Non-Invasive Methods for Laboratory Animal Health Assessment During Tick-Borne Encephalitis

Anna Alzheeva; Andrey Belov; Anastasia Rogova; Alena Andrianova; Lidiya Romanova; Magomed Gadzhikurbanov; Anastasia Averyanova; Galina Karganova

doi:10.20944/preprints202605.0602.v1

Submitted:

08 May 2026

Posted:

09 May 2026

You are already at the latest version

Abstract

Non-invasive methods for monitoring the condition of laboratory animals play a key role in ensuring animal welfare and improving the reliability of scientific data. This study evaluates the effectiveness of two non-invasive approaches - daily body weight measurement and urine analysis by qPCR - for monitoring the health of BALB/c mice infected with tick-borne encephalitis virus (TBEV). Calculation of the first derivative of body weight change allowed precise determination of disease onset, which correlated with clinical symptoms and detection of viral RNA in urine. Mathematical analysis of body weight change dynamics (first derivative with type 2 cubic spline smoothing, rh = 1) showed that a derivative threshold value of ≤ −0.6 reliably distinguishes infected BALB/c mice from healthy ones (AUC = 1 in ROC analysis). Urine analysis by qPCR allowed for the detection of viral RNA as early as the second day after infection, with a peak on the seventh day. The mathematical model was further tested on C57BL/6, CBA, and BALB/c mice of different ages and proved to be suitable. The threshold values of the derivative were found to be dependent on the mouse strain. The proposed non-invasive methods offer a humane and accurate alternative to invasive procedures, contributing to higher ethical standards and quality of research in virology.

Keywords:

tick-borne encephalitis virus

;

non-invasive monitoring

;

body weight dynamics

;

urine qPCR

;

laboratory mice

;

mathematical modeling

Subject:

Biology and Life Sciences - Virology

Introduction

The use of laboratory mice to model various diseases is indispensable for studying pathogenesis, immune response development, and evaluating the efficacy of therapeutic and preventive agents. In pharmacological and toxicological studies, laboratory mice play a key role as model organisms. Such studies often require long-term animal observation, necessitating reliable and objective methods for monitoring their overall condition. Particular challenges arise when studying viral infections.

Traditional invasive methods, such as in vivo blood sampling, can be painful and cause stress in animals, potentially affecting study results (Guzman et al., 2024; Hylander et al., 2022). Invasive procedures also require highly qualified personnel and specialized equipment, as well as specific conditions for handling material from infected animals. This limits their availability, and they are practically impossible to use when studying behavior and cognitive abilities (Bartelik et al., 2024). Therefore, there is a need for non-invasive methods to monitor the general condition of laboratory animals. Such methods are more accessible and cost-effective, making them preferable for widespread use in routine laboratory practice. They improve animal welfare and increase the quality of scientific research (Kahnau et al., 2023). Non-invasive methods can be used to monitor animal condition, assess treatment efficacy, and predict disease outcomes.

In studies of physiology and biochemistry, changes in body weight may indicate alterations in metabolism, energy balance, or other physiological processes (Gaukler et al., 2015; Yang et al., 2014). Changes in body weight are often used as an indicator of stress or toxic exposure. Weight loss has been reported during acute stress and transport, particularly in rats (average 3.71%) and C57BL/6 mice (4.69%) (Lee et al., 2012). It has also been shown that significant body weight loss in mice indicates chronic stress and correlates with changes in corticosterone levels and hypothalamic–pituitary–adrenal (HPA) axis activity (Tran and Gellner, 2023). In preclinical studies, body weight monitoring allows assessment of the toxicity of potential drugs (Bourgeois et al., 2016; Malik et al., 2022; Saiyed et al., 2015). These data confirm that body weight can serve as an integral indicator of physiological status under various experimental conditions.

Over the course of a single day, body weight may fluctuate for various reasons (Bray et al., 2013). In long-term experiments, a natural increase in body weight may occur due to animal growth(Gargiulo et al., 2014) or as a result of external factors that influence weight (Bray et al., 2013). Thus, it is important to identify critical thresholds of body weight change at which quantitative changes become qualitative. This is particularly important in infectious diseases to determine the incubation period or the onset of recovery (Rai et al., 2024). Standard weight dynamics analysis often fails to address this issue, since the criteria for determining a pathological state are frequently based on a threshold arbitrarily set by the researcher.

The aim of our study was to develop a mathematical framework for objective, non-invasive monitoring of animal condition based on body weight data. The study was conducted using a mouse model of tick-borne encephalitis in BALB/c mice. The infectious process was assessed by the presence of the virus in urine over time and in the brain at the end of the experiment. The mathematical model was validated using C57BL/6 and CBA mice of both sexes, as well as BALB/c mice of different ages.

Results

Measurement Error in Mouse Body Weight

We assessed whether the mouse's behavior during the weighing procedure could affect the result. We also sought to determine whether weighing the animals once daily was sufficient or if more measurements were required to construct a mathematical model. To address this, we weighed five female BALB/c mice of different initial weights (10-16 g) every hour for seven hours and then analyzed the data obtained. The statistical parameters obtained during the experiment are presented in Figure 1, Table 1. For each animal, we determined mean body weight, analyzed the coefficient of variation, and calculated the relative standard error (relative SE).

The results demonstrate high reproducibility of the method for measuring body weight in laboratory mice, as indicated by the relative standard error values (relative SE < 0.5%). Thus, the weighing method employed is reliable, and the position of the mouse in the container does not affect the result.

