples from three mice individuals who had participated in previous laboratory trials between 2016 and 2018. The mice (F1 generation born in captivity, parents were captured in Prague, Czech Republic) survived the no-choice feeding tests with bromadiolone-based baits for an extended period of 18 days (data not shown) and were suspected to carry some VKORC1 mutation.

Then, in 2018–2023, extensive sampling was conducted throughout the Czech Republic to explore the frequency of mutations in the VKORC1 gene causing resistance in various house mouse populations. The aim was to investigate mouse populations in agricultural and food industry environments, including farms, grain/seed stores and facilities for processing primary agricultural products, where rodents pose a serious threat to both livestock and human health. Two sampling points were also located on waste dumps. All mice were captured using snap traps (Snap-E MouseTrap, Kness Mfg. Co., Inc.; USA) baited with cheese that were set for a 24-hour interval to ensure fresh tissue for subsequent DNA extraction. The carcasses were then transported to the laboratory and frozen until tail sampling. In total, we collected 86 individuals from 50 localities, with one or two individuals per sample point.

The tail tips of all individuals were cut using sterile scissors, placed in Eppendorf tubes with 96% ethanol and stored at − 20 °C until analysis.

DNA extraction and VKORC1 sequencing

The DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) was used to extract genomic DNA from tail samples following the manufacturer’s protocol. The quality of the DNA was assessed by agarose gel electrophoresis. Independent PCRs were performed to amplify each of the three exons of the VKORC1 gene using primer pairs specifically designed for Mus musculus: musVKORC1-ex1F and musVKORC1-ex1R for exon 1, musVKORC1-ex2F and musVKORC1-ex2R for exon 2 and musVKORC1-ex3F and musVKORC1-ex3R for exon 3 (primer sequences were obtained on request from S. Rost³³). The PCR reaction volume was 25 µl and included 1 µl of genomic DNA; 1 µl of each primer (10 µM); 9.5 µl of PCR H₂O; and 12.5 µl of PPP Master Mix (Top-Bio, Vestec, Czech Republic) containing Taq DNA polymerase, deoxyribonucleotides, reaction buffer components and additives. The amplification process consisted of an initial denaturation step at 95 °C for 3 min; followed by 32 cycles at 95 °C for 30 s, 57 °C for 30 s, and 72 °C for 30 s; and a final extension step at 72 °C for 3 min (S. Rost, pers. comm.). The amplified PCR products were analysed by agarose gel electrophoresis to confirm the presence of the target band, purified using ExoSAP-IT PCR Product Cleanup Reagent (Thermo Fisher Scientific, Waltham, MA, USA) and sequenced on both strands via Sanger sequencing at the service sequencing laboratory OMICS-Genomics Laboratory, BIOCEV, Faculty of Science, Charles University, Czech Republic.

Sequence data analysis

The sequences were analysed using BioEdit software version 7.0.5.3³⁴. Only high-quality sequences were included in the analyses after visual evaluation of the sequencing electrophoretograms. Reverse complement sequences were generated from the original sequences obtained with a reverse primer. All sequences of each exon were aligned using Clustal W software³⁵ integrated into BioEdit. Protein-coding sequences were identified by comparison to the published M. musculus VKORC1 gene sequence (GenBank accession number: GQ905715) and trimmed. The protein-coding sequences from all three exons were merged and compared to the protein-coding sequence of the M. musculus VKORC1 gene (GenBank accession number: NM178600), which is widely used as the wild-type reference for this gene (e.g.²⁰). This comparison enabled the detection of homozygous mutations. Heterozygous mutations were identified through a detailed visual examination of sequencing electrophoretograms, which revealed the presence of double peaks. The mutation’s synonymity or non-synonymity was assessed after translation to the protein sequence and comparison with the wild-type reference.

In individuals displaying rare genotypes characterized by unique or infrequent nonsynonymous mutations, PCR amplification of the exon containing the mutation was repeated, and the samples were subsequently resequenced to verify the findings. All rare and unique mutations were successfully confirmed.

Laboratory choice feeding tests

We further conducted palatability and efficacy tests in the laboratory to evaluate the functional response of a VKORC1 mutation. Choice feeding tests were conducted to compare two groups of house mice with different VKORC1 genotypes: resistant (VKORC1 mutation Tyr139Phe, homozygous) and susceptible (wild-type genotype) individuals.