The observed inter-animal differences in the coefficient of variation (CV) may be related to individual behavioral characteristics during weighing. However, it is important to note that even the maximum CV value (2.55%) is within the acceptable range for biological measurements.

For long-term studies, weighing can be performed at different times of the day without significant loss of accuracy, but for studies with small changes in weight (<3%), standardization of measurement time is recommended.

Non-Invasive Method for Monitoring Animal Condition by Body Weight Measurement

All animals used in the experiment (n = 30) were divided equally into three groups: G1 – control group, which received an injection of saline solution; G2 and G3 – test groups, which received intraperitoneal injections of 2.3 and 4.3 log₁₀ PFU of tick-borne encephalitis virus strain (TBEV) strain EK-328.

During daily weight measurements of the mice, we noticed that not all animals, even in the control group, gained weight at the same rate. A slight decrease in weight over several days was observed in a few control animals, which was then followed by a return to gradual weight gain. The effect can most likely be explained by the establishment of hierarchical relationships among the animals, as they were kept in groups of 10 in cages.

To analyze the data on body weight changes presented in Supplementary, Table 2, Table 3 and Table 4, we examined the main statistical parameters for each group of mice, generated group graphs of animal body weight changes (Figure 2A), and constructed survival and morbidity curves (Figure 2, B and C). The morbidity curves were constructed based on data processing using the experimental results processing method proposed below.

Body weight changes of BALB/c mice from three groups throughout the experiment;
Survival curves for the same groups of mice;
Morbidity curves.

(G1 – control group, which received saline solution; G2 and G3 – experimental groups, which received 2.3 and 4.3 log₁₀ PFU of tick-borne encephalitis virus intraperitoneally)

The control group (G1) showed relatively stable body weight measurements with moderate variability. We attribute the minor fluctuations in body weight observed in this group to individual characteristics of the animals rather than external factors.

The infected group G2 showed a slight increase in variability compared to the control group G1. The survival curve also shows that two animals from this group died during the experiment. The first symptoms of the disease were recorded 9-12 days after infection. Among all other surviving mice from G2, no visual changes in condition were observed during the experiment, and no disease was detected. Despite the death of two animals, group G2 was statistically similar to group G1 and had no virus detected in the central nervous system (CNS) on day 30 of the experiment.

In group G3, the median body weight is different from the mean in some cases, indicating asymmetry in the data distribution. This group exhibited pronounced fluctuations in body weight, reflecting the effect of the virus on the organism. Episodes of rapid weight loss were also odserved. Of the 10 infected animals, 3 survived until the end of the experiment. These mice, like the rest of the infected individuals in group G3, showed the first signs of disease on days 10-12 post-infection. These signs included decreased activity and weakened muscle tone in the forelimbs and/or hind limbs. Over time, we observed forepaw paresis hind limb paralysis in these mice. Between days 20-24, the condition of the animals began to improve noticeably, and they gained weight. At the end of the experiment, viral RNA was detected in brain tissue from the surviving animals.

Based on the data obtained, we chose to analyse individual changes in body weight for each mouse, since group-level comparison between the control group (G1) and the group infected with a low dose of the virus (G2) did not reveal any significant differences in descriptive statistics. Specifically, these data do not provide information about the day of the disease onset in animals.

We decided to analyze individual body weight curves and their first derivatives. Figure 3 shows representative curves of body weight change and the corresponding first derivative, which reflects the rate of weight change over time. Analysis of the first derivative allows identification of key events in body weight dynamics. Where the first derivative is negative, body weight decreases; where it is positive, body weight increases. The most negative value of the derivative corresponds to the day of the fastest weight loss, indicating the peak of disease, whereas the most positive value corresponds to the day of the fastest weight gain, indicating the peak of health. Points where the derivative equals zero correspond to critical turning points in body weight change. For instance, when the derivative shifts from positive to negative, the zero point corresponds to the day of maximum body weight and can be interpreted as the onset of disease. A shift from negative to positive corresponds to the day of recovery onset. Thus, the minimum of the derivative, for example, indicates the day of fastest weight loss (disease peak). In recovering animals, this is followed by a period of decelerated weight loss until the day of minimum body weight (a second critical point), after which body weight is restored (recovery). This analysis provides a quantitative basis for assessing the dynamics of a mouse’s condition throughout the experiment.

However, when analyzing curves without additional processing, even minor fluctuations in the body weight can lead to marked fluctuations in the derivative curve, complicating data interpretation. To address this effect, we applied cubic spline smoothing (type 2) to reduce random variation in the measurements. Smoothing coefficients were assumed to be equal, and measurement the time point were treated as sequentially increasing by one unit. This approach reduced the influence of random fluctuations and noise while preserving the main trends in body weight change.

To determine the optimal smoothing coefficient (rh), we tested different values, where rh = 0, 1, 5, and 20. Figure 4 shows the curves obtained without smoothing (rh=0) and with coefficients of 1 and 20. Smoothing with a coefficient of 1 was found to more accurately reflect the actual dynamics of body weight change, minimizing noise without significant details.

А,B. Mouse from uninfected group G1;

C,D. Mouse from infected group G2 that showed no clinical signs of disease;

E,F. Mouse from infected group G3 that developed clinical signs but survived until the end of the experiment.

Thus, applying cubic spline smoothing (type 2) with a coefficient of 1 to body weight data yields more reliable and interpretable results. This approach enables accurate assessment of the animal condition during the experiment, including the identification of key events such as disease onset, minimum body weight, and recovery initiation.