Adult individuals of both sexes were used in the tests. The animals were kept under standard conditions, with a temperature of 20–22 °C, a relative humidity of 50 ± 5%, and a light regime of 12L:12D. The resistant mice (N = 5 males, mean body weight of 20.4 ± 4.5 g; N = 5 females, mean body weight of 17.9 ± 3.9 g) were captured by livetrapping in grain stores in Nebanice (West Bohemia, Czech Republic) and Prague (central Bohemia, Czech Republic). Their genotype was confirmed by analysing the VKORC1 gene as described above. The susceptible mice were captured by livetrapping in grain stores in Chocen (East Bohemia, Czech Republic) and Opava (North Moravia, Czech Republic), their genotype was also assessed by sequencing the VKORC1 gene. The experimental animals were adult individuals of the first-generation mice born in captivity (N = 5 males: mean body weight 20.6 ± 1.6 g, N = 5 females: mean body weight 15.8 ± 2.3 g).

The experimental procedures followed the protocol recommended for choice feeding trials by the European Chemicals Agency³⁶. The test consisted of a pre-test diet assessment (five days), followed by a test period of five days and a post-treatment observation period.

The animals were individually housed in experimental wire mesh cages (26 × 17 × 17 cm) three days before the start of the pre-test phase for habituation. At the beginning of the procedure, the standard laboratory diet (Hapeko; Biokron, Ltd., Blucina, Czech Republic) was placed in two pots, and daily food intake was monitored for five days. Mice were checked daily, and the remaining food was weighed (including spillage and crumbled food) and replenished. The test period was followed when a choice between the laboratory diet and rodenticide bait (0.005% bromadiolone, Hubex, Ltd., Benesov, Czech Republic) were allowed for a period five days. The position of the pots was rotated daily and consumption of the diet and bait was monitored. Then, the animals were returned only to the laboratory diet, and the observation period started. All mice were checked and observed daily, and any signs of intoxication (changes in the appearance or behaviour of the mice) and mortality were recorded. The experiment was terminated after 21 days of the observation period, and the surviving individuals were euthanized.

Analysis of feeding tests

Survival data were evaluated by survival analysis and the log-rank test. Comparisons between the sexes of susceptible mice were performed using the Mann‒Whitney U test (all resistant mice survived the test; thus, sex differences were not evaluated). Individual bait consumption was expressed per kilogram of individual body weight at the start of the test period and was further evaluated by ANOVA with post-hoc comparisons using the Tukey HSD method. All calculations were performed using Statistica 14.0.0.15 (TIBCO Software Inc., Palo Alto, CA, USA; 2020).

Ethics approval

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Laboratory tests were carried out in strict accordance with Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 on the protection of animals used for scientific purposes and Czech national legislation (Act No. 246/1992 Coll., on the protection of animals against cruelty, as amended). The experimental protocols were approved by the local Institutional Animal Care and Use Committee and by the Ministry of the Environment of the Czech Republic (permit numbers MZP/2020/630/243 and MZP/2023/630/2219) and were in compliance with the ARRIVE guidelines.

Results

VKORC1 sequence analysis

DNA was extracted from a total of 89 tail samples, and all three exons of the VKORC1 gene were sequenced (available as electronic supplementary information, different genotypes are also available under GenBank accession numbers PQ789992-PQ790010). Analysis of three initial samples revealed one nonsynonymous mutation in the VKORC1 gene sequence. The mutation was located in exon 3, and after translation into protein, the mutation at position 139 replaced tyrosine with phenylalanine (Tyr139Phe). All three samples were homozygous for this mutation.

Subsequent sequencing analysis based on biological material obtained from monitoring mouse populations inhabiting farms and grain stores (N = 86) confirmed the widespread distribution of the Tyr139Phe mutation throughout the Czech Republic, as it was detected in 69 of the monitoring samples (80.2%); 42 were homozygous, while 27 were heterozygous (Table 1; Fig. 1). The monitoring samples included individuals from 50 localities with one or two samples per locality (one sample: 14 localities, two samples: 36 localities). At least one individual carrying the Tyr139Phe mutation was detected at 45 (90%) localities; for detailed results, see Fig. 1.

Table 1.