The diagnostic threshold (DT) establishes a value beyond which body weight change is highly likely to be caused by a pathological process. To determine this threshold, we calculated the minimum values of the first derivative obtained from body weight measurements to the control group. The mean (x̄) and standard deviation (σ) were calculated. The diagnostic threshold was determined as follows:

DT = x̄ - 2 × σ

The calculated diagnostic threshold for BALB/c mice was -0.5987, which we rounded to -0.6. This threshold proved to be a reliable prognostic marker—all animals with a derivative value below this threshold either died or developed signs of disease.

To validate this finding, we analyzed the distribution of the first derivative values for all animals participating in the experiment (Figure 5). The value of -0.58 represents a natural threshold separating animals with different outcomes and may therefore represent an alternative optimal cut-off point.

To confirm the results, we tested brian tissue of all surviving infected animals for the presence of viral RNA using qPCR. Viral RNA was detected in all three brain samples from surviving group G3. No viral RNA was detected in brain samples from animals in group G2. Thus, the body weight data correlated with the absence of clinical signs and TBEV in the CNS of healthy animals and the presence of the virus in the CNS of sick animals.

ROC analysis of the data from this experiment (n = 30) demonstrated excellent discriminatory power of the proposed method. The results are presented in Figure 6. The analysis indicates that the developed model has high diagnostic potential. The combination of a high area under the ROC curve (AUC = 1) and an optimal balance of sensitivity (100%) and specificity (100%) in the test sample makes it a reliable tool for differentiating between healthy and infected BALB/c mice.

A. ROC curve.

B. Distribution of values by class. The histogram shows a clear separation between the “Healthy” and “Sick” groups.

C. Confusion matrix showing 18 true negatives and 12 true positives, with no false negatives or false positives.

D. Plot of metric dynamics as a function of the classification of the classification threshold, illustrating the classic trade-off between sensitivity and specificity.

In addition to detecting disease, the proposed method was used to determine the day of disease onset for each animal, as reflected in the morbidity curves derived from the extreme points on the first derivative curves.

Non-Invasive Monitoring of Animal Condition Using Urine Analysis

In experiment 3, 13 female BALB/c mice were used for non-invasive monitoring of animal condition using urine analysis. They were divided into two groups: control (n = 5) and test (n = 8, received 5.7 log₁₀ PFU of TBEV strain EK-328). The animals were monitored daily for clinical signs, weighed, and urine was collected.

Survival and morbidity curves were generated for the animals in this experiment (Figure 7A, B). All eight animals developed disease, with onset recorded on days 3-5 for each animal using the proposed method.

Using qPCR, we quantified viral RNA levels in mouse urine pools, as shown (Figure 7С). A biphasic increase in viral RNA was observed, beginning on day 2 (7.92 log₁₀ copies/ml) and rising sharply by day 7 post-infection (11.1 log₁₀ copies/ml).

These data demonstrate that urine analysis by qPCR can serve as a non-invasive method for monitoring viral load and assessing the condition of infected animals. This method may be useful for long-term monitoring of infection dynamics and evaluating the effectiveness of therapeutic interventions.

Validation of the Method and Assessment of Its Applicability to Other Mouse Strains

In this experiment, the proposed non-invasive method of monitoring animal condition by body weight measurement was tested on three mouse strains (C57BL/6 ♂♀, BALB/c ♀ of different ages, CBA ♂♀).

Analysis of body weight dynamics confirmed its effectiveness in detecting infection caused by the tick-borne encephalitis virus. The use of different mouse strains allowed assessment of method sensitivity and specificity across various genetic and physiological contexts.

Survival and morbidity curves were generated for the animals (Figure 8), illustrating differences in disease progression among process different mouse strains.

C57BL/6 mice, females;
C57BL/6 mice, males;
BALB/c mice, females, infected with three different doses of TBEV;
CBA mice of both sexes, infected with 3.7 PFU of tick-borne encephalitis virus.

When analyzing the minimum values of derivatives for all control and test mice (shown in Table 4 supplementary), it was found that the diagnostic threshold suitable for the BALB/c breed is not optimal for all genetic lines.

To improve the accuracy and specificity of the method, an analysis of body weight fluctuations in healthy (control) animals of each subgroup studied was performed. The minimum values of the first derivative obtained for the control groups, as described above in experiment 2, were used as the basis for calculating individual thresholds. The results of the calculations are presented in Table 4 and demonstrate pronounced interline and significant intersex differences for the CBA mouse line. Nevertheless, our data coincided 100% with the data on morbidity and mortality.

Thus, establishing strain- and sex-specific critical values calculated derived from control group data for each strain and sex is necessary to improve the accuracy and reliability of the proposed non-invasive method. This enables the method to be adapted to the specific characteristics of each experimental model, thereby minimizing both false positive and false negative results.

Discussion

This study aimed to develop and validate non-invasive methods for monitoring the condition of laboratory mice infected with tick-borne encephalitis virus. We focused on two approaches: analyzing body weight dynamics using the first derivative and detecting viral RNA in urine via qPCR. The results demonstrate that both methods have significant diagnostic value, minimizing stress in animals and improving the accuracy and objectivity of the resulting scientific data. This is particularly important in the context of modern bioethical requirements emphasizing strict adherence to the 3R principles (Replacement, Reduction, Refinement) in research involving laboratory animals (Lindsjö et al., 2016).

Body weight is one of the most accessible and informative parameters for assessing the overall condition of laboratory animals (Guzman et al., 2024). However, the traditional approach, based on analysis of absolute values or group averages, often fails to accurately determine disease onset, especially when changes are subtle or masked by natural fluctuations caused by physiological processes (Ahloy-Dallaire et al., 2019) or social stress (group hierarchy). In our study, we developed a mathematical method based on analyzing the first derivative of the body weight curve for each animal individually. This approach allowed us to identify critical points corresponding to disease onset (the derivative crossing the zero line from above), peak of the infectious process (minimum derivative), and recovery onset (increase of the derivative after the minimum).

A key finding was the establishment of a derivative threshold of ≤ –0.6, which reliably distinguished infected animals from healthy controls (AUC = 1 in ROC analysis of the initial cohort). This criterion showed high sensitivity and specificity, making it a reliable tool for early diagnosis of TBEV infection in BALB/c mice.

One of the most important findings of our study is the variability in method effectiveness depending on mouse strain. Testing the method on C57BL/6 and CBA mice revealed inter-strain differences. The method demonstrated high sensitivity (100%) for C57BL/6 mice, whereas sensitivity was significantly lower (80%) for the CBA strain. This difference can be explained by the known lower susceptibility of the CBA strain to tick-borne encephalitis virus (Pletnev et al., 2000), which likely leads to less pronounced changes in body weight. These data underscore the need for strain-specific calibration and validation of threshold values across different animal strains, especially when working with models in which genetic background significantly influences disease pathogenesis and outcome (Brinkmeyer-Langford et al., 2017).

The second approach—detection of viral RNA in urine using qPCR—has proven effective as a non-invasive monitoring method. Our results showed that viral RNA was detectable as early as day 2 post-infection, with viral load peaking on day 7. Tick-borne encephalitis is characterized by biphasic viremia in both humans and laboratory animals (Shevtsova et al., 2017). Clinical studies also confirm the possibility of detecting TBEV RNA in patient urine, although the detection rate remains low: in the study by Kriha et al. (2025) only one of 52 urine samples tested positive, an observation the authors attribute to low viral concentration or rare viral shedding into urine. In a commentary on this study,Jiang et al. (2026) note that positive urine results during the late stages of infection may reflect local viral shedding by the kidneys rather than systemic viremia, and emphasize that the mechanism by which viral RNA enters the urine remains incompletely understood. Importantly, urine analysis allows for direct assessment of viral load without the need for invasive procedures such as blood collection, which minimizes animal stress and reduces the risk of data distortion (Balcombe et al., 2004). However, in our study, we analyzed urine pooled from eight mice. Further studies are needed to assess the suitability of this method for individual assessment of laboratory animals.

Traditional invasive methods, such as in vivo blood or tissue sampling, remain the "gold standard," but are associated with a number of significant limitations that can be addressed by the proposed non-invasive methods. Their use allows for multidimensional and objective assessment of individual animal conditions. Furthermore, the ability to accurately determine disease onset allows researchers to optimally schedule sampling for subsequent invasive analyses (e.g., brain collection), thereby reducing the number of animals used (Reduction) and minimizing animal suffering (Refinement).

Despite their successful application, the proposed methods have a number of limitations. First, daily weighing and urine collection are labor-intensive procedures, especially when working with large animal cohorts. Second, body weight dynamics can be influenced by external factors (temperature, humidity, housing density, social stress)(Ahloy-Dallaire et al., 2019; Gaskill et al., 2017), which must be strictly controlled. Third, as our study showed, the derivative threshold requires calibration for different mouse strains, underscoring the need for validation studies on each new model.

Looking forward, the mathematical framework developed here (derivative analysis) can be adapted for other infectious and non-infectious disease models in which changes in body weight serve as a marker of pathology (Toth, 2015). To enhance the accuracy and reliability of the method, integration into multidimensional assessment systems is recommended, incorporating additional non-invasive parameters such as body temperature, water and food intake, and behavioral tests. (Zentrich et al., 2021).

Overall, our study confirms that non-invasive methods based on body weight dynamics analysis and urine qPCR are effective, humane, and practical tools for monitoring the condition of laboratory animals. Their adoption into routine practice has the potential to significantly raise the standards of preclinical research in virology and other areas of biomedicine, yielding more accurate and ethically sound scientific results.

Conclusions

Our study demonstrates that non-invasive methods, such as daily weighing and urine analysis by qPCR, are effective for monitoring the condition of laboratory animals infected with TBEV. These methods allow for an objective assessment of the animals' condition and determination of the day of disease onset.

Analysis of the first derivative of the body weight curve allowed us to more accurately identify the days corresponding to critical turning points, such as disease onset, the peak of the infectious process (minimum derivative), and the onset of recovery. This may be useful for evaluating the effectiveness of treatment and prevention strategies.

Urine analysis by qPCR is an additional non-invasive method that can be used to determine viral load and assess the condition of animals.

Our work opens opportunities for further research aimed at developing and optimizing non-invasive methods for monitoring the condition of laboratory animals with various viral infections. Data on the dynamics of body weight change in mice may be useful for studying the mechanisms of infectious disease progression.

Materials and Methods

Virus Used in the Experiment

To infect the animals, we used the Siberian subtype of tick-borne encephalitis virus, strain EK-328, isolated in Estonia in 1972 from the tick I. persulcatus. GenBank accession number DQ486861.1. The virus underwent 12 passages in mice and 2 passages in PEC cell culture. To infect the animals, the virus was used in the form of a culture fluid of infected PEC cells in the appropriate dilution.