Overview and frequency of all mutations in the vitamin K epoxide reductase enzyme encoded by the VKORC1 gene in house mice from the Czech Republic. HomoZ = homozygous, HeteroZ = heterozygous.

Mutation	Exon	Frequency (%)	HomoZ	HeteroZ	Note
Nonsynonymous mutations
Gly2Ala	1	1.2	0	1
Ala26Pro	1	2.3	0	2
Asp36His	1	1.2	0	1
Ser52Tyr	1	1.2	0	1
Phe55Val	1	5.8	0	5
Arg58Trp	1	2.3	0	2
Trp59Arg	2	2.3	0	2
Trp59Cys	2	4.7	1	3
Arg61Trp	2	1.2	0	1
Ala115Thr	3	3.5	2	1
Tyr139Phe	3	80.2	42	27
Synonymous mutations
Leu10Leu	1	100	86	0	M. m. musculus
Glu37Glu	1	100	86	0	M. m. musculus
Cys132Cys	3	100	86	0	M. m. musculus
Ser117Ser	3	2.3	0	2

Fig. 1.

Distributions of resistant (carrying the Tyr139Phe mutation) and susceptible individuals within house mice populations in the Czech Republic (total N = 86). For a detailed list of localities, see Supplementary Table S1. Small dots represent one sample per locality, while large dots represent two samples per locality. The black colour represents Tyr139Phe homozygous individuals, the grey colour represents Tyr139Phe heterozygous individuals, and the white colour represents susceptible (wild type) individuals.

In addition to the predominant Tyr139Phe mutation, ten different low-incidence nonsynonymous mutations were identified in the VKORC1 gene across 86 samples (Table 1). Additionally, four synonymous mutations were detected in exon 1 and exon 3. Three of them had a frequency of 100%. For detailed VKORC1 genotype of all sampled individuals, see Table 1 and Supplementary Table S1.

Laboratory choice feeding tests

The results of the experiment demonstrated a significant difference in survival rates between resistant and susceptible mice (p < 0.001). All susceptible mice fed the bromadiolone-based bait died, and all resistant mice survived the experiment. The mortality of susceptible mice did not differ between the sexes (Z = −1.07, p > 0.05), and mortality occurred 4–9 days (mean = 6.9 days) after the introduction of the bait (Fig. 2). Mice exhibited various symptoms of anticoagulant toxicity, including weakness, a hunched posture and lethargy, from 4 to 7 days after bait introduction. Resistant mice exhibited almost no observable changes in appearance or behaviour for the duration of the 21-day observation period. One individual displayed a temporary decrease in activity (between 11 and 16 days after bait introduction) and then recovered completely.

Fig. 2.

The cumulative survival of resistant (N = 10) and susceptible (N = 10) mice fed bromadiolone-based bait.

Resistant and susceptible mice consumed 660.4 and 589.4 g of the bait/kg bodyweight (BW), respectively. Bait consumption differed between sexes (F_1,16 = 12.0, p < 0.01), whereas the effect of resistance (F_1,16 = 3.12, p > 0.05) and the sex*resistance interaction (F_1,16 = 0.14, p > 0.05) had no effect. Post hoc tests revealed only a single difference when resistant females consumed more of the bait than did susceptible males (p < 0.01); for the means, see Table 2. As both test groups were fed the same bait, including the concentration of bromadiolone, the same results were observed for the dose of bromadiolone ingested (i.e., resistant females ingested more bromadiolone than susceptible males). In general, resistant mice survived ingestion of comparable (slightly higher) doses of bromadiolone, which was effective at causing complete mortality in susceptible individuals (Table 2).

Table 2.

Bait intake and corresponding bromadiolone dosage in laboratory choice feeding tests. The data are presented as the means and SDs.

	Resistant mice		Susceptible mice
	Males	Females	Males	Females
Consumption of the bait (g/kg BW)	598.5 ± 137.0	722.3 ± 79.7	512.3 ± 48.6	666.5 ± 69.5
Total bromadiolone dosage (mg/kg BW)	29.9 ± 6.9	36.1 ± 4.0	25.6 ± 2.4	33.3 ± 3.5

Discussion

Anticoagulant resistance in synanthropic rodents is widespread across Europe; however, its distribution and genetic type are not monitored or documented in all countries with extensive use of anticoagulant rodenticides. This study provides not only the first record of AR resistance in house mice in the Czech Republic but also the first evidence of the differences of the VKORC1 genotype of the Eastern European house mouse subspecies (M. m. musculus).