Animals

A total of 152 laboratory mice were used in this study, comprising the following lines:

BALB/c (n = 63 females) for experiments 1-3;
C57BL/6 (n = 26 mice: 12 males and 14 females);
CBA (n = 50 mice: 30 females and 20 males).

BALB/c mice were used in all three experiments to calculate the error in measuring mouse body weight and to develop non-invasive methods for monitoring general health by measuring body weight and analyzing the urine of infected animals.

Mice of the C57BL/6, CBA, and BALB/c strains (aged 10–15 months) were used to test the proposed method based on body weight measurement.

All additional information regarding initial animal weight, age, number of groups, and group sizes is provided in (Supplementary, Table S1).

The mice were kept under standard conditions in a conventional vivarium: a light/dark cycle, 24°C, with food and water provided ad libitum. All procedures were performed in accordance with EU Directive 2010/63/EU on animal experiments and with the approval of the Scientific Center's Bioethics Committee (Protocol No. 06102023 dated October 6, 2023). The animals were monitored daily for mortality and signs of disease (1 “cross” - ruffled feathers, 2 ‘crosses’ - ruffled feathers, untidiness, low mobility, 3 “crosses” - low mobility, weakened motor function of the limbs (paresis), 4 “crosses” - severe intoxication, paresis and paralysis of the limbs). At the end of experiment, animals were euthanized by decapitation when they showed clear clinical signs of tick-borne encephalitis virus infection, upon reaching the humane endpoint. All procedures involving animals were performed by qualified and experienced personnel.

In experiment 1, five female BALB/c mice were used to determine the measurement error over a period of seven hours. The mice were kept in a single cage with an area of 330 cm². The animals were weighed on ViBRA ADAMHCB-302 scales using a laboratory weight container 5 times per hour from 11:00 a.m. to 6:00 p.m. throughout the day, and the collected data were then statistically processed. All measurements were performed in accordance with standard operating procedures to minimize stress and ensure reproducibility.

In experiment 2, 30 female BALB/c mice were used to develop a method for non-invasive monitoring of the general condition of laboratory mice based on changes in body weight during infection. The mice were divided into three groups of 10 individuals each: G1 – control group, G2 and G3 – experimental groups that received 2.3 and 4.3 log₁₀ PFU of tick-borne encephalitis virus (TBEV) strain EK-328 respectively. The mice were weighed daily for 30 days at approximately the same time each day. The general condition and presence of paralysis were assessed daily, and animal deaths were recorded. At the end of the experiment, the brains of the surviving mice from all groups were removed and stored at -70°C in Eppendorf tubes until further use (Figure 9)

To test the proposed method of non-invasive assessment of the condition of laboratory animals, we divided 104 mice of four strains into groups and infected them with different doses of the virus described above (Supplementary, Table 1).

In experiment 3, 13 female BALB/c mice were used for non-invasive monitoring of the animals' condition using urine analysis. The control group of 5 females received an intraperitoneal dose of 0.3 ml of Earle's medium, while the remaining 8 females received 5.7 log₁₀ PFU of tick-borne encephalitis virus strain EK-328. Throughout the experiment, the animals were weighed and urine samples were collected daily. Urine was collected daily in the first hour on the day by two researchers. For collection, the animal's home cage was placed on a manipulation table covered with a disposable transparent plastic bag. The mouse was removed from the cage by holding it by the tail over the bag. If the mouse did not urinate during the transfer from the cage, it was placed on the bag and gently restrained, or allowed to move around on the surface of the bag while being carefully held by the tail. Urine drops were then collected with an insulin syringe into an Eppendorf tube. If a mouse was unable to urinate, the same procedure was repeated after all other mice in the experiment. After collecting urine from one animal, the bag was replaced with a new one. Immediately after collection, urine from infected mice was pooled and stored in Eppendorf tubes at –70°C until use. (Figure 10).

Molecular Biology Methods

RNA was extracted from brain tissue homogenates using TRI Reagent LS (Sigma-Aldrich, USA) according to the manufacturer’s protocol, which included isopropanol precipitation and RNA washing with 80% ethanol. The RNA pellet was dried in an incubator at 37°C and then dissolved in water for injection (Chumakov FSC R&D IBP RAS, Russia).

Brain samples from euthanized mice were weighed and manually homogenized in 0.9% sodium chloride solution (Chumakov FSC R&D IBP RAS, Russia) to obtain 10% homogenates, which were aliquoted and stored at –70°C. RNA was extracted from brain homogenates using TRI Reagent (Sigma-Aldrich, USA) according to the manufacturer’s instructions. qRT-PCR was performed as described previously (Romanova et al., 2007; Schwaiger and Cassinotti, 2003) with minor modifications. Poliovirus type 1, Sabin 1 strain, was used as an internal control and added to samples prior to RNA extraction. Reverse transcription was carried out using M-MLV reverse transcriptase (Evrogen, Russia) according to the manufacturer’s protocol with the following primers: for TBEV—GTB1R (5′-CCATTCCGGCTCTGAACTTG-3′, nt 8056–8037); for poliovirus—PVR1 (5′-CGAACGTGATCCTGAGTGTT-3′, nt 7209–7228).

qPCR was performed using the R-412 qPCR reaction kit (Syntol, Moscow, Russia) according to the manufacturer’s instructions. Real-time PCR was carried out in a C1000 Thermal Cycler (Bio-Rad, Hercules, CA, USA), and fluorescence detection was performed using a CFX96 Real-Time System (Bio-Rad, Hercules, CA, USA). The oligonucleotides and probes used were: for TBEV—TBEL1 (5′-TCTGAGGGAGACACACTTGG-3′, nt 7672–7691), TBER1 (5′-GTGCGCCTGTAAACAAAGAA-3′, nt 7754–7735), and probe TBEP1 (5′-FAM-TCCTTGGTGCAGCTGTTCAGCC-BHQ-1-3′, nt 7730–7709); for poliovirus—PVL1 (5′-GGCAGACGAGAAATACCCAT-3′, nt 7121–7140), PVR1 (5′-CGAACGTGATCCTGAGTGTT-3′, nt 7209–7228), and probe PVP1 (5′-R6G-TTGATTCATGAATTTCCTTCATTGGCA-BHQ-1-3′, nt 7185–7159).

TBEV RNA levels were measured in duplicate and expressed as log₁₀ viral RNA equivalents per ml of sample after comparison with standards prepared as described in Litov et al. (2023) using TBEV primers:BHT7Kgg44(5′-ATGACTGGATCCTAATACGACTCACTATAGACGAATATAGACATCCAGCC-3′) and Kgg45 (5′-TTGTCCTCAGCCAGAACCAC-3′). For TBEV universal detection, the following primers were used: BHT7Kgg23 (5′-ATGACTGGATCCTAATACGACTCACTATAGTACTTTCTGAATGACATGGC-3′) and Pow/TBE3′ (5′-AGCGGGTGTTTTTCCGAGTC-3′).

For RNA extraction from pooled urine, 125 µL of urine was collected, and 1 µg of SPEV RNA and internal control were added. Extraction was performed using the same protocol as for brain homogenates. Reverse transcription was carried out using primers Pow/TBE3′ and PVR1. The oligonucleotides and probes used were: for TBEV—F-TBE (5′-GGGCGGTTCTTGTTCTCC-3′), R-TBE (5′-ACACATCACCTCCTTGTCAGACT-3′), and TBE-probe (5′-FAM-TGAGCCACCATCACCCAGACACA-BHQ1-3′) as described by Schwaiger and Cassinotti (2003). Viral RNA levels were expressed as log₁₀ genomic copy particles (GCP) per ml (log GCP/ml).

Statistical Analysis

Statistical analysis of mouse weight in all experiments was performed using R (version 4.3.0).

In Experiment 1, measurement accuracy was assessed using the coefficient of variation (CV), calculated as (SD/mean) × 100%, and the relative standard error (Relative SE), calculated as (SE/mean) × 100%.

Morbidity and survival curves were generated using GraphPad Prism 10.

The script for calculating the first derivative and generating graphs is available on GitHub at:https://github.com/annakakyanova/Mouse-Weight-Analysis-with-Health-Monitoring Access to the online application:https://annakalynova.shinyapps.io/mouse_weight_app/

The script for ROC analysis using the sample described in this study is available at:https://github.com/annakakyanova/ROC-Analysis-

Illustrations were created using the online service Illustrae.co.