Currently, the identification of nonsynonymous mutations in the sequence of the VKORC1 gene is a routine tool for assessing functional genetic resistance in synanthropic rodents. To date, dozens of different mutations have been described in house mice. The three most common resistance mutations, Leu128Ser, Tyr139Cys or the spretus genotype (Arg12Trp/Ala26Ser/Ala48Thr/Arg61Leu), are widely distributed in European countries.

The analysed samples from Czech populations of house mice showed the presence of a nonsynonymous mutation in the VKORC1 gene resulting after translation into the Tyr139Phe mutation. This mutation confers bromadiolone resistance in house mice and has not been previously described in this species in the scientific literature. An unpublished report from Austria is available on the RRAC website, in which the mutation was identified in a single individual¹⁰. The mutation was initially identified in our samples from mice that had survived laboratory feeding tests with bromadiolone-based baits (our three initial samples). Furthermore, monitoring of Czech house mouse populations across 50 farms and grain/seed storage facilities revealed that this mutation is prevalent throughout the country. The mutation was found in 80.2% of the 86 samples, with the majority having a homozygous genotype (60.9%).

The detection of a previously almost undetected mutation in combination with the absence of other frequently occurring mutations (i.e., Leu128Ser, Tyr139Cys or the spretus genotype) was somewhat surprising. We expected the detection of at least one of these widespread mutations, as they were documented in Germany²⁰, a neighbouring country of the Czech Republic.

In addition to the predominant Tyr139Phe mutation, 10 other nonsynonymous mutations were detected in our study. Some of them have been previously described (Trp59Arg and Trp59Cys in black rats³⁷) without the functional response, some confer only moderate resistance to FGARs (Ala26Pro, Arg61Trp and Ala115Thr in black rats^37–39, or, to our knowledge, have not been previously reported, and their resistance status is unknown (Gly2Ala, Asp36His, Ser52Tyr, Phe55Val, Arg58Trp). As these nonsynonymous mutations were detected in only 1–5 samples, we do not consider them to be significant sources of resistance in Czech house mouse populations. On the other hand, we detected three synonymous mutations (Leu10Leu, Glu37Glu and Cys132Cys) in all of our samples that have not been previously reported in studies involving VKORC1 mutations in house mice. As most of the studies in Europe were conducted with M. m. domesticus (Western European house mouse), we can conclude that the genotype with these three synonymous mutations is specific to the M. m. musculus subspecies (Eastern European house mouse). Unfortunately, no reports on the detection of any VKORC1 synonymous mutations have been published in studies analysing house mice from countries covering the area of distribution of this subspecies, i.e., Serbia²⁶, Russia²⁷ and Finland²⁸.

Resistance is most frequently monitored in Norway rats, while house mice and black rats are less studied species⁹. However, recent studies from several European countries have repeatedly shown that resistance may be quite widespread in mouse populations, specifically in 95% of 134 analysed samples from Great Britain⁴⁰, 100% of 111 samples from Spain¹², 84% of 50 samples from Ireland⁴¹ and 80% of 86 samples from the Czech Republic (this study). Goulois et al. suggested that the difference in the incidence of resistance between widespread Norway rats and mice (black rats have only limited distribution in Europe) could be caused by different management practices for both species. While mice are essentially controlled by non-professionals, the management of rats is more frequently carried out by pest control professionals or trained persons²¹.