References

Ahloy-Dallaire, J.; Klein, J.D.; Davis, J.K.; Garner, J.P. Automated monitoring of mouse feeding and body weight for continuous health assessment. Lab. Anim. 2019a, 53, 342–351. [Google Scholar] [CrossRef]
Ahloy-Dallaire, J.; Klein, J.D.; Davis, J.K.; Garner, J.P. Automated monitoring of mouse feeding and body weight for continuous health assessment. Lab. Anim. 2019b, 53, 342–351. [Google Scholar] [CrossRef]
Balcombe, J.P.; Barnard, N.D.; Sandusky, C. Laboratory routines cause animal stress. Contemp. Top. Lab. Anim. Sci. 2004, 43, 42–51. [Google Scholar]
Bartelik, A.; Čater, M.; Cevik, Ö.S.; Franco, N.H.; Voikar, V. Focus on novel approaches: Home-cage monitoring of laboratory mice. Scand. J. Lab. Anim. Sci. 2024, 1–5 Pages. [Google Scholar] [CrossRef]
Bourgeois, T.; Delezoide, A.L.; Zhao, W.; Guimiot, F.; Adle-Biassette, H.; Durand, E.; Ringot, M.; Gallego, J.; Storme, T.; Le Guellec, C.; Kassaï, B.; Turner, M.A.; Jacqz-Aigrain, E.; Matrot, B. Safety study of Ciprofloxacin in newborn mice. Regul. Toxicol. Pharmacol. 2016, 74, 161–169. [Google Scholar] [CrossRef]
Bray, M.S.; Ratcliffe, W.F.; Grenett, M.H.; Brewer, R.A.; Gamble, K.L.; Young, M.E. Quantitative analysis of light-phase restricted feeding reveals metabolic dyssynchrony in mice. Int. J. Obes. 2013, 37, 843–852. [Google Scholar] [CrossRef]
Brinkmeyer-Langford, C.L.; Rech, R.; Amstalden, K.; Kochan, K.J.; Hillhouse, A.E.; Young, C.; Welsh, C.J.; Threadgill, D.W. Host genetic background influences diverse neurological responses to viral infection in mice. Sci. Rep. 2017, 7, 12194. [Google Scholar] [CrossRef] [PubMed]
Gargiulo, S.; Gramanzini, M.; Megna, R.; Greco, A.; Albanese, S.; Manfredi, C.; Brunetti, A. Evaluation of Growth Patterns and Body Composition in C57Bl/6J Mice Using Dual Energy X-Ray Absorptiometry. BioMed Res. Int. 2014, 1–11. [Google Scholar] [CrossRef] [PubMed]
Gaskill, B.N.; Stottler, A.M.; Garner, J.P.; Winnicker, C.W.; Mulder, G.B.; Pritchett-Corning, K.R. The effect of early life experience, environment, and genetic factors on spontaneous home-cage aggression-related wounding in male C57BL/6 mice. Lab Anim. 2017, 46, 176–184. [Google Scholar] [CrossRef] [PubMed]
Gaukler, S.M.; Ruff, J.S.; Galland, T.; Kandaris, K.A.; Underwood, T.K.; Liu, N.M.; Young, E.L.; Morrison, L.C.; Yost, G.S.; Potts, W.K. Low-dose paroxetine exposure causes lifetime declines in male mouse body weight, reproduction and competitive ability as measured by the novel organismal performance assay. Neurotoxicol. Teratol. 2015, 47, 46–53. [Google Scholar] [CrossRef]
Guzman, M.; Geuther, B.Q.; Sabnis, G.S.; Kumar, V. Highly accurate and precise determination of mouse mass using computer vision. Patterns 2024, 5, 101039. [Google Scholar] [CrossRef]
Hylander, B.L.; Repasky, E.A.; Sexton, S. Using Mice to Model Human Disease: Understanding the Roles of Baseline Housing-Induced and Experimentally Imposed Stresses in Animal Welfare and Experimental Reproducibility. Animals 2022, 12, 371. [Google Scholar] [CrossRef] [PubMed]
Jiang, X.; Cha, M.; Zhang, Q.; Yang, W.; Luo, E.; Qing, W. Comment on: Detection of tick-borne encephalitis virus RNA in patient samples at different stages of infection. J. Infect. 2026, 92, 106687. [Google Scholar] [CrossRef]
Kahnau, P.; Mieske, P.; Wilzopolski, J.; Kalliokoski, O.; Mandillo, S.; Hölter, S.M.; Voikar, V.; Amfim, A.; Badurek, S.; Bartelik, A.; Caruso, A.; Čater, M.; Ey, E.; Golini, E.; Jaap, A.; Hrncic, D.; Kiryk, A.; Lang, B.; Loncarevic-Vasiljkovic, N.; Meziane, H.; Radzevičienė, A.; Rivalan, M.; Scattoni, M.L.; Torquet, N.; Trifkovic, J.; Ulfhake, B.; Thöne-Reineke, C.; Diederich, K.; Lewejohann, L.; Hohlbaum, K. A systematic review of the development and application of home cage monitoring in laboratory mice and rats. BMC Biol. 2023, 21, 256. [Google Scholar] [CrossRef] [PubMed]
Kriha, M.F.; Kamis, J.; Dvorakova, M.; Tardy, L.; Elsterova, J.; Teislerova, D.; Chrdle, A.; Palus, M.; Ruzek, D.; Hönig, V. Detection of tick-borne encephalitis virus RNA in patient samples at different stages of infection. J. Infect. 2025, 90, 106481. [Google Scholar] [CrossRef]
Lee, S.; Nam, H.; Kim, J.; Cho, H.; Jang, Y.; Lee, E.; Choi, E.; Jin, D. Il; Moon, H. Body weight changes of laboratory animals during transportation. Asian-Australas. J. Anim. Sci. 2012, 25, 286–290. [Google Scholar] [CrossRef]
Lindsjö, J.; Fahlman, Å.; Törnqvist, E. ANIMAL WELFARE FROM MOUSE TO MOOSE—IMPLEMENTING THE PRINCIPLES OF THE 3RS IN WILDLIFE RESEARCH. J. Wildl. Dis. 2016, 52, S65–S77. [Google Scholar] [CrossRef]
Litov, A.G.; Okhezin, E.V.; Kholodilov, I.S.; Polienko, A.E.; Karganova, G.G. Quantitative Polymerase Chain Reaction System for Alongshan Virus Detection. Methods Protoc. 2023, 6, 79. [Google Scholar] [CrossRef]
Malik, M.K.; Bhatt, P.; Singh, J.; Kaushik, R.D.; Sharma, G.; Kumar, V. Preclinical Safety Assessment of Chemically Cross-Linked Modified Mandua Starch: Acute and Sub-Acute Oral Toxicity Studies in Swiss Albino Mice. ACS Omega 2022, 7, 35506–35514. [Google Scholar] [CrossRef]
Pletnev, A.G.; Karganova, G.G.; Dzhivanyan, T.I.; Lashkevich, V.A.; Bray, M. Chimeric Langat/dengue viruses protect mice from heterologous challenge with the highly virulent strains of tick-borne encephalitis virus. Virology 2000, 274, 26–31. [Google Scholar] [CrossRef] [PubMed]
Rai, P.; Marano, J.M.; Kang, L.; Coutermarsh-Ott, S.; Daamen, A.R.; Lipsky, P.E.; Weger-Lucarelli, J. Obesity fosters severe disease outcomes in a mouse model of coronavirus infection associated with transcriptomic abnormalities. J. Med. Virol. 2024, 96, e29587. [Google Scholar] [CrossRef] [PubMed]
Romanova, L.Iu.; Gmyl, A.P.; Dzhivanian, T.I.; Bakhmutov, D.V.; Lukashev, A.N.; Gmyl, L.V.; Rumyantsev, A.A.; Burenkova, L.A.; Lashkevich, V.A.; Karganova, G.G. Microevolution of tick-borne encephalitis virus in course of host alternation. Virology 2007, 362, 75–84. [Google Scholar] [CrossRef] [PubMed]
Saiyed, Z.M.; Sengupta, K.; Krishnaraju, A. V.; Trimurtulu, G.; Lau, F.C.; Lugo, J.P. Safety and toxicological evaluation of Meratrim®: An herbal formulation for weight management. Food Chem. Toxicol. 2015, 78, 122–129. [Google Scholar] [CrossRef]
Schwaiger, M.; Cassinotti, P. Development of a quantitative real-time RT-PCR assay with internal control for the laboratory detection of tick borne encephalitis virus (TBEV) RNA. J. Clin. Virol. 2003, 27, 136–145. [Google Scholar] [CrossRef]
Shevtsova, A.S.; Motuzova, O.V.; Kuragina, V.M.; Akhmatova, N.K.; Gmyl, L.V.; Kondrat’eva, Y.I.; Kozlovskaya, L.I.; Rogova, Y.V.; Litov, A.G.; Romanova, L.Iu.; Karganova, G.G. Lethal Experimental Tick-Borne Encephalitis Infection: Influence of Two Strains with Similar Virulence on the Immune Response. Front. Microbiol. 2017, 7. [Google Scholar] [CrossRef]
Toth, L.A. The influence of the cage environment on rodent physiology and behavior: Implications for reproducibility of pre-clinical rodent research. Exp. Neurol. 2015, 270, 72–77. [Google Scholar] [CrossRef]
Tran, I.; Gellner, A.K. Long-term effects of chronic stress models in adult mice. J. Neural Transm. 2023. [Google Scholar] [CrossRef]
Yang, Y.; Smith, D.L.; Keating, K.D.; Allison, D.B.; Nagy, T.R. Variations in body weight, food intake and body composition after long-term high-fat diet feeding in C57BL/6J mice: Variations in Diet-Induced Obese C57BL/6J Mice. Obesity 2014, 22, 2147–2155. [Google Scholar] [CrossRef]
Zentrich, E.; Talbot, S.R.; Bleich, A.; Häger, C. Automated Home-Cage Monitoring During Acute Experimental Colitis in Mice. Front. Neurosci. 2021, 15, 760606. [Google Scholar] [CrossRef] [PubMed]

Figure 1. Changes in mouse body weight over a 7-hour period.

Figure 2. Characteristics of experimental TBEV infection in BALB/c mice aged 3–4 weeks.

Figure 3. Body weight curve and first derivative for a mouse from infected group 3 (G3) that survived tick-borne encephalitis with pronounced clinical signs but without fatal outcome.

Figure 4. Body weight curves and first derivatives with different smoothing coefficients (rh = 1; rh = 25).

Figure 5. Analysis of minimum first derivative value distribution across all three mouse groups.

Figure 6. ROC analysis of the proposed method (n = 30). The selected critical value (-0.6) is indicated in black.

Figure 7. Survival (A), morbidity (B), and quantitative analysis of TBEV in urine (C) in mice infected with tick-borne encephalitis virus.

Figure 8. Survival (left) and morbidity (right) curves for mice of different strains and ages infected with TBEV.

Figure 9. Experimental design for Experiments 1 (А) and 2 (B).

Table 1. Statistical parameters of body weight measurement accuracy.

Animal No.	Mean body weight, g	Overall CV*, %	Relative SE, %
1	10.80	0.73	0.17
2	15.78	2.33	0.29
3	13.87	1.63	0.22
4	12.65	2.55	0.20
5	13.42	1.92	0.20

*Overall coefficient of variation (CV) = (SD/mean) × 100%.

Table 2. Minimum first derivative values for body weight change in all mouse groups.

Animal No.	G1	G2	TBEV in brain (G2), log₁₀ copies/ml	G3	TBEV in brain (G3), presence/ log₁₀ copies/ml
1	-0.257	-0.326	-	-0.625’	+
2	-0.118	-0.273	-	-0.864*	5,62
3	-0.026	-0.285	-	-0.994*	8,19
4	-0.401	-0.836’	+	-1.098’	+
5	-0.140	-0.451	-	-0.885’	+
6	-0.236	-1.536’	+	-1.023’	+
7	-0.501	-0.458	-	-1.419’	+
8	-0.578	-0.170	-	-1.389’	+
9	-0.217	-0.462	-	-1.098*	6,32
10	-0.213	-0.535	-	-1.123’	+
x̄	-0.269	-0.533		-1.052
x̄-2×σ	-0.599	-0.910		-1.277

*Animal recovered, ‘animal died; «-» < limit of detection; «+»presence of virus in the CNS of animals that died with clinical signs of disease.

Table 3. Minimum first derivative value.

Animal No.	Control group	Experimental group	Day of disease onset
1	-0.20	-1.65	3
2	-0.10	-0.85	3
3	-0.20	-1.10	4
4	-0.075	-1.50	3
5	-0.10	-0.80	4
6		-1.50	5
7		-2.10	4
8		-1.50	3

Table 4. Critical values of the first derivative for different mouse strains.

	C57BL/6 males	C57BL/6 females	BALB/c females (old)	СВА males	СВА females
DT	-0.294	-0.484	-0.50	-0.66	-0.52

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.