The presence of a nonsynonymous mutation in the VKORC1 gene does not necessarily indicate functional resistance in the context of pest control practices. Each mutation must be evaluated to determine if it causes resistance in practice. Various methods are available for determining resistance, including lethal feeding tests, blood clotting response tests, measurements of vitamin K epoxide reductase activity⁴² or field studies with resistant rodents for practical resistance^43,44. Although the Tyr139Phe mutation has not been described in house mice (with an exception of a single individual from Austria on the RRAC website), it can be concluded that it confers functional resistance to anticoagulants. This conclusion is based on the following evidence. First, the initial three sequenced samples originated from individuals who survived no-choice feeding tests with bromadiolone. Second, it has been reported that the Tyr139Phe mutation confers resistance to both FGAR and SGAR (bromadiolone) compounds in Norway rats. Its occurrence was repeatedly observed in rats from Belgium, France, Korea, the Netherlands and the United Kingdom (reviewed in⁹). Furthermore, a laboratory feeding study using bromadiolone-based baits clearly demonstrated a difference in survival between resistant (homozygotes with the Tyr139Phe genotype) and susceptible (wild-type genotype) individuals, with survival rates of 100% and 0%, respectively. Resistant mice survived the ingestion of substantial amounts of the bait and corresponding doses of bromadiolone (29.9 and 36.1 mg/kg for males and females, respectively). These values are many times greater than the published acute oral median lethal dose (the dose that is lethal to 50% of the animals in the group; LD₅₀) of bromadiolone for house mice: LD₅₀ = 0.99–1.75 mg/kg⁴⁵.

Anticoagulant rodenticides, particularly SGARs, persist in residual concentrations in the liver of rodents for an extended period of time following bait consumption⁵. Intoxicated mice and rats with anticoagulant residues in their bodies thus present a risk of secondary intoxication for potential predators and scavengers of those rodents (e.g.^19,46–48). This risk becomes even greater due to the presence of resistant individuals in rodent populations. These rodents survive longer and therefore carry anticoagulant residues for longer periods of time, which prolongs their availability as potential prey for predators and scavengers^49,50. Resistant rodents may also consume bait for an extended period of time and thus ingest a much greater dose of anticoagulant than susceptible individuals, which means that they may carry more anticoagulants in their livers; nevertheless, the available data are limited^49–51.

The confirmation of the presence and knowledge of the distribution of bromadiolone resistance in house mouse populations in the Czech Republic is a basic prerequisite for the implementation of an appropriate strategy for the use of effective anticoagulant preparations for house mice control. Our molecular and in vivo laboratory results show that bromadiolone preparations should not be further used against house mice because their efficacy in resistant populations is markedly limited. Bromadiolone and brodifacoum are the two most commonly used active substances for rodent control in the Czech Republic. Currently, 48 of the 126 (about 40%) approved SGAR preparations in our country are based on bromadiolone (and 45 are based on brodifacoum)⁵². In addition to the frequent use of bromadiolone-based preparations, there is another issue relevant for the control of resistant rodent populations. Recently, two types of SGAR preparations differing in the concentration of the active substance have become available on the Czech (and EU) markets. The traditional 50 ppm preparations for professional users, which have been used for many decades, and the “new” below 30 ppm preparations, which are intended for the general public and were introduced to the market in 2018⁵³. However, the potential adverse effects of the use of low-dose (below 30 ppm) anticoagulant baits on anticoagulant resistance remain unknown. There is a concern that their application may facilitate the spread of existing resistance to less potent SGARs across rodent populations or, more undesirably, accelerate the development of resistance to more potent SGARs.

In conclusion, our study offers new insights into the resistance status of synanthropic rodents in the Czech Republic, especially shedding light on the relatively understudied European house mouse subspecies (M. m. musculus) in this regard. Through genetic analysis of the VKORC1 gene, we revealed specific synonymous mutations that are characteristic of this subspecies. Furthermore, we identified a nonsynonymous mutation, Tyr139Phe, which is found in 80.2% of 86 individuals and confers resistance to bromadiolone. Standard laboratory feeding experiments further validated the physiological response of the Tyr139Phe mutation, confirming its crucial role in the survival of mice following exposure to bromadiolone-based baits.

By utilizing the results of this study and those already published by other research teams, a comprehensive understanding of the scope of rodent resistance and its genetic variations across different regions can be gained. This knowledge will enable the development of tailored rodent control strategies, which are crucial for safeguarding agricultural and public health interests across Europe and other countries.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: M.F., Z.S., R.A, V.S.; Formal Analysis: M.F., Z.S.; Funding Acquisition: M.F., R.A, V.S.; Investigation: M.F., Z.S., R.A.; Visualization: M.F., Z.S.; Writing Original Draft: M.F., Z.S., V.S.; Review & Editing: M.F., Z.S., R.A, V.S.

Data availability

The datasets generated and/or analysed in the current study are available as electronic supplementary information and in the GenBank database under accession numbers PQ789992-PQ790010.

Declarations

Competing interests

The authors declare no competing interests.